:34!) 11::03: ::TS '

2.2 22:2: 2’21: 22-22222
222222: 222' 22.2 22" 22222

2.2322 2'22 23225352 22322: . 22 22222:

2:22 2:2 222:2 2

D sen 2222 for the Degree of Ph. 7!)
WOMEN‘S ST 93 U1! ’ERSIT"
FREDFIIC MiCHAEL SERCHUK

‘ 1975

 

 

ll lll'lll

93 1 59 0528

 

 

 

 

 

 

 

 

7|

 

 

 

 

 

 

 

 

 

 

 

 

a,

2222 2:2"

This is to certify that the

thesis entitled

THE EFFECTS OF THE LUDINGTON PUMPED
STORAGE POWER PROJECT ON FISH
PASSAGE THROUGH PUMP-TURBINES

AND ON FISH BEHAVIORAL PATTERNS

presented by

Fredric Michael Serchuk

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in Fisheries & Wildlife

«412:9. a'z/<

Major professor

Date June 8‘ 1976

0-7639

 

 

 

 

6/27 2}»

ABSTRACT
THE EFFECTS OF THE LUDINGTON PUMPED STORAGE POWER PROJECT

ON FISH PASSAGE THROUGH PUMP-TURBINES
AND ON FISH BEHAVIORAL PATTERNS

By

Fredric Michael Serchuk

PART 1. AN EVALUATION OF MORTALITY INCURRED BY FISH PASSING THROUGH
PUMP-TURBINES AT THE LUDINGTON PUMPED STORAGE POWER PLANT - Passage
success of fish through pump-turbines at the Ludington Pumped Storage
Power Plant was studied in l97h and 1975. Methods were developed for
introducing fish to the turbines using weighted paper sacks and for
recapturing individuals using Jaw-fastened styrofoam floats. The
passage of various sized wooden boards was also studied to provide
additional information on object size and mechanical damage.

Twenty fish tests were conducted in l97h and 1975 using rainbow
trout, chinoox anc coho salmon, and yellow perch. A total of 27h2 fish
were used; 1017 in 197A and 1725 in 1975. Control groups of fish were
used in 1975 to assess handling and recovery losses.

PUmping mode mortality estimates were derived from five tests in
197A and six tests in 1975. Mortalities in 197A varied from 33 - 63%
and averaged 56.5% i 13.3%. During 1975, adjusted mortalities ranged
from 5% - 75% and averaged 67.7% i 7.2%.

Damage rates for fish killed during pumping passage were 37.2%
in l97h and 61.5% in 1975. Most inJured fish displayed lacerations
or decapitation implying that mechanical contact and shearing forces

were causative agents.

 

Fredric Michael Serchuk

Size-selective mortality could not be unquestionably established
in pumping eXperiments. The narrow size range of fish used may have
obscured detection of this relationship.

Estimates of generating mortality were obtained from one experiment
in 197A and four experiments in 1975. Mortality was 67.2% in 197A and
averaged h0.7% t 27.1% in 1975. Adjusted experimental estimates varied
from 35 - 75% in 1975.

About half (h7.8%) of the fish killed during generation were
damaged. Nearly all of the injuries were lacerations or decapitations.
Releases of known dead fish showed similar damage rates and injury forms.

Evidence for size-selective mortality in generating trials was
absent. This was expected due to the relatively wide wicket gate
settings during generating tests. Presumably, the large difference in
wicket gate setting between pumping and generating modes also effected
the observed differences in recovery and mortality rates between the
two modes.

Seven pumping and two generating wooden board tests were performed
in l97h using 1h02 boards. Board size ranged from 6 - 26 inches. In
the pumping tests, recovery and damage rates increased with board size.
Damage was low for the smaller boards (6-inch — 3.9%) but was nearly
complete for the largest boards (26-inch - 97.1%). In the generating
tests, damage rates also increased with board size although these were
much less than pumping estimates for boards greater than 12 inches.

The differential wicket gate setting between the operating modes
probably effected these damage rate differences.

Board damage rates were less than the mortality rates of fish.

Several explanations are advanced for these discrepancies. The most

Fredric Michael Serchuk

plausible is that factors other than mechanical contact contribute
significantly to fish death during passage.
Unresolved problems in applying the mortality estimates derived

from these studies are identified.

PART II. MOVEMENT PATTERNS AND ORIENTATION OF FISHES IN THE LUDINGTON
PUMPED STORAGE RESERVOIR AS REVEALED BY TRACKING STUDIES - Movement
patterns, activity levels, and residence periods of carp and trout were
studied in the Ludington Pumped Storage Reservoir during 197A and 1975
using ultrasonic telemetry and float-tracking procedures. Sixty-five
fish were monitored for daily tracking periods of l - 2h hr, for a
total of 159 tracks covering 1159 hr.

Most fish movements were restricted to areas adjacent to the
reservoir embankment. The shoreline appeared to serve as a reference
marker for locomotory activity. Significant differences in movement
pattern were not detected between species, between sonic and float-
tagged individuals, or between day and night.

Mean swimming speed for carp was 0.17 body lengths/sec (range 0.01 -
0.6h body lengths/sec) and 0.63 body lengths/sec (range 0.01 - 1.30
body lengths/sec) for trout. Movement rates were similar between sonic
and float-tagged fish and between years for carp. Rates of movement
differed between pumping and generating power plant modes but this
trend was inconsistent between years. Mean swimming speeds were lower
than those reported from laboratory studies.

Both carp and trout exhibited a crepuscular rhythm in swimming
speed. Both groups of fish remained active at night and this behavior

may account for their passage into the reservoir.

Fredric Michael Serchuk

Factors affecting swimming Speed were analyzed for their relative
importance by a stepwise linear multiple regression analysis. The
independent variables explained 83 percent of the variation in speed
for trout, and 0 to 65 percent of the variation in speed for carp.
Behavioral, environmental, and power plant features influenced
trout activity, while water currents and water-level drawdown were
most important in affecting carp swimming rates.

Minimum residence period of fish in the Ludington reservoir was
assessed for 62 individuals during 1975. Carp appeared to remain in
the reservoir longer than any of the other species examined, although
the data from each species were variable.

The need for future studies related to the impact of hydroelectric

development on fish behavior and population dynamics is indicated.

THE EFFECTS OF THE LUDINGTON PUMPED STORAGE POWER PROJECT
ON FISH PASSAGE THROUGH PUMP-TURBINES

AND ON FISH BEHAVIORAL PATTERNS

PART I. AN EVALUATION OF MORTALITY INCURRED BY FISH
PASSING THROUGH PUMP-TURBINES AT THE

LUDINGTON PUMPED STORAGE POWER PLANT

PART II. MOVEMENT PATTERNS AND ORIENTATION OF FISHES IN THE
LUDINGTON PUMPED STORAGE RESERVOIR AS REVEALED

BY TRACKING STUDIES

Fredric Michael Serchuk

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

1976

ACKNOWLEDGEMENTS

I wish to express my gratitude to Dr. P. I. Tack for his
encouragement, guidance, enthusiasm and ideas throughout my graduate
program, and in all aspects of my research. Special credit is also
due to Drs. E. W. Roelofs, T. G. Bahr, K. W. Cummins and C. R. Liston
for their counsel, suggestions and sustained interest in my studies.

I thank Dr. J. L. Gill for his assistance and advice on the statistical
aspects of the research.

The successful performance of the turbine passage and fish
behavioral studies required much assistance and cooperation. Many
individuals gave generously of their time, energy, and knowledge in
the execution of the field work. The devotion of these people to the
many tasks involved in the field activities was most gratifying and I
remain deeply indebted to them for their aid. Accordingly, deepest
appreciation is expressed to the following individuals: B. Anderson,
W. Duffy, J. Gulvas, R. Hauer, D. Lawson, D. Lechel, and C. Liston.
Similarly, special thanks are extended to D. Brazo, P. Cascorelli,

M. Chaffee, L. Gaylord, B. Hauer, K. Helmreich, D. Huber, K. Hunter,
M. John, B. Kendall, F. Koehler, G. Peterson, B. Rasher, D. Rondell,
and M. Simons.

I especially thank L. Yeck and G. Yeck for their aid in constructing
many of the materials used in the mortality work and for their sound
advice and moral support. Gratitude is also expressed to D. Rondell for

the use of his vessel in most of the 1975 Passage experiments.

ii

Acknowledgment is due too to Dr. C. Liston and D. Brazo for
integrating my research with other project studies, and allowing me
to freely rely on their cooperation in providing personnel, laboratory
facilities and equipment for the field tests. The generosity of
C. Liston in affording computer access is also appreciated.

I am grateful to R. Gerkowski and R. Sequin for organizing the
turbine outages and providing information and equipment.

I thank J. Boger for typing this manuscript and J. Church for
preparation of the graphics and keypunching of the data.

Finally, acknowledgment is made to Consumers Power Company for
their funding of this research and the Department of Fisheries and
Wildlife, Michigan State University for providing this unique research

opportunity.

iii

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . . . . . . . . . . . . .

LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . .

GENERAL INTRODUCTION . . . . . . . . . . . . . . . . . .
DESCRIPTION OF THE STUDY AREA . . . . . . . . . . . . .
PART I. AN EVALUATION OF MORTALITY INCURRED BY FISH
PASSING THROUGH PUMP-TURBINES
AT THE LUDINGTON PUMPED STORAGE POWER PLANT
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . .
REVIEW OF FISH PASSAGE STUDIES . . . . . . . . . . . .
METHODS AND MATERIALS . . . . . . . . . . . . . . . . . . .
Methodology Development . . . . . . . . . . . . . . .

R ESULTS O O O O I O O O O O O O O O 0 O O O O 0 O 0

Fish Passage - Pumping Mode . . . .

Fish Passage - Generating Mode . . . . . . . . .
Board Passage - Pumping and Generating Modes . . . . .
Pumping Mode . . . . . . . . . . . . . . . . . .
Generating Mode . . . . . . . . . . . . . . . . .
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . .
CONCLUSIONS 0 o o o o o o o o o o o o o o o o a o o o o o 0
SUMMARY . . . . . . . . . . . . . . . . . .

LITERATURE CITED . . . . . . . . . . . .
PART II. MOVEMENT PATTERNS AND ORIENTATION OF FISHES
IN THE LUDINGTON PUMPED STORAGE RESERVOIR
AS REVEALED BY TRACKING STUDIES

INTRODUCTION . . . . . . . . . . . . . . . . .

iv

vi

ix

60

TABLE OF CONTENTS (Cont'd)

METHODS AND MATERIALS . . .

Equipment . . . . . .
Capture and Tagging of
Tracking Procedures .
Data Analysis . . . .

RESULTS . . . . . . . . .
Movement Patterns .
Angular Change . . . .
Swimming Speed . . .
Retention Time .

DISCUSSION . . . . . . . .
Movement Patterns . .
Angular Change . . . .
Swimming Speed . . . .
Retention Time .

CONCLUSIONS . . . . . .

SUMMARY . . . . . . . . .

LITERATURE CITED . . . . .

61
61

6h
65

67
76
9o
91
105
112
112
11h
115
119
121
123

126

Table

10

ll

12

13

LIST OF TABLES

PART I. AN EVALUATION OF MORTALITY INCURRED BY FISH

PASSING THROUGH PUMP—TURBINES
AT THE LUDINGTON PUMPED STORAGE POWER PLANT

Mortality test procedures for 197A and 1975 fish
passage studies at the Ludington Pumped Storage
Power Plant.

Damage test procedure for 197A board passage study
at Ludington Pumped Storage Power Plant.

Turbine passage tests conducted in 197A.

Fish passage turbine tests conducted in 1975.
Summary of 197A fish passage experiments.

Summary of 1975 fish passage experiments.

Summary of physical data on live fish passage
experiments conducted on the pumping mode in 197k.

Only tests using standard procedures are included.

Summary of physical data on fish passage experiments
conducted in 1975.

Comparison of recapture and mortality data derived
from fish passage experiments performed on the pumping
and generating modes in 1975.

Damage data from 197A rainbow trout passage experiments.

Summary of damage data from 1975 fish passage
experiments.

Comparison of damage data from fish passage experiments
performed on the pumping and generating modes in 1975.

Incidence of immediate and delayed mortality of
recaptured fish from 1975 fish passage experiments.

vi

Page

12

1h

16
17
18

19

22

23

25

26

27

29

3O

LIST OF TABLES (Cont'd)

Table

lh

15

16

17

18

Table

Comparison of immediate and delayed mortality data
of recaptured fish from fish passage experiments
performed on the pumping and generating modes in

1975.

Analysis of variance of body lengths of fish for four
possible recapture groups (recaptured live; recaptured
dead or died later; float-only recaptured; not
recaptured at all) in pumping mortality tests performed
in l97h.

Analysis of variance of body lengths of fish for four
possible recapture groups (recaptured live; recaptured
dead or died later; float-only recaptured; not
recaptured at all) in pumping mortality tests performed
in 1975.

Summary of adjusted mortality rates for generating fish
passage experiments performed in 1975.

Summary of l97h board passage experiments.

PART II. MOVEMENT PATTERNS AND ORIENTATION OF FISHES

IN THE LUDINGTON PUMPED STORAGE RESERVOIR
AS REVEALED BY TRACKING STUDIES

Summary of 197h tracking activities.
Summary of 1975 tracking activities.

Quantitative description of individual tracks
accomplished in the Ludington Reservoir, 197h.

Quantitative description of individual tracks
accomplished in the Ludington Reservoir, 1975.

Qualitative description of movement patterns of fish
tracked in the Ludington Reservoir, l97h.

Qualitative description of movement patterns of fish
tracked in the Ludington Reservoir, 1975.

Results of analyses of variance of specific swimming

speed for fish tracked on multiple occasions in the
Ludington Reservoir (only generating-mode tracks).

vii

Page

31

33

3h

h0

AZ

Page
68
69

70

73

77

79

92

LIST OF TABLES (Cont'd)

Table

8

10

ll

l2

13

Results of analyses of variance of tag type (sonic
or float) on specific swimming speed of fish tracked
in the Ludington Reservoir (only generating-mode
tracks).

Frequency distribution of hourly specific swimming
speed estimates used in diel periodicity analysis.

Independent variables used in stepwise linear multiple
regression analyses of swimming speed of fish in the
Ludington Reservoir.

Results of stepwise linear multiple regression of
independent variables on speed of movement (body
lengths/sec) of carp tracked in the Ludington
Reservoir.

Results of stepwise linear multiple regression of
independent variables on speed of movement (body
lengths/sec) of trout tracked in the Ludington
Reservoir, 1975.

Summary of fish released in the Ludington Reservoir

carrying ultrasonic transmitters or float tags
(retention time analysis) in 1975.

viii

Page

93

99

101

102

106

107

Figure

PART

Figure

PART

Figure

LIST OF FIGURES

Aerial view of the Ludington Pumped Storage Project
including the offshore Jetties and breakwall.

I.

AN EVALUATION OF MORTALITY INCURRED BY FISH

PASSING THROUGH PUMP-TURBINES

AT THE LUDINGTON PUMPED STORAGE POWER PLANT

Relationship between fish size and mortality rate
for pumping tests, l97h and 1975.

Relationship between board size and damage rate for
pumping and generating tests, 1975.

II.

(A)

(B)

(A)

(B)

MOVEMENT PATTERNS AND ORIENTATION OF FISHES
IN THE LUDINGTON PUMPED STORAGE RESERVOIR
AS REVEALED BY TRACKING STUDIES

Behavioral patterns of a sonic—tagged carp
(Fish 75-01) on three different dates in
June 1975.

Behavioral patterns of a sonic-tagged carp
(Track 75-15) and a float-tagged carp (Track
75-30) on two separate dates in 1975.

Behavioral patterns of four float-tagged carp
tracked simultaneously on September 9, l97h.

Behavioral pattern of a sonic-tagged brown trout

(Fish 75-33) tracked on October 2, 1975.

ix

Page

Page

37

AS

Page

82

8h

LIST OF FIGURES (Cont'd)

Figure

3

(A) Behavioral patterns of three float-tagged carp
tracked simultaneously on November 7, l97h.

(B) Behavioral patterns of a sonic—tagged carp
(Track 75-25) and a float-tagged carp (Track

75-26) simultaneously tracked on July 21, 1975-

(A) Behavioral patterns of two sonic—tagged brown
trout on two separate dates in August 1975.

(B) Behavioral patterns of three float-tagged carp
tracked on two separate dates (Track 75-h5,
August A, 1975; Tracks 75-33 and 75—5h, August
1h. 1975).

Diel patterns of Specific swimming Speed for carp

carrying ultrasonic transmitters and float tags in
l97h and 1975. Hourly means from 0000 to 0600 hr

are used again for the 2&00- to 3000-hr period to

Show more clearly the trend at 2h00.

Diel pattern of specific swimming speed for trout

carrying ultrasonic transmitters and float tags in
1975. Hourly means from 0000 to 0600 hr are used

again for the 2h00— to 3000-hr period to Show more
clearly the trend at 2h00.

Page

86

88

96

98

GENERAL INTRODUCTION

The development of pumped storage facilities for the generation
of hydroelectric energy has rapidly increased during the last decade
in the United States (Schoumacher, 1976). These installations have
been incorporated into electric power networks Since they improve
production reliability by providing peaking power and supply an
immediately available reserve capacity in the event of system failures.
Also, pumped storage plants are functionally attractive because they
can absorb or generate large energy loads almost instantaneously.

Pumped storage units operate similar to storage batteries.
Low-valued, off-peak electric energy is used to pump water from a
lower to an upper storage reservoir from which it is released to flow
through reversible pump—turbines and generate electric power. Pumping
is normally performed at night and during weekends while power genera-
tion usually occurs during the mornings and evenings of weekdays.

Although the physical and economic aspects of pumped storage
development have been widely studied (Woodruff, 1971; Velz et al.,
1968; Ley and Loane, 1962; Salzman, 1962) much remains to be learned
about the biological impact of these projects on aquatic resources,
particularly fishes. Detailed studies are needed on the mortality
rates sustained by fish passing through pump-turbines and the associated
causative agents (Hauck and Edson, 1976). Also, a critical need exists

to examine the behavior of anadromous and resident fish species in areas

affected by plant-induced flows. Through these studies, it should be
possible to gain an understanding of the relationships between plant
design and operation and fish survival and activity. Such information
is helpful in developing effective methods to protect fishery resources
within habitats affected by pumped storage systems.

The objectives of the present research studies were: (1) to
evaluate passage success of fish through pump-turbines at the Ludington
Pumped Storage Power Project near Ludington, Michigan, and (2) to assess
the movement patterns and behavioral activity of fish species residing
in the Ludington Reservoir. The experimental findings of both studies
were examined in relation to power plant characteristics and biological

and environmental parameters.

DESCRIPTION OF THE STUDY AREA

The Ludington Pumped Storage Power Plant, located four miles
south of Ludington, Michigan on the eastern Shore of Lake Michigan,
is the largest pumped storage project in existence with a maximum
generating capacity of 1,872 Mw (Comninellis, 1973). The upper, man-
made reservoir is 2.5 mi long (h.02 km), approximately 0.75 mi wide
(1.21 km), and has a total surface area of 8A2 acres when full
(3.1a km2). Total capacity at full pond is 27 billion gallons (1.02 x
108 m3) with 63% of this volume available for power generation. Maximum
water depths range from 97 ft (30 m) in the south to 112 ft (3h m) in
the north end. During plant operations, water levels can fluctuate a
vertical distance of 67 ft (20 m).

Transfer of water between the upper reservoir and Lake Michigan,
which serves as the lower basin 370 ft below (113 m), is accomplished
through six large penstocks, each 1300 ft long (396 m). The lower
end of each penstock leads to a Francis-type reversible turbine capable
of pumping 11,100 cfs (31h m3/S) at minimum head and 7,000 cfs (196
m3/s) at maximum head. When all turbines are operable, water can be
transferred at a.maximum flow of 75,960 cfs during generation (2151
m3/s) and 66,600 cfs during pumping (1886 m3/s).

Lakefront facilities to reduce wave action on the power house
include two 1600 ft long (A90 m) jetties and an 1850 ft long (565 m)

breakwall constructed of large rock boulders. The jetties rise 10 ft

(3 m) above the water surface and are separated from each other by a
1100 ft (335 m) channel, dredged to a minimum depth of 28.5 ft (8.7 m).
The outer breakwall, also 10 ft (3 m) above the water surface, is
positioned parallel to the shore about 2700 ft (825 m) from the power
house. Water currents between the jetties are estimated to average
2.2 ft/s (0.67 m/S) while those between the jetties and the breakwall
are estimated at 1.5 ft/s (0.h6 m/s).

An aerial view of the pumped storage reservoir and adjacent Lake

Michigan waters is Shown in Figure 1.

.Hawaxdonn can
mofiupon oaonmwgo map wsficsaosfi powwoam owmAOQm oomefim sovmsficsq one mo 30w> Hawam< .H Tasman

 

 

AN EVALUATION OF MORTALITY INCURRED BY FISH
PASSING THROUGH PUMP-TURBINES

AT THE LUDINGTON PUMPED STORAGE POWER PLANT

Fredric Michael Serchuk

A DISSERTATION

Part I

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

Department of Fisheries and Wildlife

1976

INTRODUCTION

Since 1971, the Department of Fisheries and Wildlife, Michigan
State University, under contract with Consumers Power Company, has
been conducting field studies to assess the effects of the Ludington
Pumped Storage Power Plant on the aquatic resources adjacent to the
plant and in the Ludington Reservoir. Research efforts have been
focused on documenting the temporal and spatial patterns of the
biological communities in these waters and in characterizing the
limnological features of the lake and reservoir environments through
time (Liston and Tack, 1975, 197A).

During 1973, environmental studies indicated that nearly all of
the fish Species in the inshore Lake Michigan waters had entered the
Ludington Reservoir. Also, visual observations and dead fish surveys
taken during this period (and afterwards) showed that fish passage
through the pump-turbines resulted in some physical damage and
mortality. Consequently, a two-year study of fish passage was under-
taken in 197k to assess the magnitude of this mortality and delineate,

if possible, the causative physical and biological agents.

REVIEW OF FISH PASSAGE STUDIES

Field evaluation of fish passage success through turbines at
pumped storage sites, other than Ludington, has not yet been accomplished.
Data are available, however, on fish passage at conventional hydro—
electric installations. Though much of this information was derived
from the passage of small, migrant salmonids at mainstem dams in the
Pacific Northwest, many of the findings appear relevant to pumped
storage Situations.

Early studies on fish and hydraulic turbines (Cramer and Oligher,
196k, 1960; Monten, 1963; Lucas, 1962; Schoeneman et aZ., 1961, l95h;
Von Guten, 1961; U. S. Army Corps of Engineers, 1960; Muir, 1959)
established that a variety of factors produced mortality during passage.
Of these, four general categories are recognized: (1) mechanical damage
due to contact with fixed or moving equipment; (2) pressure-induced
damage due to exposure to low pressure conditions within the turbine;
(3) shearing action damage due to passage through areas of extreme
turbulence or boundary conditions; and (h) cavitation damage due to
exposure to regimes of partial vacuum. Although specific injuries
are often characteristic of each factor, similar forms of damage can
result from the different sources (Bell, 1973; Bell et aZ., 1967).

Experiments with model turbines to assess the physical features
inducing mortality has enabled comparative studies to be performed

under a variety of turbine operating conditions (Bell, 197k;

Cramer, 1965, 1960; Cramer and Oligher, 196A, 1961; Von Guten, 1961;
Muir, 1959; Von Raben, 1957). These efforts have Shown that passage
success is related to turbine efficiency which itself is influenced

by wicket gate opening, water head, and runner Speed. Also, turbine
position relative to the tail-water and clearance distance between the
runner blades and wicket gates were shown to be important for fish
survival. Corroboration of these results has been accomplished in
field tests at conventional hydroelectric plants in the United States
and Europe (Bell et aZ., 1967). Similar verification at pumped storage
facilities is lacking.

Research on the biological factors contributing to passage
mortality has not been extensive. Little is known about the relation
between fish size and passage success in field situations. Some
studies have found that the spatial distribution of fish in the
forebay of the turbine is a critical factor in turbine kills (Long,
1968; Schoeneman et aZ., 1961) but confirmatory data for pumped storage

systems is absent.

METHODS AND MATERIALS

Methodology Development

Determination of procedures to successfully introduce and recover
fish at the Ludington site was the principal objective of the initial
mortality tests conducted in 197k (Serchuk et aZ., 1975; Tack and
Liston, 1973). Emphasis was placed on developing an experimental
design that would yield good statistical precision in the mortality
estimates with a limited number of fish and a good recapture rate.
Standard fish recovery methods were considered impractical because
of the large investment of equipment, time, manpower, money, and
fish required for statistical confidence. Many of the techniques and
sampling gear used in mortality studies at conventional plants were
similarly discounted due to the physical features of the Ludington
plant and the tremendous velocity and discharge of water at the site.
Accordingly, the use of buoyant tag devices was considered the most
productive approach since these tags bring fish to the water surface
rapidly after turbine passage. Previously, buoyant styrofoam floats
had been successfully used in tests at the Bonneville Hydraulics
Laboratory (Bell, 1973) and provided good recovery data in field
trials at conventional power plants on the Connecticut River (Johnson,
1970).

A variety of methods of float attachment and introducing fish

into the draft tubes were evaluated. Jaw attachment of the float

lO

11

proved most efficient as was the use of 2—inch bell-shaped styrofoam
tags. The most successful fish introduction technique was with a
weighted paper sack placed in front of the draft tube opening. In
this procedure, tagged fish are placed in a sack containing a small
sandbag and about a gallon of water. The sack is then lowered by

line into the water where subsequently it becomes saturated and
disintegrates, thereby releasing the enclosed fish into the draft tube.

Modifications of the initial procedures and adoption of new
techniques were implemented based on early test results. The finalized
procedure consisted of: (1) the use of commercially procured rainbow
trout as test specimens; (2) anesthetization of fish prior to tagging
with both floy tags and floats; (3) length measurements of all fish;
(h) retention of recaptured fish in a holding facility for 72 hours
after a test to assess delayed mortality; and (5) use of a control
group of fish for the evaluation of handling losses (only in 1975
tests). Also, fish were placed individually in mesh bags after tagging
to enhance the recapture rate of damaged individuals and reduce recovery
loss of fish caused by float removal during turbine passage.

In l97h, passage studies of various-sized wooden boards (1 x 3
inches) were conducted in a supplementary effort to determine the
relation between object Size and mechanical damage. Each board was
numbered and weighted with a small bag of sand secured to the wood
by several rubber bands. These sandbags were sufficient to displace
the naturally buoyant boards from the water surface to the turbine
intake.

Complete descriptions of the procedures used in the fish and board

turbine tests are presented in Tables 1 and 2, respectively.

Tmflel.

l2

Mortality test procedures for 197A and 1975 fish passage studies
at the Ludington Pumped Storage Power Plant.

 

Preparatory Activities:

5.

Pre—trial

6.

10.
11.

Procurement of fish from commercial game fish dealer.

Transport of experimental fish in hatchery tank to Ludington
fisheries laboratory.

Retention of fish overnight in transport tank at laboratory
Site. Constant aeration and water flow through tank maintained
during this holding interval.

Preparation (labeling) of float tags and mesh bags for test
trial.

Establishment of holding tank to maintain recovered Specimens
from turbine test for 72 hours for determing delayed losses.
Processing of Fish:

Anesthetization of fish with MS—222 for marking and handling.
Usually done several hours prior to the field trial.
Measurement of body length and weight for each individual fish.

Attachment of float tag to fish jaw and floy tags to dorsal
musculature.

Data recording of body measurements and float and floy numbers
for each test specimen.

Placement of fish individually into mesh bags tied with string.

Return of fish into hatchery tank after processing for recovery
from anesthetic.

Introduction of Fish into Draft Tube:

l2.

13.

1h.

15.
16.

Transport of marked, experimental fish in aerated hatchery site
from laboratory to pumped storage site.

Placement of several fish into a paper sack containing a sandbag
and about a gallon of water. Immediately prior to sack place-
ment, the condition of each fish is observed and recorded.

Attachment of paper sack to line and lowering of apparatus into
water in front of intake structure.

Saturation of sack and release of fish into turbine intake.

Repetition of steps 13—15 until entire lot of fish has been
processed.

13

Table l (cont'd):

Retrieval of Fish:

17.

18.

19.

20.

Post-test

21.

22.

230

2%.

25.

After turbine passage, recovery of fish and floats is accomplished
by one or two boat crews near the discharge areas.

Recaptured individuals are placed in a holding trough aboard boat.
Live Specimens are removed from the mesh bag and the float tag is
detached.

A record is maintained by each boat crew of the condition of
each fish at recapture and the float number of the float-only
recoveries.

Reconnaissance of the discharge area is conducted for approximately
2 hours after introduction of fish into the turbine to locate
all test individuals.

Activity:

After power plant activity has ceased, observations are made
at the intake structure for specimens that did not undergo
turbine passage. These individuals are recovered and recorded
appropriately.

All recaptured fish and floats from the discharge area are
transported to the fisheries laboratory for data processing.

Live recaptured fish are placed in an aerated holding tank for
three days to assess any delayed mortality resulting from
turbine passage.

Dead recaptures are examined for internal and external physical
damage.

Data analysis of the results in performed.

 

Table 2.

1h

Damage test procedure for 197A board passage study at Ludington
Pumped Storage Power Plant.

 

Preparatory Activities:

6.

Procurement of lumber (pine or spruce) from commercial dealer.

Sectioning of lumber into 1 x 3 inch boards of various lengths
(6, 8, 12, 18, 2A or 26 inches).

Numbering of boards in the middle and on both ends to permit
identification of pieces in the event of cracked or split
boards resulting from turbine passage.

Immersion of boards in polyurethane (twice) to reduce water-
logging upon release in the draft tube.

Preparation of paper bags filled with sand to serve as weights
to permit the boards to Sink from the water surface during the
turbine experiment.

Attachment of the sandbags to the boards (one per board) via
rubber bands.

Introduction of Boards into the Draft Tube:

10.

Retrieval

11.

12.

13.

Post-test

1h.

15.

16.

Transport of marked, weighted boards to the pumped-storage site.

Placement of groups of boards into a wooden box which is lowered
to the water surface by guide ropes.

Overturn of the box and release of the boards into the draft
tube opening.

Repetition of steps 8 and 9 until entire batch of boards has
been released to the turbine unit.

of Boards:

After turbine passage, recovery of boards and pieces of boards
is accomplished by boat crews near the discharge area.

Reconnaissance of the discharge area is conducted for several
hours after the board introduction to locate all pieces.

Survey of discharge area is performed for several days following
test to further retrieve boards.

Activity:

Upon shut-down of turbine unit, observations are made at the
intake structure for boards which did not undergo turbine passage.
These are retrieved and noted accordingly.

Recaptured boards are transported to the fisheries laboratory
and examined for mechanical damage.

Data analysis of the results is performed.

 

RESULTS

Nineteen turbine tests were performed between 28 April and
1h November 197A (Table 3). Ten trials were accomplished with 1017
fish, comprising three species: rainbow trout, chinook salmon, and
yellow perch. Nine experiments were conducted with 1h02 wooden boards
of the following sizes: 32h 6-inch boards, 1h5 8-inch boards, 338
12-inch boards, 291 18-inch boards, 256 2h-inch boards, and AB 26-inch
boards.

During 1975, ten fish passage tests were run between 15 June and
9 November (Table h). A total of 1725 fish were used; all but 51 fish
were rainbow trout. Control groups of fish were used in each experi-
ment except for the 19 October test in which large coho and chinook
salmon were utilized.

Details of all turbine trials in 197A and 1975 are summarized in

Tables 5 and 6, respectively.

Fish Passage - Pumping Mode

Pumping mortality estimates were derived from data from five
experiments in 197k and six experiments in 1975. The field trials
of 28 April, 3 May and 19 May 197k differed substantially in technique
from the standard procedure and have therefore been excluded from the
present analysis. Results of these tests are elaborated in Serchuk

et a1. (l97h) and are briefly listed in Table 5.

15

16

Table 3. Turbine passage tests conducted in 197h.

 

Type and Size Operational
Date of Sample Size Range Mode

 

FISH PASSAGE STUDIES

28 Apr 7h 1hh chinook salmon 130-260 mm Pumping
3 May 7h 10 chinook salmon 136-175 mm Pumping
19 May 7h 116 rainbow trout 1&0-316 mm Pumping
ll chinook salmon 1hO-l72 mm Pumping
21 June 7h 95 rainbow trout 162-320 mm Pumping
12 July 7% 101 rainbow trout 215-395 mm Pumping
1h Aug 7h 166 chinook salmon 8h-180 mm Generating
28 Aug 7h 90 yellow perch 96-270 mm Generating
6 Oct 7h 85 rainbow trout 282-390 mm Pumping
20 Oct 7h 105 rainbow trout 227-h70 mm Pumping
3 Nov 7h 9h rainbow trout 228-363 mm Pmmping

BOARD PASSAGE STUDIES

10 May 7h 99 pine boards 6 and 12 inches Pumping
21 June 7h 87 pine boards 18 and 2% inches Pumping
12 July 7h 98 Spruce boards 6 and 12 inches Pumping
1h Aug 7h 178 spruce boards 6,12,18,2h inches Generating
28 Aug 7h 190 spruce boards 6,12,18,2h inches Generating
3 Oct 7h A8 pine boards 6,12,18 inches Pumping
6 Oct 7h 179 pine boards 6,8,12,18,2h inches Pumping
20 Oct 7k 233 pine boards 6,8,12,18,2h inches Pumping

1h Nov 7h 300 pine boards 6,8,12,18,2h,26 inches Pumping

 

l7

 

 

Table A. Fish passage turbine tests conducted in 1975.

Species and Number Size Range Operational

Date of Fish (mm) Mode

15 Jun 75 196 rainbow trout 233-A62 Pumping

20 Jul 75 21h rainbow trout 215—h66 Pumping
8 Aug 75 205 rainbow trout 130-h70 Generating
25 Aug 75 183 rainbow trout 173-A77 Generating

21 Sep 75 173 rainbow trout 233-A90 Pumping
h Oct 75 171 rainbow trout 227—532 Generating
17 Oct 75 157 rainbow trout 231-510 Generating

19 Oct 75 h6 coho salmon 535-780 Pumping

5 chinook salmon 582—800 Pumping

2 Nov 75 186 rainbow trout 180-152 Pumping

9 Nov 75 189 rainbow trout 1714-510 Pumping

 

cod x vegans» can“ condoaoa gnaw adpov\nmau uouo>ooonss and mpaoau unhapndoon .aouspndoou noon «0 ‘.
ooa x auaoau use new“ couspAdoouxavaoau consanaoou can consumaoou econ no s
ooa x pawn cousundaou Hdpop\ouspnaoou neon no §

.ouaoaou ocapu3p com: anon I any
.oudoaou ocanusv non: o>aa< I 54v o

a...

 

a.am ~.om s.am n.am ”.6: «s an an x<v so assm so: m

m.om a.ms m.on o.m~ a.ns mm mm on gel nod seam poo on
n. n. s. o.oa o.om m m .. ho. oa

o.oo «.ma “.mm ».s~ «.ms as an “a Asa me send poo m

an nu nu «.mp m.mm m a u: Ana ma

m.na «.mm ~.~m o.so m.ao m as on Asa as one was on

Awupxmv map you noun
a non» ouoa cacao canon gonna ad ad

 

 

 

eoaasuoooe on voodoo opus muaaauao: «.mm nu me on A<V mod son «:4 3H
o.ow o.mm a ma us an. we
s.om m.~n a.mm o.m» o.on ma sa am Asa we mash fish «a
s.mo m.mm a.mo o.a~ e.sm mm mm as. l<v ma seam use an
n. u. .. o.os s.mm ea as u- gov sud mesa as: ma
o.os o.mm o.mm o.os 0.0: -- a m Asa ea mesa has n
-- .. -- -- .. Asa sa
m.~ n.~ -- u- s .Asv emu mesa has on
m: «2 Hz apnea“ a noun a eoao>ooom econ a o>ad< ‘ eoaaoaom one: open
huo>ooom huo>oaom hacoumpaoam swam unavduono puma

unease nauseous: u Hanna seam ea .62 saga eoso>oeom no .oz

 

.aacoaauonao omoauon noun :Pma uo anaaasm .m canoe

l9

 

 

 

 

 

0.00— >.mm F N I MAW n
m.nm m.mw >.mm v.00 N.nm om he mm 4 var
m.m« 0.00" 0.00" I m mm on oe now poo >—
o.oo. 0.002 I w I on m
¢.mn b.0m m.n¢ m.¢m m.mm an an we flaw mm—
m.m— o.ooF 0.00" I m mm A01 o¢ now «00 a
0.0 0.0 I I I Maw «
m.nm F.nm N.mm ¢.m> an.m¢ mm on v— 4 am“
F._n o.oo. .o.ooa I as an on me musm mam am
o.>> m.¢¢ em mm I now u
~.mm n.mm v.m> m.vm m.¢m am oe FF as me
m.em o.ooa so.ooa I _a ma on on one was mm
mfimo.av m.am s.ms m we I on mm
m.¢> m.—m m.mb ¢.b> H>._m Fm Fm mF Msv nnp
_.m o.mm .o.mm I n we a ov
m.>> o.mm o.mm I mm —— voV om sou wad m
m.sm o.m~ or e I Any ma
«.mm m.Fm m.Fm m.nm om.mm Fm ma op Adv mv—
0.6 0.00, 0.00. I e we gov cm auna Has om
¢._> F.>n mm mm I An mo,
m.>m >.mm >.N> m.>m pm.mm or m n A< 08 I
m.mw F.mm afi.¢m " m mm *Ao Fm pusm ooh m,
mm mm PH mpmoah @ smwm m oohmbooom noon % o>HH< % domwoamm oven mama
- I II muo>oomm hum>ooom haoolmpwoam sown omnobooom swam muwpmuopo puma
Isobar enermvn a R fleece swam mo .02 mo .0:
.mwuoswuomwo ommmmam swam mum? mo hummus .m canoe

2O

 

 

 

 

 

 

.mammauaa 04p a“ mom: poo «up .ouua omnm>oomn .nmwm F moosflouw n M
.mumhfldsm as» aw pom: pom p:p .mvma omuohooon .nmwm F mommaouw u n
.mwmmflmum on» ow vows won «an .opoH vonobooon .mem ¢F m uwflouw u w
.mflmhawsn 6:» ow eons you “on .muwH oonmboomh .mem F mmudnoua u M
.m.~h mo mums mmoa aonwmoo moans hp ombunmm mvmnflvmo u m
.mflmmamom on» ma pom: pom vsn .mpma oouo>ooop .mmum N mmusmouu n m
.mvwamuuon pomswmeSm mo mmmaoummou .ou5pmoomu oaoflm pm m>wam .omwm Ho ummmm ovmu u o
.mouoboomu havooswomQSm one omwoaon um uwmu .mem ¢ umosflouo u v
.mflmhamco on» sH owns «on «on .ovma vono>oomu .mem h mounaonw n o
.mfimmamcs on» aw poms won «an .ovma omno>ooon mumwm «F moodnonw u p
.mwmhawam on» a“ noun «on asp .opmH oonobouon .mem 0F moundonw n w
Aswan mmoH Houvuoo wsfimsv Fm oopmsnom n ma
mnmoam mmnspmmomu mum monflpmmooh H6909 ommoamu oceans» moms omen u Anv
mpmo~m «mnnvnmoon mum mmnnuawoou uwov moxw u NE mmmoflmn onflnusw moms mbwnm u u
ooF K mohsvmmoou Hdu0p\mouupmmoon 0000 no % n Fm ** nmwm Nonv:ou u 0 *
0.00 0.00 F N I Amv m
v.00 h.nm 0.00 0.mm Mm.nm F¢ Fm eF A<v mmF I
m.o o.ooF o.ooF I m we on me ousm son m
>.mm m.nm F F I An m
e.mF ~.om m.FF m.sm em.m~ an _n a As Fna
F.m o.ooF .o.ooF I a me Ao me unsm son N
0.00F 0.00F I N I an, N
I e.mm m.om m.mm m.me mm as N law as
com: maonvuoo on musm woo mF
mm NR Fm mvmoam a swam R oonm>oomm omen % obfifl<.% ummmmflom moon oven
mum>oomm anobooom hasoImpmon - swam msdvmummo «man
I swam mpm>oom .
*toudm puunevnom u mmvoa dmwh no .03 . . v x no .on
AmmfinwuuooV m candy

21

In the five l97h pumping trials, of a total of hhS live fish
released to the turbines, 200 or h5% were recovered. Known dead trout
releases of 162 individuals resulted in a 38% recapture rate (61 fish).
No significant difference (x2 = 2.29, P > 0.10) existed between the
recovery rates implying that the recovery percentage of dead fish
(those killed in passage) was equal to that for all test fish. This
is a necessary condition for use of float-recapture procedures.

Recovery rates in 1975 were similar to those in 197k. In 1975,
22k salmonids were recaptured from 639 pumping releases (35.1%). The
recovery rate of known dead releases was 36.h% (A8 fish). Again,
differences between the recovery rates were not evident (at2 = 0.0%,

P > 0.80). Similarly, yearly differences in known dead fish recovery
rates were absent (x2 = 0.01, P > 0.90). A highly Significant difference,
however, was detected in the live release recapture rates between years
(x2 = 10.69, P < 0.005). Reasons for this inequality are not known.

Variations in mortality rate were apparent among trials in both
197k and 1975. Pumping mortalities in 197A ranged from 33 - 63% while
unadjusted values (used for comparison) varied from 62 - 78% in 1975.
Inter-trial mortality differences, within each year, were not Signifi-
cant (1971I, x2 = 7.99, P > 0.05; 1975, x2 = 2.07, P > 0.70) when
mortality was assessed from recaptured fish. The large salmon mortality
estimate (91% - 19 October 1975) was not included in the 1975 inter-
trial comparison since fish used in this test were much longer than
in the other experiments (mean salmon length = 677 mm; mean trout
length = 353 mm). The homogeneity of the inter-yearly mortality
estimates and the relative constancy of turbine characteristics between
experiments (Tables 7 and 8) permitted pooling of the mortality data

within each year.

22

 

m OH mam I mam emm H dead aoz m

 

H HH osm I me Ham H sass poo om

H HH Hmm I mam emw H aspm poo 0

m oH Hmm I mam use H assm Has NH

m HH mam I 0mm use H assm use Hm

whouo pwom AoovoHSFopomsoB A.pm 0H mmommv moHsomo FHGD 0002 mama
muo>ooom mo .02 nouns 000m 0>Hpommmm 0900 pmonz ooHnuse wofipmhomo pmoe

 

.0005H00H who monsooooam ondoodpm onmd mpmop hHoo .spmH
sH woos mchsdm map so oopososoo mpoosfipomxo ommmmmm anm o>HH so open Hmonhsm mo hamsesm .N oHme

23

 

 

H NH osm smm mam N mssa soz a

H HH osm me mmm a mesa soz m
H a mam mmm use m mssa poo mH

a mH 0mm mmm ems a 0.2.N soo soo FH

N ma mNm mNm RNw N Soc 900 a
m 2H 0mm Hmm mam a mssa mom Hm

N ma wmm wmm mmw w :IH :00 m5< mN

N mH mam osm saw a mam coo mss m
a mH ssm mmm mam H mesa Hse om
N MH 02m 0mm s00 H mesa s50 mH
mamao poom noovomdpmuogaoa A.pm ow owqdmv onGoQO pHsD 000: mean
aho>oomm mo .02 ponds 060m 0>Hpoommm 0900 meOH3 mofinmsa moHpmmomo pmoa

 

.mwma 2H ooposoooo mpsoaHaomxo owmmmom smHm so 6900 Hmonmsm mo mmmsssm

.0 oHnme

2h

For 197h, pumping mortality averaged 56.5% with a 95% confidence
interval of i 13.3% based on recaptured fish (mortality rates computed
by assuming detached floats and unaccounted fish represented dead
fish are 75.1 and 81.h%, respectively). In 1975, the mean, unadjusted
pumping mortality was 69.9% (72.2% with salmon) with a confidence
belt of i 7.2% (Table 9). A corrected mortality rate of 65.1% (67.7%
with salmon) was obtained when the data were adjusted for handling losses
determined from control fish (mean handling loss = 13.8%). Adjusted
values derived by considering detached floats and unaccounted fish as
dead specimens were 85.5% (86.h% with salmon) and 90-6% (91-9% with
salmon), respectively.

Damage rates for live releases differed between years (x2 = 13.59,
P < 0.005). In 197k, only about a third (37.2%) of the fish that died
in passage exhibited physical damage (Table 10), while 61.5% of the
killed fish were injured in 1975 (Tables 11 and 12). Damaged individuals
generally displayed lacerations or suffered decapitation (73.5% of
1975 injuries) suggesting that mechanical contact and Shearing forces
may have been the causative factors. Analysis of immediate and delayed
mortalities in 1975 (Tables 13 and 1h) indicated that the majority
of deaths (61.5%) occurred during turbine passage.

In both years, known dead releases experienced less damage than
live releases (19.7% in 197A; h7.9% in 1975). The biological signifi-
cance of this discrepancy is unclear although it may indicate that
live fish are hampered by the floats. The types of injury exhibited
by the dead releases, however, were similar to those Shown by fish
killed in passage (69.6% of the 1975 damaged dead releases were cut

or decapitated).

.

on.

o... —

 

Isa~s.

a§u

 

up.

--\

v

u-s.

.0‘.

a .ma.

25

A0p0a mmOH Hompooo wsHmsv

H: oosnsnom

mp0oam 00Adpm0o0a 000 m0msvm0o0m H0pou

 

OOH x

00H x m0hdpm0o0m H090p\m0mdpm0o0a 0000 00 m

.mHth0o0 0gp 0H 0005 won 909 .0p0H 00H0>oo0m
.mHmmH0o0 0gp 0H 00m: nos #59 .0p0H 00m0>oo0m
.mHth0s0 0:9 GH 0003 poo pap .0p0a 00A0>oo0m
mHmmH0s0 0gp 0H 000: #00 pan .0F0H 00m0>oo0a

0000H0H 0sHpHdp moms 000a
0m00H0m 0oHpH:p cogs 0>HH<

mp0oam 00Hdufl0o0h 000 00m5pm0o0m 0000 00 a

.smHm m noosHocH
.anm H moosHosH
.smHm Fm moosHosH
.smHm OH moosHosH

smHm Hossooo

H

u E **

ll
€8,000

II
0410

 

 

 

 

 

N.ma m.Nm mm mm III H00 HOH

~.0: m.N~ w.Nm N.ww 0>.mm NOH HwH NOH A<v mm:
m . Fm mam omaa I-.. am am As SH 83228

s.NF H.0m as 0: III Hmv mmH

~.am m.wm N.Ne s.m> G.H.mm 0mm mmH Nm A40 0mm
m.mH N.ma 0F.mm H Hm smH *on 0mm maHmssa

m2 N2 H2 mp0oam 0 ana R 00A0>oo0m 0000 * 0>HH< * 00000H0m 0002

haosoo0m ha0>oo0m mHoOImp0oam SmHm onp0a0mo

0*0p0m thH0pHoz & H0poe smHm mo .02 anm 00m0>000m 00 .oz

.mama :H m00oa wsHp0H0o0m 000 mcwmsdm

0gp so 00Emomm0m mpsoswm0mx0 0w0mm0m nmwm seam 00>Hm00 0900 thH0phoa 000 0mdpm0o0m mo oomHm0msoo .0 0HQ0B

26

Table 10. Damage data from 197h rainbow trout passage experiments.

 

 

No. of No. of
No. of Dead Recaptured Recaptured % of Damaged
Test Fish Dead Dead Dead
Date Recaptured Undamaged Damaged Recaptured Fish
19 May h5* 37 8 17.8
21 Jun 22 21 1 h.5
12 Jul 1h 10 h 28.6
13* 10 3 23.1
6 Oct 22 16 6 27.3
3* 2 1 33.3
20 Oct 28 12 16 57.1
3 Nov 27 12 15 55.6

 

* Recaptured fish that were dead upon release into turbine.

27

 

 

 

0.0 II N Amav N
N N m.mH 4 ON A200 Om
N F N.a0 a s Asz mH soO soO FH
O.O II N Amav N
H H 0.0 N sN Hzmv ON
N N 0.0m a s AzHV O soO soO s
0.0 II NH Azmv NH
H s O O.FF sH s HEHV OH mesa mom HN
O O H.0m NH HN Hmav mm
m O.NH m NN Heal mN
H N N m.mm m OH Asz mH eoO mse mN
O.O II OH Hmav OH
H O.N H Om Hzmv Hm
H m 0.0s s O Hsz OH coo mse 0
N 0.0m N N Amav s
O.O II F meal a
N H m.mm m 0 Hsz a mesa HsO ON
a m a p.04 OH ON Hmav Om
O.O II N Heal N
s H m.mO m H *eAzHV O mesa esO mH
am mHa 0H m m ana seam sombeom Oomneoees smHa boos team
mo enHm saHm Ooom msHposomO pmme
*m0HHow0p0o 0w0E0Q 00w0a0a a 000m mo .oz @009 00 .02 Mo .oz
/

.mpc0aflh0mx0 0w0mm0m 30w.“ mbma Son.“ 900 0w0500 ho F0836 23.. 0.3
05

28

 

 

 

 

0oHpHdp cw ooHpod0oapoH com: 0000 00m00a0m H mm
mpHH0pmos 00m0H00 u 20
asHHopaoe opoHOoeeH u 2H
**
0sopxo0p :0xomm H mm
msesan eosoam u mHm
moH000Hp H0sm0ch n mH
00p0pmm0o0m u a
#00 Ho 00:00am u m
.x.
H 0.0m H H Hmav N
m m H 0.0F F m Fzmv OH
s a s 0 N.mm ON H Fsz HN mesa soz a
0.0 II H Hmav H
m H H N N.Hs F OH Hzmv FH
m s F 0.00H sH II HZHV :H mesa soz N
H 0.0m H H Hmav N
H O.mN H m Hemv s
O 0 0.0m NH m FZHV mH mesa soO OH
mm mHm 0H m m smHa seam Oommeom Oommemosa smHa oeoz ovum
No anm smHm Ooom meHsosomO mama
*m0Hmow0p00 000809 00w0a0m & 0000 mo 000a mo .02 00 .oz
//

AU.PGOUV

HH prdFH.

29

08thdp OF 00Hpos0ompmH 000: 0000 00m00H0m u 00
thH0paoa 00%0H00 n 20
mpHH0pmoa 0p0H0088H u 2H

08000000 000080 u 00
088£pmH 800000
00H000Hp H0sm0psH
00p0pH00o00

#80 Ho 00sm0Hm u m

IIIIII
000
HH
0

 

 

 

0 0 0.NN NH H: F0mv mm
s m m :.F 0H mNH F20v mMH
H F sH 0.Fs NN sN HZHV Os mcHsosmsoO
F m HH m.F: mN mN A0mv 0:
0 s H 2 0.0N mH Fm Az0v Nm
0 a :N Nm 0.H0 00 mH **F2Hv m0 08H0880
00 0H0 0H 0 m amHm 0000 0000800 000080080 ana 000:
mo emHa smHa mama msHsneomO
*w0Hmo00p00 000800 0000800 R 0000 00 .02 0000 00 .oz 00 .oz

.mFmH 8H 00008 00H9080800
080 00H0800 0:» so 008800H00 09:08Hh0080 0000000 gmHm 8080 0&00 000800 00 800H800800 .NH 0H00B

30

Table 13. Incidence of immediate and delayed mortality of recaptured
fish from 1975 fish passage experiments.

 

 

No. of No. of No. of Delayed % Immediate
Test Dead Immediately Mortality Mortality
Date Fish Killed Fish Fish of Dead Fish
15 Jun 8 6 2 75'0
20 Jul 16 9 7 56'3
8 Aug 61 10 51 16'h
25 Aug MO 15 25 37.5
21 Sep 30 18 12 60°C
II Oct 37 8 29 21°6
17 Oct A3 13 30 30'2
19 Oct 19 15 h 78'9
2 Nov 31 1h 17 h5'2

9 Nov 31 21 10 67'7

 

31

 

 

Table 1h. Comparison of immediate and delayed mortality data of

recaptured fish from fish passage experiments performed

on the pumping and generating modes in 1975.

No. of No. of No. of Delayed % Immediate

Plant Dead Immediately Mortality Mortality
Mode Fish Killed Fish Fish of Dead Fish
Pump 135 83 52 61. 5
Gen 181 h6 135 25.h

 

32

Size-selective mortality was assessed by an analysis of variance
of fish length in four possible recapture categories (recaptured live;
recaptured dead or died after capture; float—only recaptured; not
recaptured at all) for each experiment. If Size—selective mortality
existed, the mean length of the live recaptures should differ statisti-
cally from that of the fish killed by passage. Similarly, if size
influenced float retention or recapture itself, this Should be evident
in the mean length of fish comprising these groups.

Results of these analyses for 197A and 1975 are given in Tables
15 and 16, respectively. Normally, a probability level less than 0.05
would be indicative of significant variation. Hence, no size selective
mortality is indicated. A pairawise comparison of mean fish length
between all possible categories using Scheffe's interval (Gill, 1973)
further substantiated the lack of significant variation between category
means (P > 0.05 for all contrasts).

Since the fate of fish not recaptured or from which only a float
was recaptured could not be accurately established, inclusion of
these individuals in the size-selectively analyses may be misleading.
Accordingly, the l97h and 1975 experiments were reanalyzed by comparing
the mean Size of the live recaptures to that of the killed fish. In
all but one experiment, no difference existed between the two categories
(P > 0.110 for all trials). In the test of 20 October l97u, the mean
length of the live recaptures was Significantly greater than that of
the dead fish (311 vs 288 mm, P = 0.022).

The average length of fish used in the pumping trials varied
between 267 - 331 mm in 197% and 316 - 677 mm in 1975. Mean length

differed significantly between experiments within both years (P < 0.001

33

 

 

Table 15. Analysis of variance of body lengths of fish
for four possible recapture groups (recaptured
live; recaptured dead or died later; float-
only recaptured; not recaptured at all) in
pumping mortality tests performed in 1974.

Date Source df MS F P
21 Jun 74 Between groups 3 575.26 0.75 0.55
Within groups 91 767.28

12 Jul 74 Between groups 3 754.99 0.77 0.52
Within groups 72 985.23

6 Oct 74 Between groups 3 943.64 2.20 0.10
Within groups 71 429.86

20 Oct 74 Between groups 3 3711.39 2.02 0.12
Within ,groups 101 1841.84

3 Nov 74 Between groups 3 168.05 0.29 0.83
Within groups 90 574.89

 

3h

 

 

Table 16. Analysis of variance of body lengths of fish
for four possible recapture groups (recaptured
live; recaptured dead or died later; float-
only recaptured; not recaptured at all) in
pumping mortality tests performed in 1975.

Date Source df MS F P
15 Jun 75 Between groups 3 481.75 0.18 0.91
. Within groups 22 2703.99

20 Jul 75 Between groups 3 3111.97 1.40 0.25
Within groups 137 2231.29

21 Sep 75 Between groups 3 1170.97 0.50 0.68
Within groups 109 2347.50

19 Oct 75 Between groups 3 2095.50 0.59 0.63
Within groups 45 3581.23

2 Nov 75 Between groups 3 4349.09 1.19 0.32
Within groups 132 3660.15

9 Nov 75 Between groups 3 10107.32 2.61 0.05
Within groups 3871.21

 

35

for each year). Hence, if Size-selective mortality existed within the
size range of fish used, differences in size between live and dead
recaptures should have been most pronounced in those tests in which
relatively large fish were used. No evidence for this premise was
detected, however. These results suggest that pumping mortalities may
be dependent upon factors such as Spatial distribution of fish in the
intake or discharge, rather than on fish length.

The salmon experiment of 19 October was the only pumping trial in
which relatively large fish were used. Passage mortality was the
highest of any of the tests (90.5%) and indicated that, despite
previous analyses, fish Size may be decisive in passage success.
Examination of the relation between mortality and fish size (Fig. 2)
tends to illustrate this although it is obvious that most of the

estimates are derived from a narrow size range of fish.

Fish Passage - Generating Mode

Six mortality experiments were performed during power generation;
two trials in 197A and four tests in 1975. Yellow perch and chinook
salmon was used in l97h and rainbow trout were employed in 1975.

The test of 1h August 197A was procedurally different from the other
tests and was deleted from the present analyses. Generating trial
results for both years are listed in Tables 5 and 6.

Seventy-five live yellow perch were used on 28 August 197A.
Recovery of fish was the highest of any of the generating tests (81.3%)
as was the initial proportion of live recaptures (h6/76, 60.5%).

However, 26 of these fish died during the holding period resulting in

36

.mFmH 0:0 FmH .0000» 0
CHQESQ .HO H d O
2 . 0 0p0m A»
.H PM E 6G0 wNHm 30.2“ G00 .0 0
. . 39.09. @HSmQOHP
o o H m

.N 930E

37

of.

22.2. :9“. .._o :5sz 255.
com 08 0? com com on.

.389:8§ a.

5...; 3:; $0.8 3.9...
mub<¢ >P.J<.Eoz omkuummou 2.2-o
awkdm >._._._EOE ouhummmooz: ##2- 0

IO—

ION

won

50¢

uom

 

r3.

°/o

A 1.1 '1 VJ. HOW

38

a total mortality rate of 67.2%. Only three of the hl dead recoveries
exhibited physical damage (7.3%).

Fifteen dead perch were also released on 28 August. Of these,
eight were recaptured (53.3%) with two fish sustaining damage (both
slashed in half). Recapture rates differed between the live and dead
releases (x2 = h.02, P < 0.05) when based on recovered fish but were
not significantly different (x2 = 0.38, P > 0.50) when recaptured floats
were also considered. This disparity is not unexpected with the low
sample sizes involved.

In 1975, h55 live and 101 dead rainbow trout were used in generating
trials. As in 197h, recapture rates based on recovered fish (63.7%
live releases, 52.5% dead releases) differed between the two groups

2 = 3.97, P < 0.05) but were non-significant (x2 = 2.59. P > 0.10)

(x
when recovered floats were included. Recovery rates between years
differed significantly for the live releases (x2 = 8.1h, P < 0.0005)
but did not differ for the dead releases (x2 = 0.0h6, P > O.h0). Since
inter-trial recovery rates of live releases varied only slightly in
1975 (61.7 - 65.9%), the discrepancy between the one l97h run and the
four 1975 tests is considered biologically meaningless.

Adjusted generating mortality rates varied from 35.h — 7h.8% in

1975 (Table 6, M estimates). Although mean fish size (Table h) and

l
turbine operating conditions (Table 8) differed little between tests,
significant inter-trial differences in mortality existed (x2 = 25.32,

P < 0.0005). The two summer experiments (8 and 25 August) had much
higher mortality values than the two fall runs (h and 17 October). This

probably resulted from the high summer water temperatures and prolonged

handling inducing stress and mortality. Control group losses were 3 - 6

39

times greater in August than in October. A striking consequence of the
high control losses in the 8 August test was that the adjusted survival
rate assumed an impossible value (> 100% survival). Accordingly, the
summer mortality estimates may not be representative of generating
passage success.

The pooled 1975 generating data resulted in a mean unadjusted
mortality rate of 62.8% with a 95% confidence interval of : 27.lh%
(Table 9). Rates derived by considering detached floats and unrecovered
individuals as dead fish were 72.6 and 76.h%, respectively. Adjusted
mortality estimates (using a mean handling loss of 37.3%) were h0.7%
for recaptured fish, 56.3% for recaptured fish and floats, and 60.8%
assuming all fish not recaptured alive were killed.

Nearly half (h7.8%) of the fish killed during passage were
physically damaged (Table 12). Lacerations and decapitati ns accounted
for almost all (95.5%) of the injuries. Similarly, these damage types
were the only forms displayed by injured dead releases. Damage rates
differed significantly (x2 = 35.79, P < 0.0005) between the immediate
and delayed mortality groups of fish but no difference was observed

2 = 0.38, P > 0.50) in damage incidence between the live and dead

(x
releases (17.7% vs 22.6%).

Evidence for size selective differences between live and dead
recaptures in each generating trial was lacking (P > 0.065 for all
runs). Also, no trend was discerned among experiments between mean
fish length and mortality rate (Table 17), although mean fish size
differed little in the four tests. The absence of size selective

mortality during generating passage may be a consequence of the

relatively wide wicket gate setting used on this mode during the tests

MO

 

 

 

Table 17. Summary of adjusted mortality rates for
generating fish passage experiments performed
in 1975.
No. of Live Fish Mean Adjusted
Date Released Length Mortality
8 Aug 75 131 347 mm 74.8% a
25 Aug 75 79 358 mm 65.2%
4 Oct 75 129 358 mm 35.4%
17 Oct 75 114 380 mm 53.9%
a .
Based on a handling loss of 6.1% (Table 6)

hl

(approximately 82% open during generation vs 65% open during pumping).
This larger opening may allow a wide size range of fish to pass by

the turbine blades unimpaired. Presumably, the difference in wicket
gate settings between operational modes also accounts for the differences
in recovery and mortality rates between the pumping and generating

experiments (Table 9).

Board Passage — Pumping and Generating Modes

PumpinggMode

In l97h, seven board passage turbine tests were performed during
pumping (Table 18). In these trials, 22h 6-inch boards, 1h5 8-inch
boards, 2h2 l2-inch boards, 190 l8-inch boards, 185 2h-inch boards and
h8 26-inch boards were used. 0f the lO3h boards, 636 were subsequently
recovered (61.5% recapture rate). Recovery and damage rate generally
increased with board size in each experiment. The pooled data indicate
that board mutilation was minimal for the smaller sizes but was nearly
100% for the larger boards (Fig. 3). In most cases, damaged boards
were split in two or more pieces. Hence, the recovery rates of the
larger boards (more susceptible to damage) were expected to be higher
than the smaller ones since more than one piece per board (after

mechanical contact) was usually available for recapture.

GeneratingiMode

Generating mode board passage trials were performed on 1h and 28
August 197M. A total of 368 boards were used comprising four categories:
100 6—inch boards, 96 l2-inch boards, 101 l8-inch boards, and 71 2h-inch

boards (Table 18). Recovery rates were not significantly different

112

 

 

m.em mm m H.am Ha ma :N
m.mm ON mH e.mm mm ma mH
e.mH H m 0.0m m NH NH
2.3” : mm m.mm mm P: w moo: wcaasm
0.0 o mH m.Hm mH Hm m a» poo m 1 aHa
0.00H H o 0.0N H m mH
m.mm w mH m.oe aN am NH moo: waHaaoa
0.0 o m m.mm m a m a» poo m 1 mHa
H.mm mN wH N.ew H: e: :N
m.oe NH mN m.em N: we mH
:.mH e Hm m.Nm mm we NH oooz waHuoaoaoo
m.e N am e.mm H: a: m 4» m:< wN 1 NH%
m.am NH OH m.Hm NN aN aN
e.we HN 2N 0.2m m: mm mH
H.mN NH am O.Nm we on NH moo: weraaoaoo
N.N H as N.mw ma Hm m ae mo< 2H 1 OHa
p.0N m MN N.mm mN a: NH moo: maHaeam
e.m H mN H.mm eN a: w ae sHse NH 1 ma
a.mm mN m H.Ne Hm ma aN moo: weHaaaa
H.:e ON e :.Hm eN a: NH a» maze HN 1 mu
o.em HH w m.mm 0H ma NH moo: maHaeaa
m.mH N mH 0.0m mH cm . m a» has OH 1 ma
mcnwom 033.95 owdada oz R cmhm>ooom couscouch Ammnoefi moo: stofipmpmmo
cmhmueoomm .8 3m x 8.6ch 3. 3960mm mvudom mohdom muwm one
on» mo mo .oz mo .oz oaoom memo was pmm
omwdada x 3.30m cmuoaroomm \ &

 

 

69:25.8me mmwmmdm @300. iba mo hhwgm .m— 0.3.95

M3

 

 

 

.mmu .mmm 0.00 ONO Omm
O.OO Om ON 0.00 mO He ON
O.OO Om OO H.OO PO HOH OH
O.NN OH OO O.OO OO OO NH
O.O O OO 0.00 OO OOH O OaHooamaoO
mmm .O1Hm O .HO mmm OOOH
H.OO Om H O.N» Om OO ON
H.OO OOH OH 0.00 OOH OOH ON
O.OO HO OO 0.00 OOH OOH OH
O.NO HO OO H.NO ONH NON NH
O.OH NH HO O.OO me OOH O
0.0 O OO O.OO OOH ONN O OaHaesO
nOHEOB
H.OO Om H O.NO Om OO ON
O.OO OO O O.OO mO OO ON
O.He ON O 0.00 NO OO OH
0.0N O OH O.OO ON OO NH
0.0N O OH O.OO ON OO O OOoz OoHaeaa
O.O O HN O.NO HN OO O OO >oz OH 1 OHO
0.00 OO O 0.00 OO OO ON
O.HO ON OH O.OO OO OO OH
0.00 O OH 0.00 ON OO NH
H.HH m ON H.OO ON OO O OOoz OeHaeaa
0.0 H OH O.NO OH OO O O» poo ON 1 OHO
mouwom wmxowho mmdawa oz & omgo>oowm cmosdoman Ammsosfiv moo: stoOvdummo
wopm>oomm no paw * poech * hum>oomm muhwom mcaOom mufim wad
OMMWBMM “ mwhdom Oohm>ooom mo .02 mo .oz onwom open one * pmme

 

(l

M4

. a
Ohma mpmop mchmumsmm was msfimasm Mom mpwa mmOEOO cow ouflm whoop coozpon QHsmsoprHmm

.m madman

(*5

whiz—hm“ @z_._.<¢mzu0 I
m.—<2.kmm 0252.3. 0

32.2.5 mN_m 9.200

0.
p

m.
_

N.
_

m
p

”(O

-l'0

1O.
rON
OOO
.OO

100

tom

 

H8.

%

OBSVWVO

h6

between board sizes (x2 = 0.356, P > 0.90) and averaged 87.0% overall.

Damage rates, however, increased with board size (Fig. 3).
Comparison of the pumping and generating damage rates revealed

no significant difference for the 6-inch (x2 = 0.059, P > 0.80) and

2 = 0.59, P > 0.h0), but a marked difference for the

12-inch boards (x
larger boards (l8-inch, x2 : 9.09, P < 0.005; 211-inch, x2 = 26.63,

P < 0.0005). As in the fish passage trials, these operational mode
differences are probably due to the wider wicket gate opening on the

generating mode permitting more larger sized boards to pass unmarred

than was possible during pumping.

DISCUSSION

The turbine passage experiments conducted at Ludington during
l97h and 1975 represent the first intensive field assessment of fish
mortality at a pumped storage facility. As such, the procedures and
methodologies developed may prove suitable for similar studies at
other pumped storage sites. The use of float tags and net bags to
conduct these tests is believed to offer great potential for evaluating
passage success given a modest amount of equipment, personnel, and
money.

Although the float and bag gear have great utility, little infor-
mation exists on the effects of float attachment or bag enclosure on
the orientation and survival of fish in fish passage situations.
Because of the confining nature of the bag to fish movement, fish
survival during passage may be adversely affected. If this is true,
derived mortality rates using this equipment may be too liberal; that
is, mortality may be over-estimated (or conversely, survival under—
estimated). Although this question warrants further research, the
effect of the float and bag technique may never be adequately resolved.
Hence, interpretation of results will vary depending on how the
methodology is perceived as affecting fish behavior.

Derivation of mortality quotients from each of the Ludington
experiments was accomplished in three ways: (1) by considering only

recaptured fish; (2) by considering the float-only recaptures as dead

h?

h8

fish and combining these with the recaptured fish results; and (3) by
considering all fish not recaptured alive as mortalities. Method 1 and
2 provide the most realistic estimates of turbine mortality. Method
3 assumes that fish release to the turbines is 100% successful (i.e.,
no fish are caught in the trash slots or missed penstock entry) and
that all fish surviving passage are recaptured. Fulfillment of these
criteria was not always accomplished at Ludington as demonstrated by
the recapture of test fish many miles from the plant (missed turbine
entry) and the recovery of fish surviving passage for several days
after the initial recapture efforts. Observations at the intake of
turbine units after shut-down further substantiated that turbine
releases were seldom complete.

The recapture rates of fish, and fish and floats, for the pooled
Ludington data, were 35.7 and 73.3% during pumping and 61.3 and 8h.2%
during generation (combined data, both years, live releases). Hence,
an average of 26.7 and 15.8% of the introduced fish were unrecovered.
Similar mean recapture rates (78.6% fish and float recovery, 21.h% not
retrieved) were obtained in Connecticut River passage tests (Johnson,
1970) and illustrate that incomplete fish recovery is not unique to
the Ludington studies. Power plant features, release methodology and
intensity of recapture effort may all influence the recovery rate.

The average adjusted pumping mortality in 1975 (67.7%) was greater
than the unadjusted rate in 197% (56.5%). Although the mean size of
fish was generally slightly larger in 1975, both morta ity values are
believed to estimate the same overall parameter. The clustering of the
point estimates shown in Figure 2 appears to support this premise.

For generating experiments, the mean 1975 adjusted mortality value

h9

(h0.7%) was much lower than the one l97h estimate (67.2%). Interpreta-
tion of this difference is difficult, however, because of the large
differences in sample size between years and the use of different
species. This latter aspect is especially important since yellow

perch are physoclistous and may be more prone to pressure-related
injury than the physostomous trout (Beck et aZ., 1975; Tsvetkov et aZ.,
1972; Foye and Scott, 1965).

The disparity between pumping and generating fish mortality esti-
mates is presumably a function of turbine characteristics, particularly
wicket gate opening. Wooden board damage rates (between modes)
paralleled the operating mode differences found in fish and further
substantiated the importance of engineering factors in passage success.
These data agree well with laboratory results of mortality in relation
to the dynamic characteristics of turbines (Bell et aZ., 1967).

Although an obvious relation between board size and damage existed,
a similar pattern was not clearly apparent in either the pumping or
generating fish trials. Several explanations can be advanced for the
inconsistency between the two sets of results. Most plausible, perhaps,
is that mechanical damage is not the sole factor in causing fish
passage mortality. Rather, shearing forces and cavitation may also
be operable. Existence of these factors in the tests is suggested by
recapture of decapitated fish and fish with missing pieces of flesh
(shearing action), and metal pitting of the turbine blades (cavitation).
Damage data from the 1975 live releases (using both immediate and
delayed mortalities) indicated that only h3.h% of the pumping injuries
(36/83) and 53.1% of the generating injuries (17/32) were of a mechanical

nature (slashes, cuts or abrasions - Table 12). Weekly observations of

50

dead fish in the reservoir also showed that many fish lacked heads or
displayed other shearing action type damage (i.e., broken gill arches).
Hence, the finding by Long (1968) that factors other than fish size
(i.e., spatial distribution near the turbine) affect fish mortality
may apply at Ludington as well.

A second interpretation of the discrepancy between the fish and
board data is that the size range of fish used in the tests was too
narrow for size-selective mortality effects to be detected. Excluding
the salmon test in which good agreement with the board data was evident
(mean fish size = 27 inches, mortality = 90.5%; board size = 26 inches,
damage = 97.1%), the mean length of trout in any experiment never
exceeded 15 inches (381 mm) with all fish included within a range of
96 - 532 mm (3.8 - 20.9 inches). Thus, the damage rates observed for
the larger sized boards (greater than 18 inches) could not possibly
be statistically paralleled in the fish due to a lack of the appropriate-
sized individuals. Comparison of the fish mortality data with board
results of a similar size (12 inches) showed that both pumping and
generating fish mortalities were much higher than the respective board
damage rates (Table 17, Figs. 2 and 3). Again, the importance of
factors other than mechanical contact is implied.

Conceivably, handling mortalities were of sufficient magnitude as
to "mask" in the fish tests the size selective effects seen with the
boards. This probably occurred in the generating trial of 8 August 1975
and may have also taken place in the tests of 25 August and 21 September
1975. However, for all other runs, handling mortality never exceeded
16.0% and was thus well-controlled. Accordingly, the "masking" of size-

selective mortality in fish via handling losses is considered remote.

CONCLUSIONS

Mortality studies at Ludington indicated that passage success of
fish through the pump-turbines was relatively low, particularly during
pumping. Much effort was expended in the development of techniques
to provide accurate and precise estimates of turbine mortality.
Although a limited number of species was studied and a narrow size-
range of fish used, the mortality assessments are the first of their
kind for a pumped storage facility. Hence, these studies should prove
valuable as a foundation on which future investigations can be based.

Several problems remain unsolved in using the derived mortality
estimates. Foremost of these is the determination of the number of
fishes passing through the turbines. Population estimates are needed
for lake species affected by the power plant as well as for species
found within the reservoir. Only with this knowledge can yearly
estimates of fish loss be obtained.

Secondly, the significance of turbine losses to the welfare of
fish stocks is unclear. In some cases, the ecological consequences
of plant mortality are probably slight (i.e., kills of spawning-run
Pacific salmonids that would naturally die shortly after spawning).
However, data are lacking on the proportion of the lake populations
that undergo turbine passage. Consequently, an assessment of biological

impact is difficult.

Sl

52

Even if the number and proportion of each of the lake species
affected by the power plant is known, the resiliency of these species
to losses is obscure. Fishes exhibit many compensation mechanisms
in response to population removals and thus turbine mortalities may
not necessarily significantly affect either carrying capacity or
sustained yield. Further information of the dynamics of the lake
populations in relation to mortality would be helpful.

The state-of—the-art for evaluating fish passage and its impact
on fish population dynamics is admittedly in its infancy. As more
data and refinements in technique become available, a clearer under-

standing of these interactions will assuredly ensue.

SUMMARY

Passage mortality of fish through pump-turbines at the Ludington
Pumped Storage Power Plant was studied in 197h and 1975. Procedures
were developed for introducing fish to the turbines using weighted
paper sacks, and for recapturing individuals using jaw-fastened
styrofoam floats. Commercially procured fish were used in almost all
the field trials. Passage studies of various sized wooden boards were
also performed to provide supplementary data on size and mechanical
injury.

Ten fish tests were accomplished in both 197k and 1975. In 197b,
1017 fish were used comprising three species: rainbow trout, chinook
salmon and yellow perch. During 1975, 1725 fish were used with all
but 51 fish being rainbow trout. Control groups of fish were used
in each trial during 1975 except one run in which coho and chinook
salmon were tested.

Pumping mode mortality estimates were derived from five tests in
197h and six tests in 1975. Data from three 197h pumping experiments
were excluded because of procedural differences. A total of hh5 live
fish were released in the five valid 197h tests of which 200 or h5%
were recovered. Known dead fish releases resulted in a 38% recapture
rate which was not significantly different (P > 0.10) from the live
release rate. Satisfaction of this equality was requisite for using

the float-recapture technique. In 1975, 639 live fish were used during

53

5h

pumping; 35% were recaptured. The recovery rate of 132 dead releases
(36.h%) was again similar to the live fish (P > 0.70). Yearly
differences in live recapture rates could not be explained.

Unadjusted pumping mortalities ranged from 33 — 63% in 197h tests
and averaged 56.5% with a 95% confidence interval of i 13.3%. In
1975, adjusted mortality varied from 53.8 — 75.h% with a pooled corrected
mean estimate of 67.7% i 7.2%. Average handling mortality in 1975
was 13.8%. Mean adjusted mortality values derived by considering
detached floats and unrecovered fish as dead fish were 75.1 and 8l.h%
in 197%, and 86.h and 91.9% in 1975, respectively. Although differences
in pumping mortality existed between years, both of the mean mortality
rates are thought to be estimates of the same parameter.

Damage to the live releases was lower in 197% (37.2%) than in 1975
(61.5%) (P < 0.005). In 1975, 73.5% of the injuries were lacerations
or decapitations implying that mechanical contact and shearing forces
were causative agents. The majority of 1975 deaths (61.5%) were
immediate in nature.

The results of size—selective mortality analyses proved ambiguous
for pumping experiments. Most analyses showed no difference in size
between dead and live fish. However, mortality was highest (90.5%)
in one trial in which large salmon were used. The narrow size range
of fish used in most of the pumping trials may have obscured the
detection of size-related mortality.

Six generating mortality experiments were performed; two in 197h
and four in 1975. One of the l97h trials was inappropriate for
comparative analysis. Seventy—five yellow perch were used in the one

valid l97h test; 81.3% of these fish were recaptured. Eight of 15 dead

55

releases were recovered as well (53.3%). These recapture rates were
significantly different (P < 0.05) but proved non-significant when
recaptured floats were also included (P > O.h0). Mortality in this
test averaged 67.2%. Only three of the live releases and two of the
dead releases exhibited damage.

In 1975, h55 live and 101 dead rainbow trout were used in
generating tests. Recapture rates between the groups were different
(63.7 vs 52.5%, P < 0.05) but were similar when recovered floats were
included (P > 0.10). Yearly differences in the live release recovery
rates are probably biologically meaningless.

Adjusted 1975 generating mortalities varied from 35.h - 7h.8%
and averaged h0.7% i 27.1%. Average handling mortality was 37.3%.
Mean adjusted mortality values derived by considering detached floats
and unrecovered fish as dead fish were 56.3 and 60.8%, respectively.
Due to high handling losses, mortality estimates from summer generating
tests may not be representative.

Almost half (h7.8%) of the fish killed during generating passage
were damaged; nearly all of the injuries were lacerations or decapita-
tions (95.5%). Damage rates differed between fish sustaining immediate
and delayed mortality (P < 0.0005) but were similar between the live
and dead releases (17.7 vs 22.6%, P > 0.50).

Evidence for size-selective mortality in generating trials was
absent. The relatively wide wicket gate setting (82% Open) probably
allowed a large size range of fish to pass by the turbine blades
unimpaired. Presumably, the difference in wicket gate setting between
operational modes (82% generating, 65% pumping) accounts for much of
the difference in recovery and mortality rates between the pumping and

generating trials.

56

Nine wooden board passage tests were conducted in l97h; seven
pumping and two generating tests were performed. A total of 103h
boards were used (comprising six size groups) during pumping of which
61.5% were recovered. In pumping tests, recovery and damage rates
increased with board size. Mutilation was low for the smaller sized
boards (6-inch - 3.9%) but virtually complete for the largest boards
(26-inch — 97.1%). In the generating tests, 368 boards were used
(four size categories) resulting in an overall recapture rate of
87.0%. Recapture rates did not differ between size groups (P > 0.90)
although damage rates increased with board length.

Pumping and generating board damage rates were similar for size
groups 12 inches or less (P > 0.h0) but differed among the larger sized
boards (P < 0.005). These differences were attributed to the difference
in wicket gate setting between the operating modes.

Several explanations are advanced for the discrepancy in mortality
and damage between the fish and board studies. The most plausible is
that factors other than mechanical contact contribute significantly
to fish death. Also, the limited size range of fish used relative
to the board lengths may account for the differing results. Fish
losses due to handling mortality are thought not to have "masked"
size-selective mortality in the fish tests.

Unresolved problems in applying the mortality quotients derived

from these studies are delineated.

LITERATURE CITED

LITERATURE CITED

Beck, A. P., G. V. Poje, and W. T. Waller. 1975. A laboratory study
on the effects of the exposure of some entrainable Hudson River
biota to hydrostatic pressure regimes calculated for the proposed
Cornwall Pumped Storage Plant, pp. 167—20h. in S. B. Saila (ed.)
Fisheries and Energy Production: A Symposium. D. 0. Health and
Company, Toronto.

Bell, M. C. 197k. Fish passage through turbines, conduits, and spill-
way gates, pp. 251—261. in L. D. Jensen (ed.) Proceedings of the
Second Workshop on Entrainment and Intake Screening. Rept. 15,
Electric Power Research Institute, Palo Alto, California.

Bell, M. C., A. C. DeLacy, and G. J. Paulik. 1967. A compendium on
the success of passage of small fish through turbines. Rept.
to the U. 8. Army Corps of Engineers, Contract No. DA-35-O26-
CIVENG-66-16. 268 p.

Comninellis, E. 1973. Ludington pumped storage project. Journal of
the Power Division, ASCE, 99(P01): 69-88.

Cramer, F. K. 1960. Fish passage through turbines, model turbine
experiments. U. S. Army Corps of Eng., Walla Walla District,
Prog. Rept. 1: 17 p.

Cramer, F. K. 1965. Fish passage through hydraulic turbines. U. S.
Army Corps of Eng., Walla walla District, Memorandum Rept. 1:
16 p.

Cramer, F. K., and R. C. Oligher. 1960. Fish passage through turbines -
tests at Cushman No. 2 hydroelectric plant. U. S. Army Corps of
Eng., walla Walla District, Prog. Rept. 2: 26 p.

Cramer, F. K., and R. C. Oligher. 1961. Fish passage through turbines,
further model turbine experiments - Francis runners. U. S. Army
Corps of Eng., Walla Walla District, Prog. Rept. 3: 33 p.

Cramer, F. K., and R. C. Oligher. 196%. Passing fish through hydraulic
turbines. Trans. Amer. Fish. Soc. 93: 2h3-259.

Foye, R. E., and M. Scott. 1965. Effects of pressure on survival
of six species of fish. Trans. Amer. Fish. Soc. 9h: 88—91.

Gill, J. L. 1973. Current status of multiple comparisons of means in
designed experiments. J. Dairy Sci. 56: 973—977.

57

58

Hauck, F. R., and Q. A. Edson. 1976. Pumped storage: its significance
as an energy source and some biological ramifications. Trans.
Amer. Fish. Soc. 105: 158-l6h.

Johnson, F. A. 1970. A device for fish recovery during turbine-
passage mortality studies. Prog. Fish-Cult. 32: 236-239.

Ley, R., and E. Loane. 1962. Symposium on pumped storage: general
planning of pumped storage. Journal of the Power Division, ASCE,
88(PO2): 211-232.

Liston, C. R., and P. I. Tack. 197k. A study of the effects of
installing and operating a large pumped storage project on the
shores of Lake Michigan near Ludington, Michigan. 1973 Ann.
Rept. to Consumers Power Co., Vol. 1, Dept. Fish and Wildl.,
Michigan State Univ., 17h p.

Liston, C. R., and P. I. Tack. 1975. A study of the effects of
installing and operating a large pumped storage project on the
shores of Lake Michigan near Ludington, Michigan. l97h Ann.
Rept. to Consumers Power Co., Vol. 1, Fisheries Research, Dept.
Fish and Wildl., Michigan State Univ., 166 p.

Long, C. 1968. Diel movement and vertical distribution of juvenile
anadromous fish in turbine intakes. U. S. Fish Wildl. Serv.,
Fish. Bull. 66: 599-609.

Lucas, K. C. 1962. The mortality to fish passing through hydraulic
turbines as related to cavitation and performance characteristics,
pressure changes, negative pressures, and other factors, pp. l-2h.
in Symposium on Cavitation and Hydraulic Machinery. International
Association for Hydraulic Research.

Monten, E. 1963. The possibility of salmon smolt passing unharmed
through power plant turbines when descending to the sea. Trans-
lated from Swedish by the U. S. Joing Publication Service, for
the Fish Passage Research Program, U. S. Bureau of Commercial
Fisheries, Seattle, Washington. June, 1963.

Muir, J. F. 1959. Passage of young fish through turbines. Journal
of the Power Division, ASCE, 85(P01): 23—h6.

Salzman, M. 1962. Symposium.on pumped storage: site investigation
of pumped storage facilities. Journal of the Power Division,
ASCE, 88(PO2): 233-251.

Schoeneman, D. E., and C. O. Junge, Jr. l95h. Investigations of
mortalities of downstream migrants at two dams on the Elwha River.
Wash. St. Dept. Fish. Res. Bull. 3: A3 p.

Schoeneman, D. E., R. T. Pressey, and C. 0. Junge, Jr. 1961. Mortali-
ties of downstream migrant salmon at McNary Dam. Trans. Amer.
Fish. Soc. 90: 58-72.

59

Schoumacher, R. 1976. Biological considerations of pumped storage
development: introductory remarks. Trans. Amer. Fish. Soc.

105: 155-157.

Serchuk, F. M., c. R. Liston, and P. I. Tack. 197%. A study of the
effects of installing and operating a large pumped storage
project on the shores of Lake Michigan near Ludington, Michigan.
1. Mortality tests through turbines. II. Movement patterns
and orientation of fish in the Ludington reservoir as determined
by ultrasonic telemetry. Ninth Quarterly Report to Consumers
Power Co., Dept. Fish and Wildl., Michigan State Univ., 33 p.

Serchuk, F. M., C. R. Liston, and P. I. Tack. 1975. An evaluation of
fish passage through hydraulic turbines at the Ludington Pumped
Storage Facility. Tenth Quarterly Report to Consumers Power Co.,
Dept. Fish and Wildl., Michigan State Univ., 21 p.

Tack, P. I., and C. R. Liston. 1973. A study of the effects of
installing and operating a large pumped storage project on the
shores of Lake Michigan near Ludington, Michigan. Eighth
Quarterly Report to Consumers Power Co., Dept. Fish and Wildl.,
Michigan State Univ., 17 p.

Tsvetkov, V. 1., D. S. Pavlov, and V. K. Nezdoliy. 1972. Change of
hydrostatic pressure lethal to the young of some freshwater fish.
J. of Ichthyology 12: 307-318.

U. S. Army Corps of Engineers, North Pacific Division. 1960. Effect
of structures at main Columbia River and certain other dams on
downstream migration of fingerling salmon. Research Rept.,
Fisheries Engineering Research Program, U. S. Army Corps of
Engineers District, Portland. 83 p.

Velz, C. J., J. D. Calvert, Jr., R. A. Deininger, W. L. Heilman, and
J. Z. Reynolds. 1968. Pumped storage for water resources
development. Journal of the Sanitary Engineering Division,
ASCE, 9h(SA1): 159-170.

Von Guten, G. H. 1961. Fish passage through turbines. Journal of
the Hydraulics Division, ASCE, 87(HY3): 59—72.

Von Raben, K. 1957. Regarding the problem of mutiliation of fishes
by hydraulic turbines. (Translation from the German, "Zur Frage

der Beschadigung von Fischen durch Turbinen." Die Wasserwirtschaft
h: 97-100). Translated by the Fish. Res. Board Can. Transl.
Ser. hh8.

Woodruff, R. 1971. Pumped storage: state—of-the—art. Journal of
the Power Division, ASCE, 97(P03): 675-695.

MOVEMENT PATTERNS AND ORIENTATION OF FISHES
IN THE LUDINGTON PUMPED STORAGE RESERVOIR

AS REVEALED BY TRACKING STUDIES

Fredric Michael Serchuk

A DISSERTATION

Part II

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

Department of Fisheries and Wildlife

1976

INTRODUCTION

Behavioral investigations of fishes in habitats affected by pumped
storage operation are few and have been generally restricted to netting,
seining, and mark and recapture procedures (Robbins and Mathur, 1976;
Estes, 1971). Little information exists on the response of fish to
water velocity changes, alteration of habitat, and related aspects
associated with the biological impact of pumped storage projects
(Schoumacher, 1976). Data are also lacking on the environmental and
ecological components related to the passage of fish into pumped
storage systems, and the behavior of species retained within these
facilities.

Environmental studies at the Ludington Pumped Storage Power Project
showed that nearly all fish species in the adjacent Lake Michigan waters
have entered the Ludington reservoir (Liston and Tack, 197A, 1975).

The purpose of the present research was to determine movement patterns,
activity cycles, and residence periods of free-ranging fish in the
Ludington reservoir and to relate these attributes to environmental

cues and operational characteristics of the power plant. Field
observations were accomplished during 197A and 1975 with carp (prrinus
aarpio), brown trout (SaLmo trutta), and rainbow trout (SaZmo gairdheri).
These species were studied because of their availability, seasonal
abundance in the reservoir, and capability of retaining sonic and

float tags.

60

METHODS AND MATERIALS

Ultrasonic telemetry and float tracking techniques were used to
determine the behavioral patterns and orientation of fish. These
procedures have been previously employed in fish homing and movement
studies (Stasko at aZ., 1976; Kelso, 1976, 197A; Groot et aZ., 1975;
Warden and Lorio, 1975; Dodson and Leggett, 197A, 1973; Jahn, 1969,
1966; McCleave, 1967; see Stasko, 1975 for others), and are attractive
since they leave fish relatively unhindered and allow prolonged contact
under a variety of environmental conditions. Float tagging is
especially productive because a large sample size can be obtained

at low cost without sacrificing accuracy (Stasko, 1971).

Equipment

Ultrasonic transmitters and receiving equipment were procured
from commercial sources. Two models of ultrasonic transmitters were
used: Smith-Root SR 69-B units and Bayshore Systems Corporation
Acoustic T-2 tags. Both were cylindrical, polystyrene-housed,
location-type transmitters possessing constinuous pulsed output.

The SR 69-B tags were 6% mm long by 1h mm in diameter, weighed 12 g
in water, and had an operating frequency of 7h kHZ. The T-2 trans-
mitters were 51 mm long by 1h mm in diameter, weighed 9 g in water,
and possessed transmitting frequencies from 52-73 kHZ. Both models

yielded detection ranges of 500—1000 m. Tags with useful lives of

61

62

7, 1%, and 35 days were used, as were various pulse rates (l-h pulses/
sec) to permit individual fish identification.

The receiving units consisted of a Smith—Root TA-60 Sonic Receiver
and a Bayshore Systems LF-25 Receiver. Both were portable, battery-
operated, and tunable in a frequency range of 60-180 kHZ. During
operation, the receiver was coupled to a hand-held, unidirectional
hydrophone (Smith-Root SR—70-H) having a conical beam pattern of 80
with a peak sensitivity of TA kHZ. Earphones were used during sonic
operations to exclude extraneous background noises.

Float-tracking devices consisted of styrofoam spheres, 10.2 cm
in diameter, color-coded with flourescent paint, and numerically
labelled to afford specific identification of individuals. Each tag
was connected by 5.5 m of monofilament nylon line to a looped metallic

fish pin used in attaching the device to the fish.

Capture and Tagging of Fish

Adult fish were captured in the Ludington reservoir by gill nets.
Fish used for tracking were selected from individuals that lacked
extensive gill net lesions or other visible injury, and were normally
marked, at the time of capture, with two floy tags for identification.
The average total body length for tracked fish was 61 cm for carp and
62 cm for trout. In 197A, fish were placed in a holding cage in the
reservoir or a continuous flow maintenance tank at the fisheries
laboratory prior to transmitter or float attachment. In 1975, fish
were processed aboard boat soon after capture.

Two methods of ultrasonic tagging were used. for trout, trans-

mitters were administered orally and inserted into the stomach by a

63

syringe constructed of PVC piping. Care was taken not to rupture the
stomach linings or otherwise inflict injury during this process.

With carp, the sonic tags were fastened externally just below the
anterior section of the dorsal fin. Transmitters were affixed with
their axis parallel to that of the fish by pins inserted through the
uppermost back of the fish. The pins were secured to the transmitter
with a small vinyl covering wrapped tightly about the tag with
electrical tape. Once inserted through the fish, the pins were fitted
‘with a Peterson disc to retain the transmitter snugly to the body.

No anesthetics were used prior to either mode of tagging. The entire
procedure was generally accomplished in less than two minutes, after
which the fish was returned to water.

Float tags were attached by drawing the steel fish pin, knotted
to the nylon line securing the float, through the dorsal musculature
immediately anterior to the dorsal fin. The pin was then crimped
against a Peterson disc to hold the tag firmly in place. This procedure
was modified slightly with some carp where attachment was made by
tying the float line to the base of the anterior spine in the dorsal
fin.

During 197A, tagged fish were held in a live box in the reservoir
for at least 12 hr after tagging to assure recovery from handling and
to permit buoyancy adjustments due to tag weight. In 1975, fish were
observed in a tank aboard the tracking boat for 5-15 min after tagging
and subsequently released in the reservoir if activity appeared
normal. In this latter year, fish were usually not tracked on the
day of release to avoid recording possible abnormalities in behavior

induced by tagging trauma.

6h

Tagged fish were either released in the reservoir by overturning
the live box or by transporting netted fish to various randomly—selected
shoreline release sites. Time between capture and release varied
from 0.1 (immediate release) to 259.h hr (released after 11 days in

the laboratory holding tank).

Tracking Procedures

Movements of tagged fish were monitored by maneuvering the
tracking vessel close to and in line with the position of the fish.
Fish location was assessed by the intensity and directionality of the
incoming signal, or visually with float-tagged individuals. Although
boat noise appears to have little effect on fish (Peterson, 1975;
Stasko et aZ., 1973; McCleave and Horrall, 1960; Hasler et aZ., 1969;
Johnson, 1960), the direction of approach was varied to avoid biasing
the orientation of tracked specimens. A position fix was recorded
once every 10—30 min (generally 20 min for carp; 15 min for trout) and
was determined by triangulation with a marine sextant using embankment
features (20 reservoir dike-load centers) as landmarks. On several
occasions (particularly with float-tagged fish), two or more fish
were tracked simultaneously by alternating between fish locations on
a fixed time schedule. These tracks proved useful in assessing the
variability in movement pattern and speed among individuals exposed
to similar environmental conditions.

Throughout each track, data on wind strength and direction, sun
visibility, light penetration, water and air temperature, and atmos-
pheric pressure were recorded coincident with fish position. Environ—

mental information records maintained by Consumers Power Company at

65

the plant were also used when field measurements were inconsistent
or lacking. Information on the operational status of the pumped
storage facility during tracking periods (Operational mode, number
of turbines on, and water elevation changes) was also accessed from

Company records.

Data Analysis

Position fixes for each track were plotted on a large-scale map
of the Ludington reservoir by use of a three-arm protractor. The
plotted course of each fish consisted of a series of points connected
by lines best thought to represent the path of movement between
successive positions. Digitization of the mapped position plots
facilitated computer analysis of the field observations and permitted
the determination of swimming speed, angular change of movement, and
movement pattern for each interval, for selected intervals, and for
the entire track.

Swimming speed was calculated from the distance travelled along
the estimated course and is considered a "calculated speed" since
corrections for current velocity, swimming depth, and non-linear
movement could not be accurately assessed. Movement rate was usually
expressed in body lengths/second (L/s) since this is the simplest
and most useful method of comparing swimming speed performance of
fish of different sizes (Webb, 1975). Angular change was evaluated
by determining the direction change (in degrees) between the straight
line course established for an interval, and the line constructed for

the subsequent interval. Angular means, rather than arithmetic means,

66

were computed for all track-level analyses (Batschelet, 1965, 1972).
Movement patterns were appraised from graphs of tracks derived from
CalComp maps.

Periodicity in swimming speed was determined by grouping movement
rate data according to time of day (hourly intervals). Speed estimates
were averaged from all members of the same species to provide a mean
value for each time interval. Data obtained from the same fish on
different tracking days were considered independent and were included
together in the determination of each of the hourly means.

The influence of environmental, behavioral, and power plant
operating characteristics on fish movement was evaluated by stepwise
linear multiple regression procedures (Draper and Smith, 1966). An
a = .25 significance level was chosen for both the entry and deletion
criteria of a variable. A direct examination of residuals for each
analysis was performed to check for lack of linearity and to determine
whether the assumptions concerning the error components were met.

All residual examinations (graphs not included) indicated that a
linear relationship was appropriate and that the data were relatively
free of abnormalities.

The minimum retention time for individuals in the Ludington
reservoir was assessed by noting the date on which tagged fish were
last observed or detected in the reservoir. Supplementary information
on residence time was obtained from mark and recapture efforts using

gill nets.

RESULTS

In 197A, 29 carp were tracked for periods ranging from l—2h hr
(median 8 hr) for a total of 630 hr covering 80 tracking efforts
(Table 1). Thirty tracks were accomplished using 8 sonic-equipped
fish and 50 tracks were performed with 21 float-tagged individuals.
During 1975, 36 fish were tracked; 23 carp were observed from 2.6-
2h.0 hr per effort (7 hr median) for a total of 351 hr, and 13 trout
(brown and rainbow) were followed for periods of 2.h-2h.0 hr (median
7 hr) for a total of 178 hr (Table 2). Sonic tags were used with 28
of the 51 carp tracks (55%) and 23 of the 26 trout tracks (88%).

Distance of individual carp tracks in 197A ranged from 0.07-l6.h9
km (median 2.h0 km) for a total of 219.h0 km (Table 3). In 1975,
individual carp paths varied from 0.39-9.23 km (median 1.90 km) for
a total of 129.70 km. Trout tracks covered distances of 0.15-27.hh km
(median 7.90 km) for a total of 2112.110 km (Table 11).

Most tracks commenced in the morning and terminated in late
afternoon. Three carp and one brown trout were tracked through the
night. Seven tracks were conducted at dawn (6 carp and l brown trout)
and five tracks at dusk (all carp). In some cases, tracks were abbrevi—
ated because of loss of signal or equipment failure, lack of fish
movement for an extended time (several hours), or adverse weather

conditions.

67

68

 

 

Table 1. Summary of 197A tracking activities.

Track No. of Duration of
Type Species Tracks Tracks
Sonic 8 carp 30 287.6 hours
Float 21 carp 50 3h2.7 hours
Total 29 carp 80 630.3 hours

 

69

 

 

Table Summary of 1975 tracking activities.

Track No. of Duration of

Type Species Tracks Tracks

Sonic 8 carp 28 188.5 hours
8 brown trout l6 l2h.0 hours
3 rainbow trout 7 36.6 hours

Float 15 carp 25 162.6 hours
2 brown trout 3 17.2 hours

TOTALS:

Sonic 19 fish 51 3h9.1 hours

Float 17 fish 28 179.8 hours

Total 36 fish 79 528 . 9 hours

 

.453.-

I L v-u,1-"mvz CA. .fl': — .411—

"d

:—

ir-I - -Cl..\l.‘

"li'xl -

-‘-."-\u-.\:-

5..

C 9‘ - nlvil‘U.

'\Ic Inl. ‘ Ial‘ulu

70

 

 

OH HH OH.\~H. 0.0 0.0 OH.O OO.O OOO aeoHO OO1OO OO1O~ deco .\O
OH NO OO.HOO. H.O O.N O0.0 O0.0 OOO ucoHa OO1OO ON1OO auoO ..O
OH HO OH.\OH. HO O.O OO.N O0.0 OOO to: 81.; OT: and ....O
OH OHH OH.\OH. 0.0 0.0 OO.N OO.O OOO aeoHa OO-O» ON1OO dado H\O
OH OO ON.\ON. N.OH O.NH O0.0 Oo.O OOO oaoHa OO1OH ON1OO acne O\O
OH OH OH.\OH. N.O N.O OO.H OO.O OOO aaoHa OO1OO ON1OO OO.O H\O
OH OOH ON.\ON. N.OH O.NH NO.O OO.O OOO acoHa OO1OO ON1Os OO.O H\O
ON OO HO.\OO. O.ON 0.0N O0.0H O0.0H OOO oHsom OO1OO ON1OO asap O\O
OH OO ON.\OH. O.OH N.HH OO.O OO.N OOO oHeom OO1OH NN1OO arse OHO
OH NNH HO.\HO. O.O O.O OO.O NO.HH OOO oHeom OO1OO HN1OO acne H\O
OH OHH OH.\OH. 0.0 O.O NO.O OO.HN OOO uHeoO OO1OO ON1OO dado H\O
OH OO OH.\OH. 0.0 O.O NO.N OO.NH OOO oOeoO OO1OO OH1OO dado ONHH
OH HO NH.\HH. 0.0 O.» OO.O OO.ON OOO oHcoO OO1Oe OH1OH asno ONHO
OH OO OO.\OO. O.ON O.ON OO.O O0.0 OOO aHsoO OO1OH ~H1Oe aaeo ONHO
OH OOH OO.\OO. O.N N.N O0.0 NH.O OOO oHeom OO1OO OH1Os acaO ONHH
OH ONH OO.HOO. O.N H.N OO.O OO.O OOO cHeom OO1OO OH1Oe Omar ON\H
OH OO NH.\HH. 0.0 0.0 OH.N O0.0 OOO oHaoO OO1OO OH1OO mass OO.O
OH OO OH.\OH. 0.0 O.O ON.O OO.O OOO OHsom «O1Oe OH1OO Once ON\>
OH OO ON.\ON. 0.0H N.OH OO.O N0.0 OOO oHsom OO1OO NH1OO Oreo .OH;
ON OHH OO.\OO. O.N O.N O0.0 O0.0 OOO oHcom OO1O» HH1OO auao O.\O
OH OO OH.\NH. 0.0 O.O NO.N ON.OH OOO oHeom OO1OO OH1OO auaO OHOO
OH OO OO.\OO. O.O 0.0 OO.H OO.O OOO cHeom OO1OO OO1O~ Once HH\1
ON NO OH.\OH. H.O 0.0 OO.O OO.OH OOO oHaoO OO1Oe OO1OO dado Oer
OH HOH HO.\OO. O.N O.N OO.O OO.O OOO oHacO NO1Oe OO1OO auao HH\O
OH OO ON.\NN. 0.0H O.OH OO.O OO.O OOO oHaom HO1O» OO1OO aaao OH»
NH OOH OO.\OO. 0.0 O.O O0.0 O0.0 OOO oHeom HO1OO OO1OO once H\H
NH NO OH.\OH. O.O 0.0 OO.N OO.O OOO oHcom HO1OO OO1OO Oreo OHHO
O OO ~H.\OH. H.HH O.OH NO.O NO.O OOO oHcoO HO1OO OO1OO OO.O ONHO
HH Om ON.\ON. N.OH 0.0H OO.O O0.0 OOO oHsom HO1OO NO1OO dado HN\O
HH OO NN.\NN. O.OH O.OH OO.O OH.O OOO OHOoO HO1OO HO1OO aaao ON\O
.0. easy go H1. Hex-o. H.\nov Hun. Hess. Hnl. caps .on .on .oHooaO coco
eugenics-nu OHus< oouau»< schemata neon OooaO cocoa cocoa-HO soHaeasO saunas Hausa naH. noose
~33 00:05 unease 09.35 339 03.35 mocha H38.
.OOOH .sHopnouoz souuaHvsa on» eH oosaHHa-oooc axons» H¢=OH>HOcH «o aoHoaHauuoo o>HanuHasuaa .HH oHona

'71

 

NH OO ON.:N. O.OH O.OH OO.O OO.H NOO to: HN1OO 81.: 88 32
NH OO ON.\ON. H.OH O.OH OO.H OO.O OOO aeoHO ON1OO OO1OO OO.O HH\OH
HH OO ON.\OH. N.OH 0.0 OO.H OO.O OOO OooHO OH1OH OO1OO Once OHOOH
HH OHH O0.000. O.N O.N O0.0 O0.0 OOO uHsou OH1OO OO1OO NN.O OHOOH
HH NO OO.OOO. O.N O.N ON.O O0.0 ONO OuoHO OH1OO OO1OO OO.O O\OH
HH ON OO.OOO. O.ON O.ON O0.0 OO.O OOO acoHO OH1OO OO1OO auuu O\OH
HH OO NH.OOH. 0.0 N.O OO.N NO.O OOO oHeom OH1OO OO1OO OO.O OOOH
OH OO OH.\OO. H.O O.O OO.H OO.O OOO oHnoO OH1OO OO1OO OO.O OHOH
HH ONH ON.\ON. 0.0H 0.0H ON.N O0.0 ONO OeoHO OH1OO NO1OO OO.O OHOH
HH OOH NH.\NH. O.O 0.0 HO.H OO.O OOO aHeoO OH1OO HO1OO OO.O OHOH
NH On O0.000. O.OH O.OH OO.H OO.H OOO anHO OH1OO OO-OO Once N\OH
OH On NO.\NO. O.H 0.0 NN.O OO.O OOO aeoHO OH1OO OO1OO OO.O NHHO
OH OOH OH.\OH. O.O 0.0 OO.N OO.O OHO aaoHO NH1OO OO1OO Once NHxO
OH. OO OO.OOO. O.N N.N O0.0 O0.0 OOO anHO OO1OO OO1OO OO.O NHOO
OH OO NO.\HO. O.H O.O OO.O OO.N OOO acoHO OO1OO OO1OO OO.O OHHO
OH OO OO.OOO. O.OH H.OH OO.N OO.O OOO oeoHO OH1OO OO1OO OO.O OHHO
OH ON HO.\OO. 0.0N O.HN NO.O OO.O OOO acoHO OH1OO OO1OO OO.O OHOO
OH OO O0.000. H.O 0.0 HO.H OO.O OOO OuoHO OH1OO OO1OO Once OOO
OH OO HH.\HH. 0.0 0.0 OO.H O0.0 OOO 0.0HO OHaOO NO1OO undo O\O
OH ONH O0.000. O.HN 0.0H O0.0 O0.0 OOO aeoHO OH1OO HO1OO Once O\O
OH OO OH.\OH. 0.0 O.OH OO.O OO.O OHO anHO NH1OO OO1OO OO.O OHO
OH ON O0.000. H.O H.O OO.H O0.0 OOO anHO HH1OO OO1OO OO.O OOO
OH OO O0.000. N.O N.O HO.H OO.O OOO anHO OH1OO OO1OO OO.O OOO
OH OO O0.000. 0.0 0.0 OO.H O0.0 OOO anHO OO1OO OO1OO Once O\O
OH OO O0.000. 0.0 0.0 OO.H OO.O OOO O.OHO OO1OO O0.00 asea OHO
OH OO OO.OOO. 0.0 N.O OO.H OO.O OOO aooHO HH1OO OO1OO Once OOO
OH OO O0.000. O.ON 0.0H O0.0 OO.O OOO O-oHO OH1OO OO1OO OO.O O\O
OH OHH NH.\NH. O.O O.O OO.H OO.O OOO OnoHO OO1OO OO1OO OO.O OOO
OH OOH OO.OOO. O.N O.H O0.0 O0.0 OOO O.OHO OO1OO NO1OO OO.O OHO
OH ON OH.\OH. O.HH O.HH O0.0 OO.O OOO anHO OO1OO HO1OO anus OOO
O I. I Io . .
..,.O._O1....._..................:... a... On... t... ...H..... ...1... a... a... a... 2.... .2
.3 .3: 09.33 4:08... 000.35 33b. .0955 much. Hench.

 

“6.083 m Gun-B

72

.co>uoopo no: wannabes nous: nu nOOONoA on«» new and noise oHoca naon Lou H1O nnpmaoH Odom
.vovsHonH one) uc0l0>ol 20H: avoHuon oaHu anon» cho

n
a

 

 

NH OO OH.\OH. O.HH O.HH HH.O O0.0 OOO oeoHO ON1OO OO1OO aHuO O\HH
NH HO OH.\OO. 0.0 H.O NO.H O0.0 OOO anHO ON1OO OO1OO NN.O O\HH
NH OO NN.ONN. 0.0H O.OH O0.0 O0.0 NOO acoHN ON1OO OO1OO OO.O O\HH
NH OO ON.\ON. N.OH 0.0H OO.O O0.0 OOO uaoHO ON1OO OO1OO dado O\HH
NH OOH OH.\HH. 0.0 0.0 OO.H OO.O OOO o-oHO ON1OO OO1OO OO.O O\HH
OH OO OH.\OH. O.HH O.HH OO.H OO.O NOO OnoHN ON1OO OO1OO OO.O O\HH
OH On O.N.th. 0.0H 0.0H O0.0 O0.0 ONO 38m ON1OO OOJO Pic 1:
NH NO ON.\ON. 0.0H 0.0H OO.O N0.0 OOO anHO ON1OO OO1OO nude H\HH
NH ON O0.000. 0.0 H.O O0.0 O0.0 OOO anHO ON1OO NO1OO O0.0 H\HH
NH OO HN.\HN. 0.0H 0.0H OO.O O0.0 ONO oHeom ON1OO HO1OO O0.0 H\HH
OH OH NN.ONN. H.OH H.OH O0.0 O0.0 NOO pnoHO ON1OO OO1OO dado ON\OH
OH OHH O0.000. O.O 0.0 HH.H O0.0 ONO O.OH» ON1OO OO1OO O0.0 ONHOH
OH OO ON.\ON. O.OH 0.0H OH.O O0.0 NOO aaoHO ON1OO OO1OO OO.O ON\OH
OH HO OH.OOH. 0.0 0.0 OO.H O0.0 ONO anHO ON1OO OO1OO Once ON\OH
OH ON O0.0HO. 0.0N O.ON O0.0 O0.0 OOO anHO ON1OO OO1OO OO.O ON\OH
NH NH nN.\nN. 0.0H O.OH HH.O OH.O OHO oHeom NN1OO OO1OO OO.O OH\OH
NH on OH.\OH. 0.0H 0.0 OO.N O0.0 OOO OnoHO ON1OO OO1OO dado OH\OH
NH O. OH.\OH. 0.0H N.HH ON.O O0.0 OOO anHO OH1OO OO1OO nude OH\OH
NH on ON.\ON. 0.0H 0.0H OO.N O0.0 OOO anHO OH1OO NO1OO O0.0 OHHOH
NH NoH O0.000. H.O 0.0 O0.0 O0.0 OOO 0.0HO OH1OO HO1OO Open HHHOH
O . .
can“? egg-«.85 :3..qu- 88 Hooch“. HH.H-Hum. oar-urn LHOLHOB an? Hon-NOON a“. sonata .3 Sam 33
Noon: ouch-pa «sauna coca-pa House uuuauu< House Heaps

 

Ao.acoav m oHnda

73

 

9. O.H OH.\..H. N.OH H.O NH.H 8... OOO :82 8.2. O0.0N Ed: .3:
”ON OO NH.\OH. 0.0N N.ON HO.N OO.H NOO :H:oO HH.mN NN.ON nun: .nNNN
OH NH NH.\NH. O.N 0.0 O2N OO.O on“ .53.. O.H.ON ON.2. 9.8 NNz.
OH NO OH .\NH . H . N 0.0 HN.N OO.O NOm 3:8 HH.mN NN.ON 9:: NNHN
OH HOH HN.\NN. H.OH N.HH OO.H NH.N OOO «no: NH.2. ON.2. 93: HN\N
OH NNH 8.3:. H.O H.O HH.H «N.N NOO 3:8 HHAN 3.2. 8.8 HNNN
OH «OH ON.\ON. N.OH H.OH OO.N OO.O OOO 3:8 OO.2. ..N.2. EB: :22.
OH HO 3.32 O.OH O.Hm .3.NN oo...N OHO 3:8 :72. 8.2. 2.2:. :22: NHNN
.H OO $.32 0.0H O.NH No... OO.O OOO :8: OO.ON NN.ON 9:: OHNN
..H OO No.\Oo. N... O.H HH.H OO.O O.O 3:8 3.2. N.ON H3»... 56:: OHNN
NH NO NN.RN. N.OH N...H N...H NO.N OOO $3... 8.2. 8.2. 9:6 OHNN
NH 3 OH.\ON. O.OH ...HH ONH OO.H OOO 3:8 OO.ON 2.2. 93: HHNN
NH NO HN.\HN. ...NH ...NH NH.m 8... OOO .33.. OO.ON O72. :56 HHNN
NH 8 HQ}? O.OH 0.0H ON... 8... OOO 3:8 OO.ON 2.2. so HH\N
NH Om OH.\OH. O.N 0.0 OO.H NO.O 8a 2.3.. SAN OH.ON 9.8 2N
NH On HH.3H. O.O O.O H2N N0.0 OOO 3:8 O0.0N 3.2. 9.8 2N
OH N.. 2.31 O.N H.O NN.H 8.0 OOO :82 No.2. ..H..mN 93: N2.
OH R 3.3:. 2m 2m 8.: O0.0 OHN 3:8 O0.0N 2.2. 93: NNN
OH 8 HH.\OH. O.N O.N HH.H OO.NH OOO :82 8.2. 2.2. B8 N2.
OH NO OH.\OH. O.HH N.OH NN.O 8...N OOO 3:8 3.2. 2.2. Pa: NNN
OH H.. “N.N..N. O.NH 0.0H OO... O0.0 OHN :32 3.2. 2.2. E8 HNN
OH ON 3.xno. N.N N.H N46 ON.O OOO £32 3.2. 3.2. 5 H2.
OH an OO.NOO. N.ON 0.0N O0.0 N0.0 OR anoHN No.3 8.2. 9.8 OOO
OH HO OO. 20. H . H O .N ON.O mm . N OOO 3:8 3.2. 8.2. 9.8 On \O
OH NO OH.\HH. N.N N.O OO.H O0.0 can .83... O0.0N OO.2. 93: ONNO
OH NOH 8.3:. 0.0 O.H OO.O SO 8.. 3:8 HoAN 3.2 9:8 ONNO
NH 3 3 .EO. O.N N.N O0.0 9.. .O OR 33.. NO.ON ..O.2. PI: ONNO
NH HN OO.NNO. 0.0 O.H O0.0 OO.O 8.. 3:8 HO.nN O0.0N B8 ONNO
OH ONH NO.\NO. 2n 0.0 3.: OO.O :9. 3:8 8.2.. NO.nN Pl: ONNO
OH HN 2.31 O.N H.N $3 8.0 8.. 3:8 H0.0N SAN 93: OHNO
H8 92. O: H.. 3:: 38. .3: .93. .I. 8P .8 .o- .388 38
85885-8. 30:: 8:85 :2»!!— 88 :88 :88 8:33: 8395: 533 :83. .3: H95.
:38 .0985 38:9 3:85 Hole «3...»: €29 H38

 

68H {Soto-oz 8933 23 cu con-«H5800. .1095 3:213 ho Sufi—tog gaudvavg

.4 0H3

7h

 

ON O. OO.NOO. H.HO O.ON O0.0 OO.O OHO oaoHN NN.ON O0.0N »:::a ::::O NN\O
ON OOH OO.NHO. O.H 0.0 ON.O OH.O OOO OH:OO ON.ON O0.0N O:O:O ::::O NN\O
OH 0m m.H\~.H n.05 H.OO «0.0H «0.0 can cacao «N.ON an.mp 0:059 arena Omxo
OH Om NO.NON. H.OO 0.0. OH.N O0.0 OOO OH:OO ON-ON N0.0N ansa :aouO .ONNO
ON HO N.HNH.H H.ON O.NN O0.0N O0.0 OOO OHOOO .N.ON O0.0N uao:e ::::O OH\O
HN OO OO.\HO. O.NO O.OO OO.HH O0.0 OOO :H::O HN.ON O0.0N O:::N ::::O OH\O
ON OOH ON.\MN. 0.0H 0.0H OO.N N0.0 OOO OooHu mN.ON O0.0N OLOO HH\O
ON On HO.\HO. O.NN O.NN NO.O O0.0 OON O.OH: NN.ON O0.0N Ops: HH\O
OH OOH NH.\OO. ...N 0.0 OO.H 8.N OOO 0.0H... ON.ON N0.0N PS: O.NO
OH O. ON.\OO. O.NH O.NH HH.N O0.0 OOO OooHN HN.ON H0-0N NN.O NH\O
OH ON OO.NHO. N.O O.N O0.0 OO.N OOO O.OHN ON.ON O0.0N NN.O NH\O
OH OOH OO.NOO. 0.0 0.0 O0.0 O..N OOO :H:OO OH.ON OO-ON :::O HH\O
NH NO NN.\NN. O.HH N.HH O0.0 N0.0 OOO oH:8 OH.ON OO.ON Ouau N\O
HN HO ON.\NN. O.NH O.NH OO.N NO.O OOO :H:OO OH.ON NH.ON Ouao O.ONO
NN O: OO.NOO. 0.0N O.HN NH.O NO.H OOO OHOOO OH-ON O0.0N NN.O .O\O
ON ON NN.\HN. O.NH 0.0H H..O O0.0 OOO OooHN OH.ON OO.ON Ouco .\O
mm «an mo.\wo. o.” O.H am.o OO.N onw canon 0H.mh O0.0N “Lao a\c
ON ON ON.\nN. H.NH N.OH OH.O OO.N OOO O.OH» NH.ON O0.0N OO.O H\O
ON NHH OH.\HH. 0.0 N.O OO.H O0.0 ONO :H:oO OH.ON N..ON nun: HHO
ON OO OO.\HO. N.O N.N OO.O O0.0 ONO :H::O OH.ON H0.0N Nuno HO\N
ON O OO.NOO. O.HN O.HN N0.0 NO.H ONO :H::O OH.ON OO.ON O::O o.Om\N
NN an OO.NNO. N... O... O0.0 On... ONO 3:8 OH.ON O0.0N 9:5 .OONN
HN ON OO.NOH. 0.0N 0.0N OO.N NO.N OOO anHN OH.ON O0-0N Ounu ON\N
HN ON ON.\NH. N.NH 0.0H OH.H OO.O ONO :H:om OH.ON N0-0N Ouao ON\N
HN NO OH.\OH. O.NH O.NH OO.H OO.O O.O anHN OH.ON OO.ON nun: O.ON\N
HN an HO.\ON. 0.0H O.NH OO.H O0.0 ONO OHOOO OH.ON O0.0N Ouao O.ONxN
HN NOH Hm.\aN. N.OH H.HH OO.N OO.O OOO uaoHN OH.ON OO.ON O::u .ONHN
HN OHH OO.NON. N.NH N.NH OO.H OO.O ONO OHOOO OH.ON O0.0N Nuao .ONxN
ON On NO.N.O. 0.00 H.OO O0.0 O0.0 OOO OnoHN OH.ON NO.ON Ouau .NNN
ON OHH HO.\OO. O.N O.N Na.O N0.0 NOO :H::O HH.ON H0.0N :::O ON\N
E I . .
taunwwale ogafiwnui :33“... 88 Won—OOO Home“. oofiwa LN“? :HOOIfiOJ OO.HH—.9 On“... How”... .388 3::
Nov.) onduo>< duo-NB ouduo>< mucus ounuu>< mocha HOOOB

 

.:.~:cOv a OHQON

'(5

in...» noun-6

£095 03.35”
Onto-no a: ace-26- :36. OH 33.8.— !» .8» on. 1003 30:: noon .8» O-O Baud: Odom”
6.3533 0.8) it! 50: 30:!— 30 :23 OH.H-o“

 

 

HH OOH ON.\OH. 0.0H 0.0 OO.H O0.0 OOO 3:8 O0.0N ON.ON 39:. :38 OO\OH
HH NN OO.\OO. O.NO O.OO HO.OH OO.O OOO 3:8 OO.ON ON.ON 89:. En ONBH
HH NO OO<OO. N.OO N.NO NN.OH 8.N OOO 3:8 O0.0N NN.ON 829 838 NNBH
HH NO O.H\N.H O.ON H.ON NO.OH 8.N OOO 3:8 O0.0N ON.ON 26¢ 8:3. HN\OH
HH OO N.HNH.H N.ON O.OO OO.O OO.N 8O 38 O0.0N ON.ON 26.6 8:3. 83H
OH NO OO.O? O.N O.H OH.O ON.N 8O 38 O0.0N ON.ON «8:9 838 OH\OH
OH OO OH.\NH. O.OH O.O 8.: SO OOO 3:8 OO..ON ON.ON :25. 8:3. OHBH
OH ON OO.NOO. O.O O.O ON.H OO.N OOO 3:8 OO.ON NN.ON 929 8:8 22
OH ON O.HNO.H O.HO O.ON NO.O OO.O ONO 3:8 O0.0N HN.ON «.68. :8 ONOH
OH ON O.HNO.H 0.00 0.0N NN.ON OO.N ONO 38o O0.0N ON.ON 2.9:. .598 NxOH
OH ON OO.\ON. O.ON 0.0H OO.O OO.N OOO 3:8 NO.ON O0.0N «:89 8:3. ONR
OH OOH NO.NS. O.O H.O OO.O 8.O OOO 3:8 OO.ON OO.ON .529 8.8- ONE
OH NO OO.NON. 0.00 0.00 OO.HH OO.N OHO 3:8 O0.0N 5.2. g 28.! OHR
OH OO OO.NOO. O.HO 0.00 OO.OH OO.O OHO 3:8 OO.ON OO.ON $89 :8: OHNO
OH NO HO.\ON. H.OO O.OO NO.OH OO.O OHO 3:8 OO.ON OO.ON :33 :88 NHR
OH NO HN.\NH. O.NH N.OH OO.N OO.N OHO 38 8.2. OO.ON :33. En..- OHNO
OH OOH 8.2.9 0.0 O.O ON.H OO.N OOO 3:8 ON.ON OO.ON B8 OR
ON OO OO.NOO. 0.00 0.00 OH.O NN.O OOO 0.0H.— NNnON N0.0N 39F 2.98 ON\O
ON NO ON.ON. O.NO 0.00 O0.0 O0.0 OHO O83 ON.ON 3.2. «8b. 85.! SB
.8 .58 No O.NH. O.N-3 .l. .92: .I. 8D. .8 .8 .388 38
2:3 3.! .985 alarm-m. 8 O88 3.8 8:33: 832.8 £33 8.6 O.HN :88
:3... 8:85 $88 828»: O83 832:. O83. H38.

 

2.88. O 3.3.

76
Movement Patterns

Movements of the 65 fish tracked in the Ludington reservoir (159
tracks) were distinctly patterned. Each track was assigned to one of
three general categories depending on the orientation of locomotory
activity (Tables 5 and 6). Additionally, specific behavioral patterns
such as milling, passage behind and in front of the upper intake
structure, and movement across open-water were noted for each track.

Both carp and trout tended to restrict their movements to areas
adjacent to the reservoir wall (Figures 1-h). Seventy-five percent
of the carp tracks and 92% of the trout tracks exhibited this pattern.
Differences between float and sonic-equipped individuals, both within
and between years, were not significant (:1:2 analyses, P > 0.05) for
the distribution of behavioral patterns. Similarly, the frequency of
on-the-wall tracks was not significantly different between carp and
trout ($2 = 2.78, P > 0.05).

The movements of individuals tracked during the same day appeared
to be independent as evidenced by activity in different sectors of
the reservoir (Fig. 2A). Fish traversed the reservoir in both clock—
wise and counterclockwise directions (Track 70, Fig. 2B; Track 55,
Fig. hA) and showed little consistency in directionality of movement
from day to day. All parts of the reservoir were frequented although
many tracks were localized in one reservoir section or another.
Individuals tracked on consecutive days were often found in different
areas (Fig. 1A) indicating that site-preference behavior was minimal.
Diel and operational-mode differences in movement pattern were not

apparent (as determined from track maps), although statistical

77

 

 

5. Qualitative description of aoveaent patterns or
fish tracked in the Ludington Reservoir. 1974.
7::SR Descriptionl 3::fk Description3
1 ORV-3 wall 35 ONUOSB wall
2 O.H-88,3 walls 36 ONH-SE wall
3 Oil-88,3 walls 37 OlI-I,NV walls
4 DIV-8,83 walls 38 ONH-U wall; BI.
5 OHH-SB wall 39 ONH-8,8! walls
5 OHH-ll,l,lV walls 40 ONH-IB,B,SB walls
7b OPE-8U Diddle 41 Oli-l,l8,3 walls
ab wasn-sw walls 42 ouv-ss wall
Central aiddle. ‘3 Oil-33 wall
9 °"'3 "11 44 onw-v,ww walls; BI.
10 OHH-3,3V walls ‘5 ONH-8,8U '811.
11 OHH-8H wall; ailling. 46 ORV-I wall: lillinz.
12 ONH-88,8,8U walls 47 OHH-SB wall
:3 °':’33 "11 48 oxv-s wall
12b ::H-::n:::: middle; ‘9 O'V-SB '.11
milling. 50 OHH-N.HH walls
16b CPU-Central middle; 5' 03"“ "11
ailling. 52 ONH-88 wall
17 ONH-all around; BI. 53 ONH-B wall
18 0NV-H,8U.S,SB walls 54 lIXBD-U wall:
19b nIan-v wall: 3 'idd1°'
Central middle. 55 ONH—U,8V,S,SB,B walls
20 HIXBD-B wall; 56 ORV-3 wall
Central aiddle. 57 ONH-SB wall
21b ONH-V wall: ailling. 58 HIXBD-B wall; U-B;
22 CPU-Central aiddle. Central aiddle.
23 0P!-S,N a Central 59 CPU-parallel to I well
:i:::;;.3’1r‘1 60 onw-ln wall
24 ONH-8,88 walls 6‘ 03"" "11
25 ONH-88,3 walls 62 MIXED-E wall; N middle.
26 0NU-SU,8 '3113‘ 63 MIXED-SB wall; S middle.
83-33. 64 ONH-SU,S,SE walls
27 ONH-SI wall 65 0NV-SU,S,SE walls
28 ONH-SV,U walls 66 ONH-H,SH,S,SE,B,NE walls
29 ONH-S wall 67 ONH-S wall
30 ONH-SB wall 68 ONH-SU,S,SE walls
31 ONH—SE,E,NE walls 69 ONH-E wall
32 ONH-SB wall 70 ONH-33,3,SH walls
33 ONH-B wall 71 ONH-3.83.3,IB walls
34 lIXID—NU wall: BI; 72 ONH-SE wall

N middle; SE-NU.

78

 

 

 

Table 5. (continued)
Track Descriptiona Track Descri tion.
No. No. p
73 MIXED-SH wall; SI-SB: 78 MIXED-8,3! wall;
8 middle. 8 middle.
74 DIV-8H wall; B-V. 79 MIXBD-B wall;
75 CPU-Central middle 3 '1ddl"
76 ONH-SB wall 80 OPU— 8 middle
77 orw-s middle
3OH! = Movements were confined to shoreline contours (on the wall) for
greater than 80% of the track duration.
OPV = Movements were confined to open-water locations (of! the wall)
for greater than 80% of the track duration.
MIXED = Movements were distributed in both along-shore and open-water
locations.
BI 2 Movement behind upper intake structure occurred at least once

during track duration.

l-B,SE-8I,NB-SE, etc a Movement from one shore location, across open
water, to another shore location occurred at
least once during track duration.

Mill

b

ing = Localized sig-sag movements.

Fish was released at buoy in center of reservoir.

79

 

 

Table 6. Qualitative description of moveaent patterns of
fish tracked in the Ludington Reservoir. 1975.
Track Description. Track Description“
lo. lo.
1 DIV-B wall 36 OHH-E wall
2 ORV-B wall 37 OHH-S wall
3 ONH-SB wall 38 DIV-8 middle
4 onv-sw wall 39 oxw-3 'g11
5 OHH-S wall 40 ONH-B wall
6 OHH-8 wall 41 ONH-B wall
7 DIV-SB wall 42 ONH-S wall
8 onw—z wall 43 MIXED-SM wall;
9 ouw-s wall 3 '1d410-
_ 44 ONH-SE wall;
10 ONE B/wall milling.
‘1 onu-3_;38;;?nd‘ 45 ONH-33,3 walls
12 MIXED-SB wall; ‘5 0N"§§s§3"11°'
8 middIOe - .
13 ONH-SE I811 47 ONH-3,3 .8118}
88-8.
14 onw-ss 'all 48 nIxnn-s,ss walls:
15 OHH-SB .311 S liddle.
16 MIXED-SB wall: 49 onu_3 .311,
33 middle. .1111n8,
17 OPE—S middle 50 ONH-SE wall
18 MIXED-B wall; 51 OF!— parallel to
SH-SB. w wall
19 ONH-B wall 52 OHH-U wall
20 ONH-SE wall 53 OPE-R middle: V-N;
21 ONH-SB wall Spiral pattern.
54 ORV-V wall by intake;
22 MIXED-E wall;
B-H; 8 middle. BI; entered intake.
23 ONH-all around; BI. 55 on"g{:.‘r°“nd‘ 31’
24 ONH-E wall 56 ONH-all around; BI;
25 ONH-E wall H-3.
26 ONH-NE wall 57 ONH-& around; BI:
27 ONH-B wall ';'3 "11'
28 MIXED-N wall; 58 ONH-3 4 around; BI, IFI.
N middle. 59 ONH-SB wall;
29 onw-sa,s, sw walls ';1;1::'N 1
N -S
M - '
31 oxw-ss wall; 1 s middlzf 8.
milling. 6 ONH 3 ss 3 NE 11
2 - wa
32 OHH-3/4 around; Sé-SG.’ a;
SU-NB id 1
6 OFV-S d
33 nxxan-su wall; 3 " °
SH middle. 64 ONH-53,8 walls
34 onw-w wall 65 ONH-all around; BI.
35 ONH-8V wall; 66 onw-311 around: BI;

SB-SU.

NE-NU.

80

Table 6. (continued)

 

 

Track a Track a
No. Description No. Description
67 ouv-3/4 around 74 onw—s wall
68 ONH-SI wall 75 0N!-S,SH,NE,N walls;
69 ouv-ns,u,lw walls; 3""3'
BI. 76 ONH-all around; BI;
70 ONH-all around; BI: SE'SH‘
s—sv; ns-ss. 77 ouv-3/4 around; BI;
, SE-SU.
71 ouv-3/4 around. 31;
SU-NB. 78 ONH—H,SU,S,SE walls;
72 onw-sv wall 31°
73 ONH-N wall 79 MIXED-SE wall; 8 middle.
a OMH = Movements were confined to shoreline contours (on the wall) for

greater than 80% of the track duration.

0?! = Movements were confined to open-water locations (off the wall)
for greater than 80$ of the track duration.

MIXED z Movements were distributed in both along-shore and open-water
locations.

BI = Movement behind upper intake structure occurred at least once
during track duration.

IPI 2 Movement in front of upper intake structure occurred at least
once during track duration.

U-B, SB-SU, NE-SE, etc 2 Movement from one shore location, across open
water, to another shore location occurred at
least once during track duration.

Milling = Localized zig-zag movements.

Figure

1.

(A)

(B)

81

Behavioral atterns of a sonic-tagged carp
(Fish 75-013 on three different dates in
June 1975.

Behavioral patterns of a sonic-tagged carp
(Track 75-15) and a float-tagged carp
(Track 75-30) on two separate dates in
1975.

Solid circle (0) represents the start of
a track. Solid rectangle (I) represents
the end of a track.

82

 

 

 

PROJECT

IRRCK8I76-OS

78—03
75-01

 

 

 

 

PROJECT

YRRCKSI7S-3O

75-15

In

15

 

0 L

30

 

 

 

.-—-. ___-. k.._ . ._ n.. ---_J

 

Figure

2.

(A)

(B)

83

Behavioral patterns of four float-tagged
carp tracked simultaneously on September

9. 1974.

Behavioral pattern of a sonic-tagged brown
trout (Fish 75-33) tracked on October 2,
1975.

Symbols as in Figure 1.

8h

 

 

 

PROJECT

TRRCK5174-39

74-38
74-37
74-36

 

39

37

 

 

 

PROJECT

TRRCK3I76-70

 

 

0L

 

 

 

~——.——_—-— ._..7..

 

 

Figure

3.

(B)

85

Behavioral patterns of three float-tagged
carp tracked simultaneously on November
7, 1974.

Behavioral patterns of a sonic-tagged carp
(Track 75-25; and a float-tagged carp
(Track 75-26 simultaneously tracked on
July 21, 1975.

Symbols as in Figure 1.

86

 

 

 

PROJECT

TRRCM8I74-80

74-79
74-78

 

 

1 an

 

 

 

PROJECT

TRRCKSI76-28

76-25

 

 

 

26

 

 

 

 

 

Figure

4.

(A)

(B)

87

Behavioral patterns of two sonic-tagged
brown trout tracked on two separate
dates in August 1975.

Behavioral patterns of three float-tagged
carp tracked on two separate dates

(Track 75-45, August 4, 1975; Tracks 75-53
and 75-54, August 14, 1975).

Symbols as in Figure 1.

88

 

 

 

PROJECT

1m: I 78-68

 

 

 

 

 

 

 

PROJECT

78-83
78-46

momma-so

 

 

 

 

 

 

 

 

89

comparisons could not be made due to the small number of observations
taken at night and during pumping operations.

During late fall, carp tended to move in the open-water areas
of the reservoir (Fig. 3A). Gill net collections accomplished during
this time supported these data as catches were higher in the bottom
gill nets (set off-shore) than in the surface gear set near the
reservoir embankment (Gulvas, 1976).

On several occasions, interactions between monitored fish occurred
(Fig. 3B). This behavior was seen when tracked fish were part of the
same school (carp) or frequented the same reservoir locality. In this
latter instance, the Juxtaposition of individuals most often occurred
at suspected feeding sites (rockprubble area adjacent to the reservoir
ramp - see Track 30, Fig. 1A; and ground-water pump outfall area on
the mid-east bank of the reservoir - see middle of Track hS, Fig. hB).

Several other types of movement activity were Observed apart
from the general on-the—wall, openawater, and patrolling patterns
(Tracks 1 and 3, Fig. 1A; Track 15, Fig. 1B). The most common of
these was movement behind the upper intake structure (Track 70, Fig. 2B;
Trace 55, Fig. hA). This occurred in five carp tracks and in 13 of
the 26 trout tracks (Tables 5 and 6). Fish approached the intake
structure from both northerly and southerly directions and normally
remained quite close to the reservoir wall. Individuals seemed little
affected by water currents in the intake vicinity as this behavior
occurred during both generating and pumping operations.

Milling (localized zig-zag movement) was noted in seven carp
tracks but only once in trout (Track 59, Fig. hA). Apparently, this

behavior indicates stress since it was observed (in 6 of the 8 tracks)

90

on either the first day after a fish was released or on the last day
tracked. Also, two carp accounted for five of the milling tracks.
These observations suggest that milling may result from either
individualized behavior or physiological disturbances caused by
handling and tagging.

A prominent spiral pattern of movement was observed in carp
tracks 7h-23 and 75-53 (Fig. hB). During both efforts, this occurred
in the north-central section of the reservoir during power generation.
Although the stimuli evoking this behavior are unknown, it is possible
that the fish were responding to current gyres in this locale.

Fish 75-23 (Track Sh, Fig. hB) was the only monitored individual
directly observed to leave the reservoir during tracking. Initially,
this fish passed behind the upper intake structure and then proceeded
in front of the intake where it was soon lost from sight. The float

tag was not recovered in either the reservoir or the lake.

Angular Change

Mean track turning angles ranged from 11-1th for carp in 197k
and 8-165o for carp during 1975. Trout monitored in 1975 exhibited
average turning angles from 211-1100 (Tables 3 and 11). Frequency of
turning (using data pooled into hSO intervals) was not significantly
different between carp in 197k and 1975 (x2 = 2.92, P > o.ho). Like-
wise, no difference was detected in the frequency of large turning
angles (> hSO) for carp between years (x2 = 0.3h, P > 0.50). Comparison
of the frequency of angular changes greater than hSO between carp and
trout indicated a significant difference (x2 = 25.93, P < 0.005). Only

19.2% of the trout tracks had large mean turning angles while the

91

corresponding frequency in carp was 73.7%. Since both species tended
to lead along the reservoir wall during most of their movements, these
results indicate that trout were more directional (in terms of straight-

ness) than carp, and seldom.meandered.

Swimming Speed

Calculated average swimming speeds of individual fish varied
from 0.01 - 0.6h L/sec (0.8 - 38.1 cm/sec) for carp and 0.01 - 1.30
L/sec (0.8 - 79.9 cm/sec) for trout (Tables 3 and h). The overall
mean rate of movement was 0.16 L/sec (9.9 cm/sec) for carp tracked
in l97h, 0.19 L/sec (11.3 cm/sec) for 1975 carp, and 0.63 L/sec
(38.6 cm/sec) for trout. No significant differences were detected in
track swim speed between individuals tracked on multiple days for
carp (Table 7) or trout (F = 1.8h, P = .201). Similarly, differences
in swimming rates between brown trout and rainbow trout were not
evident (F = 0.h6, P > 0.75). These anaLyses indicate that, for
generatingemode observations, pooling of individual track swim data
by fish group (carp; trout) is statistically appropriate. The low
number of tracks accomplished during pumping operations precluded their
analysis by parametric procedures.

Comparison of swimming speed between sonic and float-tagged carp
(Table 8) showed no difference in l97h but a highly significant
difference in 1975 (P = 0.008). Surprisingly, in both years, the mean
speed for float-tracked organisms was greater than that of sonic-
equipped fish (197h - 0.18 vs 0.17 L/sec; 1975 - 0.22 vs 0.10 L/sec).
Results of a two-way analysis of variance of carp mean track speed

(Table 8) revealed that, overall, neither tag type nor year differences

92

Table 7 . Results of analyses of variance of specific
swimming speed for fish tracked on multiple
occasions in the Ludin ton Reservoir (only
generating-mode tracks?

 

 

Source df MS F P
1974 Sonic Fish, Carp
Between fish 4 .0159 1.050 0.408
Within fish 19 .0151
1974 Float Fish, Carp
Between fish 16 .0152 1.394 0.212
Within fish 29 .0109
1975 Sonic Fish, Carp
Between fish 5 .0089 1.643 0.229
Within fish 11 .0054
1975 Float Fish, Carp
Between fish 4 .0313 1.628 0.269
Within fish 7 .0193
1974 Sonic & Float Fish, Carp
Between fish 21 .0148 1.179 0.311
Within fish 48 .0126
1975 Sonic & Float Fish, Carp
Between fish 10 .0189 1.756 0.143
Within fish 18 .0108
1974 & 1975 Sonic & Float Fish, Carp
Between fish 32 .0166 1.373 0.138

Within fish 66 .0121

 

93

 

 

Table 8 . Results of anal sea of variance of tag type
(sonic or float on specific swimming speed
of fish tracked in the Ludin ton Reservoir
(only generating-mode tracks
Source df MS F P
1974 Carp
Between tag types 1 .0023 0.153 0.693
Within tag types 75 .0149
1975 Carp
Between tag types 1 .1317 7.920 0.008
Within tag types 38 .0166
1974 & 1975 Carp
Between tag types 1 .0694 4.379 0.039
Within tag types 115 .0158
2-Way ANOVA, 1974 & 1975 Carp, Sonic & Float Tags
Main Effects 2 .0430 2.773 0.065
Year .0500 3.199 0.073
Tag Type 1 .0020 0.148 0.999
Interactions 1 .0680 4.389 0.036
Year X Tag Type 1 .0680 4.389 0.036
Residual 113 .0160

 

9b

were significant (P > 0.05). However, a significant interaction
between year and tag type was apparent. This was effected by the
relatively low rate of movement of 1975 sonic carp compared to the
197M sonic fish (41% slower) and the 1975 float-equipped carp (55%
slower). Reasons for this low rate of swimming speed are not known.
For trout, swimming speed was similar between float and sonic-tagged
individuals (F = 0.008, P = 0.93).

No consistent relation was found between swimming speed of carp
and operational mode. In l97h, 8 of 9 fish tracked through pumping
and generating periods showed a higher movement rate during generation
than during pumping activity. Generating speeds were significantly
higher than pumping speeds (Wilcoxon Signed Ranks Test (Conover, 1973),
P < 0.05). In contrast, 10 of 12 1975 multi-mode tracked fish exhibited
higher swimming speeds during pumping than generating. Again, the
difference was significant (Wilcoxon Signed Ranks Test, P < 0.05).
Interpretation of the discrepancy between these data is hampered by
the absence of current measurements, and the confounding of pumping
and generating activities with light levels (i.e., pumping occurring
at night, generation during the day).

A distinct diel cycle of swimming speed was demonstrated in both
carp and trout (Fig. 5 and 6). In the two species, movement occurred
regularly during both daylight and nighttime periods but speeds
were highest during dawn and dusk. Crepuscular activity patterns
were similar for float and sonic-tagged individuals and was evident
in both years. Accordingly, swim speed data were pooled from all
individuals of each group (Table 9). Maximum mean carp speeds

occurred between 0530 - 0730 hr and, to a lesser degree, at 2100 hr.

95

Figure 5. Diel pattern of specific swimming speed for carp
carrying ultrasonic transmitters and float tags
in 1974 and 1975. Hourly means from 0000 to
0600 hr are used again for the 2400- to 3000-hr
period to show more clearly the trend at 2400.

 

 

(935/1) 033dS SNIWWIMS

1400 1800 ”0200022002400020004000800

TIME OF DAY

020004000600“ DOC 1200

Figure

6.

97

Diel pattern of specific swimming speed for trout
carrying ultrasonic transmitters and float tags
in 1975. Hourly means from 0000 to 0600 hr are
used again for the 2400- to 3000-hr period to
show more clearly the trend at 2400.

 

 

 

 

 

 

 

OOOOOEOOMOOWOCDIWO ”MOO” sooaooouoozeooocooosoo-

 

('3”/'1) 033d5 ONIWWIMS

TIME OF DAY

99

Table 9 . Frequency distribution of hourly specific swimming
speed estimates used in diel periodicity analysis.

 

 

Hour of No. of Observations No. of Observations
Day (Carp) (Trout)
0100 4 1
0200 4 1
0300 3 1
0400 3 1
0500 3 2
0600 9 2
0700 8 3
0800 13 5
0900 57 15
1000 94 23
1100 108 23
1200 113 23
1300 111 22
1400 107 20
1500 104 20
1600 98 16
1700 79 5
1800 48 1
1900 10 1
2000 12 1
2100 9 1
2200 9 1
2300 5 1
2400 4 1

Total 1015 190

 

100

Minimum.values (except for three observations at 0500) occurred about
mid-day. For trout, maximum.mean speed values occurred from 1830 -
2130 hr, with a secondary peak of activity at 0830 hr. Rate of
movement was generally higher during the night than in the day for
carp but the reverse of this pattern was observed in trout. Sun
visibility did not appear to have any effect on the daytime swimming
speeds of either group of fish (Carp: F = 0.939, P = 0.39h; Trout:

F = 0.317, P = 0.732).

Multiple regression analyses of mean track swimming speed (gener-
ating-mode tracks only) for carp and trout were accomplished using
12 independent variables (Table 10). Track-level analyses were used
for analysis because they encompass "lag responses" to external
stimuli and meet the statistical assumption of independence of
observations.

Seven regression analyses were performed on the carp swim speed
data to account for possible differences in the importance of variables
between study years and among sonic and float-tagged fish (Table 11).
In most cases, the proportion of variance explained by the significant
variables was small; only in the 1975 sonic track analysis was more
than 36% of the variability in swim speed accounted for (R2 = 0.65).
Of the significant parameters, water-level and current variables
(Resfluct, Fluxrate, and Noturbon) appeared in five of the seven
analyses. The addition of several quadratic variables (interaction
terms) as independent factors did not significantly alter the multiple
correlation coefficients (data not included).

The analysis of trout movement rate resulted in the inclusion

of seven variables into the multiple regression equation and a

101

Table 10. Independent variables used in stepwise linear
multiple regression analyses of swimming speed
of fish in the Ludington Reservoir.

 

 

Variable Variable Name Units
Air Temperature Tempair OC
Water Temperature Tempsurf oC
Atmospheric Pressure Barpress Inches Hg
Wind Direction Winddir Degrees
Wind Velocity Wind1 Knots
Light Penetration Lightpen Meters
Holding Time of Fish

From Capture to Holdtime Hours

Release

Days Fish Was at Liberty

in Reservoir After Daysfree Days

Release

Daysfree2 Days2

(Quadratic Term)
Turbines in Operation Noturbon Units (0-6)

Fluctuation in Reservoir
Water Level from Start Resfluct Feet
to End of Track

Average Rate of Water
Level Fluctuation Fluxrate Meters/Second
During Track

 

102

 

omp. row. who. Fo¢. oww. Fo¢. who.

mxomha oaqom era?

sameness

 

 

woe. emr.l moo. smw.l Nor. Moe. w—m. nwucnflz
NOF. mmm. moo. oom. mes. own. who. mswpvaom
POF. man.l moo. ooh.l omo. oom. moo. nopuspoz
exodus pmoam a canon .wwmp
omo. mm—. FPO. omw. mmo. mmm. mmw. mswwvaom
omo. mpm.l mwo. MNN.I Omo. mmm. mpo. podammom
excess seeds s eflsom .mems s seas
ovom m o monaowmwcmwm m m m monwOHMHQMfim oswz
Henna pudendum p m nowmmonmom mamswm N mamfiwasz mapwwnm> manmwnm>
.nwo>nomom
nopmnflusq on» as coxoonp undo mo moom\mspwnoa aconv paosobos mo comma no
moanmwnmb pneunomoch mo noflmmon on mamwpass Hooded omwzmopm mo mpasmom .FF manna

103

 

 

 

smF. me. Foo. mmF. me. pom. moo. nonHSpoz
emF. eem.u moo. mme.u mus. mme. moo. assesses
exodus ofiqom mwme
mmAm<Hm¢> ez<onszHm oz
excess esoae s eesem .msms
mmw. oom.l Foo. Fo¢.I an. oom. mFF. pedammmm
mmm. mmm. Foo. hmo. nun. mom. who. Neonmhmq
_em. eoo._n eoo. mmo.u «mm. Nem. mes. mosamsse
mm.. mam. moo. mmm. _mm. ose. emo. seepeaom
ms_. mmm.u moo. mae.u ne_. mas. moo. meesssfla
mMomna pecan enm—
mpom «pom eoanfiMHsMfim m m m ooqmowmwsmfim msmz
Henna unmuqmvm scammonmom mamafim N mamfipass mapwwhd> manwflnmb
Aemssepsoov . s_eanse

10h

 

 

 

owN. 0mm.l *nr. ¢m0.I OON. mew. New. wosHmmmm
3N. NE. of. mNm. mos. mNn. of. sopngoz
excuse pmoam mum?
duom spam moswoNMNsmam m m m ooddoemanmfim mamz
Hosea chacqmpm scammonmom manawm N mamfipasz mapwflnm> mapwwnw>
Aeossspsoov .a_ oases

105

multiple R2 = 0.83 (Table 12). Time at liberty in the reservoir
(Daysfree) exhibited the highest correlation with swimming speed

(R = 0.605). Surface water temperature, however, possessed the
largest standardized regression coefficient (Beta = 0.8h6) and hence
can be considered the most relatively important factor influencing
fish speed. A broad range of water temperature-fish speed observa—
tions (ll - 21°C) was incorporated in the regression analysis and
thus the statistical results should be biologically valid. Reservoir
water—level drawdown (Resfluct) was also an important parameter as
indicated by its moderately large standardized coefficient and the
substantial increase in R2 (17%) gained when this variable entered
the regression equation. Although air temperature (Airtemp) had a
relatively high standardized coefficient (Beta = -0.82h), the biological

significance of this variable in affecting swimming rates is obscure.

Retention Time

In 1975, 62 fish comprising four species (32 carp, 25 brown trout,
h rainbow trout, and l lake trout) were observed for their residence
time in the Ludington reservoir (Table 13). Thirty individuals were
equipped with sonic transmitters; 32 fish carried float tags. Although
determination of precise residence times was constrained by tag
characteristics (battery-life, shedding of tag, and tag failure) and
meteorological conditions (resulting in failure to check reservoir
specimens due to adverse weather), "minimal retention periods" were
ascertained, nevertheless.

Individual variability in minimum residence time was large.

Values ranged from 125 days to less than one day. Carp appeared to

106

 

 

¢m—. mm—. Foo. ebo. mNm. F—m. msN. mswpwaom
NNF. emN. Foo. mmN. mFm. Nom. mmo. vqfl3
m¢—. wwm. Foo. owp. ¢>>. 0mm. Noo. posHmmem
.m_. em@.: .00. sec.: mom. owe. mmo. sssesme
was. mew. moo. nso. oem. mme. ewo. assesses
Nor. mp¢. Noo. NFN. Om¢. mvm. who. Havana;
ems. Fee. moo. mom. men. mow. moo. assesses
mpom epom ooqmoNMfiqmflm m m m oonmowmwdmfim osmz
gonna unmeasum noflmmonmom manawm N mamfipadz manwflnop mapwwmw>

 

.mbm— .nfio>nommm
noanflvsq ens a“ uoxomnp psonw mo Momm\mnpmnoa hconv pnoSobos mo comma no
moanmflnmb unaccomocqfi mo scammon on oamwpass amonfia mmwsmmpm mo mpHSmom . NFeHnme

107

 

 

 

mmummonm omwa mow when m —\m ¢N\~ szrv0fiaom new when
"m\h no nonmeoown man when NF mN\> >w\m uuoam mow mumo
¢N\~ so nonmeoomn mum when m NN\~ >F\> peoflm new when
mN\> no «mnmhooon mam ammo b ¢N\n Pw\h umoam mow mumo
”mnemono mmufl mum mama r ¢N\w hw\w Axspv “new mm» mnmo
wmwmmoum mmwm mew whee m ¢N\> mw\w AMspvownom mow «scum .m
nm\e so use menus» may ease NF nm\e sa\s “seam new muse
¢w\h no umho>oomu mam w w m\m pmomm on mnuo
s o m: «82 S. 93.0
m\Fw no “mumboomn :mam whee mNF m\ww .M\v Axsmvowaom new when
¢\m no ”onebooon mem name mm v\m m\w pmoam mow mnwo
wmxn no wwnmsooon nmwm mhmu FN ¢N\p n\~ Axarvownom new menu
m\v no eono>ooon may name 0 N\> mN\m pdoflm mow mnmo
vmnomono OMflM MMB mhdv mm ¢\m mw\m Axsmvoanom mow mumo
mN\m no «mumpoomn mma m ¢N\m ¢N\m pmoam om memo
hmsw no amnmeooon mum w w ¢N\m awoam om mnuo
name m N\b ¢N\m «moam mm» ammo
¢N\> no nouo>oomh was when m N\> ¢N\w pmoam new mnuo
«see m mmxe ~N\m Assevowsom on psosm .n
.mmwomouo omwa was mam” m >N\m m—\m Axsfivownom mew mumo

oownmm nm>ummno ommmflmm voxomnm
mmanmm mowpuowmm mamasww puma even no spam mmmm mma nmwm an? moaommm

.mbmw aw Amflmmawna snap cowunopmnv mmuu uuoflm no unoppwumueuv

ouuomwnvas muwhunuo afio>nmmmm noumcuemn one «w mommmamn nmwm mo mamnnmm .n« oanmm

108

 

 

¢\m no umnm>ooou mom has F >N\m mN\m paoah ooh uuonm .m

. use” N mm\m mm\m ”seam new "seem .m

hmo F hN\w mN\m Axe—Vowuom no» psouu .m

«has a mm\o am\m “seam on name

name m om\m sm\m Amsmvowsom mm» uses“ .m

ease m oN\m ma\m “selveseom mm» «some .m

mw\w no umvomwmu nah «has 0 ¢w\m ¢F\m Amsmvofimom om pAonm .m

name a F.\m s\m Assnvossom m muse

ammo a esxm e\m “seam mm» muse

omuwnw omnmvum umwm when 5 ¢w\m m\m pmoah mow mnmU

when m ms\w e\m useam .mm» mnso

"N\m no umno>oomn mum «hue NF m~\m h\m paoam me when

N«\w no vouo>oooa may whee ow «~\m F\m pecan on menu

h\m do nono>ooon woa whee e m\@ F\m awoah no» menu

nouomoxo omaa may «has 0— F—\m "\m Auahvowaom mmh memo

couscous om.H men whee m e\m mm\> szsvohsom 0% pzosm .m

¢\m no omnohooon mow mhmu m OM\h ¢N\h pmoah oh mamo

mN\m no monopoomh mow nhmo m OM\h ¢N\> pmoah ow memo

m\m no omnm>oomu men mhmo FF o\m ¢N\h waoam mmh mnoo

«\m no mmum>oomn woe when w wN\h ¢N\h “scam new mnuo
ooflne- omeammno omwmaox omxoans

mMnmnm: nowunmwmm unmade» whoa mpmn mo open emha may nmwm mmh newcomm

 

.nn manna

109

 

mmxow no “mumeeoen seam mm\oF Fm\ofi Ass—venue” m “noun .m
omxov no omnomuom pom s ms\oF Auspv “sow on «menu .m
omenmoum emefl may m mm\os es\os AmssVOfinom new psons .m
mass m ms\os axon Axsmvefleom new “use“ .m
m\ow no assesses pom mass 0 e\o. e\ow Assmvoflsem on “noun .m
when m «\ow om\m Axsmvohuom on when“ .m
b\o— no mmnm>oomu meh has m h\oh on\m Ausmwownom mmh. “noun .m
mass m mm\m nm\m Asssvessom mm“ psosm .m
mhso m m\op .nm\m Amswvossom new “sons .m
mass m mm\m wrxm «seam on reuse .m
ma u m m_\m ofi\m. Anem,ousom .0” “none .m
mass m ml\m ml\m AsswVOflsom mew “sons .m
mF\ no eouopoons sash "has NF m.\m . e\m psoam a ”seam .m
m\m no empoeeoe pom uses 0 e\m ¢\m AMaFVoHsom. a pussy .m
m\m no ombnmmno pom whoa o ¢\m v\m pmoam on wuonm .m
m\m no emsnmmpo pom «has 0 e\m ¢\m pecan on “nose .m
m\m no omenemno.»om mass 0 «\m e\m «seam oh scone .m
use _ smxm mm\m “seam om esoum .m
when m mm\o mm\m seaflm on usonu .m
name mm m\m m.\m Axsmvowsom new mass

vownom uohnompo omomaom quoocu
mxuunmm scannepom nvuwnwm pmma open mo ovum omhm mom mem mu; newcomm

 

“woznfiwaoOV .m— manna

 

110

 

0)

.51
I I

‘D

C

m~\os mm\op Axsmvofisom an essay .m
0n\ow mN\oF AxamvowMOm nah wacky .m
om>nmmno mmmmflom umuoonm
men even mo oven mmha mam mem we» mmwoomm

 

Aeessspsoov .n_ manna

lll

remain in the reservoir longer than any of the other species examined
(carp median retention = 8.5 days vs 2 days for brown trout, h days
for rainbow trout, and 3 days for lake trout). Mark and recapture
data (Gulvas, 1976) tended to corroborate these findings.

Little information exists on the residence time of fishes in
the Ludington reservoir during winter due to a lack of sampling.
However, eight fish (7 carp and 1 white sucker) were recaptured in
reservoir gill nets during 1975, after having been released in the
reservoir in 197h. Presumably, these individuals over-wintered in
the reservoir, although it is conceivable (but highly improbable due
to the mortality probabilities associated with turbine passage) that

these fish left the reservoir and later returned.

DISCUSSION

Tracking investigations of carp and trout in the Ludington
Reservoir afforded the unique opportunity to examine fish behavior,
through intensive surveillance, in relation to both environmental
stimuli and pumped storage activity. The results represent the
initial effort to assess fish movement and orientation in a "pure"
pumped-storage situation. Equally, the Ludington findings represent
the first major field evaluation of fish in the absence of vegetative
cover and daily water temperature gradients. While seemingly
unnatural, these latter conditions may become more prevalent with

the continued development of hydroelectric installations.

Movement Patterns

The most conspicuous behavioral pattern exhibited by tracked
fish was the tendency to remain near the reservoir embankment during
most of their movements. Excursions into open waters were generally
brief, and usually resulted in movement away from one shore location
to another. Apparently, the embankment-water interface served as a
reference marker for fish orientation and movement. Such shoreline
behavior is relatively common in fishes and has been documented for
a variety of species (yellow perch, white sucker - Kelso, 1976;
largemouth bass, smallmouth bass, spotted bass - Peterson, 1975; brown

bullhead - Kelso, 197h; steelhead trout - Falter and Ringe, 197h;

112

113

coho salmon - Scholz et aZ., 1973; cutthroat trout - McCleave and
Horrall, 1970)-

Diel differences in movement patterns for both carp and trout
were not evident. Individual patterns were similar in all reservoir
sectors during both plant operational modes, and during day and
night. Fish movement patterns seemed little affected by magnitude or
direction of water currents.

Patterns of movement were similar for sonic and float-tagged
fish implying that float-tracking can be a profitable technique in
behavioral studies. Corroboration of float-tracking results by
telemetry has previously been accomplished (Hasler et aZ., 1969;
McCleave and Horrall, 1970) and suggests that the inclusion of both
techniques in field studies may provide a financially-attractive and
reliable approach for documenting fish movements.

Although seasonal differences in trout patterns were not apparent
(perhaps because of the limited temporal nature of the trout studies),
carp exhibited more open-water movements during the late fall than
in the warmer months. This was especially obvious in November l97h
when seven of the ten tracks exhibited off-shore components. This
pattern was consistent with reservoir gill net catches and with the
seasonal distribution records noted by other investigators (McCrimmon,
1968; Mackay, 1963; Sigler, 1958; Adams and Hankinson, 1928; Tracy,
1910).

The sensory mechanisms involved in the intimate orientation of
carp and trout with the reservoir embankments are not known. The
display of similar patterns of movement during both daylight and

darkness suggests that optical stimuli are not necessary, although

11h

visual clues may be used when available. Water currents (rheotaxis)
may be important but, here too, differential directional movements
under identical conditions as well as similar movements under different
current regimes (operating modes) indicate the existence of other
sensory cues. Of these, thigmotactic stimuli mediated through the
acoustico-lateralis system seem most probable. A "distant touch
orientation" (Lowenstein, 1957) between fish and the reservoir wall
may exist in which the embankment is perceived from vibrations of

the water waves against the wall created by swimming motions.

This response has previously been observed in brown trout (DeVore,
1975; Baldes and Vincent, 1969) and may account, at least partially,
for the "leading activity" noted when fish encounter fishing nets

or other diversionary structures (Leggett and Jones, 1971; Hunter and
Wisby, 196%). Most probably, fish in the Ludington reservoir (and
elsewhere) utilize combinations of sensory modalities simultaneously

and, in the absence of one environmental cue, readily switch to others.

Angular Change

Although angular change has been used as an important behavioral
parameter in tracking studies by others (Kelso, 197k; Dodson and
Leggett, l97h, 1973; Madison et aZ., 1972), this statistic was accorded
limited use in the Ludington study. This was due to the propensity
of fish to orient in proximity to the Ludington embankment, thereby
restricting movement and directional changes to those imposed by the
physical features of the reservoir. As a result, the probability
distribution of directional changes seldom approximated a circular

distribution upon which the analysis is based (Batschelet, 1965).

115

Hence, the use of angular analysis for the Ludington data may not have
been entirely appropriate.

The relation between angular change in direction and fish physio-
logical processes remains obscure. Until more is known about this
relationship, the use of angular change as a significant behavioral

parameter should be constrained.

Swimming Speed

The average swimming speeds for tracked individuals (0.17 L/sec -
carp; 0.63 L/sec - trout) are much less than values from laboratory
studies. For Cdrassius, Fry and Hart (l9h8) reported a mean swim
speed of 6.h L/sec; Radcliffe (1950) cited a value of 3.h L/sec; and
Bainbridge (1960) observed speeds of greater than 5.0 L/sec. Labora-
tory findings on rainbow and brown trout speeds provide mean values
of 3.3 L/sec (Paulik and DeLacy, 1957), 10.0 L/sec (Blaxter and
Dickson, 1959), 5.0 L/sec (Reimers, 1956), and 1.9 L/sec (Bainbridge,
1962). Even the fastest observed swimming speeds for Ludington fish
(1.1 L/sec, carp 7h-5 during a sixéminute interval in Track 7h-l7;
2.7 L/sec, brown trout 75-25 during a four—minute interval in Track
75-58) are low relative to the laboratory results. Similar low field
fish swim speeds (Young et aZ., 1972; Holliday et aZ., l97h) suggest
that fish seldom exhibit sustained activity levels (Webb, 1975)
comparable to those obtained in experimental situations. This
discrepancy may result from the short temporal nature of most perfor-
mance tests, as well as the inability to incorporate into laboratory
designed studies important behavioral aspects such as foraging,

schooling, or territoriality known to influence swimming activity.

116

Activity levels of carp exhibited a diel periodicity. Dawn and
dusk movement rates were higher than those observed during daylight
(0900 - 1700 hr) and nighttime speeds were greater than those in the
day. Similar fluctuations in carp activity were surmised from angling
records by Marlborough (1970) and noted by Gibbinson (1968) and Cole
(1905). Although winter movement observations of carp at Ludington
are lacking, Johnsen and Heitz (1975) reported that carp in Lake
Mendota only moved at night during this time of the year. They also
reported that instrumented fish tended to move in the company of
other monitored individuals. Similar behavior was observed on several
occasions with Ludington carp. This behavior is not atypical of
species which aggregate (school) for feeding and spawning.

Swimming speeds of trout also demonstrated a crepuscular rhythm.
This pattern has been previously seen in sonic-tracked brown trout
(Holliday et aZ., l97h; Young et aZ., 1972) but was absent in steelhead
trout tracked in the Snake River, Idaho-Washington (Falter and Ringe,
l97h). Unfortunately, almost all of the nighttime swim speed data
for Ludington trout were obtained from one brown trout during a 2h hr
track (Track 75-23). While conclusions based on these data are
tentative, they do affirm earlier accounts that lake brown trout are
both day and night active and exhibit peak activity at dusk (Brynildson
et aZ., 1973; Young et aZ., 1972).

The tendency for both carp and trout to remain active at night
may be responsible for the passage of these species into the Ludington
reservoir. Pumping activities normally occur at night and presumably
affect those species that are night-active. Hence, the reservoir fish

composition may be different from that in the lake because of

llT

differential behavioral rhythms (and consequently passage susceptibility)
of the lake fish species. This may explain the relatively low numbers
of yellow perch in the reservoir (Gulvas, 1976) since, although this
species is abundant in Lake Michigan, it is inactive at night (Eddy
and Underhill, l97h; Scott and Crossman, 1973), and thus will have a
low probability of being drawn into the reservoir during pumping.
Multiple regression analyses of movement speed of trout and carp
revealed that swimming speeds were affected by both environmental
factors and plant operation. Trout swimming speeds were significantly
influenced by climatic variables (wind direction and velocity; air
and water temperature), power plant conditions (water-level drawdown),
and behavioral features (days at liberty). The most important para-
meter was Daysfree indicating that rate of movement increased with
trout residence time. While this correlation may be spurious (due
to low sample size of N = 23), the relation could reflect physiological
adjustments to tag attachment and handling, or acclimation to the
reservoir environment. Such disturbances in activity and orientation
have been recorded in other behavioral studies (McCleave and Stred,
1975; Holliday et aZ., 197k; Hart and Summerfelt, 1973; Shepard, 1973;
Gallepp and Magnuson, 1972; Black, 1958, Spoor, l9hl), although
similar evaluations of post-handling effects are generally lacking
in field-oriented, fish-tracking investigations. The swimming
performance analysis of the Ludington trout suggests that these
delayed locomotor responses may be substantial and should be assessed
in future behavioral research. Without these data, the results of

activity level studies should be evaluated with caution.

118

The response of carp movements (in terms of specific swimming
speed) to external variables differed from that of trout in two
respects: (1) the environmental and power plant factors used in the
regressions accounted for only a small percentage of the variation in
movement rate, and (2) the significant variables influencing movement
rate were different between years, and between sonic and float-tracked
individuals. In all cases, the extreme variation in swimming speed
between tracks (within and.among fish) resulted in low correlation
coefficients. It is apparent, however, that carp speed varied
inversely with changes in the reservoir water-level elevation (Resfluct
and Fluxrate) during power generation. The significant negative
correlation of movement rate with the number of turbines in operation
(Noturbon) in the l97h pooled analysis further implies that the inverse
relation between movement and the magnitude of water drawdown was a
real phenomenon rather than a mathematical artifact. Although water
level manipulation has been a widely used management technique in
carp control (McCrimmon, 1968; Jester, 1971; Sigler, 1958), previous
documentation of this relationship is lacking.

A.major constraint in the interpretation of the multiple regression
analyses for both groups was the absence of biological factors in
the predictive models. The inability to parameterize biological
features such as competition, spawning, hunger, predation, and homing
necessitated their exclusion from a mathematical treatment. The
presence of these interactions may possibly be inferred from the data
(i.e., diel patterns suggesting feeding activity), but their importance
in affecting fish movements remains unclear. A more sophisti>ated
field-experimental approach to resolve the influence of these factors

on swimming speed is plainly warranted.

119
Retention Time

Retention studies of selected fish during 1975 revealed that
minimum residence periods in the reservoir were variable, ranging
from h - 125 days for carp and 0 - 12 days for trout. Homing behavior
may account for the apparent rapid departure of trout from the reser-
voir, although other factors (attraction to currents; increased move-
ment activity with water level drawdown — note positive Resfluct
results in the regression analysis) offer equally plausible explanations
for this phenomenon.

The passage of fish out of the Ludington reservoir is known from
the recapture of reservoir-tagged individuals in Lake Michigan.
However, the tracking studies indicated only one instance (out of 159
tracks) in which an individual was actually observed to leave the
reservoir. Though the tracking data seem inconsistent with the
recapture findings (and gill-net studies which indicate that fish
population abundance has not changed significantly in three years
(Gulvas, 1976), this disparity may be resolved by considering the
probability of a fish leaving the reservoir. If all members of a
species in the reservoir are susceptible to removal, and hence loss
rate a function of population size, then the probability of loss of
any one fish will be small relative to the daily removal percentage
of the population. Since normally only one or two fish were tracked
each day, the likelihood of these fish being removed (especially during
the restricted part of the day in which they were observed ) was rather
low. Furthermore, the departure rate of fish assuredly varies with

the intensity and duration of generating activity, and thus the

120

probability of recording the exit of a tracked individual on any one

day must have fluctuated widely. From this perspective (and considering
that most of the fish tracked were carp which appear to have a long
residence time), it is not startling that only one of the 65 tracked
fish was witnessed in its exit from the reservoir.

The turnover rate (loss rate) of the reservoir fish populations
coupled with reservoir abundance estimates are necessary in assessing
passage mortality during power generation. Thus, more intensive
efforts to precisely delineate population size and residence time may

be required in the future.

CONCLUSIONS

The behavioral studies conducted at Ludington during 197h - 1975
indicate the importance of considering fish behavior in the siting,
design, and operation of hydroelectric projects, particularly pumped-
storage installations. An understanding of fish movements including
swimming depths, activity cycles, and the influence of water flows
will aid in constructing and operating power plants to minimize fish
attraction, entrainment, and passage mortality.

Further information at Ludington would augment this study and
help in the final impact analysis. The behavior of species near the
plant and in the reservoir could be better defined by further analysis
of existing field data or expansion of ultrasonic telemetry. Recent
developments with automatic monitoring and multichannel transmitters
(which relay data on depth, temperature, and location) offer great
potential for interpretation of fish movements. Water currents near
the plant and in the reservoir should also be better defined and
compared with data from pre—operational modelling efforts for
verification.

Knowledge thus gained should be imparted to design engineers so
that placement and operation of power plants will minimally affect
the normal activities and dynamics of fish populations. Detailed
fish behavior data may lead to the development of efficient guidance

barriers (including lights, water and air jets, sound, louvers, and

121

122

conduits) which elicit attractive or avoidance responses from fish,
thus reducing impact. Distinct near-shore movements noted in this
study indicate the possibility for designing bypass channels in the
upper reservoirs of pumped storage systems. This would permit an
alternate fish pathway to the lower basins and thus possibly reduce
overall fish passage mortality.

Modelling efforts of power plant impact and ecosystem dynamics
should be developed which incorporate behavioral phenomena into the
simulation framework. The inclusion, initially, of such behavioral
aspects as spawning periods and seasonal migration patterns would be
useful in developing a basic perspective from which to assess environ—
mental events. More refined behavioral parameters (light-temperature-
water current-locomotory relationships) should be integrated into

these models as the data become available.

SUMMARY

Movement patterns, activity levels, and residence periods of
carp and trout were investigated in the Ludington Pumped Storage
Reservoir in 197h and 1975 using ultrasonic telemetry and float-
tracking procedures. These species were studied because of their
availability and abundance, tag retention capabilities, and biological
and recreational importance.

Sixty-five fish (52 carp, 10 brown trout, 3 rainbow trout) were
monitored for a total of 1159 hr, spanning 159 tracks. Tracking
periods ranged from 1—2h hr and were generally accomplished during
the daytime. Four fish were tracked through the night and 12 individuals
were monitored during dawn and dusk. Tracking sessions were terminated
for a variety of reasons including equipment failure, expiration of
the transmitter battery, cessation of fish movement, and adverse
weather conditions.

The most common fish movement pattern was a straight path orienta-
tion parallel and adjacent to the reservoir embankment. The shoreline
appeared to serve as a reference for locomotory activity. Open-water
excursions occurred but were normally brief. Obvious seasonal
differences in movement pattern were not evident in trout, although
carp displayed more offshore movements in late fall than in the warmer
months. Patterns of movement were similar between sonic— and float—

tagged fish, and during both daylight and darkness.

123

12h

Length of individual tracks ranged from 0.07 - 16.h9 km for
carp and 0.15 — 27.hh km for trout. Average swimming speed for carp
varied from 0.01 - 0.6h L/sec (0.8 - 38.1 cm/sec), and from 0.01 - 1.30
L/sec (0.8 - 79.9 cm/sec) for trout. No significant differences
(P > 0.05) were detected in movement rates between sonic and float
tagged individuals, between brown and rainbow trout swim speeds, and
between carp swimming rates for l97h and 1975. A significant two-way
interaction (P < 0.05) was evident, however, between year and track
type for carp speeds. Rates of movement determined for carp during
pumping and generating operations differed significantly (P < 0.05)
for each year but this trend was inconsistent between years. Overall,
the mean swimming speed for carp was 0.17 L/sec (10.5 cm/sec) and
0.63 L/sec (38.6 cm/sec) for trout. These values are lower than most
activity levels reported from laboratory studies.

Pronounced diurnal activity levels were displayed by the tracked
species. Trout exhibited activity peaks during dawn and dusk and
remained active throughout the day and night. Carp displayed a
similar crepuscular rhythm in swimming speed, but were much more
active at night than during the day. The tendency for both groups
of fish to remain night-active may account for their passage into
the reservoir.

Factors affecting swimming speed were analyzed for their relative
importance by stepwise linear multiple regression analysis. For
trout, the independent variables explained 83% of the variation in
movement rate. Behavioral features (days at liberty), environmental
factors (water temperature), and power plant operations (reservoir

drawdown) were influential in trout movement. For carp, the multiple

125

correlation coefficients were low and had a range of 0 - 65%. Although
different variables were significant in each of the carp analyses,
water current parameters and power plant factors appeared in a majority
of the regression equations. Biological factors were not evaluated

in the regressions for either species and, hence, their importance

on fish movement rate was not determined.

Residence period of fish in the Ludington reservoir was assessed
from observations on 62 individuals accomplished during 1975. Minimum
residence intervals ranged from several hours to 125 days. Carp seemed
to remain in the reservoir longer than any of the trout species
examined. Median retention period for carp was 8.5 days and 2 days
for the combined trout species. Greater accuracy of residence time
estimates can only be obtained by expanding this aspect of the
Ludington research.

The need for future studies related to the impact of hydroelectric

development of fish populations is indicated.

LITERATURE C ITED

LITERATURE CITED

Adams, C. C., and T. L. Hankinson. 1928. The ecology and economics
of Oneida lake fish. Roosevelt Wildlife Ann. 1(3 & h): 319-333.

Bainbridge, R. 1960. Speed and stamina in three fish. J. Exp.

Bainbridge, R. 1962. Training, speed and stamina in trout. J.

Exp. Biol. 39: 537-555-

Baldes, R. J., and R. E. Vincent. 1969. Physical parameters of
microhabitats occupied by brown trout in an experimental flume.
Trams. Amer. Fish. Soc. 98: 37-hl.

Batschelet, E. 1965. Statistical methods for the analysis of problems
in animal orientation and certain biological rhythms. Amer.
Inst. Biol. Sci. Monogr. l: 57 P.

Batschelet, E. 1972. Recent statistical methods for orientation
data, pp. 61-91. in S. R. Galler, K. Schmidt-Koenig, G. J.
Jacobs, and R. E. Belleville (eds.) Animal Orientation and Navi—
gation. National Aeronautics and Space Administration, Washington,
D. C.

Black, E. C. 1958. Hyperactivity as a lethal factor in fish. J.
Fish. Res. Board Can. 15: 573-586.

Blaxter, J. H. S., and W. Dickson. 1959. Observations on the swimming
speeds of fish. J. Cons. int Explor. Mer. 2h: h72-h79.

Brynildson, 0. M., V. A. Hacker, and T. A. Klick. 1973. Brown trout:
life history, ecology and management. Wisc. Dept. Nat. Resour.
Publ., 23h: 16 p.

Cole, L. J. 1905. The German carp in the United States. Rept. U. S.
Bur. Fish 190h: 525—6hl.

Conover, W. J. 1971. Practical nonparametric statistics. John
Wiley and Sonc, New York. A62 p.

DeVore, P. W. 1975. Daytime behavioral responses of adult brown

trout (Salmo trutta) to cover stimuli in stream channels. M.S.
Thesis, Michigan State Univ., East Lansing. 38 p.

126

127

Dodson, J. J., and W. C. Leggett. 1973. The behavior of American
shad (Alosa sapidissima) homing to the Connecticut River from
Long Island Sound. J. Fish. Res. Board Can. 30: l8h7-1860.

Dodson, J. J., and W. C. Leggett. l97h. Role of olfaction and vision
in the behavior of American shad (Alosa sapidissima) homing to
the Connecticut River from Long Island Sound. J. Fish. Res.
Board Can. 31: 1607-1619.

Draper, N., and H. Smith. 1966. Applied regression analysis. John
Wiley and Sons, New York. h07 p.

Eddy, S., and J. C. Underhill. 19TH. Northern fishes (3rd edition).
University of Minnesota Press, Minneapolis. hlh p.

Estes, R. D. 1971. The effects of the Smith Mountain Pumped Storage
Project on the fishery of the lower reservoir, Leesville,
Virginia. Ph.D. Thesis, Virginia Polytechnic Institute and
State Univ., Blacksburg. 151 p.

Falter, C. M., and R. A. Ringe. l97h. Pollution effects on adult
steelhead migration in the Snake River. Ecol. Res. Ser., EPA-660/3—
73-017. 101 p.

Fry, F. E. J., and J. S. Hart. l9h8. Cruising speed of goldfish in
relation to water temperature. J. Fish. Res. Board Can. 7: 169-
175.

Gallepp, G. W., and J. J. Magnuson. 1972. Effects of negative buoyancy
on the behavior of the bluegill, Lepomis macrochirus Rafinesque.
Trans. Amer. Fish. Soc. 101: 507-512.

Gibbinson, J. A. 1968. Carp. MacDonald, London.

Groot, C., K. Simpson, I. Todd, P. D. Murray, and G. A. Buxton. 1975.
Movements of sockeye salmon (Oncorhynchus nerka) in the Skeena
River estuary as revealed by ultrasonic tracking. J. Fish. Res.
Board Can. 32: 233-2h2.

Gulvas, J. A. 1976. An evaluation of.the composition and relative
abundance of fishes pumped into the Ludington Pumped Storage
Reservoir from Lake Michigan, 1972-1975. M.S. Thesis, Michigan
State Univ., East Lansing.

Hart, L. G., and R. C. Summerfelt. 1973. Homing behavior of flathead
catfish, Pylodictis olivaris (Rafinesque) tagged with ultrasonic
transmitters. Proc. 27th Ann. Conf. S.E. Assoc. Game and Fish
Comm: 520-531.

Hasler, A. D., E. S. Gardella, R. M. Horrall, and H. F. Henderson.
1969. Open-water orientation of white bass, Rbccus chrysops,
as determined by ultrasonic tracking methods. J. Fish. Res.
Board Can. 26: 2173-2192.

128

Hauck, F. R., and Q. A. Edson. 1976. Pumped storage: its significance
as an energy source and some biological ramifications. Trans.
Amer. Fish. Soc. 105: 158-l6h.

Holliday, F. G. T., P. Tytler, and A. H. Young. l97h. Activity levels
of trout (salmo trutta) in Airthrey Loch, Stirling, and Loch
Leven, Kinross. Proc. R.S.E. (B) 7h: 315-331.

Hunter, J. R., and W. J. Wisby. 196k. Net avoidance behavior of
carp and other species of fish. J. Fish. Res. Board Can. 21:
613-633.

Jahn, L. A. 1966. Open-water movements of the cutthroat trout (Salmo
clarki) in Yellowstone Lake after displacement from spawning ,,i
streams. J. Fish. Res. Board Can. 23: 1h75-1h85.

Jahn, L. A. 1969. Movements and homing of cutthroat trout (SaZmO
clarki) from open-water areas of Yellowstone Lake. J. Fish. 1,
Res. Board Can. 26: l2h3-126l.

Jester, D. B. 1971. Effects of commercial fishing, species intro-
ductions, and drawdown control on fish populations in Elephant
Butte Reservoir, New Mexico, p. 265-285. in G. E. Hall (ed.)
Reservoir Fisheries and Limnology. Amer. Fish. Soc. Spec.
Publ. No. 8.

Johnsen, P. B., and J. G. Heitz. 1975. Beep! Beep! Beep! Where
are the carp? Wisc. Cons. Bull. (Nov-Dec 1975): 12—13.

Johnson, J. H. 1960. Sonic tracking of adult salmon at Bonneville
Dam, 1957. U. S. Fish Wildl. Serv. FiSh. Bull. 176: h71-h85.

Kelso, J. R. M. 197k. Influence of a thermal effluent on movement
of brown bullhead (ICtaZurus nebulosus) as determined by ultra-
sonic tracking. J. Fish. Res. Board Can. 31: 1507-1513.

Kelso, J. R. M. 1976. Movement of yellow perch (Perca flavescens)
and white sucker (Chtostomus commersoni) in a nearshore Great
Lakes habitat subject to a thermal discharge. J. Fish. Res.
Board Can. 33: h2-53.

Leggett, W. C., and R. A. Jones. 1971. Net avoidance in American
shad (Alosa sapidissima) as observed by ultrasonic tracking
techniques. J. Fish. Res. Board Can. 28: 1167-1171.

Liston, c. R., and P. I. Tack. 197k. A study of the effects of
installing and operating a large pumped storage project on the
shores of Lake Michigan near Ludington, Michigan. 1973 Ann.
Rept. to Consumers Power Co., Vol. 1, Dept. Fish and Wildlife.,
Michigan State Univ., 17h p.

129

Liston, C. R., and P. I. Tack. 1975. A study of the effects of
installing and operating a large pumped storage project on the
shores of Lake Michigan near Ludington, Michigan. 197A Ann.
Rept. to Consumers Power Co., Vol. 1, Fisheries Research, Dept.
Fish and Wildl., Michigan State Univ., 166 p.

Lowenstein, 0. 1957. The sense organs: the acoustico-lateralis
system, p. 155-209. in M. E. Brown (ed.) The Physiology of
Fishes. Academic Press, New York.

Mackay, H. H. 1963. Fishes of Ontario. Publ. Ontario Dept. Lands
Forests.

Madison, D. M., R. M. Horrall, A. B. Stasko, and A. D. Hasler. 1972.
Migratory movements of adult sockeye salmon (Oncorhynchus nerka)
in coastal British Columbia as revealed by ultrasonic tracking.
J. Fish. Res. Board Can. 29: 1025-1033.

Marlborough, D. 1970. The behavior of common carp (Gyprinus carpio L.)
and crucian carp (Corassius carassius L.) as indicated by angling
records. Lond. Nat. A9: 69—76.

McCleave, J. D. 1967. Homing and orientation of cutthroat trout (Salmo
clarki) in Yellowstone Lake, with special reference to olfaction
and vision. J. Fish. Res. Board Can. 2h: 2011-20hh.

McCleave, J. D., and R. M. Horrall. 1970. Ultrasonic tracking of
homing cutthroat trout (SdZmo cZarki) in Yellowstone Lake. J.
Fish. Res. Board Can. 27: 715-730.

McCleave, J. D., and K. A. Stred. 1975. Effect of dummy transmitters
on stamina of Atlantic salmon (Stho solar) smolts. J. Fish.
Res. Board Can. 32: 559-563.

McCrimmon, H. R. 1968. Carp in Canada. Bull. Fish. Res. Board
Can. 165: 93 p.

Paulik, G. J., and A: C. DeLacy. 1957. Swimming abilities of upstream
migrant silver salmon, sockeye salmon and steelhead at several
water velocities. Univ. Wash. Scool Fish. Tech. Rept. hh: l-hO.

Peterson, D. C. 1975. Ultrasonic tracking of three species of black
basses,.Micropterus spp., in Center Hill Reservoir, Tennessee.
M.S. Thesis, Tennessee Technological Univ., Cookeville. 1&2 p.

Radcliffe, R. W. 1950. The effect of fin clipping on the cruising
speed of goldfish and coho salmon fry. J. Fish. Res. Board
Reimers, N. 1956. Trout stamina. Progr. Fish-Cult. 18: 112.

Robbins, T. w., and D. Mathur. 1976. The Muddy Run Pumped Storage
Project: a case history. Trans. Amer. Fish. Soc. 105: 165-172.

130

Scholz, A. T., J. C. Cooper, D. M. Madison, R. M. Horrall, A. D. Hasler,
A. E. Dizon, and R. J. Poff. 1973. Olfactory imprinting in coho
salmon: behavioral and electrophysiological evidence. Proc. 16th
Conf. Great Lakes Res. 1973: lh3-153.

Schoumacher, R. 1976. Biological considerations of pumped storage
development: introductory remarks. Trans. Amer. Fish. Soc.
105: 155-157.

Scott, W. B., and E. J. Crossman. 1973. Freshwater fishes of Canada.
Bull. Fish. Res. Board Can. 18%: 966 p.

Shepherd, B. 1973. Transmitter attachment and fish behaviour.
Underwater Telemetry Newsletter 3: 8-11.

Sigler, W. F. 1968. The ecology and use of carp in Utah. Utah State
Univ., Agr. Exp. Sta. No. h05: 62 p.

Spoor, W. A. l9hl. A method for measuring the activity of fishes.
Ecology 22: 329-331.

Stasko, A. B. 1971. Review of field studies on fish orientation.
Ann. N. Y. Acad. Sci. 188: 12-29.

Stasko, A. B. 1975. Underwater biotelemetry, an annotated bibliography.
Fish. Mar. Serv. Res. Dev. Tech. Rep. 53h: 31 p.

Stasko, A. B., R. M. Horrall, and A. D. Hasler. 1976. Coastal move—
ments of adult Fraser River sockeye salmon (Oncorhynchus nerka)
observed by ultrasonic tracking. Trans. Amer. Fish. Soc. 105:
6h-71.

Stasko, A. B., R. M. Horrall, A. D. Hasler, and D. Stasko. 1973.
Coastal movements of mature Fraser River pink salmon (Oncorhynchus
gorbuscha) as revealed by ultrasonic tracking. J. Fish. Res.
Board Can. 30: 1309-1316.

Tracy, H. C. 1910. Annotated list of the fishes known to inhabit
the waters of Rhode Island. Fortieth Ann. Rept. Comm. Inland
Waters Rhode Island: 35h—360.

Warden, R. L. Jr., W. J. Lorio. 1975. Movements of largemouth bass
(Micropterus salmoides) in impounded waters as determined by
underwater telemetry. Trans. Amer. Fish. Soc. th- 696-702.

Webb, P. W. 1975. Hydrodynamics and energetics of fish propulsion.
Bull. Fish. Res. Board Can. 190: 158 p.

Young, A. H., P. Tytler, F. G. T. Holliday, and A MacFarlane. 1972.
A small sonic tag for measurement of locomotor behaviour in
fish. J. Fish. Biol. A: 57-65.

"I71111111111111.1111“