F

 

WIHIIIHHWIWIWIWWWHWHHHUIUIWHUI L

  

- \Iflllllllllfllflllllwfllml\\\\,jl\l|flll\jj|

' LIBRARY

Michigan State
University .

 
       

This is to certify that the

thesis entitled 1

\

ABUNDANCE, DISTRIBUTION AND COMMUNITY INTERACTIONS
0F DEMERSAL FISHES INHABITING A NEW PUMPED STORAGE RESERVOIR 1
ON LAKE MICHIGAN NEAR LUDINGTON, MICHIGAN

presented by

Joseph R. Bohr

has been accepted towards fulfillment
of the requirements for

Master of Science ngree in Fisheries and Wildlife

 

,/ q 7

Major professor

Date 22 February 1980

 

0-7639

ABUNDANCE, DISTRIBUTION AND COMMUNITY INTERACTIONS
OF DEMERSAL FISHES INHABITING A NEW PUMPED STORAGE RESERVOIR

ON LAKE MICHIGAN NEAR LUDINGTON, MICHIGAN

By

Joseph R. Bohr

A THESIS

Submitted to

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

MASTER OF SCIENCE

Department of Fisheries and Wildlife

1980

 

ABSTRACT

ABUNDANCE, DISTRIBUTION AND COMMUNITY INTERACTIONS
0F DEMERSAL FISHES INHABITING A NEW PUMPED STORAGE RESERVOIR
ON LAKE MICHIGAN NEAR LUDINGTON, MICHIGAN
BY

Joseph R. Bohr

The abundance, spatial distribution and interactions of sculpins
(Coitub spp.), johnny darter (Etheoatoma nigaam) and trout—perch
(Peacopéia omiAcomaycuA) captured by trawls during 1974 through 1977
in the Ludington Pumped Storage Reservoir were studied. Monthly johnny
darter density peaked at 4.4 fish/100 m2 in August 1976. Sculpin
density peaked at 4.5 fish/100 m2 in the same month. Trout—
perch peaked at 2.0 fish/100 m2 in July 1976.

The food habits of sculpins, johnny darters, trout—perch and burbot
(Lota iota) captured in 1976 were studied. Sculpins fed on juvenile fish
and benthic macroinvertebrates. Johnny darters and trout—perch fed
mainly on benthic macroinvertebrates. Burbot fed primarily on fish and
were the most important predator in the demersal community.

Reservoir catch per effort of sculpins, darters and trout—perch was
dependent on temperature. Johnny darter and trout-perch fluctuation

were highly dependent on total invertebrate densities.

 

ACKNOWLEDGMENTS

I wish to acknowledge Michigan State University, Department of
Fisheries and Wildlife for providing a research assistantship and the
facilities for this study. I am grateful to the Consumers Power and
Detroit Edison Companies for making the study site available and for
providing the necessary funds.

I thank Dr. Charles R. Liston, major professor, for providing the
opportunity and the necessary data, and I thank Dr. Peter I. Tack for
his encouragement.

I am grateful to committee members Dr. Stanley Zarnoch and Dr.
Thomas Burton for their assistance and support.

The support, knowledge and good nature of Dan Brazo, Field Director
at Ludington, was of great help and I am grateful.

Rich O'Neal kindly provided trout—perch food habit data and I
thank him.

Many thanks to co—workers Bob Anderson, Joan Duffy, Walt Duffy,
Fred Koehler, Rick Ligman, Rich O'Neal and Greg Peterson for doing more
than their share while this thesis or other duties kept me busy elsewhere.

The following graduate students are thanked for their efforts in
gear development and in data collection earlier in the history of the
project: John Armstrong, Tom Chiotti, Larry Green, John Gulvas, Rick
Hauer, Dan Lawson, Dave Lechel, Greg Olson, and Fred Serchuk.

Leo Yeck provided much in the way of overall knowledge, safety
consciousness and general philosophy for which I am grateful.

7L7;

 

 

A very big thank you is due Ms. Joan Tims for her typing ability,
demand for perfection and all around good humor.
Lastly, a special thank you to my parents, Del and Lucille Bohr.

I hope they realize my gratitude though it is not often displayed.

‘q
NA-
V‘.

 

TABLE OF CONTENTS

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

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

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

DESCRIPTION OF THE STUDY AREA

POPULATION SIZE, SPATIAL DISTRIBUTION AND
POPULATTION COMPARISONS .

Methods .
Fish Collections . .
Trawl Measurements . . . . . . . . . .
Statistical Methods . . . . . . . . .

Results and Discussion
Trawl Measurements . . . . .
Catch Per Effort, Population
Estimations and Spatial Distribution

Sculpins
Johnny Darter .
Trout— —perch . . .

Reservoir — Lake Michigan Comparisons

FOOD HABITS . . . . . . . . . . . .

Methods .

Results and Discussion
Sculpin Food Habits .
Johnny Darter Food Habits
Trout—perch Food Habits
Comparative Food Habits

REGULATING FACTORS . . . . . . . . . . . . . . .
Predation . . . . . .
Parasitism . . .
Spawning Habitat . . . . . .

Water Quality .
SUMMARY AND CONCLUSIONS

LITERATURE CITED . . . . . . . . . . . . . . .

iv

 

 

Table

10

Total monthly trawl samples from stations Tl through

LIST OF TABLES

T6 in the Ludington reservoir 1974 through 1977

Estimated monthly average densities of total inverte-

brates as determined by Ponar grab samples in the

Ludington reservoir from 1974 through 1977 (adapted

from Duffy and Liston 1978).

Trawl widths and estimated tow speed at several depths

in Lake Michigan.

Comparison of average catch per unit of effort of
sculpins in the north (T5, T6) central (T1, T4) and
south (T2, T3) sections of the Ludington reservoir.

Comparison of average catch per unit effort of johnny

darters in the north (T5, T6), central (T1, T4) and
south (T2, T3) sections of the Ludington Reservoir.

Comparison of average catch per unit of effort of
trout—perch in the north (T5, T6), central (T1, T4)

and south (T2, T3) sections of the Ludington reservoir.

Comparisons of 1977 average monthly catch per unit of
effort between trawls in the Ludington reservoir and

in nearby Lake Michigan.

A comparison of relative numbers of sculpins, johnny

darters and trout—perch in trawl samples from the
Ludington reservoir and two areas of Lake Michigan.

Calculations used in estimating volumes of certain
organisms in food habit studies. Adapted from

Kenaga (1975).

Components of the major food groups in the Ludington

reservoir based on stomach analysis of sculpins,

johnny darters,

trout—perch and burbot.

18

22

25

29

31

34

35

 

LIST OF TABLES (Continued)

Table

11

12

13

14

15

l6

l7

18

Percent of total volume (Z V), percent of total
numbers (% N), percent frequency of occurrence (Z F0)
and percent relative value (2 RV) of food types eaten
by sculpins (51—100 mm TL) in the Ludington reservoir
in 1976.

Percent of total volume (Z V), percent of total numbers
(2 N), percent frequency of occurrence (2 F0) and
percent relative value (Z RV) of food types eaten by
johnny darters (51-100 mm TL) in the Ludington reservoir
in 1976.

Percent of total volume (Z V), percent of total numbers
(2 N), percent frequency of occurrence (2 F0) and
percent relative value (Z RV) of food types eaten by
trout-perch in the Ludington reservoir in 1976 (adapted
from O'Neal 1977).

Percent relative value of food items and major food
types found in sculpins, johnny darters and trout—perch
between 51 and 100 mm TL captured in the Ludington
reservoir during summer 1976.

Resource overlap values calculated using relative values
of major food groups for specimens between 51 and 100 mm
TL taken in summer 1976.

Relative mouth sizes of demersal fish inhabiting the
Ludington reservoir. Ten fish of each species were
measured.

Average volume (mm3) of the major prey common to
sculpins, johnny darters and trout—perch taken in the
Ludington reservoir in 1976.

Percent of total volume (Z V), percent of total numbers
(Z N), percent frequency of occurrence (Z F0) and
percent relative value (Z RV) of food types eaten by
burbot in the Ludington reservoir in 1976.

vi

40

42

43

45

47

48

52

 

LIST OF FIGURES

Figure

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

2 The Ludington Pumped Storage Reservoir showing trawl‘
path positions.

3 The location of the Ludington Pumped Storage Reservoir,
gill net sampling stations in the reservoir and Lake
Michigan trawl stations.

4 Seasonal reservoir trawl catch per unit effort of
sculpins, 1974 through 1977.

5 Seasonal reservoir trawl catch per unit effort of
johnny darters, 1974 through 1977.

6 Seasonal reservoir trawl catch per unit effort of

trout—perch, 1974 through 1977.

12

19

23

26

 

INTRODUCTION

The Ludington Pumped Storage Reservoir provided a unique oppor—
tunity to study the development of and interactions within a new aquatic
system. Filling of the reservoir has exposed the new environment to
colonization by benthic invertebrates and various fish species. Colon—
ization by benthic macroinvertebrates from 1973 through 1977 has been
described by Duffy and Liston (1978). Gulvas (1976) investigated fish
colonization in the Ludington Reservoir through gill net and trawling
efforts and found that alewife (Afioba paeudohaaengua [Wilson]), smelt
(Obmehué mohdax Linne), spottail Shiner (Nothopib hudboniub [Clinton]),
sculpin (Cottub cognaZuA Richardson and C. batndi Girard), johnny darter
(Etheoatama Mgwm Rafinesque), trout-perch (Pump/512 omacomaycu/s
[Walbaum]) and carp (Cypninua cadpto Linne) were abundant. Steelhead
(Saflmo gaiadneni Richardson), brown trout (S. Inuita Linne), coho
(Oncathynchua hLAchh [Walbaum]), and chinook salmon (0. tahawytAcha
[Walbaum]) were reported to be seasonally abundant in the reservoir.
Notably absent in significant numbers in reservoir catches were yellow
perch (Pmca filiauezscem [Mitchill]), white (Caioatomws comma/50M
[Lacepede]) and longnose suckers (C. catabtomua [Forster]), round white—
fish (17/1050me cyund/Laceum Pallas) and lake trout (Salve/Mum
namaycubh [Walbaum]). Burbot (Loia Kata [Linne]) were not consistently
abundant from 1972 through 1975 (Gulvas 1976).

Interests in the Ludington plant have focused primarily on species

of direct importance to commercial and sport fisheries such as the sal—

 

 

monids and yellow perch. Much work has been done and is continuing on
estimation of fish losses due to operation of the Ludington plant (Liston
1977). The idea that the reservoir may actually add a measure of pro—
duction in the form of forage fish has not been adequately investigated.
The present study offers preliminary estimates of that returned production.

This study concentrates on the demersal community of sculpins,

 

johnny daters, and trout—perch in the Ludington reservoir. These species
are not overly active swimmers and are, therefore, more likely to form a
resident, though open, reservoir population. Other demersal species
such as the ninespine stickleback (Pungiiiub pungitiué [Linne]) and
longnose dace (Rhinichthyb catahactae [Valenciennes]) were not captured
in adequate numbers in the reservoir to allow study. Alewives, smelt
and spottail shiners, while captured frequently and in large numbers,
are more pelagic in habit and are believed to be too transient to allow
meaningful conclusions. Burbot were abundant in 1976 collections and
their food habits were analyzed to determine their importance as a
predator in the demersal community. Salmonids are seasonally abundant
in the reservoir and since they may also be important predators, a

number of stomachs were examined.

DESCRIPTION OF THE STUDY AREA

The Ludington Pumped Storage Plant was constructed by the Consumers‘
Power and Detroit Edison Companies of Michigan on the eastern shore of
Lake Michigan about 6.4 km south of Ludington. The plant has operated
commercially since January 1973, after initial filling in October 1972.
The reservoir is 4.0 km long, averages 1.2 km in width and has a total
surface area of 3.4 km2 when full (Figure 1). The inside wall is paved
with asphalt down to a clay berm which functions as a service road
during reservoir drawdowns. The reservoir bottom and sides up to the
service road are lined with compacted clay. Total clay bottom area is
approximately 3.2 kmz. A limestone rock area 365 m by 244 m lines the
bottom in front of the intake structure and serves as protection from
scour. Maximum water depth is approximately 29 m in the south and 34 m
in the north end (Liston and Tack 1975).

The Ludington Power Plant operates as a method of storage for surplus
electrical energy. Water is pumped up to the reservoir from Lake Michigan
with six Francis—type reversible pump—turbines during off—peak demand
periods, generally at night and on weekends. During peak electrical demand
periods water is allowed to flow back through the turbines to produce
electricity. Each turbine revolves at a maximum of 112.5 rpm during
generation and can discharge up to 358 m3/sec. A maximum of 314 m3/sec
per unit can be passed during pumping. With all six units operating,
maximum water flow during generation is about 2,151 m3/sec and about
1,886 m3/sec during pumping. Water levels in the reservoir can fluctuate
a maximum of 20.4 m under normal operation and the weekly turnover rate

is about 2.4 (Liston and Tack 1975).

 

 

 

Figure 1. Aerial view of the Ludington Pumped Storage Project including
the offshore jetties and breakwall.

 

.1, “1.1.11“ we,

3

 

 

 

Physical and chemical parameters of the reservoir were investigated
by Liston et a1. (1976). The following values were reported for 1973
and 1974: Turbidity, 0.4—36.0 FTU; water clarity measured by secchi
disc, 0.3—4.5 m; dissolved oxygen, 9—12.6 ppm; pH, 8.0-8.5; alkalinity,
102—123 ppm; and dissolved solids, 159—190 ppm. The reservoir was nearly
always homeothermous and temperatures rarely exceeded 22 C in the summer.
Updwellings in Lake Michigan during summer resulted in occasional precipi-
tous drops in reservoir temperatures, followed by rapid recoveries. In
general, no appreciable differences in chemical or physical parameters
have been noted between the reservoir and adjacent Lake Michigan (Liston
et a1. 1976). .

The Lake Michigan bottom near the Ludington plant has some inshore
sandy areas but is largely clay and stone. Certain areas have outcroppings

of large rocks presenting difficulties in sampling.

  
  

'...-

.1. "y "t _
:._ A

9;) :m" REEF? 3 ‘1

 

!
i .
- ';.I -' find-Hit": fi-i': 3:3";

  

   

. .r' I;. :7" .l‘,‘ ...
I“ I” i a t.u'ot .' '. - =1.

in: :nm ESE-EN .- ‘

 

L? ,flm's') 1.9V
.-._2 $1.2 ' -.- I'-.'.-"...-E."

gel-“3:... I

 

POPULATION SIZE, SPATIAL DISTRIBUTION
AND POPULATION COMPARISONS

Methods

Fish Collections

During 1974 through 1977 sculpins, johnny darters and trout-perch
were collected with a semi—balloon otter trawl having a 7.6 m head rope,
38.1 mm stretch mesh body and a 3.1 mm cod end liner. Five-minute hauls
at approximately five knots were made after dark at each of six stations
around the reservoir perimeter (Tl through T6, Figure 2) and at three
stations through the reservoir center (T7 through T9). Table 1 presents
the monthly number of trawl samples taken.

Fish were relatively rare in samples from stations T7 through T9
and it has since been decided that trawl lines were not always of sufficient
length to adequately sample the bottom in these center stations. Therefore,
comparisons have been limited to the six perimeter stations.

Burbot and salmonid specimens used in this study were captured in
bottom and surface gillnets. Bottom nets were 106.6 m in length, 1.8 m
deep, and consisted of 15.2 m panels of 25, 51, 64, 76, 102, 114, and
117 mm stretched mesh. These were set approximately weekly at stations
7, 8 and 9 (Figure 3) during 1973 through 1977. Surface gillnets were
used consistently since August 1974. Two nets were used initially, one
of 127 mm stretch mesh and the second of 177 mm mesh. Both were 91 m long
and 31.1 m deep, and were set perpendicular to the reservoir walls. A
third surface net with mesh identical to the bottom nets was first used
in April 1975. All gillnets were set for approximately 24 hours. Com—

plete details of gear and sampling dates are presented by Gulvas (1976).

 

 

 

 

Figure 2. The Ludington Pumped Storage Reservoir showing trawl .path
positions. .

 

 

Table 1. Total monthly trawl samples from stations Tl through T6
in the Ludington reservoir 1974 through 1977.

Station

 

Month

 

1974

April
May

June

July
August
September

HHNNNH

P—‘l—‘NNNH

l—‘l—‘NNNH

l—‘F—‘NNNH

HIP-"NNNH

t—‘I—‘NNNH

 

1975

May

June

July
August
September
October

Nl—‘Ob—‘NN

NI—‘OI—‘NN

H H o H N N

f-‘l—‘Ol—‘NN

l—‘l—‘OONN

l-‘l-‘OONN

 

1976

April
May

June

July
August
September
October

H H N N Mia H

H H N N MIA H

H H N N NI—l H

H H N N H H H

H H N N H H H

H H N NIH H H

 

1977

April

May

June

July

August
-September

l--“l\>l\)l\>l\>|-J

|-—-‘l\)r\)f\)l\Jl>-I

HNNNNH

l—‘NNNNH

l—‘NNNNH

l—‘NND—‘NH

 

 

11

Figure 3. The location of the Ludington Pumped Storage Reservoir, gill
net sampling stations in the reservoir and Lake Michigan
trawl stations.

 

 

 

 

 

 

 

 

 

 

    
  

 

12

m
v.4
J

a

Z
4
o
L I
o
_
E

 

 

 

13

Regular bottom trawling was accomplished in Lake Michigan near the
plant beginning in 1977. Two trawling stations were established, station
one approximately 4.8 km south of the power plant and station two just
north of the plant jetty structures (Figure 3). Gear and methodology
were identical to those in the reservoir. Data were used for comparative
purposes.

Fish collected in trawls were placed on ice overnight and processed
the following morning. Processing consisted of recording total length
(mm), weight (g), sex, maturity, and gonadal condition. If large numbers
of any species were collected, a subsample of 20 specimens was processed.
In all cases, total numbers and weight were recorded for each species at
each station. Fish were preserved in 10% formalin and later changed to
40% isopropyl achohol.

Fish captured in gillnets were tagged and released when possible,
but otherwise were taken to the laboratory and processed on the day of
capture in a manner identical to the trawl samples.

Catch per unit effort data were analyzed seasonally. Spring was
considered to be from ice—out through 15 June, summer from 16 June through
31 August, and fall from 1 September to the end of sampling (about mid-
November). The designated seasons follow roughly the changes in the local

environment.

Trawl Measurements

It is necessary to know the bottom area covered by a trawl sample
in order to estimate the size of the populations present. Many factors
affect trawl sample size and the relationships are often quite complex

(Wathne 1959). Boat speed, depth at the sample site and the length of

 

 

14

line to the trawl are all important variables, and these factors were
measured in the following manner.

Small floats were attached to the trawl doors with sufficient line
to allow the buoys to reach the water surface at all depths to be
encountered. The trawl was then deployed in a manner identical to normal
sampling operations. The towing vessel was followed by a second boat
from which the distance between the floats was measured. Measurements
were made at two towing speeds for depths of 1.5, 3.0, 4.6, 6.1 and 9.1 m.
Actual boat speed was estimated through timed General Oceanic current
meter readings for the top speed only. All measurements were made on the
same day near the Ludington plant in Lake Michigan, just north of the
north jetty.

An average of the measurements taken at the 3 through 9 m depths
was used in estimating reservoir populations. Trawl paths around the
reservoir perimeter were usually within this depth range. An estimate of
the bottom area covered was then calculated using the average trawl
width and the average boat speed. These data were used in estimating

population densities.

Statistical Methods

Reservoir trawl data were compared across years using the Friedman
two—way layout non—parametric statistical test (Hollander and Wolfe 1973).
Monthly average catch per unit of effort (CPE) data were used in the comp—
utations. When significant differences were observed, a multiple
comparison test based on the Friedman rank sum (Hollander and Wolfe 1973)
was used to determine where yearly differences were significant. An

a—level of 0.05 or less was considered significant.

15

The Friedman test was also used to analyze distributional patterns
of the demersal species in the reservoir. Average catch per unit effort
data were blocked by year and the north (stations T5 and T6), central
(stations T1 and T4) and south (stations T2 and T3) sections of the
reservoir were compared. Again, an a—value of 0.05 or less was considered
significant.

The Kendall test for independence (Hollander and Wolfe 1973) was
used as a measure of correlation between catch and temperature in the
reservoir. Catch per unit effort of a given species on sample dates when
all six stations were sampled was compared with bottom temperature.
Analysis was performed on data from the year of greatest catch (1976)
and young—of—the—year fish were excluded. This prevents complications
that could arise due to annual population fluctuations and varying
spawning success.

It was suspected that fish numbers in the reservoir may fluctuate
annually in response to invertebrate abundances. Again the Kendall test
was used. Monthly catch per unit effort trawling data were compared
with total invertebrate density estimates derived from Ponar grab sample
data presented by Duffy and Liston (1978). Total invertebrates included
oligochaeta, chironomidae, amphipoda, hydracarina, isopoda and others of
minor importance. Data used are presented in Table 2.

The Kendall test was also used to compare 1977 Lake Michigan trawl

data with reservoir trawl data of the same year.

 

l6

 

 

 

 

 

Table 2. Estimated monthly average densities of total invertebrates
as determined by Ponar grab samples in the Ludington reservoir
from 1974 through 1977 (adapted from Duffy and Liston 1978).
Total Grand
Invertebrates Number of Monthly Average Annual Average
(no./m2) Samples (no./m2) (no./m2)
1974 12 ' 3,195
April 38,345 12 3,195
May 5,464 12 455 1,816
June 31,623 12 2,635
August 11,723 12 977
1975
May 64,995 11 5,909
July 86,547 12 7,212 6,305
September 85,452 12 7,121
October 59,344 12 4,945
1976
April 44,784 9 4,976
August 69,671 9 7,741 7,175
October 79,264 9 8,807
1977
April 34,017 8 4,252
July 127,202 9 14,134 7,715
November 39,374 9 4,375

 

 

17

Results and Discussion

Trawl Measurements

Measurement of trawl widths did not yield easily interpretable
results. As shown in Table 3, there was not a regular progression of
trawl widths from shallow to deep water. These observations are similar
to those reported by Wathne (1959) concerning larger trawling gear. In
that work, DeBoer (1957) was cited to have found that trawl door spread
is largely determined by the degree of contact between the doors and the
bottom. Wathne (1959) suggests that in shallow water the line angle is
such that the otter—boards are pulled up, thus there is poor bottom
contact and small door spread. As bottom contact increases, the trawl
opening increases. At the 1.5 m depth in Lake Michigan, very good contact
between the boards and the bottom was apparent and may explain the rela—
tively wide trawl opening. The measurement at the shallow depth (1.5 m)
was not used in population estimates as reservoir trawls were always in
deeper water. It is believed that the average of the other measurements
is a reasonable first estimate of trawl width. More study of the gear
used and sites sampled at the Ludington project is recommended to

determine the causes of variations.

Catch Per Effort, Population Estimations and Spatial Distribution
Sculpins. Sculpin numbers in the reservoir increased progressively
from 1974 through 1976, and declined slightly in 1977 (Figure 4). Catch
in 1976 was determined to be greater than that in 1974 (a = .026), but
other annual differences were not significant.
Numbers caught were consistently lowest in spring for each year.

This may be caused by movement to the lake in the fall with return

 

18

Table 3. Trawl widths and estimated tow speed at several depths in
Lake Michigan.

 

 

Depth Trawl Width Estimated Speed
(m) (m) (km/hr)
1 5 6.6 5 6
3 0 5.0 6 O
4 6 6.2 5 l
6 l 5.3 3 O

 

 

= SPRING

Q =$UMMER '

‘- =FALL

  

I977

   

 

per unit effort of sculpins,

gig/gig A
v40{94049.549404040494040404O<94Q104649464O494.40404040404949404040 9401...

|976

catch

 

 

 

 

5 m1

7 m

.. n3 r.

.u. I r

.. .m

. m

4 1....

7 1..

E m.

O O O a.
3 2 I O a

Hmoumm .523 mma IUH<U

Figure 4.

3974 through 1977.

 

2O

migrations in early summer. A second possibility is that sculpins burrow
in the bottom mud or under rocks and become inactive in cold water, thus
escaping the trawl. This is substantiated somewhat by the Kendall test
comparing temperature and catch. Catch of sculpins was somewhat dependent
on temperature (a = 0.08), thus as temperature increased catch also tended
to increase. This could be due to increased activity within the reservoir
and increased immigration from the lake leading to a greater vulnerability
to capture.

Frequency of occurrence of sculpins in trawl samples increased
throughout the years of the study. For 1974, sculpins were caught in
only 28% of the samples. This increased to 64% in 1975, 88% in 1976 and i
95% in 1977. Increasing occurrence and numbers of sculpins substantiates
the theory of a resident population buildup in the reservoir, but the
buildup could not be attributed to increases in invertebrate abundance as
sculpin numbers were not dependent on that prey (Kendall test, a = 0.44).
Sculpins feed primarily on amphipods and juvenile fish. Amphipods are
not captured in Ponar grab samples as easily as more sedentary invertebrates
and thus may have been underestimated in the analysis. Data on juvenile
fish abundances were not available.

An estimate of the maximum number of sculpins present in the reservoir
for any month was made for August 1976, and the estimated number present
at that time was 143,458 (f 108,327, 90% C.I.). This translated to 4.5
sculpins/100m2 (f 3.4). The high catch per effort in August 1976 may be
due to the high water temperature of 200 C at the time of sampling. In—

creased activity of a large population probably caused the peak.

 

21

No significant pattern of spatial distribution was observed for
sculpins. This may be due to the even distribution of amphipods and isopods,
major prey of the sculpin, through the reservoir (Duffy and Liston 1978).
Sculpins did tend to occur in greater numbers in the north end in 1976 and
1977 (Table 4), but this observation was not statistically significant.

Johnny Darters. The johnny darter population in the Ludington reser-
voir has fluctuated (Figure 5). Significant differences were observed
between years 1975 and 1977 (a = .02). This is so even though peak catch
occurred in summer 1976, and is apparently due to the method of comparison
used. Non-parametric methods do not take into account the magnitude of
differences as much as the consistency of the direction of those differences.

Except for in 1974, the spring catch was always the lowest, otherwise
no consistent seasonal trends were apparent. Johnny darter catch was even
more dependent on temperature than sculpin catch (a = .002). As with the
sculpins, increased activity may have contributed to increased catch per
effort.

Johnny darters were relatively infrequent in trawl samples from 1974
and 1975 (33% and 36% frequency of occurence, respectively) but occurred
more than twice as often in samples taken during 1976 and 1977 (74% and
73%, respectively). Darter numbers were quite dependent on total inverte—
brate abundance in the reservoir (a = .021). Invertebrate densities have
generally increased over the study years (Table 2) and may explain the
apparent increase in johnny darter population density.

The highest catch of johnny darters occurred during August 1976,
when the mean catch per unit effort was 98.8. The estimated number present
in the reservoir at that time was 141,383 (f 79,993, 90% C.I.), or 4.4

darters/100m2 (f 2.5). The August 1976 peak may be due to the relatively

 

22

 

 

Table 4. Comparison of average catch per unit of effort of sculpins
in the north (T5, T6), central (T1, T4) and south (T2, T3)
sections of the Ludington reservoir.

Year North Central South

1974 1.2 0.9 1.4

1975 3.8 2.1 3.7

1976 23.7 22.9 7.3

1977 19.2 16.0 4.6

 

 

 

23

SPWNG

[XI

SUMMER

.=FALL

 

|977

 

 

 

 

 

 

 

Z??? 6
v........... .9... 9.99.90.10.00“ .10“ ....W
5
7
. B
4
7
% OJ
1 _ _ . _ _ l
O O O O
3 2 l

Hmouum .H_ZD mun. Iquo

Seasonal reservoir trawl catch per unit effort of johnny

Figure 5.

1974 through 1977.

darters,

 

24

warm water (20 C) on the sample date.

Johnny darters were distributed fairly evenly between the north,
central and south sections of the reservoir. In 1974, the south end held
substantially more darters than the rest of the reservoir, but no signifi-
cant trends were apparent in the following years (Table 5). Friedman
analysis did not show any significant differences. Duffy and Liston (1978)
found that chironomids tended to be highest in the north and central
stations in the reservoir. It had been expected that the darter distri-
bution would correlate with that of the chironomid larvae, its major prey.
While the differences were not significant, johnny darters occurred in
slightly higher numbers in the north end in 1976 and 1977 (Table 5).

Trout—perch. The trout-perch population in the Ludington reservoir
increased from 1974 through 1976 and declined slightly in 1977 (Figure 6).
The catch in 1976 was significantly greater than that in 1974 (a = 0.026).
Increasing spring catches suggest that trout—perch may overwinter in the
reservoir rather than migrate to the lake. As with both sculpins and
johnny darters, trout-perch catch was dependent on temperature (Kendall
test, a = .03).

Trout—perch occurred in only 17% of the trawl samples in 1975. They
were caught in over half the samples in 1975 (57%) and occurred in 86% of
the trawls in both 1976 and 1977. Trout-perch catch per effort was highly
dependent on total invertebrate density (a = .001). Thus annual increases
in invertebrate densities (Table 2) may explain increases in the trout-
perch population.

The highest catch of trout—perch occurred in July 1976. The mean
catch in that month was 44.7 and the estimated number present in the

reservoir was 63,966 (f 40,640, 90% C.I.), or 2.0 trout—perch/lOOm2

 

25

 

 

Table 5. Comparison of average catch per unit effort of johnny darters
in the north (T5, T6), central (T1, T4) and south (T2, T3)
sections of the Ludington reservoir.

Year North Central South

1974 6 1 2.1 22 7

1975 1.0 0.9 1.0

1976 27.1 22.2 22.0

1977 15.4 11.7 7.6

 

26

I: SPRING
Q = SUMMER

= FALL

 

 

 

 

     

 

 

 

 

 

 

 

30 ‘
*—
a:
E 20 —
U- "1
L” ésiséss
#—
z
D
o:
w
n.
I —<
U IO
'—
<[
o
O m {—— _ :_

|974 |975 l976 | 77

Figure 6. Seasonal reservoir trawl catch per unit effort of trout—perch,
1974 through 1977.

 

27

(f 1.3). This peak in the catch per effort coincides with peak water
temperatures for 1976.

Trout-perch were consistently and significantly more abundant in the
south end of the reservoir (Table 6). No consistent patterns in prey
distribution can explain the spatial distribution of trout—perch.
Oligochaetes were most abundant in the south in 1976 (Duffy and Liston

1978), but no patterns were apparent in other years.

Reservoir — Lake Michigan Comparisons

 

Comparisons were made by Gulvas (1976) between species composition
at the Ludington plant and several other power plant sites around Lake
Michigan. His comparisons, however, combined all collection methods used
making conclusions difficult. Adequate trawl data from Lake Michigan in
the Ludington area were not available at that time so that appropriate
comparisons with the reservoir also were not possible.

Beginning in 1977, regular bottom trawling was accomplished in Lake
Michigan in the plant vicinity so that direct comparisons could be made
with the reservoir. A report by Jude et a1. (1975) provides adequate
data for simple comparisons between a sampling area in southern Lake
Michigan and samples taken from the Ludington reservoir and the near
vicinity in Lake Michigan.

Trawl samples from Lake Michigan near the Ludington plant were
qualitatively similar to samples taken in the reservoir. A complete
description of the samples is presented by Brazo and Liston (1979). A
comparison of the 1977 monthly catch per unit of effort of sculpins,

johnny darters and trout—perch is presented in Table 7.

28

 

 

Table 6. Comparison of average catch per unit of effort of trout-perch
in the north (T5, T6), central (T1, T4) and south (T2, T3)
sections of the Ludington reservoir.

Year North Central South

1974 0.1 0.2 0.3

1975 2.1 5.0 7.9

1976 8.8 16.5 40.6

1977 9.8 11.0 26.4

 

 

 

Table 7-

Comparisons of 1977 average monthly catch per unit of effort

between trawls in the Ludington reservoir and in nearby

Lake Michigan.

 

 

 

Month Sculpin Johnny darter Trout—perch
Reservoir Lake Reservoir Lake Reservoir Lake
April 6.0 1.2 0.2 0.0 7.7 0.0
May 23.2 2.5 2.2 8.8 32.4 4.5
June 15.0 1.8 7.1 19.0 23.0 11.8
July 12.7 3.8 13.2 18.8 10.1 0.7
August 19.3 16.7 25.0 33.0 6.2 2.5
September 30.8 26.5 38.2 2.0 9.0 11.2

 

 

30

Sculpin numbers were consistently higher in the reservoir than in the
lake. It was determined through the Kendall test for independence
(Hollander and Wolfe 1973) that the lake and reservoir numbers were some!
what positively correlated (T = .60). The latter observation suggests
that migrations between the lake and reservoir are quite high and there
may not be distinct populations present in each environment. However,
sculpins may simply move to deeper unsampled water both in the reservoir
and lake in response to environmental changes (e.g. temperature) that are
common to both areas. Sculpins are known to leave the nearshore area in
late fall and presumably move to deep water (M.S.U. Ludington Research

Lab, unpublished data). A distinct reservoir population is therefore

 

still possible.

Johnny darters were generally more abundant in Lake Michigan trawls
(Table 7). Reservoir and lake catches were determined to be independent,
that is, no significant correlation was noted ( T: .33). This may be
considered as evidence of separate populations.

Trout—perch were consistently and substantially more abundant in
reservoir trawl catches compared to samples taken in nearby Lake Michigan
(Table 7). Reservoir and lake catches were not correlated ('r= .33) and,
again, this may be evidence for separate populations.

Seasonal fluctuations of the demersal fish if not due to migrations
between the reservoir and Lake Michigan, may be caused by movement to
deeper water within the reservoir and varying activity rates due to
temperature.

A simple demersal community comparison is presented in Table 8. The
relative numbers of sculpins, johnny darters and trout—perch were calcu—
lated using data collected by Jude et a1. (1975) near Warren Dunes State

Park in southern Lake Michigan and data collected near Ludington and in

 

31

Table 8- A comparison of relative numbers of sculpins, johnny darters
and trout—perch in trawl samples from the Ludington reservoir
and two areas of Lake Michigan.

 

 

 

. 1 Lake Michigan1 Lake Michigan2
Reservo1r .
Lud1ngton Warren Dunes
Sculpin 37.7% 32.3% 1.3%
Johnny darter 28.5% 49.2% 3.4%
Trout-perch 33.7% 18.5% 95.3%

 

11977 data

2Data adapted from Jude et a1. 1975

 

 

32

the Ludington reservoir. Overall qualitative species composition was
similar at all three sites. There were strong contrasts, however, in
relative numbers of the three major demersal species caught in the three-
areas. Sculpins and johnny darters were relatively unimportant in trawl
catches at Warren Dunes. The demersal community there was dominated by
trout—perch, while trout-perch were much less important in samples from
either the Ludington reservoir or nearby Lake Michigan. Relative abundances
of all three demersal species were nearly equal in the reservoir in 1977,
but johnny darters were relatively more abundant in the lake.

The causes of the differences between the northern and southern

Lake Michigan sites are not easily discerned. Trawling gear and methodology

 

were quite similar. The Ludington samples were taken four years after
those at Warren Dunes but yearly fluctuations are not likely to explain
the substantial difference. Temperature regimes were substantially
different at the two sites. In 1973 the area near Warren Dunes warmed
earlied than at Ludington, attained a higher maximum, and maintained a

2 to 40C higher temperature through the sampling period (Jude et a1. 1975;
Liston et a1. 1976). Such temperature differences probably occur annually
and may be the major cause of community differences. Higher temperatures
along with possible differences in bottom types may affect food types and

availability leading to the differences in community structure.

FOOD HABITS

The food habits of sculpins, johnny darters, and trout—perch
captured in the Ludington reservoir during 1976 were analyzed. Burbot
and salmonid food habits were also studied for comparative purposes since

they were believed to be important as predators in the demersal community.

Methods

Stomachs were removed after the fish had been preserved by cutting
at the pyloric sphincter and at the beginning of the esophagus. When
necessary, stomach contents were observed under a binocular dissecting
scope. Food items were counted and when possible, volumes were taken by
water displacement using an automatic burette and a graduated centrifuge
tube. Small items were measured with a 150 mm rule and volumes were
estimated through use of formulas for similarly shaped geometric objects.
Formulas used are presented in Table 9 and are similar to those used by
Kenaga (1975).

Freshwater fish are basically opportunists and are not likely to
select food on the basis of taxonomic groups. More likely prey are chosen
by size or location. Therefore, food items were grouped into what is
hoped are appropriate categories for this analysis. Table 10 presents
a list of the components of the major food groups as used in this study.

Frequency of occurrence, percent of total numbers,and percent of
total volume were calculated for each food item for each fish species.

No one of these figures is sufficient to represent the importance of a
particular food type to a population of fish. It is also difficult to
compare two populations using any one of these figures alone. Percent

of total volume is probably the most important statistic, as it approximates

33

 

 

 

Table 9. Calculations used in estimating volumes of certain organisms
in food habit studies. Adapted froanenaga (1975).

 

 

 

Equation Organism

W'R2°L Chironomid larvae and pupae
Oligochaeta

L'W-D Amphipodal
Isopoda2
Mysis

l/6W°L°W'D Cladocera
Ostracoda

1/2{H°R2°L + l/3(h'R2'L)} Copepoda

Where: R = radius

L = body length
W = body width

D = body depth

 

1D = lateral depth

2D = dorso—ventral depth

 

 

35

TablelO. Components of the major food groups in the Ludington reservoir
based on stomach analysis of sculpins, johnny darters, trout-
perch and burbot.

 

 

Major Food Group Components

Benthic Microinvertebrates Cladocerans (Chydoridae),
harpacticoid copepods, nematodes,
ostracods

Benthic Macroinvertebrates Immature insect forms (Chironomidae,

Trichoptera, Ephemeroptera, Coleop—
tera), Oligochaetes, amphipods,
isopods, gastropods, Hydracarina

 

Larval Fish Larval, post—larval, and young of the
year fish, primarily smelt and alewives.

Fish Juvenile and adult fish, primarily
forage fish such as smelt, alewives,
sculpins and darters. Also ninespine
stickleback and spottail shiners

Fish Eggs Unidentified to species

 

 

 

36

the caloric value of a particular food item to the population. Used
alone, however, it is subject to bias. Large prey may have been taken
by only a few fish, and thus its importance may be over—emphasized.
Conversely, small prey may be taken consistently, but due to low overall
volume will be judged unimportant if only percent of total volume is con-
sidered. Frequency of occurrence is an important statistic but again it
cannot be used alone, and since values do not add to unity, certain com-
parisons (e.g. overlap values) are impossible. A more appropriate
statistic is necessary for comparative purposes.

A new statistic will be used in this study. The product of the
actual volume and the frequency of occurrence for each food item, herein
called the relative value (RV), allows easier comparisons. The relative
value is obtained for each food item, all relative values are added, and
a percent of total relative value is calculated for each food item. The
method tends to moderate extreme values. If, for example, a food item
occurs in few stomachs the percent relative value would be less than the
percent total volume, even though the item may account for a large per—
centage of the total volume. The calculation is an attempt to determine
the actual value of a food item to a population of fish.

Analysis was limited to the summer season (16 June through 31 August)
since nearly all specimens were caught in this period.

Most fish (sculpins, johnny darters and trout-perch) were between
51 and 100 mm in total length (TL) and analysis was limited to this size
range.

Certain morphological traits can limit the prey types taken by a
fish. In comparing the morphology of sculpins, trout—perch and johnny

darters, mouth size is perhaps the most important characteristic. The

 

 

37

width of the mouth aperture was measured across the open mouth at the jaw
angles to the nearest 0.5 mm using a divider and rule. The gape width
was then expressed as a percentage of total body length.

A resource overlap equation was used to measure the similarity of
diets of the three species. The formula, adapted from Colwell and
Futuyama (1971) and Sale (1974) is as follows:

Cih= 1 -1/2}:Ipij - phjl

where Ci is the resource overlap between species i and species h and

h

pij and phj are the proportions of the food resource j used by species
1 and species h, respectively. Overlap values may range from 0, indicating

completely different food habits, to 1, indicating identical food habits.

Results and Discussion

Sculpin Food Habits. The results of the analysis of stomachs from
sculpins between 51 and 100 mm TL are presented in Table 11. Of the 50
stomachs examined, all but one contained food. Sculpins fed primarily on
benthic macroinvertebrates and larval fish. Amphipods and Chironomid
larvae were the major components in the macroinvertebrate group. Larval
fish, mainly smelt and alewives, were eaten less often (lower frequency
of occurrence). Due to the greater volume, however, this group was
probably more important to the sculpin population overall (greater
relative value). Benthic microinvertebrates occurred in about one—fourth
of the stomachs examined, but due to their low volume were not very
important in the sculpin diet.

Other studies of sculpin food habits display similar findings.
Roster (1937) found that immature insects were of major importance to

SCulpins in New York streams. Dineen (1951) also noted the importance of

 

 

38

Table 11. Percent of total volume (% V), percent of total numbers
(% N), percent frequency of occurrence (% F0) and percent
relative value (% RV) of food types eaten by sculpins
(51—100 mm TL) in the Ludington reservoir in 1976.

 

 

% V % N % F0 % RV

Benthic Microinvertebrates 0.6 14.7 26.5 0.3
Benthic Macroinvertebrates 24.0 .71.1 91.8 43.1
Larval Fish 75.2 13.1 38.8 56.6
Fish 0 0 0 0
Fish Eggs trace 1.0 2.0 trace

Number of stomachs 49

Total volume (cc) 11.257

Total number of 505

food items

 

 

 

 

39

insects in the sculpin diet. All reports reviewed had found some level

of predation on fishes by sculpins (Koster 1937; Dineen 1951; Bailey 1952;
Patten 1962; Phillips and Claire 1966; Clary 1972), but no other population
was as piscivorous as the sculpins in the Ludington reservoir. This is
most likely due to environmental differences. The previous studies con-
centrated on stream populations where small fish are probably not as
available. The Ludington reservoir, being essentially a part of Lake
Michigan, likely has a larger source of young—of—the~year fish.

Use of benthic microinvertebrates presents another interesting
contrast. Microcrustaceans were reported to be taken only rarely in one
study (Koster 1937) and none of the other studies observed them. Over
25% of the sculpins in the Ludington reservoir had taken microinvertebrates.
Again this difference is due to differences between a stream environment
and what is essentially an extension of Lake Michigan.

Johnny Darter Food Habits. A total of 56 stomachs from johnny
darters between 51 and 100 mm TL were examined and all but 2 contained
food. Results of the analysis are presented in Table 12. Johnny darters
relied heavily on benthic macroinvertebrates, with Chironomid larvae
being the major component. Microinvertebrates occurred frequently and
in relatively large numbers, but due to low volume did not constitute
a major part of the diet.

In a study of johnny darters from nearby Lake Michigan, Bringold
(1978) also found benthic macroinvertebrates to be the most significant
part of the diet. Again, Chironomid larvae were the most important prey.
Other prey occurred in proportions similar to those in the reservoir,
although amphipods, ostracods, nematodes and eggs were taken more often

in the lake.

 

40

Table 12- Percent of total volume (% V), percent of total numbers
(% N), percent frequency of occurrence (% F0) and percent
relative value (% RV) of food types eaten by johnny darters
(51—100 mm TL) in the Ludington reservoir in 1976.

 

 

zv %N 7. F0 %RV
Benthic Microinvertebrates 3.8 29.5 68.5 2.7 ‘
Benthic Macroinvertebrates 96.2 70.3 96.3 97.3
-1

Larval Fish 0 0 0 0
Fish 0 0 0 0
Fish Eggs trace 0.2 1.9 trace

Number of stomachs 54

Total volume (mma) 452.54

Total number of 532

food items

 

 

41

All other studies reviewed reported diets similar to the darters
inhabiting the reservoir (Turner 1921; Karr 1963; Lotrich 1973) in that
johnny darters feed mainly on benthic macroinvertebrates and tend to

specialize in preying on chironomid larvae.

Trout—perch Food Habits. The food habits of the trout—perch in the

 

Ludington reservoir were analyzed by O'Neal (1977). Trout-perch relied
mainly on benthic macroinvertebrates (Table 13). Although larval fish
accounted for the greatest prey volume, they were taken only rarely, and
thus were not as important overall. Microinvertebrates were taken often,
but were not overly important in the diet because of low volume.

Relatively few published reports of trout—perch food habits are
available. Anderson and Smith (1971) described the diet of Lake Superior
trout—perch as consisting mainly of benthic macroinvertebrates such as
amphipods, Chironomid larvae, and MyAiA sp. Entomostracans were also
eaten, but no fish were reported in the diet. Other studies on the ecology

of trout—perch in the Great Lakes include Kinney (1950) and Bostock (1967).

Comparative Food Habits. A comparison of relative values of the
major food types taken by sculpins, trout—perch and johnny darters is
presented in Table 14. Note that the sums of individual items within a
major group do not necessarily equal the percent relative value of the
major group. This is due to characteristics of the frequency of occurrence
statistic which is used in calculating the relative value. Frequency of
occurrence values are not necessarily additive, thus differences may occur
when groups are combined;

.Sculpins had the most distinct food habits, relying mainly on larval

fish with benthic macroinvertebrates (primarily amphipods) making up nearly

 

Table 13. Percent of total volume (% V), percent of total numbers
(% N), percent frequency of occurrence (% F0) and percent
relative value (% RV) of food types eaten by trout—perch
in the Ludington reservoir in 1976 (adapted from O'Neal 1977).

 

 

% V % N % F0 7 RV

Benthic Microinvertebrates 2.4 47.3 61.7 4.2
Benthic Macroinvertebrates 35.9 51.6 86.7 87.2
Larval Fish 61.7 1.2 5.0 8.6
Fish 0 0 0 0
Fish Eggs 0 0 0 0

Number of stomachs 60

Total volume (cc) 2.271

Total number of 846

food items

 

 

 

 

43

Table 14. Percent relative value of food items and major food types
found in sculpins, johnny darters and trout—perch between
51 and 100 mm TL captured in the Ludington reservoir during
summer 1976.

 

 

 

Scul in Johnny Trout—
p darter perch
Benthic
Microinvertebrates 0.3 2.7 4.2
Cladocera 0.4 2.8 4.8
Copepoda 0 trace ' trace
Nematoda 0 trace 0
Ostracoda 0 . 0 0
Benthic
Macroinvertebrates 43.1 97.3 87.2
Chironomid
larvae 4.2 93.0 70.6
Chironomid
pupae 1.9 3.6 3.6
Amphipoda 23.6 trace 9.5
Isopoda 1.1 trace trace
Oligochaeta 0.1 0.5 0.3
Mysis trace trace trace
Other 0 trace 0
Larval Fish 56.6 0 8.6

Fish Eggs trace trace 0

 

 

44

all of the remaining diet. Johnny darters fed almost exclusively on
benthic macroinvertebrates, and chironomid larvae were the most important
component of this group. The trout—perch diet was quite similar to that'
of the darters, although larval fish were taken and benthic microinverte-
brates were somewhat more important.

Competition in the Ludington demersal fish community cannot be
properly evaluated by the use of food overlap values. Colwell and
Futuyama (1971) stated that any overlap value can be evidence either for
or against the existence of competition between species. If food overlap
is high, competition may be present, but due to the availability of only
limited food types niche displacement is limited. Conversely, high
overlap may suggest limited competition in the presence of an oversupply
of food. Low overlap could indicate competition for a limited resource
with niche displacement taking place. On the other hand, low values may
simply be the manifestation of morphological and behavioral differences
in the absence of actual competition. Although actual competition therefore
cannot be evaluated, overlap values are a means of quantifying the
similarities in diets.

Food overlap values for sculpins, johnny darters and trout—perch,
calculated using percent relative values, are presented in Table 15.

The greatest overlap occurred between johnny darters and trout—perch.
Sculpin food habits overlapped relatively little with either trout—perch
or johnny darters. While Table 15 is based on the relative values of the
major food groups, overlap was also calculated using the relative values
of the individual food items (i.e. taxonomic groups). Again, trout—perch/
darter overlap was highest (0.77) and sculpin/trout—perch was higher than

sculpin/darter (0.27 and 0.07, respectively). The sculpin values are much

 

 

 

45

Table 15. Resource overlap values calculated using relative values of
major food groups for specimens between 51 and 100 mm TL
taken in summer 1976.

 

 

Sculpin Trout—perch Johnny darter
Sculpin l 0.52 0.43
Trout—perch l 0.90

Johnny darter 1

 

 

46

lower than those presented in Table 15. This points out that although
sculpins fed to an extent on food types common to the other species
(benthic macroinvertebrates), the primary components of those food groups
differ substantially.

The differences in the diets of sculpins, johnny darters and trout-
perch (Table 14) may be due in part to morphological differences between
the species. Table 16 presents a comparison of the relative gape width of
these three demersal forage fish. Sculpins possess a much wider mouth than
either the trout—perch or johnny darter, allowing capture of larger prey
by the former. This would help explain the relatively high importance
of fish in the diet of sculpins, while piscivory was rare in trout-perch
and not observed at all in darters. In addition, amphipods, a fairly large
prey, were important to sculpins, but were taken only rarely by either
trout—perch or johnny darters. Johnny darters and trout—perch possess
similar size mouths (Table 16) and have noticeably similar feeding habits
(Table 14).

Mouth size may affect the types of foods eaten, and the size of
foods common to all the predators may differ also. Table 17 presents the
average volumes of the major foods eaten by sculpins, johnny darters and
trout—perch. Amphipods and isopods taken by sculpins were much larger
than those eaten by johnny darters or trout—perch. Chironomid larvae
taken by darters averaged half the volume of those taken by either
sculpins or trout—perch, but pupae taken by sculpins and darters averaged
nearly the same size. Lack of variation in pupae size may explain this
observation. Sculpins preyed upon much larger Cladocerans than either of

the other predators, but results with the other invertebrate prey are

 

 

47

 

 

Table 16. Relative mouth sizes of demersal fish inhabiting the
Ludington reservoir. Ten fish of each species were
measured.

, Range of Relative Gape Width

Spec1es Lengths (7 of T t 1 L th)

(mm TL) . o a eng

Sculpin 36—85 11.4

Johnny

darter 34—66 5.3

Trout—perch 62—139

 

 

48

Table 17- Average volume (mma) of the major prey common to sculpins,

johnny darters and trout-perch taken in the Ludington

reservoir in 1976.

 

 

€222; 232:2;

Chironomid
larvae 2.82 1.11 2.36

Chironomid
pupae 2.67 2.05 1.35
Amphipoda 22.57 3.14 1.98
Isopoda 9.93 0.05 2.20
Oligochaeta 0.76 1.11 0.15
Cladocera 0.93 0.11 0.15
Copepoda — 0.05 0.05
Larval Fish 128.22 — 140.00

 

 

49

inconclusive. Although the relationship is not clearcut, mouth size is
quite likely a major factor in determining food habits.

Behavioral traits can determine to an extent the types of food eaten.
The three species under discussion tend to display different predatory
behavior. The sculpin is mainly an ambush predator (Phillips and Claire
1966), lunging after its prey through the use of oversized pectoral fins.
Cottids rely primarily on viSual stimuli although the use of olfaction
has been suggested (Clary 1972). The johnny darter relies mainly on
visual stimuli (Daiber 1956; Roberts and Winn 1962; Daugherty et a1. 1976;
personal observation) and tends to stalk its prey (personal observation).
Darters are not adept swimmers and would probably be unable to capture
rapid prey such as amphipods or young fish even if the mouth size were
large enough. Information on trout—perch feeding behavior is lacking,
but judging from general morphology (presence of swim-bladder, lack of
oversize pectoral fins), trout—perch probably feed in a different manner
than sculpins or johnny darters. While sculpins and darters lack swim—
bladders and tend to feed along the bottom, trout—perch probably are more
likely to swim up in the water column in search of prey.

Food is probably not limiting in the Ludington reservoir due to high
organic inputs from Lake Michigan (M.S.U. Ludington Research Lab, unpublished
data). If food is not limiting, competition is not an important force
and feeding habits are likely an expression of preference caused by
morphological and behavioral characteristics of the predator. The sculpin,
for example, being basically an ambush predator (Phillips and Claire 1966)
is not likely to encounter relatively sedentary prey. The sculpin is also

able to feed on a wider size range of prey since it possesses a relatively

 

 

 

50

wide mouth. These types of factors probably have greater effect on the
food habits of these species than does the availability of food in the

Ludington reservoir.

 

 

 

 

REGULATING FACTORS

A fish population may be regulated by competition for food and
space, predation, parasitism and disease, available spawning habitat, and
by general water quality. As discussed earlier, competition is difficult

to quantify and was not attempted in this study.

Predation

The primary predator in the Ludington reservoir demersal community
appears to be the burbot. Fish made up nearly all of the food ration of
burbot captured in the reservoir in 1976 (Table 18). Alewives were the
most important prey by volume and relative value, but sculpins and stickle—
backs were also taken frequently.

Johnny darters were not taken very frequently but it is probable that
predation will increase if an observed decline in area alewife numbers
(Brazo and Liston 1979; M.S.U. Ludington Research Lab, unpublished data)
continues. Although trout-perch were not observed in any burbot stomachs
in 1976, Van Oosten and Deason (1938) reported extensive use of this prey
by burbot in Lake Michigan. The piscivorous habit of burbot is supported
by numerous reports including McCrimmon and Devitt (1954), Beeton (1956),
Bonde and Maloney (1960) and Muth and Smith (1974).

Burbot numbers were notably high in the reservoir in 1976, such that
predatory pressure may have been somewhat exaggerated in that year. This
predatory pressure may explain in part the observed decline in the popula-
tion of demersal forage fish in 1977.

Several other species are potential predators on the sculpins,
johnny darters and trout—perch in the Ludington reservoir. Sculpins may
act as major population regulators in the community by taking larval

and juvenile fish.

51

 

 

52

Table 18. Percent of total volume (% V), percent of total numbers (% N),
percent frequency of occurrence (% F0) and percent relative
value (% RV) of food types eaten by burbot in the Ludington
reservoir in 1976.

 

 

 

Z V % N % F0 % RV

Benthic Microinvertebrates 0 0 O 0
Benthic Macroinvertebrates 4.1 76.5 56.0. 2.5
Larval Fish 0 0 _0 0
Fish 95.2 15.7 72.0 97.5

Sculpin 14.0 I 5.2 16.0 9.4

Johnny darter 2.4 1.3 4.0 0.4

Trout—perch 0 0 0 0

Stickleback 4.0 2.0 16.0 2.7

Alewife 37.9 2.0 12.0 20.0

Unidentified fish 36.7 4.6 40.0 61.2
Fish Eggs 0.6 7.8 4.0 0

Number of stomachs 25

Total volume (cc) 70.632

Total number of 153

food items

 

 

 

 

53

Steelhead and brown trout, and coho and chinook salmon are seasonally
abundant in the reservoir (Gulvas 1976; Brazo and Liston 1979) but predation
is suspected to be low. Of 83 of these salmonids examined and found to
contain food, only four (5%) had eaten sculpins, and one (1%) had taken a
johnny darter. Primary foods were alewives and smelt. These salmonids are
probably mainly pelagic in habit and do not often encounter demersal fish.

Lake trout have been found to feed on sculpins in the Ludington area
of Lake Michigan, although alewives were more important in the diet
(Chiotti 1973). Sculpins were most important to the lake trout in the
early part of the year and were rarely taken after August. Since lake
trout are not common in the reservoir until the onset of spawning in the
fall, it is not likely they are important predators in the community.

Round whitefish were reported to feed on fish eggs by Armstrong
(1973), but it is doubtful this would have much effect on the reservoir
populations of sculpins, johnny darters or trout-perch. Both sculpins and
darters protect the eggs to some extent (Savage 1963; Winn 1958; personal
observation). More importantly, round whitefish are not abundant in the
nearshore area or in the reservoir until well after the spawning periods

of these species.

Parasitism
Although intensive work has not been done on the parasites infecting
the reservoir community, preliminary observations suggest the presence of
only relatively low numbers. Sculpins from the Ludington reservoir were
infected fairly heavily with acanthocephalans in the gastrointestinal
tract, but other parasites were rarely noted. Parasites were not observed

in stomachs of either the trout—perch (R. P. O'Neal, personal communication)

 

 

 

 

54

or johnny darters. The regular occurrence of acanthocephalans in sculpins
presents an interesting contrast. Acanthocephalans utilize crustaceans
such as amphipods, isopods or ostracods as intermediate hosts (Barnes 1974).
Perhaps the fact that iatpods and especially amphipods are quite important

in the sculpin diet explains the rate of infestation.

Spawning Habitat

Young—of—the—year sculpins and johnny darters have been captured
fairly regularly in the Ludington reservoir, but it was unknown whether
successful spawning had occurred in the reservoir until observations made
in 1979. In late June 1979, the reservoir water level was drawn down past
a service road, allowing a walking survey around the perimeter. Two—meter
wide transects were made at regular intervals along the bottom from the
wall to the water's edge. Transects over the protective rock layer near
the reservoir intake structure revealed several sculpin and johnny darter
egg masses, most developed to the near hatching stage. Although data are
meager, estimates of number of sculpin egg masses ranged between 5 and 132
(90% confidence interval) in the rocky area. Data for johnny darters did
not allow estimation of egg mass numbers, but a total of seven spawning
sites were observed in one transect. In addition, several spawning sites
were observed on an exposed clay pile in the reservoir center. It is
believed, therefore, that some successful spawning occurs in the reservoir.

Trout—perch young—of—the—year are rare in both the Ludington reser—
voir and the power plant vicinity. In addition, few larval trout—perch
have been observed after recent extensive larval fish sampling effort
(Dan Brazo, personal communication). Trout—perch have been observed

moving upstream in spawning condition in Manitoba, Canada (Lawler 1954)

 

55

and it is suspected that similar use is made of streams in the Ludington
area. Only one young—of—the—year trout—perch was observed in reservoir
trawl catches over the study period, while young of both sculpins and

johnny darters have been captured quite frequently. It is unlikely that

significant trout—perch spawning occurs in the Ludington reservoir.

Water Quality

In an analysis of the sediments in the Ludington reservoir, Duffy
and Liston (1978) reported generally good conditions for benthic produc-
tion. They noted, however, that occasional samples emitted a sulfurous
odor characteristic of anaerobic decomposition. It is possible that a
continued influx of organic material from Lake Michigan would lead to
increasing anaerobic conditions near the reservoir bottom, causing re—
duction in demersal fish populations. Anaerobic conditions will probably
continue, but will likely be only localized due to the continual exchange
of large quantities of water with Lake Michigan. Future monitoring of

bottom conditions is advisable.

 

SUMMARY AND CONCLUSIONS

The objectives of this study were to estimate population size,
describe spatial distribution patterns and to investigate interactions
between species in the demersal fish community of the Ludington reservoir.
Trawling was the primary method of collection. The reservoir was sampled
from 1974 through 1977 and nearby Lake Michigan was sampled in 1977 for
comparative purposes.

Sculpin numbers increased in the reservoir from 1975 to a peak of
4.5 fish/100m2 in August 1976. The population declined slightly in 1977.
Johnny darters fluctuated annually, reached a maximum density of 4.4
fish/100m2 in August 1976, and declined slightly thereafter. Trout—perch
increased to a high of 2.0 fish/100m2 in July 1976, and declined in 1977.
The consistent decline of the three species remains unexplained, but may
be due in part to predation by an especially large burbot population in
1976.

Seasonal fluctuations can be explained in part by temperature variation.
The catch per effort for all three demersal species was dependent to some
extent on water temperature.

Annual fluctuations of johnny darter and trout-perch numbers were
found to be dependent on total invertebrate densities. Sculpin numbers
were not dependent, but results may have been biased somewhat by the
method of invertebrate sampling.

Trout—perch were found to have a significant spatial distribution
in the reservoir being much more abundant in the southern end. This
pattern is probably due to habitat preferences. Neither sculpins nor
johnny darters exhibited any significant spatial distribution patterns

within the reservoir.

 

 

 

57

Sculpins were found to be more abundant in the reservoir than in
nearby Lake Michigan. Overall 1977 reservoir catch per unit effort (CPE)
was 17.8 while CPE in the lake was 7.6. Monthly catch in the reservoir
was determined to be dependent of the Lake Michigan catch, thus, it is
possible that a high degree of migration occurs between the two environments.

Johnny darter CPE in Lake Michigan during 1977 was generally higher
than that in the reservoir, but not substantially so. Catch in the
reservoir was determined to be independent of that in nearby Lake Michigan,
suggesting that the populations may be independent.

Trout—perch were caught in much greater numbers in the reservoir
than in Lake Michigan. Overall 1977 reservoir CPE was 15.9, while the
lake CPE was only 4.4. Reservoir and lake catches were determined to be
statistically independent, thus, separate populations are possible. Since
no spawning has been observed within the reservoir, however, the meaning
of the statistical result is unclear.

Trawling data from Jude et a1. (1975) allowed general community
comparisons between Lake Michigan near Warren Dunes, Lake Michigan near
Ludington, and the Ludington reservoir. Trout—perch were very abundant
relative to sculpins and darters in the Warren Dunes area, making up over
95% of the catch of those species. Trout-perch were much less important in
samples from either the Ludington reservoir or nearby Lake Michigan in
1977. Johnny darters dominated the catch in Lake Michigan near Ludington
(49.2%) followed by sculpins (32.3%) and trout—perch (18.5%). All three
species were nearly equal in reservoir samples in 1977. Differences
between the northern (Ludington) and southern (Warren Dunes) sampling
sites in Lake Michigan are possibly due to the different temperature

regimes in the two areas.

 

58

Sculpins in the Ludington reservoir fed mainly on larval fish and
benthic macroinvertebrates (primarily amphipods). Macroinvertebrates were
the major food source for both johnny darters and trout—perch. Johnny
darters fed almost exclusively on Chironomid larvae and pupae while
trout—perch had a slightly more varied diet.

Predation on the demersal forage fish in the reservoir appears to
be quite low. The primary predator is the burbot but this species is not
overly abundant. Predation by large salmonids is probably only occasional
and minor.

None of the populations were parasitized to any significant degree.
No major die—offs were observed during the period of study.

Spawning habitat and water quality may act as reservoir population
regulators. Suitable spawning habitat for johnny darters and sculpins is
limited, and trout—perch are unlikely to spawn in the reservoir at all.
Water quality is generally good at present, but the potential exists for
localized anaerobic conditions which may induce mortality.

A certain amount of Lake Michigan production is lost through plant—
induced fish mortalities. Production is undoubtedly returned through
migrations of sculpins, darters, trout—perch and other fish which were
spawned or have fed in the reservoir. Due to the apparently low predation
on the demersal fish community, however, little reservoir production is

returned to the lake as salmonid food.

 

 

 

 

LIST OF REFERENCES

 

LITERATURE CITED

Anderson, E. D. and L. L. Smith, Jr. 1971. A synoptic study of food
habits of 30 fish species from western Lake Superior. Agric.
Expt. Sta. Univ. Minn. Techn. Bull. 279:199 pp.

Armstrong, J. W. 1973. Age, growth and food habits of the round white—
fish, Phobopium cyzindnaceum (Pallas), in central Michigan.
M. S. Thesis. Mich. State Univ. Dept. of Fish. & Wildl. 76 pp.

Bailey, J. E. 1952. Life history and ecology of the sculpin Cottua
baihdi punctulatub in southwestern Montana. Copeia 1952 (4):
243-55.

Barnes, R. D. 1974. Invertebrate zoology. W. B. Saunders Co. Phila—
delphia. 3rd Edition. 870 pp. ’

Beeton, A. M. 1956. Food habits of the burbot (Lota iota ZacuAIWXA)
in the White River, a Michigan trout stream. Copea 1958—60.

Bonde, T. and J. E. Maloney. 1960. Food habits of burbot. Trans. Am.
Fish Soc. 89(4):374-76.

Bostock, A. R. 1967. The ecology of the trout—perch, PedcopALA
omiscomagcua, in Lake Superior. M. A. Thesis, Univ. Michigan.

40 p.

Brazo, D. C. and C. R. Liston. 1979. The effects of five years of
operation of the Ludington pumped storage power plant on the fishery
resources of Lake Michigan (1972—1977). Mich. State Univ., Dept.
Fish. Wildl. 1977 Ann. Rep. (Fish. Res.) to Consumers Power Co.

405 pp.

Bringold, M. 1978. A study of the seasonal food habits of the johnny
darter, Ethcoatoma nigdum (Rafinesque) in Lake Michigan near
Ludington, Michigan. Unpublished manuscript. Mich. State
Univ. Ludington Res. Lab. 14 pp.

Chiotti, T. L. 1973. Food habits, reproductive biology, and lamprey
scarring of planted lake trout (Safucfiinué namaycubh) in the
inshore waters of Lake Michigan at Ludington, Michigan. M. S.
Thesis, Mich. State Univ. 62 pp.

Clary, J. R. 1972. Predation on the brown trout by the slimy sculpin.
Prog. Fish. Cult. 34(2):9l—95.

Colwell, R. K. and D. J. Futuyama. 1971. On the measurement of niche
breadth and overlap. Ecology 52:567—576.

Daiber, F. C. 1956. A comparative analysis of the winter feeding

habits of two benthic fishes. Copeia, 1956(3):141—151.

59

_;

 

 

 

60

Daugherty, C. H., L. B. Daugherty and P. P. Blair. 1976. ViSual and
olfactory stimuli in feeding behavior of darters (Ethcobtoma)
inhabiting clear and muddy waters. Copeia l976(2):380-382.

DeBoer, P. A. 1957. Trawl gear measurements obtained by underwater
instruments. FAO International Fishing Gear Congress 1957,
Hamburg, Germany, Paper No. 25(b), 21 pp.

Dineen, Clarence F. 1951. A comparative study of the food habits of
COIIuA baihdii and associated species of Salmonidae. Am.
Midl. Nat. 46(3):640-645.

Duffy, W. G. and C. R. Liston. 1978. The appearance and colonization
of benthic macroinvertebrates in the new Ludington pumped
storage reservoir during the first five years of power plant
operation. Mich. State Univ., Dept. Fish. Wildl. 1977 Ann.
Rep. (Limnol. Res.) to Consumers Power Co. 61 pp.

Gulvas, J. A. 1976. An evaluation of the composition and abundance
of fishes pumped into the Ludington Pumped Storage Reservoir
from 1972 to 1975. M. S. Thesis, Mich. State Univ. 107 pp.

Hollander, M. and D. A. Wolfe. 1973. Nonparametric statistical methods.
John Wiley and Sons, N. Y. 503 pp.

Jude, D. J., F. J. Tesar, J. A. Dorr 111, T. J. Miller, P. J. Rago and
D J. Stewart. 1975. Inshore Lake Michigan fish populations
near the Donald C. Cook Nuclear Power Plant, 1973. Special
Report No. 52 of the Great Lakes Research Div., Univ. of
Mich., Ann Arbor. 267 pp.

Karr, J. R. 1963. Age, growth, and food habits of johnny, slenderheaded
and blacksided darters of Boone County, Iowa. Proc. Iowa
Acad. Sci. 70:228—336.

Kenaga, D. E. 1975. Food selection and feeding relationships of yellow
perch, Pe/Lca Maw/scene (Mitchill), white bass, Mononc ch/Lyeope
(Rafinesque), freshwater drum, Aptodtnotue gnunniené (Rafin—
esque) and goldfish, Caflflabbiub auaaiue (Linnaeus) in western
Lake Erie. M. S. Thesis, Mich. State Univ., Dept. Fish & Wildl.
50 pp.

Kinney, E. C. Jr. 1950. Life history of the trout-perch, Pctcopeia
omlecomaycue (Walbaum), in western Lake Erie. M. S. Thesis,
Ohio State Univ. 75 p.

Koster, W. J. 1937. The food of sculpins (Cottidae) in central New
York. Trans. Am. Fish. Soc. 66:374-82.

Lawler, G. H. 1954. Observations on the trout-perch Pehcopbtb
omiecomaycue (Walbaum) at Heming Lake, Manitoba. Jour. Fish.
Res. Ed. Canada ll:l~4.

 

 

61

Liston, C. R. 1977. A study of the effects of installing and oper—
ating a large pumped storage project on the shores of Lake
Michigan near Ludington, Michigan. 1976 Annl. Rep. Fisheries
Res. Vol. I, No. 1. Estimates of salmonid mortalities occurring
at the Ludington Pumped Storage Facility during 1975 and 1976,
and related studies. Submitted to Consumers Power Co. Dept. of
Fish. and Wildl. Michigan State Univ. 102 pp.

Liston, C. R., D. C. Brazo and P. I. Tack, 1976. A study of the effects
of installing and operating a large pumped storage project on
the shores of Lake Michigan near Ludington, Michigan. Mich.
State Univ., Dept. Fish. Wildl. 1974 Annl. Rep. to Consumers
Power Co. Vol. II (Physical—chemical aspects). 65 pp.

Liston, C. R. and P. I. Tack. 1975. A study of the effects of install-
ing and operating a large pumped storage project on the shores
of Lake Michigan near Ludington, Michigan. 1974 Annl. Rep.,
Vol. 1. Fisheries Research. Submitted to Consumers Power Co.
Dept. of Fish. & Wildl. Michigan State Univ. 166 pp.

Lotrich, V. A. 1973. Growth, production and community composition of
fishes inhabiting a first—, second-, and third—order stream
of eastern Kentucky. Ecol. Monogr. 43:377—397.

McCrimmon, H. R. and O. E. Devitt. 1954 Winter studies on the burbot,
Lora (CIR Kacuarnia, of Lake Simcoe, Ontario. Can. Fish.
Cult. 16:35—41.

Muth, K. and L. L. Smith, Jr. 1974. The burbot fishery in Lake of the
Woods. Ag. Exp. Sta., Univ. of Minn. Tech. Bull. 296.

O'Neal, R. P. 1977. Food habits of the trout-perch, PehcopAiA OWMACO-
magcue (Walbaum), in a pumped-storage reservoir near Ludington,
Michigan. Unpublished manuscript. Mich. State Univ. Ludington
Res. Lab. 17 pp.

Patten, B. G. 1962. Cottid predation upon salmon fry in a Washington
stream. Trans. Am. Fish. Soc. 91:427—429.

Phillips, R. W. and E. W. Claire. 1966. Intragravel movement of the
reticulated sculpin COIIuA pcnpflcxue, and its potential as a
predator on salmonid embryos. Trans. Am. Fish. Soc. 95:210—212.

Roberts, N. J. and H. E. Winn. 1962. Utilization of the senses in
feeding behavior of the johnny darter, Ethcoetonu nignwn.
Copeia 1962:567-570.

Sale, P. F. 1974. Overlap in resources use and interspecific compe-
tition. Oecologia (Berlin) 17:245—256.

Savage, T. 1963. Reproductive behavior of Cottub baindii. Copeia
1963:317-325.

 

62

Turner, C. L. 1921. Food of the common Ohio darters. Ohio J. Sci.
22(2):41-62.

Van Oosten, J. and H. J. Deason. 1938. The food of the lake trout
(0114211me manuycwsh) and the lawyer (Lo/ta macuA’Lo/sa) of
Lake Michigan. Trans. Am. Fish. Soc. 67:157-177.

Wathne, F. 1959. Observations on the trawl—door spread and a discus—
sion of influencing factors. U.S. Fish. Wildl. Serv. Comm.
Fish. Rev. 21(10):7-15.

Winn, H. E. 1958. Observations on the reproductive habits of darters
(Pisces—Percidae). Amer. Midl. Nat. 59(1):l90—212.

 

 

MICHIGAN STQTE UNIV. LIBRARIES
llHIUWIIIHWWWIHHIHHWIWWWW
31293100643208