mm

This is to certify that the
dissertation entitled

THE EFFECTS OF ZEBRA MUSSELS (DRE/SSENA
POLYMORPHA) ON INLAND LAKE ECOSYSTEMS

 

presented by

CARRIE ELIZABETH HULT SCHEELE

LIBRARY
Michigan State
University

 

 

has been accepted towards fulﬁllment
of the requirements for the

 

Doctoral degree in Zoology, and Program in
Ecology, Evolutionary Biology
and Behavior

 

 

 

/-’ /ajor Professor’s Signature

August 15, 2007

 

Date

MSU is an afﬁnnative—action, equal-opportunity employer

‘-:—-~--- .l-'-‘-'-'-'-l-O-n-l-I-I-I-l-a-I-l-D-O---l-I-I--o-l-l-l-I-I-'-I-i-I-¢-b-- . _

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

I"
C‘.
CF)

 

 

 

 

 

 

 

 

 

 

 

6/07 p:/CIRC/DateDue.indd-p.1

 

THE EFFECTS OF ZEBRA MUSSELS (DREISSENA POL YMORPHA) ON INLAND
LAKE ECOSYSTEMS

By

Carrie Elizabeth Hult Scheele

A DISSERTATION
Submitted to
Michigan State University

in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Zoology, and Program in Ecology, Evolutionary Biology and Behavior

2007

ABSTRACT

THE EFFECTS OF ZEBRA MUSSELS (DREISSENA POL YMORPHA) ON INLAND
LAKE ECOSYSTEMS

By
Carrie Elizabeth Hult Scheele

The zebra mussel (Dreissena polymorpha) has drastically transformed large
freshwater ecosystems since its invasion of the Laurentian Great Lakes in 1988. Zebra
mussels are now invading small inland lakes, but less is known about their effects on
these systems. Previous studies in well-mixed, shallow systems have shown that zebra
mussels negatively affect primary producers and small primary consumers through
predation. However zebra mussel effects on larger primary and secondary consumers
have varied greatly, likely due to resource competition, and other indirect pathways
through the food web.

I conducted a survey of 50 thermally stratiﬁed lakes with similar nutrient
concentrations and morphometries in southem Michigan to examine the direct and
indirect impacts of zebra mussels on the biomass of microzooplankton, the biomass and
species composition of macrozooplankton, and the grth rates and diet composition of
bluegill sunﬁsh (Lepomz's macrochirus). Twenty-ﬁve lakes were infested with zebra
mussels (invaded), while 25 lakes were zebra mussel free (uninvaded).

Phytoplankton biomass was 24% lower in lakes with zebra mussels, and water
clarity was 21% greater in invaded lakes. Total microzooplankton biomass was 44%
lower, with ciliate and rotifer biomass 39% and 45% lower, respectively, in invaded
lakes. Total macrozooplankton biomass was 33% lower, cladoceran biomass was 43%

lower, yet, copepod biomass was not signiﬁcantly different. Daphnia spp., a large

macrozooplankton and the preferred prey of bluegill sunﬁsh, was 40% lower in invaded
lakes. My study is the ﬁrst to document signiﬁcantly lower biomass of large
macrozooplankton, including Daphnz'a spp. in invaded lakes. Although
macrozooplankton biomass was signiﬁcantly lower, macrozooplankton community
composition did not change in invaded lakes. Zebra mussels are most likely affecting
zooplankton directly through predation and indirectly through resource competition.
Because zooplankton biomass was signiﬁcantly reduced in invaded lakes, I
expected bluegill sunﬁsh grth to be lower. Surprisingly, only growth in ﬁrst year
bluegill was lower, while juvenile and adult growth was greater by 1.8-4.2 mm per year
in invaded lakes. Contemporary and historical bluegill mean length at age studies conﬁrm
that zebra mussels are affecting bluegill, as there were no differences in mean length at
age prior to 1988, the zebra mussel invasion in North America, yet after the invasion
adults were signiﬁcantly larger in invaded lakes. Stomach content and stable isotope
analyses show that adult bluegill may have switched their diet in invaded lakes to include
more benthic macroinvertebrates, providing an important potential mechanism for the
unexpected growth patterns. I attribute lower ﬁrst year grth to low microzooplankton
abundance while in the larval grth stage, and the inability for growth to catch up once
they move back into the littoral zone later in the smnmer. Higher juvenile grth likely
results from increased benthos promoted by zebra mussels. Finally, higher adult growth
most likely results from diet shifts from Daphnia spp. to benthos. This study indicates

that zebra mussels are not contributing to bluegill sunﬁsh stunting.

ACKNOWLEDGEMENTS

I would like to thank my advisor, Dr. Donald Hall for his guidance, support and
mentoring throughout my degree program. Thanks to my graduate committee members,
Dr. Orlando “Ace” Samelle, Dr. Daniel Hayes, Dr. Stephen Hamilton and Dr. Gary
Mittelbach, all of whom contributed greatly to my research. I would like to offer heartfelt
thanks to my graduate student collaborator, Lesley Knoll, for her support and expertise,
and for helping make our project a success. Thanks also to Alan Wilson for help with
stable isotope analyses, Aaron Jubar for training me to age bluegill scales, Dr. Mary
Bremigan for the use of her laboratory and equipment, and for providing the historical
bluegill mean length at age data, and Dr. Patricia Soranno for helping train me to count
zooplankton, and providing historical limnological data. I would also like to thank
everyone that helped in the ﬁeld and laboratory: Chelsea Stephen, Kendra Cheruvelil,
Kevin Pangle, Sherry Martin, Drew Kramer, Tom Alwin, Geoff Horst, Konrad Kulacki,
Scott Kissman, Andrea Stoddard, Mary Martin, Michelle Smith, Katherine Gurnin, Craig
Wynne, Allison Rober and Stacey Nellis. Thanks to the members of the Hall Lab, Nancy
Seefelt, Mary Martin and Jeff Winston for your friendship and guidance. Lastly, I would
like to thank my friends and family for their love and support throughout my program.
Funding was secured from: Graduate Women in Science Eloise Gerry Fellowship,
Graduate School Dissertation Completion Fellowship, John R. Shaver Research
Fellowship, George H. Lauff Research Award, Graduate Research Enhancement Award,
the MSU Ecology, Evolutionary Biology and Behavior Program, and the MSU Zoology

Department.

iv

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................................. vi
LIST OF FIGURES ......................................................................................................... viii
CHAPTER 1

THE EFFECTS OF ZEBRA MUSSELS (DREISSENA POL YMORPHA) ON
MICROZOOPLANKTON AND MACROZOOPLANKTON IN INLAND MICHIGAN

LAKES ................................................................................................................................. 1
Abstract .................................................................................................................... 2
Introduction .............................................................................................................. 4
Methods .................................................................................................................. 10
Results .................................................................................................................... l 5
Discussion .............................................................................................................. 16
Literature Cited ...................................................................................................... 30

CHAPTER 2

THE EFFECT OF ZEBRA MUSSELS (DREISSENA POL YMORPHA) ON THE
GROWTH RATE AND DIET COMPOSITION OF BLUEGILL SUNFISH (LEPOMIS

MACROCHIR US) IN INLAND MICHIGAN LAKES ..................................................... 36
Abstract .................................................................................................................. 37
Introduction ............................................................................................................ 39
Methods .................................................................................................................. 41
Results .................................................................................................................... 53
Discussion .............................................................................................................. 55
Literature Cited ...................................................................................................... 82

APPENDIX ........................................................................................................................ 91

LIST OF TABLES

CHAPTER 1

Table 1. Average, range and standard error of physical and biological variables in
zebra mussel invaded and uninvaded lakes in 2002-2003. * indicates signiﬁcance at
a = 0.05 .............................................................................................................................. 23

CHAPTER 2

Table 1. Effects of zebra mussels on ﬁsh grth in several lakes, rivers, and experimental
enclosures. (- = decrease, 0 = no change, + = increase) .................................................... 64

Table 2. Multiple-regression estimates for average weight per prey item (g wet weight) in
bluegill sunﬁsh stomachs from 20 of the study lakes (N uninvaded = 10, N invaded = 10)
in 2002-2003 ...................................................................................................................... 66

Table 3. Average, range and standard error of physical and biological variables in
zebra mussel invaded and uninvaded lakes in 2002-2003. "‘ indicates signiﬁcance at
a = 0.05 .............................................................................................................................. 67

Table 4. Fixed effects from a linear mixed-effects model of 2467 observations of
individual bluegill grth in the ﬁnal year from 50 lakes (25 uninvaded, 25 invaded) in
the years 2002-2003 ........................................................................................................... 69

Table 5. Fixed effects from a linear mixed-effects model comparing 83 observations of
juvenile bluegill mean length at age (ages 1-2) in 24 lakes in the survey with both
historical (prior to 1988) and contemporary (2002-2003) data. Non-signiﬁcant interaction
terms were removed from the model ................................................................................. 69

Table 6. Fixed effects from a linear mixed-effects model comparing 37 observations of
juvenile bluegill mean length at age (ages 1-2) in 24 lakes only using historical (prior to
1988) data. Non-signiﬁcant interaction terms were removed from the model .................. 70

Table 7. Fixed effects from a linear mixed-effects model comparing 46 observations of
juvenile bluegill mean length at age (ages 1-2) in 24 lakes only using contemporary
(2002-2003) data. Non-signiﬁcant interaction terms were removed from the model ....... 70

Table 8. Fixed effects from a linear mixed-effects model comparing 148 observations of
adult bluegill mean length at age (ages 3-6) in 24 lakes in the survey with both historical
(prior to 1988) and contemporary (2002-2003) data. Non-signiﬁcant interaction terms
were removed from the model ........................................................................................... 70

vi

Table 9. Fixed effects from a linear mixed-effects model comparing 86 observations of
adult bluegill mean length at age (ages 3-6) in 24 lakes only using historical (prior to
1988) data. Non-signiﬁcant interaction terms were removed from the model .................. 71

Table 10. Fixed effects from a linear mixed-effects model comparing 62 observations of
adult bluegill mean length at age (ages 3-6) in 24 lakes only using contemporary (2002-
2003) data. Non-signiﬁcant interaction terms were removed from the model .................. 71

Table 11. Fixed effects from a linear mixed-effects model comparing 260 observations of
percent zooplankton in the stomach of adult bluegill in 26 lakes in 2002-2003 ............... 71

Table 12. Fixed effects from a linear mixed-effects model comparing 260 observations of
adult bluegill stomach weight in 26 lakes from 2002-2003. Non-signiﬁcant interaction

terms were removed from the model ................................................................................. 72

Table 13. Fixed effects from a linear mixed-effects model comparing 81 observations of
the change in 615N between zooplankton and adult bluegill in 29 lakes in 2002-2003 ..... 72

Table 14. Fixed effects from a linear mixed-effects model comparing 81 observations of
the change in 513 C between zooplankton and adult bluegill in 29 lakes in 2002-2003 ..... 72

vii

LIST OF FIGURES

CHAPTER 1

Figure 1. Simpliﬁed diagram of an inland lake food web including zebra mussels. Solid
lines indicate the direct effects of zebra mussels on other compartments in terms of
energy ﬂow. Dotted lines indicate indirect zebra mussel effects. Line weight represents
the strength of interactions between compartments. Note that microzooplankton and
macrozooplankton can be affected by zebra mussels through multiple pathways in the
food web. Other potential indirect pathways not included are the effects of zebra mussels
on light, nutrients, dissolved organic carbon and bacteria, all of which may in tum affect
micro- and macrozooplankton ........................................................................................... 25

Figure 2. Ciliate, rotifer, cladoceran and copepod dry biomass (ug L'l) in zebra mussel
uninvaded and invaded lakes. P- values are ﬁom t-tests for ciliates, rotifers and
cladocerans, and the Mann-Whitney U test for copepods. Error bars represent standard
error .................................................................................................................................... 26

Figure 3. Daphnia spp., calanoid copepod, and cyclopoid copepod dry biomass (pg L'l)
in zebra mussel uninvaded and invaded lakes. P-values are from a t-test for calanoid
copepods, and Mann-Whitney U tests for Daphnia spp. and cyclopoid copepods. Error
bars represent standard error .............................................................................................. 27

Figure 4. Mean relative biomass of the ﬁve most abundant macrozooplankton taxa in
zebra mussel uninvaded and invaded lakes ....................................................................... 28

Figure 5. Macrozooplankton relative biomass PCA factor scores for zebra mussel
invaded and uninvaded lakes. Factor 1 (t48 = -0.77, p = 0.45) and Factor 2
(t43 = -1.18, p = 0.25) ......................................................................................................... 29

CHAPTER 2

Figure 1. Residual plot for the multiple-regression model used to estimate the average
weight of prey items in bluegill sunﬁsh guts from 20 of the study lakes (N uninvaded =
10, N invaded = 10) in 2002-2003. R2 = 0.85, F1 1,139 = 100.95, p < 0.001. ....................... 73

Figure 2. Frequency distribution and approximate life history category of bluegill sunﬁsh
placed into 10 mm size classes in uninvaded and invaded lakes in 2002-2003. Life history
categories are separated by the bold vertical lines ............................................................. 74

Figure 3. Growth increment (m) i SE in the ﬁnal year for bluegill sunﬁsh in three life

history categories (as determined by length at the beginning of the ﬁnal year from Figure
2) in zebra mussel uninvaded and invaded lakes in 2002-2003. Fmg = 5.74, p = 0.02 ..... 75

viii

Figure 4. Comparison between historical (prior to 1988) and contemporary (2002-2003)
data of bluegill mean length at age (m) :t SE in 24 of our study lakes (N uninvaded =
13, N invaded = 11). Life history categories are separated by the bold vertical line.
Juvenile Historical F1,” = 0.33, p = 0.57; Juvenile Contemporary F131 = 2.76, p = 0.11;
Adult Historical F1452 = 0.69, p = 0.41; Adult Contemporary F137 = 6.37, p = 0.02. ........ 76

Figure 5. Percent zooplankton :t SE in the diet of adult bluegill sunﬁsh from invaded and
uninvaded lakes in 2002-2003 (N uninvaded = 13, N invaded = 13). F134 = 2.24, p =
0.15 ..................................................................................................................................... 77

Figure 6. Adult bluegill sunﬁsh stomach content biomass (g) 1: SE as a function of the
length of ﬁsh at capture (m) i SE in invaded and uninvaded lakes in 2002-2003 (N
uninvaded = 13, N invaded = 13). F134 = 0.60, p = 0.44. .................................................. 78

Figure 7. Average isotopic composition (813 C d: SE and 8 15N :1: SE) of select food web
compartments in invaded (open circles) and uninvaded (closed circles) lakes in 2002-
2003. For the bluegill and zooplankton compartments (N uninvaded = 12, N invaded =
17), for the sediment and phytoplankton compartments (N uninvaded = 1, N invaded = 5)
and for the snail and zebra mussel compartments (N uninvaded = 0, N invaded = 5). ..... 79

Figure 8. The change in 6 15N i SE between zooplankton and bluegill sunﬁsh
compartments in invaded and uninvaded lakes in 2002-2003 (N uninvaded = 12, N
invaded = 17). F137 = 6.70, p = 0.02 ................................................................................. 80

Figure 9. The change in 6 ‘3 C :t SE between zooplankton and bluegill sunﬁsh

compartments in invaded and uninvaded lakes in 2002-2003 (N uninvaded = 12, N
invaded =17).F1,27 = 2.19, p = 0.15. ................................................................................. 81

ix

CHAPTER 1

THE EFFECTS OF ZEBRA MUSSELS (DREISSENA POL YMORPHA) ON
MICROZOOPLANKTON AND MACROZOOPLANKTON IN INLAND

MICHIGAN LAKES

Carrie E.H. Scheelel, Lesley B. Knollz’ 3, and Orlando Samelle2

1Department of Zoology, Michigan State University, East Lansing, MI 48824-1222

2Department of Fisheries and Wildlife, Michigan State University, East Lansing, MI
48824-1222

3 Current address, Department of Zoology, Miami University, Oxford, OH 45056

Abstract
The zebra mussel (Dreissena polymorpha) has drastically transformed large freshwater
ecosystems since its invasion of the Laurentian Great Lakes in the late 19805. Zebra
mussels are now invading smaller inland lakes, but less is known about their effects on
these systems. We conducted a survey of 50 thermally stratiﬁed lakes in Michigan with
similar nutrient concentrations and morphometries to examine the direct and indirect
impacts of zebra mussels on the biomass of microzooplankton and the biomass and
community composition of macrozooplankton. Twenty-ﬁve lakes were infested with
zebra mussels (invaded), while 25 lakes were zebra mussel free (uninvaded). In zebra
mussel invaded lakes, phytoplankton biomass was 24% lower, and water clarity was 21%
greater. Total microzooplankton biomass was 44% lower, with ciliate and rotifer biomass
lower by 39% and 45 %, respectively, in invaded lakes. Total macrozooplankton biomass
was 33% lower, cladoceran biomass was 43% lower, and Daphnia spp. biomass was 40%
lower in invaded lakes. Copepod biomass, calanoid copepod biomass and cyclopoid
copepod biomass were not signiﬁcantly different. We also observed no difference in
macrozooplankton relative biomass, indicating that macrozooplankton community
structure did not differ in invaded lakes. Our microzooplankton results are similar to
previous studies conducted in shallow well-mixed systems, although the magnitude of the
zebra mussel effects in our study was smaller. On the other hand, we are the ﬁrst study to
document lower biomass of large macrozooplankton, including Daphnia spp., in invaded
lakes. We suggest that zebra mussels affected both microzooplankton and
macrozooplankton biomass directly through predation and indirectly through resource

competition. Zebra mussels have a bottom-up effect on inland lake ecosystems, and this

effect may be observed in higher trophic levels. Understanding how zebra mussels affect
both microzooplankton and macrozooplankton will identify potential mechanisms by

which ﬁsh are inﬂuenced by this invader.

Introduction

The zebra mussel (Dreissena polymorpha Pallas), an exotic species native to the
Ponto-Caspian region of Eastern Europe, has invaded the United States and is rapidly
spreading throughout freshwater systems. First detected in Lake St. Clair, Michigan in
1988 (Herbert et al. 1989), zebra mussels quickly established populations in all ﬁve of the
Great Lakes and several major river systems (e.g., Hudson, Mississippi, and Ohio Rivers)
(Ludyanskiy et al. 1993). Zebra mussels have also been inadvertently spread by
recreational boat trafﬁc and are colonizing smaller inland lakes (Kraft and Johnson
2000). Although most zebra mussel invasions are concentrated in the Midwestern United
States, they have been transported to water bodies as far west as Washington and
California (U SGS 2007). Zebra mussels colonize hard substrates in lakes and rivers and
can clog water-intake pipes and screens of municipal water supply facilities, industrial
facilities, electric power plants, irrigation pipes and boat engine cooling systems, and
they foul boat hulls and docks. Economic losses in the Great Lakes basin due to damage
and control costs at power plants and water supply units are estimated at $500 million
(US) each year, but there are no estimates available for the economic costs resulting from
damage to beaches, ﬁshing or boating (Pimentel 2005).

Zebra mussels are efﬁcient ﬁlter feeders that represent a new component in the
food webs of North American lakes. They ﬁlter particles as small as 1 pm in diameter but
most efﬁciently feed on particles between 5 and 45 pm in diameter (Maclsaac and Rocha
1995, Sprung and Rose 1988, Ten Winkel and Davids 1982). Zebra mussel ﬁltering leads
to decreased phytoplankton abundance and chlorophyll a (Makarewicz et a1. 1999,

Caraco et al. 1997, F ahnenstiel et al. 1995a, b, Heath et al. 1995, Nicholls and Hopkins

1993), increased water clarity (Caraco et al. 1997, Fahnenstiel et al. 1995a, b, Heath et al.
1995), and altered phytoplankton community composition (Jack and Thorp 2000,
Bastviken et al. 1998). Through selective feeding, zebra mussels can promote blooms of
the toxic colonial cyanobacterium, Microcystz's aeruginosa, in lakes with low to moderate
nutrient levels (Sarnelle et al. 2005, Knoll 2004, Raikow et al. 2004, Vanderploeg et al.
2001)

Zooplankton dynamics can be affected both directly and indirectly by zebra
mussels. Some microzooplankton (ciliates and rotifers) are small enough to be directly
consumed by zebra mussels, and their abundance usually declines in invaded systems
(Pace et a1. 1998, Maclsaac 1996, MacIsaac et al. 1995, MacIsaac et al. 1991). Zebra
mussels can also indirectly affect microzooplankton through resource competition for
phytoplankton (Idrisi et al. 2001, Caraco et al. 1997, Fahnenstiel et al. 1995 a, Heath et
al. 1995). Given these multiple interaction pathways, zebra mussel ﬁltering has the
potential to affect ciliate and rotifer biomass differently. Because zebra mussels prefer
food particle sizes of 5-45 pm (Maclsaac and Rocha 1995, Sprung and Rose 1988, Ten
Winkel and Davids 1982), they should inﬂict greater mortality on ciliates than rotifers,
since ciliate cell sizes are commonly within the preferred range, while rotifers are
generally larger (oﬁen > 100 um). Bacteria are also an important food source for
microzooplankton. However due to the small size of bacteria (< 1 um), its abundance is
typically not affected by zebra mussel presence (Findlay et al. 1998, Cotner et al. 1995).
Ciliates generally consume bacteria more effectively than rotifers. Thus, the relative

magnitude of zebra mussels’ predatory effect may be larger for ciliates, while their

competitive effect may be larger for rotifers. Therefore, it is not obvious whether mussel
invasion will have a greater overall impact on ciliates or rotifers.

Most macrozooplankton are too large to be consumed by zebra mussels
(Maclsaac et al. 1995, Maclsaac et al. 1991; but see Shevtsova et al. 1986), and therefore
are only indirectly affected through resource competition for phytoplankton and
microzooplankton. Studies documenting the effects of zebra mussels on
macrozooplankton abundance in the Great Lakes, inland lakes, and rivers are few and are
sometimes contradictory. In shallow well-mixed areas of the Great Lakes, Padilla et a1.
(1996) found no change in zooplankton biomass in L. Michigan, and MacIsaac et al.
(1991, 1995) found no change in macrozooplankton biomass in western L. Erie, post
invasion. Both J ohannsson et al. (2000) and Mills et al. (2003) suspected decreased
zooplankton production after invasion in L. Erie and L. Ontario respectively, but neither
study found direct evidence for this claim. In Saginaw Bay of L. Huron, Bridgemen
(1995) showed a decrease in macrozooplankton biomass that he attributed not only to
changes in algal food quality following zebra mussel invasion, but also to changes in
predation levels in the lake that were not related to zebra mussels. In shallow Oneida
Lake, NY, Daphnia species biomass and production did not change following invasion
(Idrisi et al. 2001). In well-mixed river systems, zebra mussels have negatively affected
the biomass (Thorp and Casper 2003) and growth (Jack and Thorp 2000) of smaller
macrozooplankton species such as Bosmina spp., Diaphanasoma spp. and Diacyclops
spp. Additionally, Pace et al. (1998) showed a non-signiﬁcant decline in
macrozooplankton biomass in the Hudson River after invasion. Although these results are

sometimes contradictory, the general trend in macrozooplankton biomass is either to not

 

change or to decrease after zebra mussel invasion. The disparities in these results are
most likely due to many unquantiﬁed indirect zebra mussel effects (Figure 1).

Macrozooplankton are the major food source for planktivorous ﬁsh. If
macrozooplankton biomass declines due to zebra mussel invasion, then the mussels may
negatively affect species in higher trophic levels, such as planktivorous and piscivorous
ﬁsh (Rutherford et al. 1999), and could potentially threaten the recreational ﬁshing
industry in the Midwestern US. Strayer et al. (2004) provided a brief review of zebra
mussel effects on ﬁsh in six different lake and river systems. These studies showed
increased, decreased, or unchanged ﬁsh growth rates or abundance at different sites, even
though 66% of the study locations underwent a signiﬁcant decline in phytoplankton and
50% of the study locations endured a decline in zooplankton biomass after invasion. The
authors attributed the varying results to the three indirect pathways in which zebra
mussels can affect ﬁsh: 1) reduced phytoplankton and edible consumers, 2) increased
biodeposits and shelter in mussel beds, and 3) enhanced littoral production. It is clear that
more research is required to untangle the importance and relative magnitude of each of
the three indirect effects.

Most studies on the effects of zebra mussels on lower trophic levels compare
communities before and after invasion in a single, shallow, well-mixed ecosystem. In a
well-mixed lake or river, water is frequently circulating, allowing zebra mussels to ﬁlter
the entire water column. This results in strong effects on the biota, such as large declines
in phytoplankton (Idrisi et al. 2001, Makarewicz et al. 1999, Caraco et al. 1997,
Fahnenstiel et al. 1995a, b, Heath et al. 1995, Nicholls and Hopkins 1993) and

microzooplankton (Pace et al. 1998, MacIsaac 1996, Maclsaac et al. 1995, MacIsaac et

al. 1991). Alternatively, in deep, thermally stratiﬁed lakes, zebra mussels may not ﬁlter
the entire water column during summer, and zebra mussels would be expected to have
weaker effects on the biota in these lakes (Noonburg et al. 2003, Maclassac 1996,
Maclsaac et al. 1991). Additionally, deeper, thermally stratiﬁed lakes have a smaller
proportion of epilimnetic area relative to zebra mussel habitat, also resulting in weaker
zebra mussel effects in these lakes. Our group has investigated the zebra mussel effect on
phytoplankton in deep, thermally stratiﬁed inland lakes and found signiﬁcantly lower
chlorophyll a (30% and 50%, respectively) and phytoplankton biomass (30% and 45%,
respectively) (Knoll et al. in press, Raikow et al. 2004), consistent with studies on
shallow well-mixed systems. However, it is still unknown how zebra mussels will affect
microzooplankton and macrozooplankton in thermally stratiﬁed inland lakes and whether
these results will be consistent with studies from shallow well-mixed systems. At this
time, no studies have investigated the effects of zebra mussels on microzooplankton and
macrozooplankton in multiple inland lakes.

A survey of multiple thermally stratiﬁed lake ecosystems with and without zebra
mussels is a more general approach to studying the effects of zebra mussels on inland
lakes, compared to before and after zebra mussel invasion studies on single systems. The
multi-lake survey approach provides additional statistical power to detect patterns
associated with the zebra mussel invasion that may not be detectable in before and after
invasion studies. Thus, a large-scale, multi-system study will provide important clues
regarding the indirect effect of zebra mussels on higher trophic levels, speciﬁcally

through the phytoplankton and edible microzooplankton consumer pathway.

We conducted an extensive survey of 50 thermally stratiﬁed inland Michigan
lakes (25 invaded and 25 uninvaded) to examine the effects of zebra mussels on
microzooplankton and macrozooplankton. We addressed four main questions. 1) Do
zebra mussels reduce microzooplankton biomass in thermally stratiﬁed inland lakes, and
will our results be consistent with results from well-mixed systems? 2) Do zebra mussels
cause a larger decline in ciliate biomass than in rotifer biomass? 3) Do zebra mussels
reduce macrozooplankton biomass in thermally stratiﬁed inland lakes? 4) Do zebra
mussels alter macrozooplankton community structure? This is the ﬁrst study to document
the effects of zebra mussels on microzooplankton and macrozooplankton abundance and
macrozooplankton species composition in multiple thermally stratiﬁed inland lakes. We
show that zebra mussels signiﬁcantly lowered microzooplankton biomass and that the
magnitude of the zebra mussel effect was similar on both ciliates and rotifers. We are the
ﬁrst study to show signiﬁcantly lower biomass of large macrozooplankton, including
Daphnia spp. in invaded lakes. Additionally, we show macrozooplankton community
structure did not change. The results presented in this manuscript are a compliment study
to Knoll et al. (2007; in press) that details the effects of zebra mussels on phytoplankton
biomass and community structure in these same lakes. A small portion of the
phytoplankton data from Knoll et al. (2007; in press) are reproduced here to facilitate

interpretation of the microzooplankton and macrozooplankton data.

Methods

Lake Selection Criteria

We selected 50 study lakes in Southern Michigan, USA, 25 invaded by zebra
mussels and 25 uninvaded references. Invaded and uninvaded lakes were nearly equally
balanced in the southwest and southeast regions of the state. All lakes had low to
moderate total phosphorus (TP) concentrations (<22 pg L'l, Knoll et a1 2007 in press,
Michigan Department of Environmental Quality). Lakes selected for this study were 2 9
m in maximum depth to ensure summer temperature stratiﬁcation, in contrast to most
previous studies investigating the D. polymorpha invasion that were conducted in shallow
well-mixed systems (Thorp and Casper 2003, Vanderploeg et al. 2001, Idrisi et al. 2001 ,
Pace et al. 1998, Caraco et a1. 1997, Bridgeman 1995, Maclsaac et al. 1995). The
uninvaded lakes were similar in pH and calcium concentrations to the invaded ones, and
thus would support zebra mussels if invaded (Raikow et al. 2004, Raikow 2002). To
ensure that treatment and control groups were of similar mean depths, we calculated
mean depth of all lakes ﬁ'om digitized bathymetric maps (ESRI 1999). Where lake maps
were unavailable, we obtained mean depth data from the Michigan Department of
Environmental Quality (unpublished data). We acquired zebra mussel presence or
absence data from the United States Geological Survey Nonindigenous Aquatic Species
database (USGS 2002, 2003) and the Michigan Sea Grant Zebra Mussel Infestation
Monitoring Program (Michigan Sea Grant 2002, 2003). Adult zebra mussel presence and
absence was veriﬁed while sampling the lake, and larval veliger presence and absence

was conﬁrmed while counting zooplankton samples (see below).

10

Lake sampling

We sampled lakes once either between 3 August and 5 September 2002, or 16
July and 23 August 2003. In 2002, we sampled 26 lakes (N uninvaded = 14, N invaded =
12) and in 2003, we sampled 24 lakes (N uninvaded = 11, N invaded = 13). Lakes were
sampled in late summer to ensure thermal stratiﬁcation and sampled over a short time
period to reduce the inﬂuence of seasonal variability among lakes. Zebra mussel presence
or absence was conﬁrmed visually by examining the boat launch and adjacent shallow
areas along the shoreline for approximately 15-20 min, looking for zebra mussels in the
sediment or attached to other organisms.
Physical and Biological Parameters

We measured temperature, dissolved oxygen, chlorophyll a, and total phosphorus
(TP) at the deepest part of each lake, as determined ﬁ'om bathymetric maps and an
echosounder. We measured temperature and dissolved oxygen at l m depth intervals
using a Hydrolab Surveyor 4a equipped with a Datasonde 4a (HACH Environmental,
Loveland CO). Photosynthetically active radiation (PAR) was measured at 0.5 m depth
intervals using a LiCor model Li-1000 quantum photometer with an attached spherical
quantum underwater sensor and corresponding deck sensor (LiCor Environmental,
Lincoln, NE). Light extinction coefﬁcients were determined as the slope of the linear
regression between 1n PAR and depth.

We collected an integrated epilimnetic water sample using a ﬂexible plastic tube
(5 cm internal diam, 10 m length). The tube was lowered to the bottom of the mixed
layer, as determined by the temperature and dissolved oxygen proﬁle, capped and hauled

into the boat. Two to four mixed layer samples were collected and pooled in a large

11

container, and sub samples were taken for phytoplankton composition, microzooplankton
composition, chlorophyll a and TP. Phytoplankton samples were preserved immediately
in Lugol’s solution (Hasle 1978). Microzooplankton were collected by passing a 10 L sub
sample of the mixed layer water through a 35 um mesh screen and rinsing organisms on
the screen into sample bottles containing glutaraldehyde (ﬁnal concentration: 2%). Water
samples for chlorophyll a and TP were placed on ice until they were processed later the
same day (~6 h). Chlorophyll a samples were ﬁltered through Gelman A/E glass ﬁber
ﬁlters (Gelman Sciences, Ann Arbor, MI) and frozen until laboratory analysis. TP
samples were frozen until laboratory analysis. Macrozooplankton samples were also
obtained from the deepest part of the lake. The zooplankton net (30 cm diameter, 100 um
mesh) was lowered to ~1 m above the lake bottom, and hauled through the water column.
Four hauls were pooled from each lake and preserved with 95% ethanol.

We quantiﬁed chlorophyll a by ﬁrst extracting it from the glass ﬁber ﬁlters with
90% ethanol and then analyzing the extract with a Turner Model 10-AU ﬂuorometer
(Welschmeyer 1994) calibrated to commercial chlorophyll a standards (Anacystis;
Sigma-Aldrich Chemical Company, St. Louis, MO). TP samples were oxidized via
persulfate digestion in an autoclave and then analyzed using the colorimetric molybdate
blue method (Langner and Hendrix 1982).
Phytoplankton Biomass

Phytoplankton biomass was assessed in a randomly-chosen subset of the surveyed
lakes (N uninvaded = 20, N invaded = 22; Knoll et al. 2007 in press). Phytoplankton
were identiﬁed to species in most cases and enumerated via the inverted microscope

technique (Hasle 1978). Biovolume was determined from measurements of cell

12

dimensions of at least ten individuals of common species in each sample at 1000x
magniﬁcation using a NH<ON model TE2000-S inverted microscope (NIKON, Melville
NY), a SPOT insight QE model 4.2 digital camera, and SPOT Advanced version 4.0.9
image-analysis software (Diagnostic Instruments Inc., Sterling Heights, MI). Biovolume
was converted to dry biomass assuming a speciﬁc gravity of 1 g cm'3 and a dry mass to
wet mass ratio of 0.10.

Microzooplankton Biomass

Ciliates were assessed in all 50 study lakes (N uninvaded = 25, N invaded = 25).
Ciliates were enumerated with the inverted microscope technique using a NIKON model
TE2000-S inverted microscope. Sub samples (sub sample volume 30-100 ml) were
settled in tubular chambers (Hydro-Bios, Kiel-Holtenau, Germany), the bottoms of which
were divided into inner and outer zones of equal area (Sandgren and Robinson 1984).
Within each zone, at least 20 random ﬁelds were counted at 100x magniﬁcation. Ciliate
cell volume was determined by measuring at least ﬁve individuals per taxon at 400x
magniﬁcation using the imaging system described above. Biovolume was converted to
dry biomass assuming a speciﬁc gravity of 1 g cm'3 and a dry mass to wet mass ratio of
0.10.

Rotifers were assessed in all 50 study lakes. Rotifers were identiﬁed to species
using a Nikon model E600 compound microscope at 100x magniﬁcation using a SPOT
insight Color model 3.2.0 digital camera and SPOT Advanced version 4.0.9 image-
analysis software (Diagnostic Instruments Inc., Sterling Heights, MI), and a Sedgwick-

Raﬂer counting chamber. For each lake, approximately 400 individuals (average = 394,

13

range = 341-475) were counted in a minimum of 2 sub samples. Individual dry biomass
for each species was estimated from established literature values (Pauli 1989).
Macrozooplankton Biomass

Macrozooplankton were assessed in all 50 study lakes. Macrozooplankton were
counted and identiﬁed to genus or species using a 10 ml clear PVC zooplankton counting
wheel and a Leica model M28 dissecting microscope (Leica Microsystems, Inc.,
Bannockbum, IL). Samples were diluted to a known volume and two to three 5 ml sub
samples were counted. A minimum of 450 individuals were tallied per lake.
Measurements of individuals were made at a magniﬁcation of 10x, using a Summa
Sketch III digitizing pad and ZoopBiom software (Roff & Hopcroft 1986). For each
sample, up to 50 individuals were measured for each large (> 1.0 mm) genus or species
(Daphnia galeata, D. pulicaria, D. retrocurva, Epischura spp., Leptadora spp., and
Mesocyclops spp.) and up to 25 individuals were measured for each small (< 1.0 mm)
genus or species (Alonella spp., Bosmina spp., Ceriodaphnia spp., Cyclops spp.,
Diaphanasoma spp., Diaptomus spp., Dreissena polymorpha veligers, Moina spp., and
nauplii). Biomass was calculated using published length-weight regressions for
individual species (Culver et al. 1985, Dumont et al. 1975).
Statistical Analyses

Student’s t-tests and Mann-Whitney U-tests were used to evaluate the effect of the
presence or absence of zebra mussels on lake physical and biological parameters, and on
phytoplankton, microzooplankton and macrozooplankton biomass. When data failed the
Kolmogorov-Smirmov and Shapiro-Wilk tests for normality (p > 0.05), data were log

transformed. If the log transformed data passed the tests for normality (p < 0.05),

14

Student’s t-tests were employed. When log transformed data failed normality tests, the
Mann-Whitney U non-parametric test was employed.

To determine the effect of zebra mussels on macrozooplankton community
composition, we used a principal component analysis (PCA) on macrozooplankton
relative biomass. To reduce the inﬂuence of zero values, only common taxa were
included in the analysis (Cyclops spp., Daphnia spp., Diaphanasoma spp., Diaptomus
spp., and Mesocyclops spp.). PCA taxa data were arcsine square root transformed to
achieve normality of the residuals. Factor scores were then compared for zebra mussel
invaded and univaded lakes with a Student’s t-test. All data were analyzed using Systat

version 11.0 (Wilkinson, 1989).

Results

Mean depth and TP in invaded and uninvaded lakes were not signiﬁcantly
different (U25,25 = 241, p = 0.17; Ln; = -0.32, p = 0.75; respectively) (Table 1), indicating
no bias in the selection of sample lakes with respect to these parameters. Light extinction
coefﬁcients were signiﬁcantly higher (t43 = -2.93, p = 0.01) in invaded lakes, indicating
increased penetration of PAR in the water column and increased water clarity. Although
chlorophyll a did not differ signiﬁcantly between the lake groups (t48 = 1.68, p = 0.10),
total phytoplankton, total microzooplankton and total macrozooplankton biomass were
signiﬁcantly lower in invaded lakes by 24%, 44% and 33%, respectively (t40 = 2.30, p =
0.03; I48 = 3.31, p = 0.002; t43 = 2.22, p = 0.03; respectively) (Table 1). Within the
microzooplankton and macrozooplankton categories, ciliate, rotifer and cladoceran

biomass were 39%, 45% and 43% lower, respectively, in invaded lakes (L48 = 2.16, p =

15

0.04; t43 = 3.06, p = 0.004; t43 = 2.79, p = 0.008; respectively), while copepod biomass
did not differ (U253; = 365, p = 0.31) (Figure 2). The magnitude of the effect of zebra
mussel on ciliates and rotifers was similar; ciliate biomass decreased by 39% and rotifer
biomass decreased by 45%. Within the cladoceran and copepod categories, Daphnia spp.
biomass was signiﬁcantly lower by 40% in invaded lakes (U535 = 422, p = 0.03), and
calanoid and cyclopoid copepod biomass were not signiﬁcantly different (t4g = -0.75, p =
0.94; U24” = 387, p = 0.08; respectively) (Figure 3).

Relative biomass of the most abundant macrozooplankton taxa was similar in
both zebra mussel uninvaded and invaded lakes (Figure 4). The PCA of
macrozooplankton relative biomass reduced the ﬁve taxa categories into two factors that
explained 71% of the overall variance. However, factors 1 and 2 were not signiﬁcantly
related to zebra mussel presence or absence (t43 = -0.77, p = 0.45; t43 = -1.18, p = 0.25;
respectively) (Figure 5), providing no evidence of a shift in macrozooplankton

community composition in invaded lakes.

Discussion
Microzooplankton
Our results suggest that zebra mussels have a strong negative effect on
microzooplankton in thermally stratiﬁed inland lakes (Table 1, Figure 2). However, the
magnitude of effect zebra mussels exert on microzooplankton is lower in stratiﬁed lakes
compared to shallow well-mixed systems. We found 39% lower ciliate biomass and 45%
lower rotifer biomass in invaded lakes. Experimental studies have shown that zebra

mussels reduce ciliate biovolume by 77% (Wilson 2003) and protozoan abundance by 70-

16

80% (Lavrentyev et al. 1995). In the Hudson River, total zooplankton biomass declined
by 70% (Pace et. a1 1998) and mean total zooplankton density (excluding ciliates) was
55-71% lower in Lake Erie following zebra mussel invasion (MacIsaac et al. 1995).
Reductions of zooplankton in both the Hudson River and Lake Erie were mainly
attributed to negative effects of zebra mussels on rotifers. The lesser reduction in our
study compared to the Hudson River and Lake Erie studies could result from differences
in mixing regime. The plankton in shallow, well-mixed systems may experience greater
zebra mussel impacts than in deep, stratiﬁed systems (Noonburg et al. 2003, Maclassac
1996, MacIsaac et al. 1991). In a shallow well-mixed system, pelagic organisms are more
likely to come into contact with benthic populations of zebra mussels because mussels
are able to colonize a greater proportion of the lake bottom and because frequent water
column mixing allows the entire benthic and pelagic zones of the lake to interrnix. It was
unclear whether zebra mussels would have a greater impact on rotifers and ciliates,
because zebra mussels should inﬂict greater predatory effects on ciliates than rotifers, and
greater competitive effects on rotifers. Overall, we found that the magnitude of the
reduction for both groups was remarkably similar (3 9% vs. 45%), even though theory
suggests that zebra mussels should ﬁlter ciliates more effectively than rotifers due to their
smaller size (MacIsaac and Rocha 1995, Sprung and Rose 1988, Ten Winkel and Davids
1982)

The similar effect of zebra mussels on ciliates and rotifers may be explained by
the fact that bacteria are not a primary food source of zebra mussels. Typically, zebra
mussels do not reduce the total abundance of bacteria in systems they invade (Findlay et

al. 1998, Cotner et al. 1995) because zebra mussels are not efﬁcient at eating naturally

l7

occurring bacteria due to their small size (< 0.1 pm). Planktonic ciliates feed primarily on
phytoplankton (Fenchel 1987) and bacteria, and sometimes rely on bacteria as a key
resource (Christoffersen et al. 1990). In contrast, rotifers have difﬁculty consuming
bacteria (Amdt 1993), and primarily consume phytoplankton between 4-17 pm (Bogdan
and Gilbert 1984, Bogdan et al. 1980). Thus, in invaded lakes, ciliates may be able to
take advantage of bacteria more easily than rotifers, allowing ciliates to compensate for
greater predation losses and phytoplankton reductions.

The relative importance of direct (predation) or indirect (competition)
mechanisms in the negative impacts of zebra mussels on microzooplankton cannot be
determined from this study. As discussed above, predation is likely an important factor.
Yet, phytoplankton biomass was signiﬁcantly lower in invaded lakes (Table l), as found
in previous studies (Idrisi et al. 2001, Caraco et a1. 1997, Fahnenstiel et al. 1995b). By
consuming phytoplankton, zebra mussels are competing with microzooplankton for
resources. Thus, it is reasonable to assume that reductions in phytoplankton, mediated
through zebra mussels, could indirectly affect microzooplankton abundance (Figure 1).
Previous experiments concluded that zebra mussel predation may be more important than
resource competition in reducing microzooplankton abundance (Thorp and Casper 2002,
Maclsaac et a1. 1995, Maclsaac et al. 1991). In these studies, small-bodied zooplankton
were primarily reduced while large-bodied zooplankton were not, even though both
compete for resources with zebra mussels. However, these experiments were either
conducted in small containers (Maclsaac et al. 1995, Maclsaac et al. 1991), or were short-
term (Thorp and Casper 2002). In small-scale experiments, predators and prey may

experience greater spatial overlap than in thermally stratiﬁed lakes, which may

18

exaggerate the importance of predation over resource competition, particularly on smaller
organisms (Sarnelle 1997). Short-term experiments may also overstate the importance of
predation on microzooplankton because the effects of resource competition often take
longer to observe than those of predation (Sarnelle 1997).
Macrozooplankton

Zebra mussels also negatively affected macrozooplankton biomass. Total
macrozooplankton biomass was 33% lower, cladoceran biomass was 43% lower, and
Daphnia spp. biomass was 40% lower in invaded lakes (Table 1, Figures 2 and 3).
Copepod biomass, calanoid copepod biomass and cyclopoid copepod biomass were not
signiﬁcantly different between uninvaded and invaded lakes (Figures 2 and 3). Our study
is the ﬁrst to document signiﬁcantly lower biomass of large macrozooplankton species,
including Daphnia spp., in invaded lakes. The two most comprehensive studies that
investigated the effects of zebra mussels on macrozooplankton were conducted before
and after invasion in Oneida Lake, NY, (Idrisi et al. 2000) and the Hudson River (Pace et
al. 1998). The authors documented no difference in Daphnia biomass and production,
and mean abundance of post-naupliar copepods and cladocerans, respectively. Idrisi et al.
(2000) attributed the lack of change in Daphnia spp. biomass and production to the lack
of change in primary production in the lake due to increased water clarity. Pace et al.
(1998) attributed the lack of change in cladocerans to high amounts of variation over
time, the ability of cladocerans to change their diet to bacteria, and to abundant residual
phytoplankton. These authors also attribute the lack of change in copepods to the ability
to change their diet to protozoa, and to abundant residual phytoplankton. Therefore, our

study shows a much greater effect of zebra mussels on macrozooplankton biomass

19

(33%), including Daphnia spp. (40%) in thermally stratiﬁed lakes, compared to shallow
well-mixed systems.

It is interesting that copepod biomass was not lower, given 24% lower
phytoplankton and 44% lower microzooplankton biomass. Copepods in the Hudson River
did not decline after invasion (Pace et al.1998). Yet, Thorp and Casper (2002)
documented a signiﬁcant increase of the calanoid copepod Eurytemora affinis density
(150%) in the presence of zebra mussels. These studies suggest that zebra mussels affect
macrozooplankton groups through multiple indirect pathways that may vary across taxa
and ecosystems (Figure l).

Decreased macrozooplankton biomass in zebra mussel invaded lakes is most
likely a result of resource competition. Zebra mussels signiﬁcantly lowered
phytoplankton and microzooplankton biomass (Table 1, Figure 2), thereby lowering food
resources for macrozooplankton. Additionally, Knoll (2004) reported a shiﬂ in the
phytoplankton community towards dominance by Microcysis aeruginosa, a toxic
cyanobacterium, in a subset of the lakes used in our current study. Her results suggest
that zebra mussels have not only lowered the overall abundance of both phytoplankton
and microzooplankton in invaded lakes, but also they have potentially made the
remaining food resources less edible for macrozooplankton. As a result of these altered
food resources in invaded lakes, it is highly probable that zebra mussels are outcompeting
macrozooplankton and negatively affecting macrozooplankton abundance. Resource
competition was also suggested to explain mortality in unionid bivalve mollusks
(Unionidae) (Schloesser et al. 1996), another primary consumer negatively affected by

zebra mussels.

2O

Finally, we found that although zebra mussels negatively affected the biomass of
most macrozooplankton groups, macrozooplankton community structure (as deﬁned by
relative biomass) did not differ between invaded and uninvaded lakes (Figures 4 and 5).
It is possible that through resource competition, zebra mussels have reduced population
sizes of macrozooplankton, yet not shifted the relative biomass of species.

Overall, our study shows signiﬁcant negative zebra mussel effects on
phytoplankton, microzooplankton and macrozooplankton in thermally stratiﬁed lakes.
The effects of zebra mussels on microzooplankton were somewhat smaller compared to
those in shallow well-mixed systems. On the other hand, the effects of zebra mussels on
macrozooplankton, including Daphnia spp., were much greater in our study compared to
shallow well-mixed systems. Ciliates and rotifers were similarly affected by zebra
mussels. Cladoceran biomass, including Daphnia spp. biomass was greatly lower, but
macrozooplankton community structure did not change. These striking changes to the
lower trophic levels resulting from zebra mussel invasions of lakes may have
consequences for planktivorous and piscivorous ﬁsh (Rutherford et al. 1999).

Lower zooplankton biomass, especially lower Daphnia spp. biomass, has the
potential to greatly affect higher trophic levels in lakes. Planktivorous ﬁsh, such as
bluegill sunﬁsh (Lepomis macrochirus), rely almost entirely on microzooplankton as
larvae (Bremigan and Stein 1994, Barkoh and Modde 1987, Siefert 1972, Werner 1967),
and Daphnia spp. are a major food resource for adults (Werner and Hall 1988; Mittelbach
1984, 1981). By reducing microzooplankton and macrozooplankton biomass, zebra
mussels may indirectly affect the survival, grth and ﬁtness of planktivorous ﬁshes. In

turn, decreases in planktivorous ﬁsh populations may negatively affect piscivorous ﬁshes

21

and threaten recreational ﬁshing in these lakes. Our study conﬁrms that zebra mussels
affect the phytoplankton and edible consumer pathway in a lake foodweb (Strayer et a1.
2004) through predation and resource competition, because we saw lower phytoplankton
and zooplankton biomass. Thus, it is possible that zebra mussels can negatively affect

ﬁsh that rely on these resources.

22

an: 33: 95.3-22: ch as

 

 

 

ES commas 3% $2 3.4 an? maeem
cougﬁaoﬁnm
38.
Gases amass. : AL m3
2d Baum: m3 mom 2.0 M? 333820
283-38-: :2.o-v-§.o-vv ATE
:od amiss :3 NS- :3 mac. £20580
aoaeaim Em:
SEES 683:; nu 3
m3 moods: 3o 3.2 33 e2: meoaamoaﬁsoe
24.2-26 Snead :5
Cd sauna: and mg .23 Ba ease 50:
Anew—Eb 3»:wa
mm owﬁo>< mm owﬁo><
Maggi cums—3m 33>“: 83:83 53883
30,—.

 

.86 n a a sogéama £865 ... 88-88 a 8E
woum>aas use pupae/E Emma: Snow E moi—mtg ﬁﬂmEoE can Remand mo coho 2353 can owed“ .owﬁo>< A 033.

23

 

§.owm-:.w:

carapace

9.4 a

 

 

 

*mod mud n 3 3.: 8&3 N53 von _ @38on
egg—mooNouomE
Eon.
€2,355 Amwmtawa: AL 9:
Reed :3 u 3 mw.m 36m a6 33v 3885
cog—aooaeomg
=33.
Aoweﬁv mega
Mm owBo>< mm owﬁo><
oﬁgd oumuﬁm @032: BREED 885me
“web

 

.Béeao a use.

24

 

I Planktivorous Fish I

0a.... l .

I ZEBRA MUSSELS

 

 

 

 

O

......o«.o.. r .O...
I Microzooplankton I >I Macrozooplankton I
'0... ...c

O O
.y..... .....O
0.. ..o‘

O O

I Phytoplankton I

Figure 1. Simpliﬁed diagram of an inland lake food web including zebra mussels. Solid
lines indicate the direct effects of zebra mussels on other compartments in terms of
energy ﬂow. Dotted lines indicate indirect zebra mussel effects. Line weight represents
the strength of interactions between compartments. Note that microzooplankton and
macrozooplankton can be affected by zebra mussels through multiple pathways in the
food web. Other potential indirect pathways not included are the effects of zebra mussels
on light, nutrients, dissolved organic carbon and bacteria, all of which may in turn affect
micro- and macrozooplankton.

 

 

 

 

 

 

 

25

.5
o

 

p = 0.04

(D

O)

 

.h

 

N

 

 

 

 

 

O

 

Uninvaded Invaded

100

 

p = 0.008

on
O

O)
O

 

A
O
r

1 .......
......

N
O

 

 

 

. .'. 3.
.. . c.a_-
$.23 ......

O

 

 

Cladoceran Biomass (pg L") Ciliate Biomass (pg L")

Uninvaded Invaded

01
O

 

p = 0.004

b
o

 

N
O

—h
0

 
 

Rotifer Biomass (pg L")
(a)
O

 

 

 

 

O

 

Uninvaded Invaded

.3
O
o

 

 

p = 0.31

m
D

 

O)
O

A
O

N
O

 

 

 

 

O

 

Copepod Biomass (pg L")

Uninvaded Invaded

Figure 2. Ciliate, rotifer, cladoceran and copepod dry biomass (pg L'l) in zebra mussel
uninvaded and invaded lakes. P— values are from t-tests for ciliates, rotifers and
cladocerans, and the Mann-Whitney U test for copepods. Error bars represent standard

error.

 

80

 

0'!
O

 

p=033 p=094

604

 

co A
O o
r--—~I

40*

 

N
O

 
 

. ;
.'. "':-‘
v. ',-. .

. '1: "fl
Uninvaded Invaded

201

.5
o

 

Calanoid Biomass (pg L")

 

 

 

 

 

 

 

 

 

O

 

Uninvaded Invaded

 

40

p = 0.08
30 4

 

20*

10‘

 

 

 

 

 

Cyclopoid Biomass (pg L") Daphnia spp. Biomass (pg L")

Uninvaded Invaded

Figure 3. Daphnia spp. , calanoid copepod, and cyclopoid copepod dry biomass (pg L")
in zebra mussel uninvaded and invaded lakes. P-values are from a t-test for calanoid
copepods, and Mann-Whitney U tests for Daphnia spp. and cyclopoid copepods. Error
bars represent standard error.

27

 

 

 

 

 
 
 
 
 

 

 

 

U) 0.8 -

(D

(U

E - Cyclops spp.

.9 0.6 - :1 Daphnia spp.

CD - Diaptomus spp.

g [:1 Diaphanasoma spp.
1.8 0-4 ‘ - Mesocyclops spp.
T)

D:

0.2 i

 

 

 

0.0 -

Uninvaded Invaded

Figure 4. Mean relative biomass of the ﬁve most abundant macrozooplankton taxa in
zebra mussel uninvaded and invaded lakes.

28

 

 

 

 

 

 

 

 

O Uninvaded
2 - O Invaded .
O . Q
1 O O. .
N COQ) O O O
r... 0 Qt) ‘
3 o - . ‘ 0
g . . O . 0.0 O
in o o 8 .
-1 — O O
o 0 0
-2 — . O
O
’3 I I I T I I
-3 -2 -1 0 1 2 3 4
Factor 1

Figure 5. Macrozooplankton relative biomass PCA factor scores for zebra mussel invaded
and uninvaded lakes. Factor 1 (Lu; = -0.77, p = 0.45) and Factor 2 (Lu; = -1.18, p = 0.25).

29

Literature Cited

Amdt, H.. 1993. Rotifers as predators on components of the microbial web (bacteria,
heterotrophic ﬂagellates, ciliates) — a review. Hydrobiologia 255/256: 231-246.

Barkoh, A., and T. Modde. 1987. Feeding behavior of intensely cultured bluegill fry.
Progressive F ish-Culturist 49: 204-207.

Bastviken, D.T.E., N.F. Caraco, and J .J . Cole. 1998. Experimental measurements of
zebra mussel (Dreissena polymorpha) impacts on phytoplankton community
composition. Freshwater Biology 39: 375-386.

Bogdan, K.G., and J .J . Gilbert. 1984. Body size and food size in freshwater zooplankton.
Proceedings of the National Academy of Sciences (USA) 81: 6427-6431.

Bogdan, K.G., J .J . Gilbert and PL. Starkweather. 1980. In situ clearance rates of
planktonic rotifers. Hydrobiologia 73:73-77.

Bremigan, M.T., and RA. Stein. 1994. Gape-dependant larval foraging and zooplankton
size: implications for ﬁsh recruitment across systems. Canadian Journal of
Fisheries and Aquatic Sciences 51: 913-922.

Bridgeman, T.B., G.L. Fahnenstiel, G.A. Lang, and T.F. Nalepa. 1995. Zooplankton
grazing during the zebra mussel (Dreissena polymorpha) colonization of Saginaw
Bay, Lake Huron. Journal of Great Lakes Research 21(4): 567-573.

Caraco, N.F., J .J . Cole, P.A. Raymond, D.L. Strayer, M.L. Pace, S.E.G. Findlay, and
D.T. Fischer. 1997. Zebra mussel invasion in a large, turbid river: phytoplankton
response to increased grazing. Ecology 78: 588-602.

Christoffersen, K., B. Riemann, L.R. Hansen, A. Klysner, and HE. Sorensen. 1990.
Qualitative importance of the microbial loop and plankton community structure in

a eutrophic lake during a bloom of cyanobacteria. Microbial Ecology 20: 253-
272.

Cotner, J .B., W.S. Gardner, J .R. Johnson, R.H. Sada, J .F . Cavaletto, and RT. Heath.
1995. Effects of zebra mussels (Dreissena polymorpha) on bacterioplankton:

evidence for both size-selective consumption and grth stimulation. Journal of
Great Lakes Research 21: 517-528.

Culver, D.A., M.M. Boucherle, D.J. Bean, and J.W. Fletcher. 1985. Biomass of
freshwater crustacean zooplankton from length-weight regressions. Canadian

Journal of Fisheries and Aquatic Sciences 42: 1380-1390.

Dumont, H.J., 1. Van de Velde, and S. Dumont. 1975. The dry weight estimate of

30

biomass in a selection of cladocera, copepoda and rotifera from the plankton,
periphyton and benthos of continental waters. Oecologia 19: 75-97.

ESRI 1999. Arc View. Version 3.2. ESRI, Redland, CA.

Fahnenstiel, G.L, G.A. Lang, T.F. Nalepa, and TH. Johengen. 1995 a. Effects of zebra
mussel (Dreissena polymorpha) colonization on water quality parameters in
Saginaw Bay, Lake Huron. Journal of Great Lakes Research 21: 435-448.

F ahnenstiel, G.L., T.B. Bridgeman, G.A. Lang, M.J. McCormick, and T.F. Nalepa. 1995
b. Phytoplankton productivity in Saginaw Bay, Lake Huron: effects of zebra

mussel (Dreissena polymorpha) colonization. Journal of Great Lakes Research
21: 465-475.

Fenchel, T.. 1987. Ecology of protozoa: the biology of free-living phagotrophic protists.
Springer-Verlag, Berlin.

Findlay, S., M.L. Pace, and D.T. Fischer. 1998. Response of heterotrophic planktonic
bacteria to the zebra mussel invasion of the tidal freshwater Hudson River.
Microbial Ecology 36: 131-140.

Hasle, GR. 1978. The inverted-microscope method, p. 88-96 in A. Soumia, editor.
Phytoplankton manual. UNESCO, Paris, France.

Heath, R.T., G.L. Fahnenstiel, W.S. Gardner, J.F. Cavaletto, and SJ. Hwang. 1995.
Ecosystem-level effects of zebra mussels (Dreissena polymorpha): an enclosure

experiment in Saginaw Bay, Lake Huron. Journal of Great Lakes Research 21:
501 -5 16.

Herbert, P.D., B.W. Muncaster, and G.L. Mackie. 1989. Ecological and genetic studies
on Dreissena polymorpha (Pallas): A new mollusk in the Great Lakes. Canadian
Journal of Fisheries and Aquatic Sciences 46: 1587-1591.

Idrisi, N., E.L. Mills, L.G. Rudstarn, and DJ. Stewart. 2001. Impact of zebra mussels
(Dreissena polymorpha) on the pelagic lower trophic levels of Oneida Lake, New
York. Canadian Journal of Fisheries and Aquatic Sciences. 58: 1430-1441.

Jack, J.D., and Thorp, J .H. 2000. Effects of the benthic suspension feeder Dreissena
polymorpha on zooplankton in a large river. Freshwater Biology 44: 569-579.

Johannsson, O.E., R. Dermott, D.M. Graham, J.A. Dahl, E.S. Millard, D.D. Myles, and J.
LeBlanc. 2000. Benthic and pelagic secondary production in Lake Erie after the

invasion of Dreissena spp. with implications for ﬁsh production. Journal of Great
Lakes Research 26 (1): 31-54.

Knoll, LB. 2004. Understanding zebra mussel (Dreissena polymorpha) impacts on

31

autotrophic and heterotrophic plankton of inland lake ecosystems. M.Sc. Thesis.
Department of Fisheries and Wildlife, Michigan State University, East Lansing,
MI.

Knoll, LB, 0. Samelle, S.K. Hamilton, C.E.H. Scheele, A.E. Wilson, J.B. Rose and
MR. Morgan. 2007. Invasive zebra mussels (Dreissena polymorpha) increase
cyanobacterial toxin concentrations in low-nutrient lakes. Limnology and
Oceanography. In press.

Kraﬁ, CE, and LE. Johnson. 2000. Regional differences in rates and patterns of North
American inland lake invasions by zebra mussels (Dreissena polymorpha).
Canadian Journal of Fisheries and Aquatic Sciences 57: 993-1001.

Langner, C.L., and RF. Hendrix. 1982. Evaluation of a persulfate digestion method for
particulate nitrogen and phosphorus. Water Research 16: 1451-1454.

Lavrentyev, P.J., W.S. Gardner, J.F. Cavaletto and J .R. Beaver. 1995. Effects of the zebra
mussel (Dreissena polymorpha) on protozoa and the phytoplankton from Saginaw
Bay, Lake Huron. Journal of Great Lakes Research 21: 545-557.

Ludyanskiy, M.L., D. MacDonald, and D. MacNeill. 1993. Impact of the zebra mussel, a
bivalve invader. BioScience 43: 533-544.

Maclsaac, H.J.. 1996. Potential abiotic and biotic impacts of zebra mussels on the inland
waters of North America. American Zoologist 36:287-299.

Maclsaac, H.J., and R. Rocha. 1995. Effects of suspended clay on zebra mussel
(Dreissena polymorpha) faeces and pseudofaeces production. Archiv fur
Hydrobiologie 135: 53-64.

Maclsaac, H.J., C.J. Lonnee, and J .H. Leach. 1995. Suppression of microzooplankton
by zebra mussels: importance of mussel size. Freshwater Biology 34: 379-387.

MacIsaac, H.J., W.G. Sprules, and J .H. Leach. 1991. Ingestion of small-bodied
zooplankton by zebra mussels (Dreissena polymorpha): can cannibalism on
larvae inﬂuence population dynamics. Canadian Journal of Fisheries and Aquatic
Sciences 48: 2051-2060.

Makarewicz, J .C., T.W. Lewis, and P. Bertram. 1999. Phytoplankton composition and
biomass in offshore waters of Lake Erie: pre- and post-Dreissena introduction

(1983-1993). Journal of Great Lakes Research 25: 135-148.

Michigan Sea Grant 2002, 2003. Michigan Sea Grant Zebra Mussel Infestation
Monitoring Program. (http://www.miseagrant.umich.edu/ais/1akes.html)

Mills, E.L., J.M. Casselman, R. Dermott, J .D. Fitzsimons, G. Gal, K.T. Holeck, J.A.

32

Hoyle, O.E. Johannsson, B.F. Lantry, J.C. Makarewicz, E.S. Millard, I.F.
Munawar, M. Munawar, R. O’Gorrnan, R.W. Owens, L.G. Rudstam, T. Schaner,
and T.J. Stewart. 2003. Lake Ontario: food web dynamics in a changing

ecosystem. (1970-2000). Canadian Journal of Fisheries and Aquatic Sciences 60:
471-490.

Mittelbach G.G.. 1984. Predation and resource partitioning in two sunﬁshes
(Centrarchidae). Ecology 65: 499-5 13.

Mittelbach G.G.. 1981. Foraging efﬁciency and body size: a study of optimal diet and
habitat use by bluegills. Ecology 62: 1370-1386.

Nicholls, K.H., and GI. Hopkins. 1993. Recent changes in Lake Erie (North Shore)
phytoplankton: cumulative impacts of phosphorus loading reductions and the
zebra mussel introduction. Journal of Great Lakes Research 19: 637-647.

Noonburg, E.G., B.J. Shuter, and RA Abrams. 2003. Indirect effect of zebra mussels
(Dreissena polymorpha) on the planktonic food web. Canadian Journal of
Fisheries and Aquatic Sciences 60: 1353-1368.

Pace, M.L., S.E.G. Findlay, and D. Fischer. 1998. Effects of an invasive bivalve on the
zooplankton community of the Hudson River. Freshwater Biology 39: 103-116.

Padilla, D.K., S.C. Adolph, K.L. Cottingham and D.W. Schneider. 1996. Predicting the
consequences of dreissenid mussels on a pelagic food web. Ecological Modeling
85: 129-144.

Pauli, HR. 1989. A new method to estimate individual dry weights of rotifers.
Hydrobiologia 186: 355-361.

Pimental, D.. 2005. Aquatic nuisance species in the New York State canal and Hudson
River system and the Great Lakes basin: an economic and enviromnental
assessment. Environmental Management 35: 692-701.

Raikow, DR, 2002. How the feeding ecology of native and exotic mussels affects
freshwater ecosystems. Ph.D. Thesis. Department of Zoology, Michigan State
University, East Lansing, MI.

Raikow, DR, 0. Samelle, A.E. Wilson, and SK. Hamilton. 2004. Dominance of the
noxious cyanobacterium Microcystis aeruginosa in low-nutrient lakes is

associated with exotic zebra mussels. Limnology and Oceanography 49(2): 482-
487.

Roff, J .C., and RR. Hopcroft. 1986. High precision microcomputer based measuring
system for ecological research. Canadian Journal of Fisheries and Aquatic
Sciences 43: 2044-2048.

33

Rutherford, E.S., K.A. Rose, E.L. Mills, J .L. Fomey, C.M. Mayer and LG. Rudstam.
1999. Individual-based model simulations of zebra mussel (Dreissena
polymorpha) induced energy shunt on walleye (Stizostedion vitreum) and yellow

perch (Percaﬂavescens) populations in Oneida Lake, New York. Canadian
Journal of Fisheries and Aquatic Sciences 56: 2148-2160.

Sandgren, CD, and J .V. Robinson. 1984. A stratiﬁed sampling approach to
compensating for non-random sedimentation of phytoplankton cells in inverted
microscope settling chambers. British Phycological Journal 19: 67-72.

Samelle, O.. 1997. Daphnia effects on microzooplankton: comparisons of enclosure and
whole-lake responses. Ecology 7 8: 913-927.

Samelle, O., A.E. Wilson, S.K. Hamilton, L.B. Knoll, and DP. Raikow. 2005. Complex
interactions between the zebra mussel, Dreissena polymorpha, and the harmful
phytoplankter, Microcystis aeruginosa. Limnology and Oceanography 50(3): 896-
904.

Schloesser, D.W., T.F. Nalepa, and G.L. Mackie. 1996. Zebra mussel infestation of
unionid bivalves (Unionidae) in North America. American Zoologist 36: 300-310.

Shevtsova, L.V., G.A. Zhdanova, V.A. Movchan, and AB. Primak. 1986. Experimental
interrelationship between Dreissena and planktonic invertebrates.
Hydrobiological Journal 22(6): 36-39.

Siefert, RE. 1972. First food of the larval yellow perch, white sucker, bluegill, emerald
Shiner, and rainbow smelt. Transactions of the American Fisheries Society 101:
219-225.

Sprung, M., and U. Rose. 1988. Inﬂuence of food size and food quality on the feeding of
the mussel Dreissena polymorpha. Oecologia 77: 526-532

Strayer, D.L., K.A. Hattala, and AW. Kahnle. 2004. Effects of an invasive bivalve
(Dreissena polymorpha) on ﬁsh in the Hudson River estuary. Canadian Journal of
Fisheries and Aquatic Sciences 61: 924-941.

Ten Winkel, E.H., and C. Davids. 1982. Food selection by Dreissena polymorpha Pallas
(mollusca: bivalvia). Freshwater Biology 12: 553-558.

Thorp, J .H., and AF. Casper 2003. Importance of biotic interactions in large rivers: an
experiment with planktivorous ﬁshes, dreissenid mussels and zooplankton in the
St. Laurence River. River Research and Applications 19: 265-279.

Thorp, J .H., and AF. Casper 2002. Potential effects on zooplankton from species shifts in

34

planktivorous mussels: a ﬁeld experiment in the St. Lawrence River. Freshwater
Biology 47: 107-119.

USGS 2007. United States Geological Survey Nonindigenous Aquatic Species Database.
(http://ngs.er.usgs.gov/taxgroup/mollusks/zebramussel/maps/current_zm_quag_m

aging)
USGS 2002, 2003. United States Geological Survey Nonindigenous Aquatic Species

Database. (http://nas.er.usgs.gov/taxgroup/mollusks/zebramusseV)

Vanderploeg, H.A., J .R. Liebig, W.W. Carmichael, M.A. Agy, T.H. Johengen, G.L.
Fahnenstiel, and T.F. Nalepa. 2001. Zebra mussel (Dreissena polymorpha)
selective ﬁltration promoted toxic Microcystis blooms in Saginaw Bay (Lake

Huron) and Lake Erie. Canadian Journal of Fisheries and Aquatic Sciences 58:
1208-1221.

Welschmeyer, NA. 1994. Fluorometric analysis of chlorophyll a in the presence of
chlorophyll b and pheopigrnents. Limnology and Oceanography 39:1985-1992.

Werner, BE, and DJ. Hall. 1988. Ontogenetic habitat shifts in bluegill: the foraging
rate-predation risk trade-off. Ecology 69: 1352-1366.

Werner, G.H.. 1967. Intralacustrine movements of bluegill fry in Crane Lake, Indiana.
Transactions of the American Fisheries Society 96: 416-420.

Wilkinson, L.. 1989. Systat: the system for statistics. Evanston, 822 pp.

Wilson A.E.. 2003. Effects of zebra mussels on phytoplankton and ciliates: a ﬁeld
mesocosm experiment. Journal of Plankton Research 25(8): 905-915.

35

CHAPTER 2

THE EFFECT OF ZEBRA MUSSELS (DREISSENA POL YMORPHA) ON THE
GROWTH RATE AND DIET COMPOSITION OF BLUEGILL SUNFISH

(LEPOMIS MA CROCHIRUS) IN INLAND MICHIGAN LAKES

Carrie E.H. Scheele', Lesley B. Knollz’ 5, Alan E. Wilson3, Orlando Sarnellez, and

Stephen K. Hamilton4

1Department of Zoology, Michigan State University, East Lansing, MI 48824-1222

2Department of Fisheries and Wildlife, Michigan State University, East Lansing, MI
48824-1222

3 Cooperative Institute for Limnology and Ecosystems Research, Ann Arbor, Michigan
48105

4WK. Kellogg Biological Station, Michigan State University, Hickory Corners, MI
49060-9516

5Current address, Department of Zoology, Miami University, Oxford, OH 45056

36

Abstract

The zebra mussel (Dreissena polymorpha) has drastically transformed large
freshwater ecosystems since its invasion of the Laurentian Great Lakes in 1988. Zebra
mussels are now invading small inland lakes, but less is known about their effects on
these systems. We conducted a survey of 50 southern Michigan lakes with similar
nutrient concentrations and morphometries to examine the indirect impacts of zebra
mussels on the grth rate and diet composition of bluegill sunﬁsh (Lepomis
macrochirus). Twenty-ﬁve lakes contained zebra mussels (invaded), and 25 lakes did not
(uninvaded). Light extinction coefﬁcients were 21% higher in invaded lakes, while
phytoplankton, microzooplankton, and macrozooplankton biomass were signiﬁcantly
lower (24%, 44%, and 40%, respectively) in the presence of zebra mussels. Zebra
mussels signiﬁcantly affected grth rates of bluegill sunﬁsh, and the effect changed
depending on the life history stage of the ﬁsh. The grth of ﬁrst year bluegill was lower
in invaded lakes likely due to low microzooplankton abundance while in the larval
growth stage, and the inability for growth to catch up once they move back into the
littoral zone later that summer. In contrast, juveniles and adults grew an additional 1.8-
4.2 mm per year in invaded lakes. Higher growth of juveniles and adults in invaded lakes
was unexpected. Through stomach content and stable isotope analyses, we show evidence
that adult bluegill may have switched their diet in invaded lakes to include more benthic
macroinvertebrates. Juveniles most likely took advantage of increased benthic
invertebrate biomass promoted by zebra mussels in invaded lakes. The most likely
mechanism for higher adult grth rates was diet supplementation with benthic

invertebrates that are promoted by zebra mussel presence in lakes. We also observed a

37

signiﬁcant positive effect of zebra mussels on adult bluegill mean length at age in a
subset of 24 lakes (11 invaded, 13 uninvaded) in 2002-2003, yet no effect was detected in
these same lakes and treatment groups prior to invasion (< 1988). These pre- and post-
invasion data support the zebra mussel effect on bluegill growth in 2002-2003. Overall,
our results provide no evidence of a negative zebra mussel effect on juvenile and adult
bluegill growth, even though we do show evidence that bluegills may have switched their

diet. Thus, zebra mussels do not contribute to bluegill stunting.

38

Introduction

The zebra mussel (Dreissena polymorpha Pallas) has invaded the United States
and is rapidly spreading throughout freshwater systems. Economic losses in the Great
Lakes basin due to damage and control costs are estimated at $500 million (US) each
year (Pimentel 2005). First detected in Lake St. Clair, Michigan, USA, in 1988 (Herbert
et al. 1989), zebra mussels quickly established populations in all ﬁve of the Great Lakes
and several major river systems (e. g. Hudson, Mississippi, and Ohio Rivers) (Ludyanskiy
et al. 1993), and soon began to invade smaller inland lakes (Kraft and Johnson 2000).

Zebra mussels are efﬁcient ﬁlter feeders that consume large quantities of algae
and small zooplankton. Phytoplankton biomass usually declines following zebra mussel
invasion and water clarity increases (Idrisi et al. 2001, Caraco et a1. 1997, Nicholls and
Hopkins 1993). At the same time, zebra mussels can promote blooms of the toxic
colonial cyanobacterium, Microcystis aeruginosa, in low nutrient lakes (Knoll et al. 2007
in press, Sarnelle et al. 2005, Knoll 2004, Raikow et al. 2004, Vanderploeg et al. 2001).

Due to their efﬁcient ﬁltering and production of feces and pseudofeces, zebra
mussels can divert energy from pelagic to benthic communities (Maclsaac 1996),
enhancing nutrient ﬂuxes to the benthos (Conroy et al. 2005), and increasing benthic
algae and benthic primary productivity (F ahnenstiel et al. 1995, Lowe and Pillsbury
1995). Additionally, the increased habitat heterogeneity created from their colonies
coupled with the production of nutrient rich feces and pseudofeces results in increased
biomass of benthic macroinvertebrates (Stewart et al. 1998a, b, Ricciardi et al. 1997,

Thayer et al. 1997, Stewart and Haynes 1994).

39

Zooplankton dynamics can be affected both directly and indirectly by zebra
mussels. Microzooplankton are directly consumed by zebra mussels resulting in
decreased biomass of this trophic level (Pace et al. 1998). Most macrozooplankton are
too large to be consumed by zebra mussels (Maclsaac et al. 1995, Maclsaac et al. 1991;
but see Shevtsova et al. 1986), but zebra mussels indirectly affect abundance (Thorp and
Casper 2003, Bridgeman et al. 1995), and fecundity (Horgan and Mills 1999) of small
macrozooplankton species by reducing food availability. Decreased macrozooplankton
abundance, in turn, affects species in higher trophic levels in the food web, including ﬁsh
(Rutherford et al. 1999).

Studies of the effects of zebra mussels on ﬁsh growth in lakes, rivers, and
experimental enclosures are few and often contradict each other. Fish growth in the
presence of zebra mussels decreased, increased, or remained the same for species in
different life stages and ecosystems (Table 1). This variation may also result from the
multitude of indirect pathways through which ﬁsh may be affected by zebra mussels in
complex aquatic food webs (Strayer et al. 2004).

Bluegill sunﬁsh (Lepomis macrochirus) is the dominant ﬁsh species in southern
Michigan inland lakes (Werner et al. 1978). Bluegills undergo ontogenetic niche shifts
twice during their life histories (Werner and Hall 1988). As larvae [< 14 mm standard
length (SL)], they forage on small zooplankton in the pelagic zone. Juveniles (20-79 mm
SL) forage in the vegetation of the littoral zone to avoid predation from largemouth bass
(Micropterus salmoides) and feed mainly on invertebrates. Adults (> 80 mm SL) are
large enough to avoid predation and feed extensively on macrozooplankton (Daphnia)

that are abundant in the water column in the pelagic zone (Werner and Hall 1988;

40

Mittelbach 1984, 1981). Reduction of macrozooplankton abundance by zebra mussels
could limit food for bluegills, substantially decrease adult growth (Osenberg et al. 1988),
and lead to lake populations dominated by stunted adults (Gerking 1962, Swingle and
Smith 1942). Bluegill population size structures skewed toward adults < 150 mm are
considered a primary management concern in the Midwestern U.S. (Drake et al. 1997).
Recently, Raikow (2004) found that zebra mussels reduced the growth rates of larval
bluegills in experimental enclosures. He attributed this to fewer microzooplankton in the
presence of zebra mussels and to competition between zebra mussels and larval bluegills
for food. Thus, if zebra mussels negatively affect larval bluegill growth, it is possible that
they could affect the growth of juveniles and adults.

We conducted an extensive survey of inland Michigan lakes and addressed 3 main
questions. 1) Are there differences in bluegill grth rates in inland Michigan lakes with
and without zebra mussels? 2) Is there a detectable change in bluegill mean length at age
in lakes containing zebra mussels compared to historical data from the same lakes prior to
zebra mussel infestation? 3) Do adult bluegill sunﬁsh alter their diet in the presence of

zebra mussels?

Methods
Lake Selection Criteria
We selected 50 lakes for study, 25 invaded by zebra mussels and 25 uninvaded
reference lakes. Lakes were located in the glacial terrain of southern Michigan, USA.
Invaded and uninvaded lakes were nearly equally balanced in the southwest and southeast

regions of the state. The lakes had low to moderate total phosphorus (TP) concentrations

41

(<22 ng"; Knoll et al. 2007 in press, Michigan Department of Environmental Quality).
All lakes selected for this study were 2 9 m in maximum depth, to ensure summer
temperature stratiﬁcation. We calculated mean depth of all lakes from digitized
bathymetric maps (ESRI 1999) to ensure that treatment and control groups were of
similar depths. Where lake maps were unavailable, we obtained mean depth data from the
Michigan Department of Environmental Quality (unpublished data). We acquired zebra
mussel presence or absence data from the United States Geological Survey
Nonindigenous Aquatic Species database (U SGS 2002, 2003) and the Michigan Sea
Grant Zebra Mussel Infestation Monitoring Program (Michigan Sea Grant 2002, 2003).
Lake sampling 2002-2003

Lakes were sampled once between either 3 August and 5 September 2002, or 16
July and 23 August 2003. In 2002, we sampled 26 lakes (N uninvaded = 14, N invaded =
12) and in 2003, we sampled 24 lakes (N uninvaded = 11, N invaded = 13). Zebra mussel
presence or absence was conﬁrmed during sampling by wading near the boat launch and
adjacent shallow areas along the shoreline for approximately 15-20 min, looking for
zebra mussels in the sediment or attached to other organisms.
Physical and Biological Parameters 2002-2003

We sampled from the deepest part of each lake, as determined from bathymetric
maps and an echosounder. We measured temperature and dissolved oxygen from the
water column at 1 m depth intervals using a Hydrolab Surveyor 4a equipped with a
Datasonde 4a (HACH Environmental, Loveland, CO). Photosynthetically active
radiation [PAR] was measured at 0.5 m depth intervals using a LiCor model Li-1000

quantum photometer with an attached spherical quantum underwater sensor and

42

corresponding deck sensor (LiCor Environmental, Lincoln, NE). Light extinction
coefﬁcients were determined as the slope of the linear regression between 1n PAR and
depth. We collected an integrated epilimnetic water sample from the surface to the
bottom of the mixed layer, using a ﬂexible plastic tube (5 cm inner diameter, 10 m
length). Two to four mixed layer samples were collected and pooled in a large container,
and sub samples were taken for phytoplankton biomass and stable isotope analysis,
chlorophyll a, TP and microzooplankton. Phytoplankton biomass samples were preserved
immediately in Lugol’s solution (Hasle 1978). Mixed-layer water for phytoplankton
stable isotope samples, chlorophyll a, and TP was put on ice and processed later in the
same day (~6 h). Seston for analysis of stable isotopes, and chlorophyll a was collected
on Gehnan A/E glass ﬁber ﬁlters (Gehnan Sciences, Ann Arbor, MI), that were frozen
until laboratory analysis. Microzooplankton were collected by passing a 10 L sub sample
of the mixed-layer water through a 35 pm mesh screen and rinsing organisms on the
screen into sample bottles containing glutaraldehyde (ﬁnal concentration: 2%).
Macrozooplankton samples were also obtained from the deepest part of the lake. The
zooplankton net (30 cm diameter, 100 pm mesh) was lowered to ~1 m above the lake
bottom and hauled through water column. Four hauls were pooled from each lake and
preserved with 95% ethanol for macrozooplankton biomass analysis. An additional two
tows were pooled from each lake, put on ice, and brought back to the lab for stable
isotope analysis. Zebra mussels, snails and sediment samples for stable isotope analysis
were collected from the littoral zone and frozen until analysis.

We measured chlorophyll a by ﬁrst extracting it from the glass ﬁber ﬁlters with

90% ethanol and then analyzing the extract with a Turner Model 10-AU ﬂuorometer

43

(Welschmeyer 1994) calibrated to commercial chlorophyll a standards (Anacystis;
Sigrna-Aldrich Chemical Company, St. Louis, MO). Total phosphorus samples were
oxidized via persulfate digestion in an autoclave and then analyzed using the molybdate
blue colorimetric method (Langner and Hendrix 1982).
Phytoplankton Biomass 2002-2003

Phytoplankton biomass was assessed in a randomly chosen subset of survey lakes
(N uninvaded = 20, N invaded = 22; Knoll et al. 2007 in press). Phytoplankton were
identiﬁed to species in most cases and enumerated via the inverted microscope technique
(Hasle 1978). Biovolume was determined from measurements of cell dimensions of at
least ten individuals of common species in each sample at 1000x magniﬁcation using a
NIKON model TE2000-S inverted microscope (NIKON, Melville NY), a SPOT insight
QE model 4.2 digital camera and SPOT Advanced version 4.0.9 image-analysis software
(Diagnostic Instruments Inc., Sterling Heights, MI). Biovolume was converted to dry
biomass assuming a speciﬁc gravity of 1 g cm'3 and a dry mass to wet mass ratio of 0. 1 0.
Microzooplankton Biomass 2002-2003

Ciliate biomass was assessed in all 50 study lakes (N uninvaded = 25, N invaded
= 25). Ciliates were enumerated with the inverted microscope technique using a NIKON
model TE2000—S inverted microscope. Sub samples (sub sample volume 30-100 ml) were
settled in tubular chambers (Hydro-Bios, Kiel-Holtenau, Germany), the bottoms of which
were divided into inner and outer zones of equal area (Sandgren and Robinson 1984).
Within each zone, at least 20 random ﬁelds were counted at 100x magniﬁcation. Ciliate
cell volume was determined by measuring at least ﬁve individuals per taxon at 400x

magniﬁcation using the imagine system described above. Biovolume was converted to

44

dry biomass assuming a speciﬁc gravity of 1 g cm'3 and a dry mass to wet mass ratio of
0.10.

Rotifer biomass was assessed in all 50 study lakes. Rotifers were identiﬁed to
species using a Nikon model E600 compound microscope (NIKON, Melville NY) at
100x magniﬁcation using a SPOT insight Color model 3.2.0 digital camera and SPOT
Advanced version 4.0.9 image-analysis software (Diagnostic Instruments Inc., Sterling
Heights, MI), and a Sedgwick-Rafter counting chamber. For each lake, approximately
400 individuals (average = 394, range = 341-475) were counted in a minimum of 2 sub
samples. Individual dry biomass for each species was estimated from established
literature values (Pauli 1989).

Macrozooplankton Biomass 2002-2003

Macrozooplankton biomass was assessed in all 50 study lakes. Macrozooplankton
were counted and identiﬁed to genus or species using a 10 ml clear PVC zooplankton
counting wheel and a Leica model MZ8 dissecting microscope (Leica Microsystems,
Inc., Bannockbum, IL). Samples were diluted to a known volume and two to three 5 ml
sub samples were counted. A minimum of 450 individuals were tallied per lake.
Measurements of individuals were made at a magniﬁcation of 10x, using a Summa
Sketch III digitizing pad and ZoopBiom software (Roff & Hopcroft 1986). For each
sample, up to 50 individuals were measured for each large (> 1.0 mm) genus or species
(Daphnia galeata, D. pulicaria, D. retrocurva, Epischura spp., Leptadora spp., and
Mesocyclops spp.) and up to 25 individuals were measured for each small (< 1.0 mm)
genus or species (Alonella spp., Bosmina spp., Ceriodaphnia spp., Cyclops spp.,

Diaphanasoma spp., Diaptomus spp., Dreissena polymorpha veligers, Moina spp., and

45

nauplii). Biomass was calculated using published length-weight regressions for
individual species (Culver et al. 1985, Dumont et al. 1975).
Bluegill Growth 2002-2003

We collected bluegill sunﬁsh from several sites in the littoral zone of each lake
using a beach seine, a cast net, or angling. We weighed and measured approximately 100
bluegill per lake, including adults [2 90 mm total length (TL)], juveniles (40 - 90 mm‘TL)
and ﬁrst year (< 40 mm TL). For age and growth determination, we obtained 5 to 10
scales from each of the ﬁrst 50 bluegills sampled. We removed the scales from the left
side of the body, just posterior of the tip of the depressed pectoral ﬁn.

To determine ﬁsh age, we pressed 6 to 10 scales from each of the 50 bluegills
sampled per lake between two microscope slides. We counted the number of annuli
present on one randomly selected scale per slide and measured the distances from the
focus to each annulus and the scale margin using a Nikon Eclipse E600 dissecting
microscope (NIKON, Melville, NY), and Optimus 6.0 software (Media Cybernetics,
Silver Spring, MD).

Bluegill growth was back calculated using the Scale Proportional Method
(Whitney and Carlander 1956):

T14- : (-a/b) + (TLc + (a/b))(S.-/Sc)

TL.- = estimated total length of a ﬁsh (mm) at the formation of the ith scale mark, for i = l,
2. . .n; TLc = total ﬁsh length (mm) at the time of capture; S, = scale radius (mm) at the
formation of the ith scale mark, for i = 1, 2. . .n; and Sc = scale radius (mm) at time of
capture. We determined the intercept (a) and slope (b) from the linear function describing

the relationship between Sc and TLC:

46

sc = 0.0228 (TLC) — 0.4266, n = 3168, R2 = 0.938.
Annual growth rates, or the change in length from one annulus to the next, were
expressed as:

ATL=TLx+1—TLx
Growth rates were expressed as a function of the total ﬁsh length (mm) at the start of the
year’s growth stanza (TLX) in the year before capture. Bluegill were then placed into 3
life history categories according to their length (mm) at the start of the year’s growth
stanza: adults (2 90 mm TL), juveniles (40 - 90 mm TL) and ﬁrst year (< 40 mm TL)
(Werner and Hall 1988; Mittelbach 1984, 1981). Note that ﬁrst year bluegill (age 1 + at
capture) encompass both the larval open water stage where they feed on small
zooplankton, and a small juvenile stage that includes feeding on macroinvertebrates in
the littoral zone.
Historical vs. Contemporary Bluegill Mean Length at Age

Historical data on bluegill growth was not available for the study lakes. However,
we examined historical bluegill mean length at age data for 24 of the lakes in our study
(N uninvaded = 13, N invaded = 11) (Michigan Department of Natural Resources
Fisheries Division, unpublished data). These data were used as a baseline to determine if
mean length at age in the lake groups changed after zebra mussel infestation. We
compared mean length at age of juvenile (ages 1-2) and adult bluegill (ages 3-6) between
lake groups and within individual time periods (< 1988 and 2002-2003) to be consistent
with the 2002-2003 growth analyses. Historical ﬁrst year bluegill data were not available
and therefore we were unable to make comparisons for that life history category. Because

actual zebra mussel infestation dates for our study lakes are not available, all historical

47

data used in our comparisons were from June-September (1956-1987), prior to the ﬁrst
zebra mussel sightings in the US in 1988. The methods and gear used to collect the
bluegill for the historical data varied from electroﬁshing, encircling gill nets, gill nets,
trap nets, fyke nets, seines, and angling. Many scientists from the MI-DNR were
involved in collecting and analyzing the historical data. In cases when there were
historical data ﬁom more than one year from the same lake, these data were averaged to

make one overall historical mean length at each age for that lake.

Contemporary bluegill mean length at age data were obtained for the same 24
lakes using the 2002-2003 bluegill dataset. The contemporary data were arranged in the
same manner as the historical mean length at age data and allowed for comparison

between the two lake groups in both the historical and contemporary time periods.
Bluegill Diet 2002-2003

Thirty adult bluegills (> 90 mm TL) were sacriﬁced per lake for stomach content
analyses. In the ﬁeld, sacriﬁced adults were killed using a 1 g/L water dose of MS-222
and then put on ice to slow down digestion. In the laboratory, the ﬁsh were thawed, and
the stomachs removed and preserved in 95% ethanol. Stomach content analyses were
performed for 26 of the 50 study lakes (N uninvaded = 13, N invaded = 13). Ten

randomly chosen adult bluegill stomachs were examined per lake.

We employed two different methods for determining the biomass of prey items in
bluegill stomachs. For both biomass methods, stomach contents were counted using a
Leica model MSS dissecting microscope (Leica Microsystems, Inc., Bannockbum, IL).
The prey were identiﬁed to family or genus [Daphnia spp., cladocerans (without Daphnia

spp.), copepods, snails, clams, zebra mussels, mites, ostracods, beetles, true bugs,

48

caddisﬂy larvae, Chironomus larvae, amphipods, springtails, damselﬂy larvae, dragonﬂy
larvae, stoneﬂy larvae, mayﬂy larvae, small worms, earthworms, moth larvae,
Chaoborous spp., mosquito larvae, and wasps]. In the ﬁrst method, measurements of up
to 50 individuals of each prey were taken under a magniﬁcation of 10x, using a Summa
Sketch III digitizing pad and ZoopBiom software (Roff & Hopcroft 1986). Biomass was
calculated using published length-weight regressions for bluegill prey items found in
three lakes habitats: 1) open water [Daphnia spp., cladocerans (without Daphnia spp.),
and copepods], 2) bare sediment (Chiromomus larvae) and 3) vegetation (all other taxa)
(Mittelbach 1981). This direct method of determining stomach content biomass was

employed for 6 of the 26 lakes.

The second method we was used was a multiple-regression technique to estimate
the mean weight of prey taxa in ﬁsh stomachs, developed by Hayes and Taylor (1991).
Stomach contents were removed from the stomach, weighed, and abundance of each prey
taxon was recorded. The multiple-regression technique regressed the total weight of the

stomach contents against the counts of the individual prey taxon present, where:
SW,-=a1x,-1+a 2132+ +a}x,-j+e,-

SW,- = total weight of stomach contents from ﬁsh i; a,- = regression coefﬁcient that
becomes an estimate of the mean weight per individual of prey taxon j; xij = number of
prey j in stomach i; and e,- = random error for stomach i. The mean weight attributable to
each taxon for the entire sample was determined by multiplying the mean count of each

taxon by the regression coefﬁcient (aj) for that group:

szaij';

49

WJ- = mean weight attributable to prey j; and 5?,- = mean number of prey item j per

stomach.
The variance of estimates of Wj was determined by:
var(ab) = [azvar(b)] + [bzvar(a)],

assuming that f,- and a,- are independent variables. The 24 prey taxa were reduced into
11 categories based on similar organism size to reduce zeros in the dataset [1. Daphnia
spp., 2. snails, clams and zebra mussels, 3. mites and ostracods, 4. beetles and true bugs,
5. arnphipods and springtails, 6. damselﬂy larvae, dragonﬂy larvae, stoneﬂy larvae and
mayﬂy larvae, 7. small worms, moth larvae, Chaoborous spp. and wasps, 8. caddisﬂy
larvae and Chironomus larvae, 9. cladocerans (without Daphnia spp.) and copepods, 10.
mosquito larvae and 11. earthworm]. The multiple-regression successﬁrlly estimated the
average weight ofthe 11 stomach content taxa (R2 = 0.85, F1 1,139 = 100.95, p < 0.001)
(Figure 1). Estimates for each taxon are reported in Table 2. The regression method was

used for 20 of the 26 lakes that were analyzed for stomach contents.

Carbon and nitrogen stable isotope analyses were performed on select food web
compartments in 29 of the 50 study lakes (N uninvaded = 12, N invaded = 19) to
determine if bluegill diet changed in invaded lakes. Twenty of these lakes (N uninvaded
= 10, N invaded = 10) were also analyzed for bluegill stomach contents. Approximately
1 cm3 of muscle tissue was removed from 3 adult bluegill (size range of 100 - 110 mm
TL), per lake. The tissue was removed from the left side of the body, just adjacent to the
dorsal ﬁn and placed into a folded weighing tin. Zooplankton was concentrated onto
Gelman A/E glass ﬁber ﬁlters (Gehnan Sciences, Ann Arbor, MI) and non zooplankton

species were removed manually. Phytoplankton samples were ﬁltered onto A/E glass

50

ﬁber ﬁlters until the ﬁlter clogged and the volume ﬁltered was recorded. Zebra mussel
and snail tissue was removed ﬁ'om the shell and placed into a folded weighing tin.
Sediment samples were also placed into a folded weighing tin. All samples were dried at

55 °C for two days.

Phytoplankton and zooplankton samples on ﬁlters were placed into tin capsules
and ground. Muscle tissues from bluegill, zebra mussels and snails, and the dried
sediment samples were also placed into tins and ground. Each ground sample was split in
half for replicate analysis, with one replicate treated with dilute HCl to remove
carbonates. Isotope measurements were made by continuous-ﬂow isotope-ratio mass
spectrometry using a Micromass Optima interfaced to a Carlo Erba NC 2500 elemental
analyzer for on-line combustion and puriﬁcation of sample nitrogen and carbon. All
isotope abundances are expressed as 515N or 513 C values relative to atmospheric nitrogen
and Pee Dee Belemnite carbon standards, respectively. We compared the acidiﬁed SN
and 813C values to those that were not acidiﬁed from a subset of the 29 lakes and
determined that all animal values were similar (A.E. Wilson unpublished data).
Additionally it has been shown that acidifying animal tissues can alter 515N and 513C
values, potentially confounding interpretations of food webs (Bunn et al. 1995). Thus, all
stable isotope values presented in this study were unacidifed, except for the
phytoplankton and sediment samples, which potentially contain inorganic carbonates
(Jackson et a1 1986, Rau et a1. 1983). Bluegill and zooplankton samples were analyzed
from all 29 lakes, sediment and phytoplankton were analyzed from 6 lakes (N uninvaded
= 1, N invaded = 5) and snails and zebra mussels were only analyzed from 5 lakes (N

uninvaded = 0, N invaded = 5).

51

Statistical Analyses

We compared physical and biological parameters for zebra mussel invaded and
uninvaded lakes using the Student’s t-test, when data were normal and had equal
variances between treatments. Whenever data did not meet the assumption of normality
(Kolmogorov-Smirrnov and Shapiro-Wilk tests for normality p < 0.05), the data were log
transformed. If the log transformation was successful to meet the normality assumption,
the log transformed data were t-tested. If the log transformation was not successful, the
untransformed data were tested using the Mann-Whitney U test (Systat®11, Wilkinson,
1989)

We used analysis of covariance (lme, R Version 2.4.1, R Development Core
Team 2007) to compare bluegill grth during the year prior to sampling between
invaded and uninvaded lakes. Length at the start of the year’s growth stanza in the year
before capture was the covariate in the model, lake was a random effect, and zebra
mussel treatment (invaded vs. uninvaded) was the ﬁxed effect of interest. We compared
contemporary and historical bluegill mean length at age using analysis of covariance
(PROC MIXED, SAS 2001) with age as the covariate, lake as a random effect, and zebra
mussel treatment and time period (< 1988 or 2002-2003) as the ﬁxed effects.

To compare the percent zooplankton found in bluegill stomachs in invaded and
uninvaded lakes, we used a linear mixed effects model (lme, R Version 2.4.1, R
Development Core Team 2007) with lake as the random factor and zebra mussel
treatment as the ﬁxed effect. In order to meet the assumptions of the model, the variable
percent zooplankton was arcsin square root transformed. We used analysis of covariance

(lme, R Version 2.4.1, R Development Core Team 2007) to compare bluegill stomach

52

weight between invaded and uninvaded lakes. Length at capture was the covariate in the
model, lake was a random effect, and zebra mussel treatment was the ﬁxed effect. To
meet the assumptions of the model, the variable stomach weight was log transformed. To
determine if the changes between zooplankton and bluegill 6'5 N or 813 C values were
different between treatments, we used linear mixed effects models (lrne, R Version 2.4.1,
R Development Core Team 2007) with lake as the random factor and zebra mussel
treatment as the ﬁxed effect. All assumptions of mixed effect models were satisﬁed in all

analyses.

Results

Physical and Biological Parameters 2002-2003

Mean depths and TP concentration in invaded and uninvaded lakes were not
signiﬁcantly different (1125,25 = 241, p = 0.17; t43 = -0.32, p = 0.75, respectively) (Table
3), indicating no bias in the selection of sample lakes with respect to these parameters.
Light extinction coefﬁcients were signiﬁcantly higher (t43 = -2.93, p = 0.01) in invaded
lakes, indicating greater penetration of PAR in the water column and greater water
clarity. Interestingly, chlorophyll (1 did not differ between the lake groups (t43 = 1.68, p =
0.10) (Table 3). Total phytoplankton, total microzooplankton, total macrozooplankton,
and Daphnia spp. biomass all were signiﬁcantly lower in invaded lakes (t4o = 2.30, p =
0.03; t43 = 3.31, p = 0.002; t4g = 2.22, p = 0.03; U25,” = 422, p = 0.03; respectively)

(Table 3).

53

Bluegill Growth in 2002-2003

A frequency distribution of bluegill length at the beginning of the ﬁnal year of
growth in invaded and uninvaded lakes revealed similar numbers of bluegill in each 10
mm size class (Figure 2). We then divided the bluegill into three life history categories
based on ontogenetic diet shifts (ﬁrst year < 40 mm, juvenile = 40 - 89 mm, and adult 2
90 mm) (Werner and Hall 1988; Mittelbach 1984, 1981). Note that ﬁrst year bluegill (age
1 + at capture) encompass both the larval open water stage where they feed on small
zooplankton, and a small juvenile stage that includes feeding on macroinvertebrates in
the littoral zone. Zebra mussels signiﬁcantly affected bluegill grth (F 1,43 = 5.74, p =
0.02), but the direction of this effect changed depending on the life history category
(F 2,2413 = 5.52, p = 0.004) (Table 4). First year grth was lower, yet juvenile and adult
growth was higher in zebra mussel lakes (Figure 3).
Historical vs. Contemporary Bluegill Mean Length at Age

Mean length at age of juvenile bluegills (ages 1-2) in the two lake groups was not
signiﬁcantly different before zebra mussel invasion (F1,14 = 0.33, p = 0.57) (Figure 4,
Table 6). Juvenile mean length at age in 2002-2003 was also not signiﬁcantly different
between invaded and uninvaded lakes (F131 = 2.76, p = 0.11) (Figure 4, Table 7).
However, mean length at age was signiﬁcantly greater in juvenile bluegill prior to 1988,
indicating that juveniles were historically larger at both ages 1 and 2 compared to
contemporary juveniles (F156 = 53.37, p < 0.001) (Figure 4, Table 5). Prior to zebra
mussel invasion, adult bluegill mean length at age (ages 3-6) did not differ signiﬁcantly
in the two lake groups (F 1,62 = 0.69, p = 0.41) (Figure 4, Table 9). Mean length at age of

adults was signiﬁcantly greater in invaded lakes in 2002-2003 (F137 = 6.37, p = 0.02)

54

(Figure 4, Table 10). Mean length at age was signiﬁcantly greater in adults prior to 1988,
indicating that before 1988, adults were larger at all ages compared to 2002-2003 (Fuzz =
253.2, p < 0.001) (Figure 4, Table 8).
Bluegill Diet 2002-2003

Stomach content analyses from 26 lakes revealed a non-signiﬁcant pattern that
bluegill in invaded lakes had 65% less zooplankton in their diets (F134 = 2.24, p = 0.15)
(Figure 5, Table 11). Stomach content weight did not differ between invaded and
uninvaded lakes, indicating that bluegill are eating similar quantities of food (F 1,24 = 0.60,
p = 0.44) (Figure 6, Table 12). Stable isotopes of SN and 513C from the food webs of 29
lakes show that bluegill have different carbon and nitrogen signatures in invaded and
uninvaded lakes (Figure 7). The 15N trophic enrichment (as determined by the change in
BUN) from zooplankton to bluegill in invaded lakes was signiﬁcantly lower (F127 = 6.70
p = 0.02) (Figure 8, Table 13), indicating that bluegill in invaded lakes eat organisms
from an isotopically distinct basal food resource. The distance from zooplankton to
bluegill in terms of 513 C is greater, yet not-signiﬁcant, suggesting that bluegill in invaded
lakes may be incorporating less 13 C ﬁ'om zooplankton and more from benthic

macroinvertebrates (F137 = 2.19, p = 0.15) (Figure 9, Table 14).

Discussion
Bluegill Growth in 2002-2003
Higher juvenile and adult bluegill grth in zebra mussel invaded lakes was
unexpected. We postulated that zebra mussels, because they are efﬁcient ﬁlter feeders,

would have a bottom-up effect on lake ecosystems. We predicted this would lead to

55

lower phytoplankton and zooplankton biomass (as seen in Table 3), and ultimately,
slower bluegill growth. Our results only supported this prediction for ﬁrst year bluegill;
grth of juvenile and adult bluegills was slightly higher in lakes with zebra mussels.
Even though our bluegill growth results were surprising, prey abundance and
bluegill feeding behavior may be responsible for the detected patterns. First year bluegill
feed on microzooplankton (copepod nauplii, rotifers, and small-bodied cladocerans) in
the pelagic zone of the lake (Bremigan and Stein 1994, Barkoh and Modde 1987, Siefert
1972, Werner 1967) for the ﬁrst portion of the summer, and then switch littoral benthic
macroinvertebrates by late summer. Because zebra mussels reduced the abundance of
microzooplankton by 44% and macrozooplankton by 33% (Table 3), ﬁrst year bluegill
would be expected to grow slower due to resource competition for the ﬁrst portion of
summer. Our ﬁrst year results are consistent with Raikow (2004) who found a 24%
decline in larval bluegill grth in the presence of zebra mussels. However, once ﬁrst
year bluegill become vulnerable to predation by bass, they move back to the littoral zone
and feed on a variety of benthic invertebrates (Turner and Mittelbach 1990, Werner and
Hall 1988, Werner et al. 1983, Mittelbach 1981). We did not sample the benthos;
however, it has been shown that zebra mussels increase benthic invertebrate biomass
(Beekey et al. 2004, Stewart et al. 1998a, b, Ricciardi et al. 1997, Thayer et al. 1997,
Stewart and Haynes 1994) and productivity (J ohannsson et al. 2000). Thayer et al. (1997)
found that adult yellow perch, which are primarily benthivores during this life history
stage, grew signiﬁcantly larger in the presence of zebra mussels due to elevated benthic
invertebrate biomass. Therefore, it is likely that ﬁrst year bluegill growth is depressed in

the beginning of the summer due to resource competition with zebra mussels for small

56

zooplankton, but after they move back to the littoral zone, their growth would be greater
due to increased benthic macroinvertebrate biomass. We still saw lower growth in ﬁrst
year bluegill in invaded lakes, which suggests that growth may have been much lower
while ﬁrst year bluegill were still in the larval stage, and the higher grth while in the
littoral zone did not completely make up for the initial slow growth.

Juveniles also feed on a variety of benthic macroinvertebrates in the littoral zone
(Turner and Mittelbach 1990, Werner and Hall 1988, Werner et al. 1983, Mittelbach
1981). Because zebra mussels promote benthic macroinvertebrate biomass and benthic
production, juvenile bluegill grth should also be greater in the presence of zebra
mussels. Thus our result of higher juvenile growth in invaded lakes due to increased
benthos is a reasonable conclusion.

Adult bluegill return to the pelagic zone of the lake once they are large enough to
avoid predation, and may feed primarily on Daphnia spp. (Mittelbach and Osenberg
1993, Werner and Hall 1988; Mittelbach 1984, 1981). We found signiﬁcantly lower
Daphnia spp. (40%) and macrozooplankton biomass (33%) in invaded lakes (Table 3)
and expected the growth of adults to be lower. Thus, the higher growth of adults in
invaded lakes was surprising. Potential reasons for higher adult bluegill growth in
invaded lakes include increased water clarity and diet shifts.

Bluegill sunﬁsh are visual predators that actively seek their prey in the water
column (Breck and Gitter 1983, Vinyard and O’Brien 1976). In dense vegetation where
their vision is limited, bluegill foraging efﬁciency declines (Harrel and Dibble 2001).
Therefore it is possible that bluegill were better able to see their zooplankton prey

because the water clarity of invaded lakes was 21% higher (Table 3); thus, swimming

57

effort needed to search for prey may decrease and prey encounter rates and foraging
efﬁciency should increase, thereby increasing energy available for growth (Werner and
Hall 1988; Mittelbach 1984, 1981; Vineyard and O’Brien 1976; but see O’Brien et al.
1976). However, because we did not see increased biomass of zooplankton in the
stomach contents of adults in invaded lakes (Figure 5), it is unlikely that water clarity is
the mechanism for higher adult growth.

Zebra mussels promoted increased abundance of benthic organisms in Lake
Ontario, Western Lake Erie, Lake Champlain, NY, USA, and enclosures in an
experimental pond adjacent to Lake St. Clair, MI, USA (Stewart and Haynes 1994,
Stewart et al. 1998a, b, Beekey et al. 2004, Thayer et al. 1997, respectively), increased
density of macrozoobenthic organisms in the shallow areas of the Hudson River (Strayer
et a1. 1998), and increased benthic production in Lake Erie (J ohannsson et al. 2000).
Adult bluegills are not obligate planktivores and will feed on many different food items
(Werner and Hall 1988; Mittelbach 1984, 1983, 1981). However, when it is not
energetically proﬁtable to eat large Daphnia, bluegill will switch to more proﬁtable items
(Mittelbach 1983). We show that zebra mussels signiﬁcantly lowered macrozooplankton
and Daphnia, the preferred food items of adult bluegill. Thus, it is possible that bluegill
were forced to switch their diets to the more abundant benthic organisms when
zooplankton prey was scarce.

Stable isotope evidence suggests that adult bluegill in invaded lakes have changed
their diet. Stable isotope analyses have been used to characterize various aquatic food
webs (Vander Zanden and Rasmussen 1999, Hamilton et al. 1992, Kling et al. 1992,

Sholto-Douglas et al. 1991). More recently, they have been used to determine food web

58

responses to exotic species invasions (Maguire and Grey 2006, Vander Zanden et al.
1999, Mitchell et al. 1996). Stable isotopes of nitrogen reﬂect the trophic position of
organisms because 15 N accumulates in tissues by approximately 3.5% per trophic
transfer, while the lighter 14N is preferentially excreted (Cabana and Rasmussen 1994,
Peterson and Fry 1987). The difference in 815N from zooplankton to bluegill in our study
was signiﬁcantly less in invaded lakes (Figure 8, Table 13). Thus, in invaded lakes,
bluegills ate organisms that were less enriched in 15N. Bluegill prey items in invaded
lakes (e. g. benthic macroinvertebrates) may therefore reﬂect a dependence on bacteria,
which have low SUN (~ 0%o, Goericke et al. 1994) compared to phytoplankton (3.7-
5.7960, Gu et al. 1996), and would result in higher trophic levels having depressed 515 N.
Carbon isotopes of consumers tend to correspond closely to their food sources, as 613C
generally changes less than 1%o per trophic transfer (Peterson and Fry 1987). Due to the
distinct 813 C isotopic ratios between aquatic and terrestrial systems, it is also possible to
determine if the food source is of allochthonous or autochthonous origin (Peterson et al.
1994, Hamilton et al. 1992). The average benthic algae 5'3 C value in lakes is enriched
compared to planktonic algae (-26%o vs. -32%o) and this difference can be reﬂected in the
513C of consumers (France 1995). The greater change in 813 C from zooplankton to
bluegill in invaded lakes, although not signiﬁcant, also shows potential evidence of a diet
shift to organisms that are enriched in 13 C. Thus, bluegill in invaded lakes may potentially
be feeding on organisms that are more associated with the benthos, because of the
enriched 513C values that were incorporated in their tissues. The shifting of zebra mussel-
invaded food webs towards enriched 513C values and thus a system more reliant on the

benthos, is consistent with the changes in 513 C values noted in Oneida Lake (NY, USA)

59

and Lough Eme (Ireland), post-invasion (Mitchell et al. 1996 and Maguire and Grey
2006, respectively).

We also show potential evidence of a diet shift in adult bluegill in invaded lakes,
based on stomach content analyses. Percent zooplankton in the stomach, although not
signiﬁcant, was 65% lower in invaded lakes (Figure 5, Table 11). In addition, overall
weight of stomach contents did not differ between lake treatments (Figure 6, Table 12).
Thus, bluegills ate similar amounts of food in invaded and uninvaded lakes, yet diets may
have consisted of more benthic macroinvertebrates in invaded lakes. Trometer and Busch
(1999) proposed prey switching from less abundant zooplankton to more abundant
benthic invertebrates as a mechanism to describe growth for age 0 ﬁsh in Lake Erie after
zebra mussel invasion. It is possible that higher adult bluegill growth in invaded lakes is
due to heightened abundance of benthos which would be consistent with the effects of
zebra mussels on adult yellow perch growth in enclosures (Thayer et al. 1997 , Rautio
1995)

Although we were surprised by the increased grth of juvenile and adult bluegill
in invaded lakes, increased grth is advantageous for bluegill populations. We show
that juveniles and adults grew approximately 1.8-4.2 mm more per year in invaded lakes
(Figure 3). This additional grth is cumulative over time, as reﬂected by the increased
difference in grth between lake groups as bluegill change from juveniles to adults.
Therefore, if a juvenile grows faster each year in an invaded lake, it should reach the
threshold size to escape bass predation sooner. In turn, faster growing bluegills will have
higher survival rates, because predation risk also declines as bluegills grow. Increased

survival will then result in populations with increased reproductive output because they

60

reach reproductive maturity sooner. Overall, these positive changes due to zebra mussels
will translate into increased ﬁtness (Werner and Gilliam 1984).
Contemporary vs. Historical Bluegill Mean Length at Age

Comparing mean length at age of bluegill prior to 1988 and data from 2002-2003
allowed us to look at the effects of zebra mussels on Michigan lakes before and after
invasion, and validated our contemporary grth analyses. We observed no differences
in the mean length at age of juvenile and adult bluegill in the two lake groups prior to
invasion (Figure 4; Tables 6 and 9). However, adult bluegill mean length at age in 2002-
2003 was signiﬁcantly greater in invaded lakes (Figure 4, Table 10). Even though using
mean length at age to examine growth may have limitations, our results were consistent
with our contemporary bluegill growth analyses (Figure 3).

The mean length at age data prior to 1988 that we obtained from the Michigan
Department of Natural Resources Fisheries Division did not include enough information
to determine size speciﬁc growth. Rather, we were limited to determining growth from
mean length at age analyses. It is possible to determine if bluegill reach a particular size
earlier or later in life from analyses of mean length at age, but it is difﬁcult to determine
if an environmental factor (i.e. presence of zebra mussels) affects all ages of bluegill in
the same way. Growth rate is derived from the slope of the line for mean length at age.
Therefore it is challenging to determine if growth rates change at different times during
the lifespan of ﬁsh because each length at age value is incorporated into the overall slope
calculation (Osenberg et al. 1988). Size speciﬁc grth is the most rigorous and accurate
way to determine ﬁsh growth, as it can distinguish between the responses of ﬁsh to

environmental factors over multiple life history stages (Osenberg et al. 198 8). While

61

comparisons of mean length at age are less rigorous grth analyses and may be difﬁcult
to interpret, we still see a signiﬁcant positive zebra mussel effect on adult bluegills in
2002-2003.

An interesting result of the pre- and post-invasion analysis was that juvenile and
adult bluegill mean length at age in all lakes, whether or not they were later invaded, was
signiﬁcantly greater prior to invasion compared to 2002-2003 (Figure 4; Tables 5 and 8).
These results indicate that bluegills were either all larger prior to invasion due to
productivity differences between time periods, or that a potential measurement bias
existed. TP was not greater in the 24 lakes prior to invasion (1124,24 = 344.5, p = 0.24) and
does not explain this pattern. An alternative explanation may be measurement variation
introduced by the many scientists who analyzed the pre-invasion ﬁsh scale data. Pre-
invasion samples were collected and analyzed between 1956-1987 by multiple scientists
from the Michigan Department of Natural Resources, while CEHS was the only scientist
to analyze all of the 2002-2003 bluegill scales. Determining the age of bluegills is a
subjective skill that requires much practice and can vary from one scientist to another.
Consistent scale reading within a time period by one person allows valid comparison.
Therefore, this source of potential variation does not affect our comparisons of mean
length at age between invaded and uninvaded lakes within the pre- or post-invasion time
period (i.e. Tables 6, 7, 9 and 10), it does only in the comparison across time periods (i.e.
Tables 5 and 8), analyses that were not paramount to our study goal.

Overall, our study shows strong negative zebra mussel effects on the primary
producers and the primary consumers (Table 3). In addition, there is a positive effect on

secondary consumer growth (Figures 3 and 4), even though we document a potential diet

62

change (Figures 5, 8 and 9). These results are consistent with the food web effects of
zebra mussels in Lake Erie. Johannsson et al. (2000) reported that zebra mussels in Lake
Erie suppressed primary production of phytoplankton and secondary production of
zooplankton in the pelagic zone, while promoting benthic production. The increased
benthic production was estimated to support increased ﬁsh biomass, and therefore the
authors found no negative zebra mussel effect on ﬁsh biomass.

Our study used a survey to search for broad-scale zebra mussel effects on bluegill
in inland Michigan lakes. The results provide no evidence of a negative zebra mussel
effect on juvenile and adult bluegill growth or bluegill mean length at age. Moreover, we
show evidence that adult bluegill have switched their diets in invaded lakes, with positive

effects on growth. Therefore, zebra mussels are not contributing to bluegill stunting.

63

 

88 a 8 8.32 o .2 .83 86:0 :22.
mg ~

098% .33 .3 Lo .2935. + moSmBMEm 98m :22

33 .3 we c0952 o 20 6di 8E N 98 ﬂ ow<
32 .3 Ho 98:80 +\.. cum 0&3 >O> AnemoauEKeuumo-v
88 a a 88.2 + >2 .83 «26:0 >o> .228 36:;
Amxazoﬂcax ﬂashed/G
VOON 46 H0 HoxmmH—m + >2 £0>a ﬁOmﬁq—m DECO>ZH +0 Hoﬂﬁm :ﬂtOQm
$3,833.. .35on5
83 .130 5802 + 20 .834 8E N 98 H .%< £2.33 coomﬁxaEdm
Eﬁuoxemmﬁo SeoMeuoUv
SON .3 Ho :o>o£om - sawEosz 8:3 mm oTN mow/w amm BE? 3:3
c.3353 BNSENEEV
woe mwﬁqaom - 855205 dxm 0:52: @50ng 305mm

33 nomtmm use cemwbﬁoa 0 85835 dxm 0:52;.
32 ._a 6 c6862 o 20 .813 83 N 98 H %< AeEEoobeE ”€on3
AVOON BOv—mdm .. mDSmoﬁoﬁD .mxm _w>.arm Smmgw =~wo~dm
85.53% 8598K Staged owaum £5 86on 5E

5.5qu

 

A8855 H + .owgno
o: u o .omeouoow u -v $55228 REoEtomxo use .263.“ £83 1296.6. 5 53on Ema :o £838 823 we Boobm ._ 033.

64

 

 

mme nomsm 93 ~808lo o mum 834 £93 .3 o oW<
VOON .3 6 535m .. >2 552 :833 0:52;. +o 5d owwﬂom
. m o :3 a o: o Aguutmﬁu 333$
32 _ H 3 0 o . m 31H >O> .3on 833
3:93—3— maximum :33qu am on: 8325 Amr—
5395

 

82.850 ._ 22S

65

Table 2. Multiple-regression estimates for average weight per prey item (g wet weight) in
bluegill sunﬁsh stomachs from 20 of the study lakes (N uninvaded = 10, N invaded = 10)

in 2002-2003.

 

 

Parameter Weight SE t-value P-value
Estimate (g)

Daphnia spp. 0.00010 0.000013 7.70 < 0.001

Snails, Clams, Zebra Mussels 0.0022 0.0018 1.22 0.22

Mites, Ostracods 0.00020 0.00083 0.24 0.81

Beetles, True Bugs 0.0040 0.0012 3.35 0.001

Amphipods, Springtails 0.001 0.00018 5.71 < 0.001

Damselﬂy, Dragonﬂy, Mayﬂy 0.0059 0.0017 3.44 0.001

and Stoneﬂy Larvae

Worms, Moth larvae, 0.0000070 0.00058 0.01 0.99

Chaoborus spp.

Caddisﬂy and Chironomus 0.00084 0.00018 4.69 < 0.001

Larvae

Clad°°emns (“’“h‘mt Daphm" 0.000013 0.000018 0.71 0.48

spp.) and Copepods

Mosquito Larvae 0.00043 0.00014 3.00 0.003

Earthworm 1.59 0.71 22.27 < 0.001

 

66

awarmwa: 393:3: p: 3:

 

 

 

.86 02%: 3% $3 a}. an? 3283
=0§§Eo§nm
Bob
38-3.8 A333. 3 n-.. ma
0; 8.73 mg mam 3o 8.... 233226
283-309: 223-210-: :5
Led 83-qu :3 N3- :3 mac- 620386
8855 am:
3.3.8.3 633:; pg 3:
m2 N363 as 3.2 $3 82: maoﬁmoﬁﬁﬁ
3.2-2.8 833$ a5
Ed Gauss: who ”3 as man 88:82
Aowqﬁv Aowqﬁu
mm owﬁo>< Mm owﬁo><
053?; 05:35 332: 33>:ch BEESE
“mob

 

.36 u a 3 858$in 33035 ... .m¢ON-NooN E moo—E
BREE: 33 @232: 33:8 333 E 8333? 33w2o£ 23 3293a mo hobo Egan; 93 owns“ .owﬁoﬁw .m 23H

67

 

 

 

 

GEw-mwe geﬂ-Ed AL we
as g. u 32: 9% 8.2 mg 3.3 £82m
dam 55}qu
aﬁomN-Sw : AESm-ﬁdu on we
.23 mg n 3 2.: mod: N52 #32 £82m
:oEcmEoBoH-EE
33
39.5.3 SEE-wwa: AL me
$86 :2 n 3 gm 2.8 3 3.9. maeam
cot—5308832
Boa
Amman: woman-e
mm. owﬁo>< mm omﬁo><
wig-m oumuﬁm @032: BREED BEESmm
“mo-H

 

32:80 m 0:3

68

Table 4. Fixed effects from a linear mixed-effects model of 2467 observations of
individual bluegill growth in the ﬁnal year from 50 lakes (25 uninvaded, 25 invaded) in

the years 2002-2003.

 

 

Effect N um Den F-value P-value
df df

Intercept 1 2413 2593 < 0.001

Zebra Mussel

Treatment 1 48 5.74 0.02

”6 mm” 2 2413 1038 < 0.001

Stage

Interaction:

23b” 2 2413 5 52 0 004

Mussel*Life ' ’

History Stage

 

Table 5. Fixed effects from a linear mixed-effects model comparing 83 observations of
juvenile bluegill mean length at age (ages 1-2) in 24 lakes in the survey with both
historical (prior to 1988) and contemporary (2002-2003) data. Non-signiﬁcant interaction
terms were removed from the model.

 

 

Effect Num Den F—value P-value
DF DF

Age 1 56 80.02 < 0.001
Time Period
(Historical or 1 56 53.37 < 0.001
Contemporary)
Zebra Mussel
Treatment 1 56 2.48 0.12
Interaction:

:3
Age zebra 1 56 4.26 0.04
Mussel
Treatment

 

69

Table 6. Fixed effects from a linear mixed-effects model comparing 37 observations of
juvenile bluegill mean length at age (ages 1-2) in 24 lakes only using historical (prior to
1988) data. Non-signiﬁcant interaction terms were removed from the model.

 

 

Effect Num Den F-valne P-value
DF DF

Age 1 14 65.78 < 0.001

zebra Mussel 1 14 0.33 0.57

Treatment

 

Table 7. Fixed effects from a linear mixed-effects model comparing 46 observations of
juvenile bluegill mean length at age (ages 1-2) in 24 lakes only using contemporary
(2002-2003) data. Non-signiﬁcant interaction terms were removed from the model.

 

 

Effect Num Den F—valne P-value
DF DF

Age 1 21 370.1 < 0.001

Zebra Mussel 1 21 2.76 0.1 1

Treatment

 

Table 8. Fixed effects from a linear mixed-effects model comparing 148 observations of
adult bluegill mean length at age (ages 3-6) in 24 lakes in the survey with both historical
(prior to 1988) and contemporary (2002—2003) data. Non-signiﬁcant interaction terms
were removed from the model.

 

 

Effect Nnm Den F-value P-value
DF DF

Age 1 122 430.9 < 0.001

Time Period

(Historical or 1 122 253.2 < 0.001

Contemporary)

Zebra Mussel

Treatment 1 122 2.04 0.16

 

7O

Table 9. Fixed effects from a linear mixed-effects model comparing 86 observations of
adult bluegill mean length at age (ages 3-6) in 24 lakes only using historical (prior to
1988) data. Non-signiﬁcant interaction terms were removed from the model.

 

 

Effect Num Den F-value P-value
DF DF

Age 1 62 525.0 < 0.001

zebra Mussel 1 62 0.69 0.41

Treatment

 

Table 10. Fixed effects from a linear mixed-effects model comparing 62 observations of
adult bluegill mean length at age (ages 3-6) in 24 lakes only using contemporary (2002-
2003) data. Non-signiﬁcant interaction terms were removed from the model.

 

 

Effect Num Den F-value P-value
DF DF

Age 1 37 238.5 < 0.001

Zebra Mussel

Treatment 1 37 6.37 0.02

 

Table l 1. Fixed effects from a linear mixed-effects model comparing 260 observations of
percent zooplankton in the stomach of adult bluegill in 26 lakes in 2002-2003.

 

 

Effect N um Den F-value P-value
DF DF

Intercept 1 234 15.56 < 0.001

Zebra Mussel 1 24 2 24 O 15

Treatment ' '

 

7l

Table 12. Fixed effects from a linear mixed-effects model comparing 260 observations of
adult bluegill stomach weight in 26 lakes from 2002-2003. Non-signiﬁcant interaction
terms were removed from the model.

 

 

Effect Num Den F-value P-value
DF DF

Intercept 1 232 2375.44 < 0.001

Length 1 232 28.64 < 0.001

Zebra Mussel

Treatment 1 24 0.60 0.44

 

Table 13. Fixed effects from a linear mixed-effects model comparing 81 observations of
the change in 5‘5 N between zooplankton and adult bluegill in 29 lakes in 2002-2003.

 

 

Effect Num Den F-value P-value
DF DF

Intercept l 52 46.59 < 0.001

Zebra Mussel 1 27 6.70 002

Treatment

 

Table 14. Fixed effects from a linear mixed-effects model comparing 81 observations of
the change in 613 C between zooplankton and adult bluegill in 29 lakes in 2002-2003.

 

 

Effect Num Den F-value P-value
DF DF

Intercept 1 52 8.57 0.005

Zebra Mussel

Treatment 1 27 2.19 0.15

 

72

 

0.6

 

 

 

 

 

o
0.4 -
A
3’ 0.2 -
E
3
E
3 0.0 - ' e
a: o
-0.2 A ’
'0-4 I l l I l I I l I I

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Predicted Stomach Content Weight (9)

Figure 1. Residual plot for the multiple—regression model used to estimate the average
weight of prey items in bluegill sunﬁsh guts from 20 of the study lakes (N uninvaded =
10, N invaded = 10) in 2002-2003. R2 = 0.85, Fmgg = 100.95, p < 0.001.

73

300

 

FIRST YEAR JUVENILE ADULT Uninvaded

200 .
150 -

100 -

Number of Fish

 

 

 

 

 

 

 

 

o :;.= VIE; .:..: . g ! a H H mm .4.

300

 

 

FIRST YEAR JUVENILE ADULT Invaded

250 - m
200 -

150 -

Number of Fish

100 -

50“

0 ,,2 , ,jﬂmgmn

I j I ' I 1
Q5105?” ’9 0559 {Pioyﬁ igbibﬁg‘b 09909::0 '99 (9 '39 33’0" >99 (9% 56°" '39 '39

 

 

 

 

 

 

 

 

 

 

 

o

.9 00’ (5° 091,30 «90 «o’ (be
Length at the beginning of the ﬁnal year (mm)
Figure 2. Frequency distribution and approximate life history category of bluegill sunﬁsh

placed into 10 mm size classes in uninvaded and invaded lakes in 2002-2003. Life history
categories are separated by the bold vertical lines.

74

 

50

- Uninvaded

1:1 Invaded
40 -

30-

20-

  

     

 

 

 

 

 

 

 

 

Growth Increment in the Final Year (mm)

FIRST YEAR JUVENILE ADULT

Life History Stage at the Beginning of the Final Year
Figure 3. Growth increment (mm) :1: SE in the ﬁnal year for bluegill sunﬁsh in three life

history categories (as determined by length at the beginning of the ﬁnal year from Figure
2) in zebra mussel uninvaded and invaded lakes in 2002-2003. F 1,43 = 5.74, p = 0.02.

75

240

 

 

 

 

 

 

JUVENILE ADULT
220 -
200 .
E 180 -
E.
4: 160 -
‘6':
c 140 -
3
c 120 -
8
100 -
E
30 .. + Uninvaded Historic
-O— Invaded Historic
60 a + Uninvaded Contemporary
—v— Invaded Contemporary
40 I l l
0 1 4 5 6 7

Age

Figure 4. Comparison between historical (prior to 1988) and contemporary (2002-2003)
data of bluegill mean length at age (mm) :1: SE in 24 of our study lakes (N uninvaded =
13, N invaded = 11). Life history categories are separated by the bold vertical line.
Juvenile Historical F1,” = 0.33, p = 0.57; Juvenile Contemporary F131 = 2.76, p = 0.11;
Adult Historical F1452 = 0.69, p = 0.41; Adult Contemporary F137 = 6.37, p = 0.02.

76

25

 

20-

 

15‘

10-

 

Percent Zooplankton in Diet

 

 

 

 

 

 

Uninvaded Invaded

Figure 5. Percent zooplankton :1: SE in the diet of adult bluegill sunﬁsh from invaded and
uninvaded lakes in 2002-2003 (N uninvaded = 13, N invaded = 13). F134 = 2.24, p =
0. 1 5.

77

(120

 

(118 -
(116 a

(114 -
Jr.
(112 ‘

 

 

(110 -

(108 -

 

(106 -

 

(104 a

 

 

o Uninvaded
o Invaded

 

Stomach Content Biomass (g)

(102 ‘

 

 

 

 

 

(100 - u
90 100 110 120 130 140

Length of Fish at Capture (mm)

Figure 6. Adult bluegill sunﬁsh stomach content biomass (g) 3: SE as a function of the
length of ﬁsh at capture (mm) :1: SE in invaded and uninvaded lakes in 2002-2003 (N
uninvaded = 13, N invaded = 13). F134 = 0.60, p = 0.44.

78

14

Bluegill
12 - +
.T EBluegill .
10 _ Zooplankton i
[Snail

Zooplankton Zebra MusselI I 1
6 d c
Phytoplankton i Sedlment

4 a a Phytoplankton

 

 

8 15N
00

e Sediment

 

 

 

-36 -34 -32 -30 -28 -26 -24 -22 -20
8 13C

Figure 7. Average isotopic composition (513 C :1: SE and 5 15N i SE) of select food web
compartments in invaded (open circles) and uninvaded lakes (closed circles) in 2002-
2003. For the bluegill and zooplankton compartments (N uninvaded = 12, N invaded =
17), for the sediment and phytoplankton compartments (N uninvaded = l, N invaded = 5)
and for the snail and zebra mussel compartments (N uninvaded = 0, N invaded = 5).

79

 

 

   

 

 

 

 

 

6
5
u?
24*
.s
813*
E
(92'
1-
0

Uninvaded Invaded

Figure 8. The change in 8 15N 3: SE between zooplankton and bluegill sunﬁsh
compartments in invaded and uninvaded lakes in 2002-2003 (N uninvaded = 12, N
invaded = 17). F137 = 6.70, p = 0.02.

80

 

 

 

 

 

U
2
GO
c: 3 ‘ q-
0-
Q
2"
a 2 ~
.:
U
1 _
0 4....;<:_.l.:’.‘_.mi

 

 

 

 

 

Uninvaded Invaded

Figure 9. The change in 8 '3 C t SE between zooplankton and bluegill sunﬁsh
compartments in invaded and uninvaded lakes in 2002-2003 (N uninvaded = 12, N
invaded =17). F137 = 2.19, p = 0.15.

81

Literature Cited

Barkoh, A., and T. Modde. 1987. Feeding behavior of intensely cultured bluegill fry.
Progressive Fish-Culturist 49: 204-207.

Beekey, M.A., D.J. McCabe and J .E. Marsden. 2004. Zebra mussel colonization of soft
sediments facilitates invertebrate communities. Freshwater Biology 49: 535-545.

Breck, J .E., and M.J. Gitter. 1983. Effect of ﬁsh size on the reactive distance of bluegill
(Lepomis macrochirus) sunﬁsh. Canadian Journal of Fisheries and Aquatic
Sciences 40: 162—167.

Bremigan, M.T., and RA. Stein. 1994. Gape-dependant larval foraging and zooplankton
size: implications for ﬁsh recruitment across systems. Canadian Journal of
Fisheries and Aquatic Sciences 51: 913-922.

Bridgeman, T.B., G.L. Fahnenstiel, G.A. Lang, and T.F. Nalepa. 1995. Zooplankton
grazing during the zebra mussel (Dreissena polymorpha) colonization of Saginaw
Bay, Lake Huron. Journal of Great Lakes Research 21(4): 567-573.

Bunn, S.E., N.R. Loneragan, and MA. Kempster 1995. Effects of acid washing on stable
isotope ratios of C and N in penaeid shrimp and seagrass: Implications for food-
web studies using multiple stable isotopes. Limnology and Oceanography 40(3):
622-625.

Cabana, G., and J .B. Rasmussen. 1994. Modeling food chain structure and containment
bioaccumulation using stable nitrogen isotopes. Nature 372: 25 5-257.

Caraco, N.F., J .J . Cole, P.A. Raymond, D.L. Strayer, M.L. Pace, S.E.G. Findlay, and
D.T. Fischer. 1997. Zebra mussel invasion in a large, turbid river: phytoplankton
response to increased grazing. Ecology 78: 588-602.

Conroy, J .D., W.J. Edwards, R.A. Pontius, D.D. Kane, H. Zhang, J .F. Shea, J .N. Richey
and DA. Culver. 2005. Soluble nitrogen and phosphorous excretion of exotic
freshwater mussels (Dreissena spp.): potential impacts for nutrient
remineralisation in western Lake Erie. Freshwater Biology 50: 1146-1162.

Culver, D.A., M.M. Boucherle, D.J. Bean, and J .W. Fletcher. 1985. Biomass of
freshwater crustacean zooplankton from length-weight regressions. Canadian
Journal of Fisheries and Aquatic Sciences 42: 1380-1390.

Drake, M.T., J .E. Claussen, D.P. Philipp and D.L. Pereira. 1997. A comparison of
bluegill reproductive strategies and growth among lakes with different ﬁshing
intensities. North American Journal of Fisheries Management 17: 496-507.

Dumont, H.J., 1. Van de Velde, and S. Dumont. 1975 . The dry weight estimate of

82

biomass in a selection of cladocera, copepoda and rotifera from the plankton,
periphyton and benthos of continental waters. Oecologia 19: 75-97.

ESRI 1999. Arc View. Version 3.2. ESRI, Redland, CA.

Fahnenstiel, G.L., T.B. Bridgeman, G.A. Lang, M.J. McCormick, and T.F. Nalepa. 1995.
Phytoplankton productivity in Saginaw Bay, Lake Huron: effects of zebra mussel

(Dreissena polymorpha) colonization. Journal of Great Lakes Research 21: 465-
475.

France, R.L.. 1995. Differentiation between littoral and pelagic food webs in lakes using
stable carbon isotopes. Limnology and Oceanography 40(7): 1310-1313.

Gerking, SD. 1962. Production of food utilization in a population of bluegill sunﬁsh.
Ecological Monographs 32:31-78.

Graham, D.M., Sprules, W.G., and SJ. Nepszy. 1999. Growth and condition of juvenile
yellow perch (Percaﬂavescens) and white perch (Morone americana) during
zebra mussel establishment in western Lake Erie (1988-1991). Pages 337-349, in
M. Munawar, T. Edsall and IF. Munawar editors. State of Lake Erie (SOLE) —
past, present and ﬁrture. Backhuys Publishers, Leiden, The Netherlands.

Goericke, R., J .P. Montoya, and B. Fry. 1994. Physiology of isotopic fractionation in
algae and cyanobacteria, pp. 187-221 in K. Lajtha and RH. Michener editors.
Stable isotopes in ecology and environmental science. Blackwell Scientiﬁc
Publications, London.

Gu, B., C.L. Schelske, and M.V. Hoyer. 1996. Stable isotopes of carbon and nitrogen as
indicators of diet and trophic structure of the ﬁsh community in a shallow
hypereutrophic lake. Journal of Fish Biology 49: 1233-1243.

Hamilton, S.K., W.M. Lewis Jr., and SJ. Sippel. 1992. Energy sources for aquatic
animals in the Orinoco River ﬂoodplain: evidence from stable isotopes. Oecologia
89(3): 324-330.

Harrel, S.L., and ED. Dribble. 2001. Foraging efﬁciency of juvenile bluegill, Lepomis
macrochirus, among different vegetated habitats. Environmental Biology of
Fishes 62(4): 441-453.

Hasle, GR. 1978. The inverted-microscope method, p. 88-96 in A. Soumia, editor.
Phytoplankton manual. UNESCO, Paris, France.

Hayes, DB, and W.W. Taylor. 1991. A multiple-regression technique for analyzing
stomach contents of ﬁsh. Transactions of the American Fisheries Society 120:
396-400.

83

Herbert, P.D., B.W. Muncaster, and G.L. Mackie. 1989. Ecological and genetic studies
on Dreissena polymorpha (Pallas): A new mollusk in the Great Lakes. Canadian
Journal of Fisheries and Aquatic Sciences 46: 1587-1591.

Horgan, M.J., and EL. Mills. 1999. Zebra mussel ﬁlter-feeding and food-limited
production of Daphnia: recent changes in lower trophic level dynamics of Oneida
Lake, New York, USA. Hydrobiologia 411: 79-88.

Idrisi, N., E.L. Mills, L.G. Rudstam, and DJ. Stewart. 2001. Impact of zebra mussels
(Dreissena polymorpha) on the pelagic lower trophic levels of Oneida Lake, New
York. Canadian Journal of Fisheries and Aquatic Sciences. 58: 1430—1441.

Jackson, D., D.D. Harkness, C.F. Mason and SP. Long. 1986. Spartina angelica as a
carbon source for salt-marsh invertebrates: a study using 813 C values. Oikos
46(2): 163-170.

Jennings, C.A.. 1996. Effects of zebra mussel (Dreissena polymorpha) density on the
survival and growth of juvenile fathead minnows (Pimephales promelas):
implications for North American river ﬁshes. Hydrobiologia 324: 157-161.

Johannsson, O.E., R. Dermott, D.M. Graham, J .A. Dahl, E.S. Millard, D.D. Myles, and J.
LeBlanc. 2000. Benthic and pelagic secondary production in Lake Erie aﬁer the
invasion of Dreissena spp. with implications for ﬁsh production. Journal of Great
Lakes Research 26 (1): 31-54.

Kling, G.W., B. Fry, and W.J. O'Brien. 1992. Stable isotopes and planktonic trophic
structure in Arctic lakes. Ecology 73(2): 561-566.

Knoll, LB. 2004. Understanding zebra mussel (Dreissena polymorpha) impacts on
autotrophic and heterotrophic plankton of inland lake ecosystems. M.Sc. Thesis.
Department of Fisheries and Wildlife, Michigan State University, East Lansing,
MI.

Knoll, L.B., O. Sarnelle, S.K. Hamilton, C.E.H. Scheele, A.E. Wilson, J .B. Rose and
MR. Morgan. 2007. Invasive zebra mussels (Dreissena polymorpha) increase
cyanobacterial toxin concentrations in low-nutrient lakes. Limnology and
Oceanography. In press.

Kraft, OE, and LE. Johnson. 2000. Regional differences in rates and patterns of North
American inland lake invasions by zebra mussels (Dreissena polymorpha).

Canadian Journal of Fisheries and Aquatic Sciences 57: 993-1001.

Langner, C.L., and RF. Hendrix. 1982. Evaluation of a persulfate digestion method for
particulate nitrogen and phosphorus. Water Research 16: 1451-1454.

Lowe, R.L., and R.W. Pillsbury. 1995. Shifts in benthic algal community structure and

84

ﬁmction following the appearance of zebra mussels (Dreissena polymorpha) in
Saginaw Bay, Lake Huron. Journal of Great Lakes Research 51(4): 558-566.

Ludyanskiy, M.L., D. MacDonald, and D. MacNeill. 1993. Impact of the zebra mussel, a
bivalve invader. BioScience 43: 533-544.

Maclsaac, H.J.. 1996. Potential abiotic and biotic impacts of zebra mussels on the inland
waters of North America. American Zoologist 36:287-299.

Maclsaac, H.J., C.J. Lonnee, and J .H. Leach. 1995. Suppression of microzooplankton
by zebra mussels: importance of mussel size. Freshwater Biology 34: 379-3 87.

Maclsaac, H.J., W.G. Sprules, and J .H. Leach. 1991. Ingestion of small-bodied
zooplankton by zebra mussels (Dreissena polymorpha): can cannibalism on
larvae inﬂuence population dynamics. Canadian Journal of Fisheries and Aquatic
Sciences 48: 2051-2060.

Maguire, C.M., and J. Grey. 2006. Determination of zooplankton dietary shift following a
zebra mussel invasion, as indicated by stable isotope analysis. Freshwater Biology
51: 1310-1319.

Mayer, C.M., A.J. VanDerValk, J .L. Fomey, L.G. Rudstam, and EL. Mills. 2000.
Response of yellow perch (Percaﬂavescens) in Oneida Lake, New York, to the
establishment of zebra mussels (Dreissena polymorpha). Canadian Journal of
Fisheries and Aquatic Sciences 57: 742-754.

Mercer, J .L., M.G. Fox, and CD. Metcalfe. 1999. Changes in benthos and three littoral
zone ﬁshes in a shallow, eutrophic Ontario Lake following the invasion of the

zebra mussel (Dreissena polymorpha). Journal of Lake and Reservoir
Management 15: 310-323.

Michigan Sea Grant 2002, 2003. Michigan Sea Grant Zebra Mussel Infestation
Monitoring Program. (http://www.miseagrant.umich.edu/ais/lakes.html)

Mitchell, M.J., B.L. Mills, N. Idrisi and R. Michener. 1996. Stable Isotopes of nitrogen
and carbon in an aquatic food web recently invaded by Dreissena polymorpha
(Pallas). Canadian Journal of Fisheries and Aquatic Sciences 53: 1445-1450.

Mittelbach G.G.. 1984. Predation and resource partitioning in two sunﬁshes
(Centrarchidae). Ecology 65: 499-5 1 3.

Mittelbach G.G.. 1983. Optimal foraging and growth in bluegills. Oecologia 59: 157-162.

Mittelbach G.G.. 1981. Foraging efﬁciency and body size: a study of optimal diet and
habitat use by bluegills. Ecology 62: 1370-1386.

85

Mittelbach G.G., and CW. Osenberg. 1993. Stage-structured interactions in bluegill:
consequences of adult resource variation. Ecology 74: 2381-2394.

Nicholls, K.H., and OJ. Hopkins. 1993. Recent changes in Lake Erie (North Shore)
phytoplankton: cumulative impacts of phosphorus loading reductions and the
zebra mussel introduction. Journal of Great Lakes Research 19: 637-647.

O’Brien, W.J., N.A. Slade and G.L. Vinyard. 1976. Apparent size as the determinant of
prey selection by bluegill sunﬁsh (Lepomis macrochirus). Ecology 57: 1304-
1310.

Osenberg, C.W., E.E. Werner, G.G. Mittelbach, and DJ. Hall. 1988. Growth patterns in
bluegill (Lepomis macrochirus) and pumpkinseed (L. gibbosus) sunﬁsh:
environmental variation and the importance of ontogenetic niche shifts. Canadian
Journal of Fisheries and Aquatic Sciences 45: 17-26.

Pace, M.L., S.E.G. Findlay, and D. Fischer. 1998. Effects of an invasive bivalve on the
zooplankton community of the Hudson River. Freshwater Biology 39: 103-116.

Pauli, HR. 1989. A new method to estimate individual dry weights of rotifers.
Hydrobiologia 186: 355-361.

Peterson, B.J., and B. Fry. 1987. Stable isotopes in ecosystem studies. Annual Review of
Ecology and Systematics 18: 293-320.

Peterson, B., B. Fry, M. Hullar, S. Saupe, and R. Wright. 1994. The distribution and
stable isotopic composition of dissolved organic carbon in estuaries. Estuaries 17:
1 11-121.

Pimental, D.. 2005. Aquatic nuisance species in the New York State canal and Hudson
River system and the Great Lakes basin: an economic and environmental
assessment. Environmental Management 35: 692-701.

Pothoven S.A., T.F. Nalepa, P.J. Schneeberger and SB. Brandt. 2001. Changes in diet
and body condition of lake Whiteﬁsh in southern Lake Michigan associated with

changes in benthos. North American Journal of Fisheries Management 2 1 (4):
876-883.

R Development Core Team. 2007. R: A language and environment for statistical
computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-
900051-07-0, URL http://wwwR-proiectorg.

Raikow, DR. 2004. Food web interactions between larval bluegill (Lepomis

macrochirus) and exotic zebra mussels (Dreissena polymorpha). Canadian
Journal of Fisheries and Aquatic Sciences 61: 924-941.

86

Raikow, BF, 0. Sarnelle, A.E. Wilson, and SK. Hamilton. 2004. Dominance of the
noxious cyanobacterium Microcystis aeruginosa in low-nutrient lakes is

associated with exotic zebra mussels. Limnology and Oceanography 49(2): 482-
487.

Rau, G.H., A.J. Mearns, D.R. Young, R]. Olson, H.A. Schafer and LR. Kaplan. 1983.
Animal ‘3 C/ 12C correlates with trophic level in pelagic food webs. Ecology 64(5):
13 14-13 1 8.

Rautio, SA. 1995. Zebra mussel effects on diet and grth of adult yellow perch and
predatory impact of freshwater drum on zebra mussels in pond enclosures.
Michigan Department of Natural Resources Fisheries Division Research Report
No.201 5.

Ricciardi, A., F.G. Whoriskey, and J .B. Rasmussen. 1997. The role of the zebra mussel
(Dreissena polymorpha) in structuring macroinvertebrate communities on hard
substrata. Canadian Journal of Fisheries and Aquatic Sciences 54: 2596-2608.

Richardson, W.B., and LA. Bartsch. 1997. Effects of zebra mussels on food webs:
interactions with juvenile bluegill and water residence time. Hydrobiologia
354:141-150.

Roff, J .C., and RR. Hopcroft. 1986. High precision microcomputer based measuring
system for ecological research. Canadian Journal of Fisheries and Aquatic
Sciences 43: 2044-2048.

Rutherford, E.S., K.A. Rose, E.L. Mills, J .L. Fomey, C.M. Mayer and LG. Rudstam.
1999. Individual-based model simulations of zebra mussel (Dreissena
polymorpha) induced energy shunt on walleye (Stizostedion vitreum) and yellow
perch (Percaﬂavescens) populations in Oneida Lake, New York. Canadian
Journal of Fisheries and Aquatic Sciences 56: 2148-2160.

Sandgren, CD, and J .V. Robinson. 1984. A stratiﬁed sampling approach to
compensating for non-random sedimentation of phytoplankton cells in inverted
microscope settling chambers. British Phycological Journal 19: 67—72.

Sarnelle, O., A.E. Wilson, S.K. Hamilton, L.B. Knoll, and DP. Raikow. 2005. Complex
interactions between the zebra mussel, Dreissena polymorpha, and the harmful
phytoplankter, Microcystis aerugz'nosa. Limnology and Oceanography 50(3): 896-
904.

SAS Institute. 2001. The SAS System Inc., Cary, NC.
Shevtsova, L.V., G.A. Zhdanova, V.A. Movchan, and AB. Primak. 1986. Experimental

interrelationship between Dreissena and planktonic invertebrates.
Hydrobiological Journal 22(6): 36-39.

87

Siefert, RE. 1972. First food of the larval yellow perch, white sucker, bluegill, emerald

Shiner, and rainbow smelt. Transactions of the American Fisheries Society 101:
219-225.

Sholto- Douglas, A. D. J. G. Field, A. G. James, and N. J. van der Merwe.1991.'3C/12C
5N/IN14 isotope ratios in the Southern Benguela Ecosystem: indicators of food
web relationships among different size-classes of plankton and pelagic ﬁsh;
differences between ﬁsh muscle and bone collagen tissues. Marine Ecology
Progress Series 78: 23-31.

Stewart, T.W., and J .M. Haynes. 1994. Benthic macroinvertebrate communities of
southwestern Lake Ontario following invasion of Dreissena. Journal of Great
Lakes Research 20: 479-493.

Stewart, T.W., 1G. Miner and R.L. Lowe. 1998a. Quantifying mechanisms for zebra
mussel effects on benthic macroinvertebrates: organic matter production and

shell-generated habitat. Journal of the North American Benthological Society
17(1): 81-94.

Stewart, T.W., J .G. Miner and R.L. Lowe. 1998b. Macroinvertebrate communities on

hard substrates in Western Lake Erie: structuring effects of Dreissena. Journal of
Great Lakes Research 24(4): 868-879.

Strayer, D.L., K.A. Hattala, and A.W. Kahnle. 2004. Effects of an invasive bivalve
(Dreissena polymorpha) on ﬁsh in the Hudson River estuary. Canadian Journal of
Fisheries and Aquatic Sciences 61: 924-941.

Strayer, D.L., L.C. Smith, and DC. Hunter. 1998. Effects of the zebra mussel (Dreissena
polymorpha) invasion on the macrobenthos of the freshwater tidal Hudson River.
Canadian Journal of Zoology 76: 419-425.

Swingle, HS, and E.V. Smith. 1941. The management of ponds with stunted ﬁsh
populations. Transactions of the American Fisheries Society 71: 102-105.

Thayer, S.A., R.C. Hass, R.D. Hunter, and RH. Kushler. 1997. Zebra mussel (Dreissena
polymorpha) effects on sediment, other zoobentlros, and the diet and grth of

adult yellow perch (Perca ﬂavescens) in pond enclosures. Canadian Journal of
Fisheries and Aquatic Sciences 54: 1903-1915.

Thorp, J .H., and AF. Casper 2003. Importance of biotic interactions in large rivers: an
experiment with planktivorous ﬁshes, dreissenid mussels and zooplankton in the

St. Laurence River. River Research and Applications 19: 265-279.

Trometer, ES, and W.D.W. Busch. 1999. Changes in age-0 ﬁsh growth and abundance

88

following the introduction of zebra mussels Dreissena polymorpha in the western
basin of Lake Erie. North American Journal of Fisheries Management 19: 604-
609.

Tumer, A.M., and G.G. Mittelbach. 1990. Predator avoidance and community structure:

interactions among piscivores, planktivores, and plankton. Ecology 71: 2241 -
2254.

USGS 2002, 2003. United States Geological Survey Nonindigenous Aquatic Species
Database. (http://ngs.er.usgs.gov/taxgroup/mollusks/zebramusseV)

Vanderploeg, H.A., J .R. Liebig, W.W. Carmichael, M.A. Agy, T.H. Johengen, G.L.
Fahnenstiel, and T.F. Nalepa. 2001. Zebra mussel (Dreissena polymorpha)
selective ﬁltration promoted toxic Microcystis blooms in Saginaw Bay (Lake

Huron) and Lake Erie. Canadian Journal of Fisheries and Aquatic Sciences 58:
1208-1221.

Vander Zanden, M.J., and J .B. Rasmussen. 1999. Primary consumer 513 C and 8‘5 N and
the trophic position of aquatic consumers. Ecology 80(4): 1395-1404.

Vander Zanden, M.J., J .M. Casselman, and J .B. Rasmussen. 1999. Stable isotope
evidence for the food web consequences of species invasions in lakes. Nature
401: 464-467.

Vinyard, G.L., and W.J. O’Brien. 1976. Effects of light and turbidity on the reactive
distance of bluegill (Lepomis macrochirus). Journal of the Fisheries Research
Board of Canada 33, 2845—2849.

Welschmeyer, N.A.. 1994. Fluorometric analysis of chlorophyll a in the presence of
chlorophyll b and pheopigrnents. Limnology and Oceanography 39:1985-1992.

Werner, BE, and J .F. Gilliam. 1984. The ontogenetic niche and species interactions in
size-structured populations. Annual Review of Ecology and Systematics 15: 393-
425.

Wemer, BE, and D]. Hall. 1988. Ontogenetic habitat shifts in bluegill: the foraging
rate-predation risk trade-off. Ecology 69: 1352-1366.

Werner, E.E., J .F . Gilliam, D.J. Hall, and G.G. Mittelbach. 1983. An experimental test of
the effects of predation risk on habitat use in ﬁsh. Ecology 62: 1540-1548.

Werner, E.E., D]. Hall, and MD. Werner. 1978. Littoral zone ﬁsh communities of two
Florida lakes and a comparison with Michigan Lakes. Environmental Biology of
Fishes 3:163-172.

Werner, R.G.. 1967. Intralacustrine movements of bluegill fry in Crane Lake, Indiana.

89

Transactions of the American Fisheries Society 96: 416-420.

Whitney, RR, and K.D. Carlander. 1956. Interpretation of body-scale regression for
computing body length of ﬁsh. Journal of Wildlife Management 20:21-27.

Wilkinson, L.. 1989. Systat: the system for statistics. Evanston, 822 pp.

90

APPENDIX

91

Appendix A. Individual lake survey data. An electronic ﬁle containing: lake name,
county, keycode, year sampled, mean depth, total phosphorus (TP), chlorophyll a, total
microzooplankton biomass, total macrozooplankton biomass, and average bluegill growth
can be obtained by emailing C.E.H. Scheele at scheelec@msu.edu.

92

l11.111111111111111191111111