A Thesis
Entitled
A Fine-scale Analysis of Spatial and Temporal Population Genetic Patterns
in the Yellow Perch (Perca flavescens)
By
Timothy J. Sullivan
Submitted to the Graduate Faculty as partial fulfillment of the requirements for
The Master of Science Degree in Biology
___________________________________
Dr. Carol A. Stepien, Committee Chair
___________________________________
Dr. Daryl Dwyer, Committee Member
___________________________________
Dr. Patrick Kocovsky, Committee Member
___________________________________
Dr. Patricia R. Komuniecki, Dean
College of Graduate Studies
University of Toledo
May 2013
Copyright © 2013, Timothy James Sullivan
This document is copyrighted material. Under copyright law, no parts of this document may be
reproduced without the expressed permission of the author.
An abstract of
A Fine-scale Analysis of Spatial and Temporal Population Genetic Patterns
in the Yellow Perch (Perca flavescens)
by
Timothy J. Sullivan
Submitted to the Graduate Faculty as partial fulfillment of the
requirements for the Master of Science Degree in Biology
The University of Toledo
May 2013
The genetic structure of a species encompasses the distribution of genetic
diversity and composition among its component populations, providing important insight
for conservation and management. This knowledge can be used to evaluate life history,
gene flow, recruitment dynamics, and responses to exploitation and habitat changes.
Discerning the changes or consistencies in population genetic patterns over time can
provide important insights into the mechanisms that regulate genetic resiliency.
Ultimately, analyses of spatial and temporal population genetic patterns may be used to
conserve genetic diversity, unique variability, and adaptive potential. The yellow perch
Perca flavescens (Percidae: Teleostei) provides an opportunity to investigate these
patterns, as its population groups have experienced variable annual recruitment, high
exploitation as a popular fishery in the Laurentian Great Lakes, and have not been
evaluated previously for temporal consistency in genetic patterns. The objective of this
thesis is to analyze the spatial and temporal genetic diversity and divergence of yellow
perch spawning groups in order to better understand its life history responses and
advance knowledge aiding its management. Population genetic patterns of yellow perch
spawning groups are assessed across the Huron-Erie Corridor (HEC) and from locations
iii
in Lakes St. Clair, Erie, and Ontario using 15 nuclear DNA microsatellite loci. Results of
this thesis research indicate that yellow perch spawning groups have appreciable genetic
diversity and are distinguished from one another by considerable genetic differences. For
example, the group spawning at the Belle Isle restoration site in the Detroit River has
relatively high genetic diversity, with an appreciable number of alleles and private alleles.
Yellow perch spawning at sites in Lakes St. Clair, Erie, and Ontario also show substantial
genetic diversity whose levels are consistent over time. However, the genetic
composition of yellow perch spawning at some given locations varied among different
sampling years. Some age cohorts born in specific years who spawned together at
Dunkirk NY (1980-2008) and Monroe MI (1997-2004), genetically varied across age
groups. This pattern did not correspond to a pattern of isolation by time (i.e., there was
not a consistent trend). The effective population size of yellow perch spawning at the
Dunkirk, NY location is relatively modest and appears to have remained relatively
consistent in size over the past 30 years. These spatial and temporal patterns likely are
linked to life-history characters, such as kin-aggregation, natal site fidelity, and/or a
sweepstakes model of reproduction. Genetic monitoring and development of long-time
data sets like those assembled here are recommended to provide an important
management assessment tool for monitoring and conserving fishery populations.
iv
Acknowledgements
I would like to thank Dr. Carol Stepien, not only for giving me this opportunity,
but for her constant support through the course of my project. I would like to thank my
committee members Dr. Daryl Dwyer and Dr. Patrick Kocovsky, whose suggestions have
greatly improved this thesis. I would like to thank the University of Toledo, the Lake
Erie Center and the Department of Environmental Sciences, members of the Great Lakes
Genetic Labortory (GLGL) including: Amanda, Lindsey, Doug, Shane, Vrushalee,
Hillary, Susanne, Jhonatan, and Carson. I would like to thank other members of the Lake
Erie Center: Merideth, Rachel, and especially Pat, Jeremy, Mark, Kristen, Jenn, Jason,
and Betsy. I would like to thank M. Bagley, B. Beckwith, J. Boase, A. Bowen, A. Cook,
J. Chiotti, S. DeWitt, J. Diemond, D. Einhouse, D. Fielder, A. Ford, K. Glomski, A. M.
Gorman, T. Hartman, C. Knight, R. Knight, P. Kocovsky, R. Kuhaneck, C. Murray, E.
Roseman, M. Sanderson, D. Sek, M. Thomas, J. Tyson, M. Werda, D. Zellar, R. Zimar,
the OMNR, MDNR, ODNR, USGS, and USFWS, the NYSDEC, and the PFBC for
providing the samples.
I would also like to take the time to thank my Mom and Dad and my sister
Lorraine for supporting me over the years. Lastly and most importantly I want to thank
Rachel. Without you I would have left this place a long time ago. You make every day
worth living and I can’t wait to continue our lives together.
v
Table of Contents
Abstract .............................................................................................................................. iii
Acknowledgements ..............................................................................................................v
Table of Contents ............................................................................................................... vi
List of Tables ................................................................................................................... ix
List of Figures ......................................................................................................................x
Preface................................................................................................................................ xi
1
Introduction…. .........................................................................................................1
2
Genetic Diversity and Divergence of Yellow Perch Spawning Populations Across
the Huron-Erie Corridor, from Lake Huron through Western Lake Erie ................6
2.1 Abstract… .........................................................................................................6
2.2 Introduction… ...................................................................................................7
2.2.1 History of the Huron-Erie Corridor ....................................................8
2.2.2 Yellow Perch Populations, Life History, and Previous Genetic
Investigation ...................................................................................10
2.3 Materials and Methods… ................................................................................13
2.3.1 Sample Collection, DNA Extraction, and Amplification .................13
vi
2.3.2 Microsatellite Data Analyses ............................................................14
2.4 Results..… .......................................................................................................17
2.4.1 Genetic Diversity of Yellow Perch Spawning Populations along the
HEC................................................................................................17
2.4.2 Genetic Divergence and Connectivity ..............................................19
2.5 Discussion..… .................................................................................................21
2.5.1 Genetic Divergence of Yellow Perch ...............................................21
2.5.2 Genetic Composition, Divergence, and Connectivity of Yellow
Perch Stocks.. .................................................................................23
2.5.3 Genetic Isolation is not Explained by Geographic Distance ............25
2.5.4 Significance of the Belle Isle Restoration Site .................................25
2.5.5 Conclusions .......................................................................................26
2.6 Acknowledgements..… ...................................................................................27
3
Temporal Population Genetic Structure of the Yellow perch Perca flavescens
within a complex lakescape ...................................................................................36
3.1 Abstract… .......................................................................................................36
3.2 Introduction… .................................................................................................37
3.3 Materials and Methods… ................................................................................42
3.3.1 Sample Collection, DNA Extraction, and Amplification .................42
3.3.2 Data Analyses ...................................................................................44
3.4 Results..… .......................................................................................................48
3.4.1 Genetic Diversity of Spawning groups, Collection Years, and Age
Cohorts ...........................................................................................48
3.4.2 Spatial and Temporal Population Genetic Structure .........................49
vii
3.4.2 Temporal Effective Population Size of the Dunkirk NY Spawning
Group .............................................................................................52
3.5 Discussion..… .................................................................................................53
3.5.1
Spatial and Temporal Patterns in Genetic Diversity ......................53
3.5.2 Spatial and Temporal Patterns in Genetic Divergence and Stock
Structure.. ...................................................................................55
3.5.3 Temporal Effective Population Size of the Dunkirk NY Spawning
Group .........................................................................................58
3.5.4 Conclusions and Management Implications .....................................58
3.6 Acknowledgements..… ...................................................................................60
4
Conclusions and Future Research ..........................................................................80
4.1 Conclusions… .................................................................................................80
4.2 Future Research… ..........................................................................................82
References ..........................................................................................................................85
viii
List of Tables
2.1
Summary statistics across Huron-Erie Corridor spawning populations ................28
2.2
Geographic and genetic parameters of yellow perch spawning populations .........29
2.3
Pairwise genetic divergences between yellow perch spawning populations .........30
2.4
Estimated effective number of migrants per generation ........................................31
3.1
Genetic parameters of yellow perch spawning populations ..................................61
3.2
Genetic parameters from (A) yellow perch collection years and (B) age cohorts at
Dunkirk NY and (C) yellow perch age cohorts at Monroe MI .............................63
3.3
Pairwise genetic divergence of spawning yellow perch ........................................66
3.4
Pairwise genetic divergence of (A) collection years and (B) age cohorts at
Dunkirk NY and (C) age cohorts at Monroe MI....................................................68
3.5
Analysis of molecular variance result under 4 possible scenarios .........................72
ix
List of Figures
2-1
Map of yellow perch spawning populations sampled ............................................33
2-2
(A) Estimated yellow perch population structure and (B) Results of ΔK
computation............................................................................................................34
2-3
Mantel (1967) regression for yellow perch spawning populations ........................35
3-1
Map of yellow perch sampling sites surveyed .......................................................75
3-2
Results of individual assignment to spawning groups ...........................................76
3-3
Estimated population structure using Bayesian STRUCTURE analysis ...............77
3-4
Mantel (1967) regression in (A) 2001-5 and (B) 2009 ..........................................78
3-5
Mantel (1967) regression between genetic distance (θST/1-θST) and time .............79
x
Preface
The chapters of this thesis are organized in order of their completion as individual
studies. Each chapter is identical to the published/submitted manuscript version.
Chapter 2 has been previously published as:
Sullivan, T.J. and Stepien, C.A. (2013) Genetic diversity and divergence of
yellow perch spawning populations across the Huron-Erie Corridor, from Lake
Huron through western Lake Erie. Journal of Great Lakes Research
Chapter 3 is in review for potential publication in Conservation Genetics:
Sullivan, T.J. and Stepien, C.A. (in review) Temporal population genetic
structure of the yellow perch Perca flavescens (Percidae: Teleostei) within a
complex lakescape.
This is publication #2013-11 from the Lake Erie Research Center.
Grant Awards to Dr. Carol A. Stepien from the NOAA Ohio Sea Grant R/LR-13, USEPA
CR-83281401-0, and GLRI project #70 supported this research. Further funding was
provided through a 2 year fellowship in the NSF GK-12 program DGE#0742395
“Graduate fellows in STEM High School Education: An Environmental Science Learning
Community at the Land Lake Ecosystem Interface” and by a research assistantship.
xi
Chapter 1
1.1 INTRODUCTION
Many fresh and saltwater fishes worldwide are in some way impacted by human
interaction and environmental changes (Vitousek et al. 1997, Jackson et al. 2001, Wolter
et al. 2004). Habitat modifications such as armoring of shorelines, damming, dyking,
bulkheading, dredging, channelization, and watershed development for industry and
agriculture act to change the availability, quantity, and quality of aquatic habitats (Leslie
and Timmins 1991, Leach 1991, Saalfeld et al. 2012). These alterations, along with
fishery exploitation and the introductions of competitive invasive species, often stress
native populations (Di Pippo et al. 2006, Devine et al. 2012, Rush et al. 2012). Survival,
recruitment, and population sizes of native organisms frequently vary spatially and
temporally, with anthropogenic influences often increasing the risk of local extinction,
reducing genetic diversity, and lowering adaptive potential (Allendorf et al. 2008,
Leirmann et al. 2012, Kopp et al. 2012).
Conservation management of populations may be aided through evaluating their
spatial and temporal patterns in genetic diversity, composition, and divergence. This
genetic information may shed light on the impacts of exploitation, habitat change, and
1
life history factors on populations (Allendorf and Luikart 2007). It also may identify
those populations with unique genetic variation and or local adaptations, thus allowing
management to focus conservation action at a biologically relevant scale. Actions that are
on too broad or narrow a scale may be unsuccessful and reduce population stability
(Stephenson 1999, Rice et al. 2012, Prachiel et al. 2012).
The yellow perch Perca flavescens (Percidae: Teleostei) comprises an important
fishery resource, for which an understanding of its genetic stock structure may aid
management and conservation. A genetic stock, as defined by Hallerman et al. (2003), is
a population subunit that shares a common gene pool, freely interbreeds, and is
genetically distinguishable from other such groups. The largest abundances of yellow
perch are found in the Great Lakes, including Lake Huron, the Huron-Erie Corridor
(which runs from Lake Huron through the St. Clair River, Lake St. Clair, and the Detroit
River into northwestern Lake Erie), Lake Erie, and Lake Ontario, whose stocks are
analyzed here. Lake Erie contains its largest fishery, averaging 180 million estimated
individuals from 1990-2012 (Minimum=43 million, Maximum=510 million, YPTG
2012), whose annual value is estimated by Kocovsky et al. (2013 in press) at hundreds of
millions of US dollars. Most Great Lakes' yellow perch populations presently are in a
period of inconsistent recovery, due to low recruitment, competition with invasive
species, overexploitation, habitat degradation/loss, and poor water quality (Munawar et
al. 2005, Guzzo et al. 2013). These factors make understanding the genetic stock
structure of yellow perch an important fish community goal and objective in Great Lakes'
fishery management (Ryan et al. 2003, GLFC 2008).
2
Stocks of yellow perch in the Great Lakes are monitored and managed by
international and interstate cooperation through the Great Lakes Fishery Commission
(YPTG 2008, 2011, 2012). However, under the current management structure, many
distinct population genetic stocks are grouped together into management units (MUs),
which do not match their true biological scale (Sepulveda-Villet and Stepien 2011,
Kocovsky and Knight 2012, Kocovsky et al. 2013 in press). Harvest limits for MUs
likely are not evenly distributed among the various genetic stocks, which may lead to
reductions of small local populations, loss of diversity, and reduced adaptive potential
(Stephenson 1999, Allendorf and Luikart 2007). Therefore yellow perch spatial genetic
relationships should be evaluated, with the goal of identifying distinct genetic stocks and
characterizing their relative genetic composition and diversity. Discerning the
consistency or dynamic nature of genetic stock structure over time also may be important
in that changes in genetic diversity and composition can indicate responses related to life
history or external selective forces (Allendorf and Luikart 2007, Allendorf et al. 2008).
Past studies of yellow perch population genetic patterns evaluated
interrelationships across their native range and identified glacial origins of contemporary
populations; however, most were limited to genetic markers having relatively lowresolution power. For example, Todd and Hatcher (1993) used allozyme protein
electrophoretic variants, which described relatively low genetic variability in Great Lakes
populations yet identified some broad-scale differences, which allowed them to trace
population founding origins to the Atlantic and Mississippian glacial refugia. Billington
(1993, 1996) analyzed mtDNA restriction fragment length polymorphisms (RFLPs) of
yellow perch, also finding relatively low levels of genetic variation, and discerning some
3
distinctive broad scale haplotype patterns across the lakes. Those studies lacked the
resolution power to discern differences in genetic composition among populations within
a lake or fine-scale system, as are analyzed here.
Fine scale studies of spatial genetic variation among Lake Erie yellow perch
spawning groups more recently have been conducted using mtDNA control region
sequences (Ford and Stepien 2004, Sepulveda-Villet et al. 2009, Sepulveda-Villet and
Stepien 2012) and with 15 higher-resolution nuclear DNA microsatellite loci (SepulvedaVillet and Stepien 2011, and 2012). Using mtDNA sequences, Sepulveda-Villet et al.
(2009) and Sepulveda-Villet and Stepien (2012) found lower genetic variation than was
characteristic of several other percid fishes, including walleye Sander vitreus (Stepien
and Faber 1998, Stepien et al. 2004), greenside darter Etheostoma blennioides (Haponski
and Stepien 2008), and rainbow darter Etheostoma caerulum (Haponski et al. 2009). In
addition to broad scale patterns, Sepulveda-Villet et al. (2009) discerned some fine scale
variation in the frequencies of mtDNA control region sequence haplotypes among
spawning groups within lakes. Sepulveda-Villet and Stepien (2011) then analyzed fine
scale genetic variation among yellow perch spawning groups across Lake Erie using 15
nuclear DNA microsatellite loci, discerning that almost all were genetically distinct. That
study also found that collection years explained some temporal genetic variation, but that
spatial structure explained considerably more of the overall patterns of variation. This
interplay between spatial and temporal patterns is more comprehensively tested here.
This thesis examines the genetic diversity, divergence, and structure of yellow
perch spawning groups, on a spatial and temporal basis. Genetic diversity and population
genetic structure are evaluated by analyzing 15 nuclear DNA microsatellite loci of adult
4
yellow perch spawning groups along the Huron-Erie Corridor (HEC) and at multiple time
points from given spawning sites in Lakes St. Clair, Erie, and Ontario. Specific questions
tested are:
1) Do spawning groups of yellow perch differ in levels of genetic diversity along the
HEC? (Chapter 1)
2) Does genetic composition differ among spawning groups, and what are the patterns of
spatial connectivity or divergence? (Chapter 1)
3) Do spawning groups of yellow perch differ in levels of genetic diversity spatially or
temporally across Lakes St. Clair, Erie, and/or Ontario? (Chapter 2)
4) Does genetic composition differ among spawning groups in Lakes St. Clair, Erie,
and/or Ontario, and what are their patterns of spatial connectivity or divergence?
(Chapter 2)
5) Is there evidence for temporal genetic consistency or differences among sampling
years or among age cohorts at a given spawning location? (Chapter 2)
6) Has estimated effective population size remained consistent at a spawning site?
(Chapter 2)
5
Chapter 2
Genetic diversity and divergence of yellow perch spawning populations across the
Huron-Erie Corridor, from Lake Huron through western Lake Erie
Previously published as Sullivan, T.J. and Stepien, C.A. (2013) Genetic diversity and
divergence of yellow perch spawning populations across the Huron-Erie Corridor,
from Lake Huron through western Lake Erie. Journal of Great Lakes Research
2.1 ABSTRACT: The yellow perch Perca flavescens supports one of the largest Great
Lakes fisheries, whose populations have varied due to environmental changes, including
exploitation and habitat degradation. The Huron-Erie Corridor (HEC) connects the upper
and lower Great Lakes, running from Lake Huron through the St. Clair River, Lake St.
Clair, and Detroit River to western Lake Erie; it serves as an essential fish migration
corridor, and contains key spawning and nursery grounds. Its shipping importance led to
its extensive channelization and dredging, destroying and degrading habitats. Since 2004,
the HEC Initiative has restored some fish spawning and nursery grounds. Our objective is
to assess the genetic diversity, connectivity, and divergence of yellow perch spawning
populations along the HEC to provide a baseline for assessing future patterns, including
responses to improved habitat. Genetic variation of seven spawning populations (N=248),
four in the HEC, one in Lake Huron, and two in western Lake Erie, are analyzed at 15
6
nuclear microsatellite loci. Results showed appreciable genetic diversity of the seven
spawning populations (mean observed heterozygosity =0.637±0.020, range 0.568-0.699),
which significantly differed in genetic composition (θST=0.011-0.099, p<0.0001-0.0007),
suggesting a history of genetic isolation; relationships did not follow a pattern of genetic
isolation by geographic distance. Notably, some nearby spawning populations were very
genetically distinctive, with high genetic diversity and high proportions of private alleles,
as characterized the Belle Isle restoration site in the Detroit River. Our study provides a
genetic benchmark to assess ongoing and future habitat restoration efforts across the HEC
and beyond.
2.2 INTRODUCTION
Maintaining genetic distinctiveness and diversity of populations may be important
for conserving their long-term stability and ability to respond to changing environmental
conditions (Allen et al., 2010; Keller et al., 2011; Miller et al., 2012). Habitat loss and
fragmentation can reduce population sizes and impede the movement of individuals
among locations, increasing the potential for inbreeding and fitness decline (Mills and
Smouse, 1994; Lande, 1998; Sato, 2006). Population genetic diversity and structure also
may be influenced by behavioral processes such as natal homing and spawning site
fidelity (Stepien and Faber, 1998; Stepien et al., 2009; Miller et al., 2012), which may
enhance specialization of reproductive groups and increase genetic divergence. In
aquatic ecosystems, the rehabilitation of habitat in natural connecting channels can be an
7
effective means to restore population structure and preserve locally adapted population
groups (Bini et al., 2003; Isaak et al., 2007).
Landscape genetics examines the role of landscape ecology on the spatial
distribution of genetic variation (Manel et al., 2003; Storfer et al., 2007). An
understanding of these patterns may guide conservation and management decisions to
restore or enhance habitat, thereby retaining or increasing population genetic diversity
and local adaptations. Here we employ a landscape genetics approach to analyze fishery
stocks, which are defined as population subunits that share a common gene pool, freely
interbreed, and are genetically distinguishable from other such groups (Hallerman et al.,
2003). We test the genetic diversity and connectivity among spawning populations of an
important fishery – the yellow perch Perca flavescens (Teleostei: Percidae) – along the
Huron-Erie Corridor (HEC) that links the upper and the lower Great Lakes. The overall
aim is to understand the genetic variation, divergence, and similarity of yellow perch
stocks within a complex and highly disturbed connecting channel.
2.2.1 History of the Huron-Erie Corridor
The HEC is one of four connecting channels within the Great Lakes, linking Lake
Huron with Lake Erie through the St. Clair River, Lake St. Clair, and the Detroit River
(Fig. 1). This area encompasses some of the Great Lakes most diverse wetlands and
contains over 65 fish species, of which 16 are threatened or endangered (Manny et al.,
2004, www.huron-erie.org). The HEC comprises the major shipping corridor between the
upper and the lower Great Lakes (US Army Corps of Engineers, 2004), where large
8
channelization projects have restructured much of its habitats (Bennion and Manny,
2011). Many of these modifications occurred within the Detroit River, leading to 96.5 km
of shipping channel dating from the 1874 construction of the Livingston Channel through
the 1968 completion of its modifications (Bennion and Manny, 2011). Fish habitats of the
Detroit River have been subjected to continuous dredging (~46,000,000 m3 removed in
all; Moulton and Theime, 2009) and sediment deposition (>41 km2; Bennion and Manny,
2011). HEC habitats were altered by increased industrialization, levels of contaminants
(Manny and Kenaga, 1991), and human population growth, along with shoreline
armoring, bulkheading, and dyking (Leslie and Timmins, 1991; Leach, 1991; HTG,
2009). Today less than 3% of its original coastal wetland areas remain (Bennion and
Manny, 2011). These habitat losses and alterations likely affected populations of yellow
perch and other fishes along the HEC.
The Huron-Erie Corridor Initiative was formed in 2004, with the goal of
rehabilitating fish spawning habitat in the Detroit and St. Clair rivers (www.huronerie.org), when 1,080 m2 of rock-cobble and ash cinders were placed at the head of Belle
Isle (site E; Fig. 1) in the Detroit River (HTG, 2009). In 2008, Fighting Island in the
middle Detroit River (Ontario) was similarly enhanced with 3,300 m2 of habitat (HTG,
2009). Assessment by federal and state biologists has concluded that these two spawning
habitats successfully attract large numbers of fishes, increasing species diversity and
abundances (Manny et al. 2007; HTG, 2009).
Although the extent of spawning habitat and size of yellow perch populations in
the HEC have not been explicitly documented, Goodyear et al. (1982) described many
regional spawning and nursery habitats (Fig. 1; Hatching). In Lake Huron, most yellow
9
perch spawning and nursery habitats are located in Saginaw Bay (site A; Fig. 1), with
additional nearshore spawning in southern Lake Huron. Along the HEC, spawning has
been documented in and above the St. Clair River delta, throughout most nearshore areas
of Lake St. Clair, including Anchor Bay (site C; Fig. 1) and L’anse Creuse Bay (D; Fig.
1), and along Belle Isle (E; Fig.1), Crystal Bay, and Grosse Ile in the Detroit River
(Goodyear et al., 1982). An estimated ~2-6.5 million yellow perch spawn in western
Lake Erie near Monroe, Michigan (F; Fig. 1; Thomas and Haas, 2000); other large
numbers spawn at Leamington/Colchestor, Ontario (G; Fig. 1) and throughout the Lake
Erie Islands (HTG, 2009).
Tagging studies of yellow perch indicate that the HEC is important for allowing
passage of individuals between riverine and lacustrine habitats, and between
overwintering grounds and spawning sites (Haas et al., 1985). The genetic diversity,
divergence, and connectivity of yellow perch spawning populations (stocks) along Lake
Huron, the HEC, and western Lake Erie are analyzed here and compared to those
throughout the geographic range. The genetic variability of these HEC spawning stocks
likely will provide a foundation for assessing the effects of present and future restoration.
2.2.2 Yellow perch populations, life history, and previous genetic investigations
Yellow perch populations reach their greatest abundances in the Great Lakes
watershed, where they support economically important commercial and sport fisheries
(Clapp and Dettmers, 2004; YPTG, 2006). Population sizes of yellow perch in western
Lake Erie were estimated at ~16-64 million throughout the 1990s (Thomas and Haas,
10
2000), with ~130 million living in Lake Erie as a whole today (YPTG, 2011). Yellow
perch stocks likely have been influenced by exploitation, pollution, habitat degradation,
and competition with exotic species (Trautman, 1981; Marsden and Robillard, 2004;
YPTG, 2011).
Cued by gradual changes in water temperature and photoperiod in late spring
(Jansen et al., 2009), yellow perch aggregate to spawn on shallow reef complexes or in
slow-moving tributaries 0.5-8 m in depth (Kreiger et al., 1983; Craig, 2000). Males move
into the nest areas first (Scott and Crossman, 1973), followed by females who drape egg
masses on submerged macrophytes or rock, which are fertilized by 2-5 males (Robillard
and Marsden, 2001; Mangan, 2004). Males generally linger post-spawn, potentially
fertilizing eggs from several females, with neither sex providing parental care (Craig,
2000). A study of yellow perch tag returns determined that post-spawning movements are
moderate; individuals tagged at Lake Erie spawning sites did not move upstream through
the HEC, whereas some of those tagged in Lake St. Clair migrated to nearby tributaries
(Haas et al., 1985).
Kin recognition and aggregative homing of yellow perch during reproduction may
lead to genetic divergence of spawning populations over time. Kin recognition has been
implicated in the closely related European perch P. fluviatilis; chemical and physical cues
are used to recognize relatives, with whom individuals preferentially associate (Gerlach
et al., 2001; Behrmann-Godel et al., 2006). Studies of yellow perch spawning in Nova
Scotia, Canada showed that removal of egg masses from a spawning site led to
significantly fewer egg masses at that site in subsequent years, as compared to control
locations (Aalto and Newsome, 1990). Those results revealed that yellow perch did not
11
follow a pattern of random spawning site selection, but likely returned to given spawning
sites (Aalto and Newsome, 1990).
Previous genetic studies examined diversity and divergence among yellow perch
spawning populations in Lake Erie, the Great Lakes, and across their native range, using
these same 15 nuclear microsatellite loci (Sepulveda-Villet and Stepien, 2011, 2012
accepted), providing an important comparison to the present study. Notably, Lake Erie
spawning populations (Sepulveda-Villet and Stepien, 2011) had appreciable genetic
diversity (mean HO=0.533±0.010; range=0.479-0.593) and their genetic compositions
significantly differed among sites (mean θST=0.233±0.020; 0.000-0.665). Diversity levels
were similar among spawning populations across the Great Lakes (mean HO
=0.551±0.013; 0.478-0.635) and somewhat lower in other areas of their native range
(0.533±0.016; 0.333-0.670), especially in isolated populations (Sepulveda-Villet and
Stepien, 2012 accepted). Sepulveda-Villet and Stepien (2012, accepted) also identified
high genetic divergence among Great Lakes spawning populations (θST=0.127±0.007;
0.008-0.282) as well as across the North American native range (θST=0.235±0.006;
0.008-0.472); divergence levels were especially pronounced among isolated populations.
Those results indicate that high divergence and moderate diversity characterizes yellow
perch spawning populations, which might also be predicted across the HEC.
Our study compares the genetic diversity, divergence, and connectivity of yellow
perch spawning populations in Lake Huron, the HEC, and western Lake Erie. Specific
hypotheses tested are: (1) genetic diversity levels of spawning populations significantly
differ, (2a) their levels of divergence significantly vary, and (2b) genetic divergence
follows an isolation by geographic distance pattern. The present investigation provides a
12
fine-scale analysis of yellow perch genetic diversity, divergence, and connectivity along
an extensively altered connecting channel.
2.3 MATERIALS AND METHODS
2.3.1 Sample collection, DNA extraction, and amplification
Adult spawning-condition yellow perch (N =248) were collected by state and
federal agency biologists and via hook-and-line fishing by us under permits issued to our
laboratory from seven spawning sites in Lake Huron, the HEC, and western Lake Erie
(lettered A-G in Fig. 1). Each collection was made from a single spawning location and
year, except for samples from Saginaw Bay (site A) where samples from throughout the
bay from two collection years were tested for difference, none was found, and thus were
pooled (Table 2). Pectoral fin clips were preserved in 95% EtOH in the field and stored at
room temperature prior to DNA extraction.
Genomic DNA was extracted and purified from the fin clip using DNeasy
Quiagen kits (QUIAGEN, Inc., Valencia Ca.), whose aliquots were frozen, labeled, and
archived. Genetic variation was analyzed using 15 nuclear DNA microsatellite loci
following Sepulveda-Villet and Stepien (2011) including: Svi2, 3, and 7 from Eldridge et
al. (2002), Svi4, 17, and 33 from Borer et al. (1999), YP13 and 17 from Li et al. (2007),
and Mpf1-7 from Grzybowski et al. (2010).
Polymerase chain reactions (PCR) consisted of 50 mM KCl, 1.5 mM MgCl2, 10
mM Tris-HCl, 50 μM of each deoxy-nucleotide, 0.5 μM each of the forward and reverse
primers, 2% dymethyl sulfoxide (DMSO), 5-30 ng DNA template, and 0.6-1.2 μM of Taq
13
polymerase per 10 μL of reaction volume. Positive and negative controls were included
in each reaction. An initial cycle of 2 min at 94°C was used for strand denaturation,
followed by 40 cycles of denaturation (94°C, 30 s), primer annealing (1 min) at a primerspecific temperature (TA; Table 1), and polymerase extension (72°C, 30 s). A final
extension at 72°C for 5 min was included to minimize partial strands.
Forward primers were synthesized with one of four 5' fluorescent labels, allowing
pool-plexing during analysis (grouped as follows: Svi2+7, Svi3+33, Svi4+17, YP13+17,
Mpf1+2+5+6, and Mpf3+4+7). Amplification products were processed for allelic length
determination by diluting at a ratio of 1:50 with ddH2O, with a 1 μL aliquot added to 13
μL of a formamide and ABI GeneScan-500 size standard solution, loaded onto 96-well
plates, and denatured for 2 min at 95°C. The denatured products were analyzed on our
ABI 3130XL Genetic Analyzer with GENEMAPPER 3.7 software (Applied Biosystems
Inc., Foster City, CA). We reviewed output profiles manually to confirm correct
identification of allelic size variants.
2.3.2 Microsatellite data analyses
Population samples were tested for conformance to Hardy-Weinberg Equilibrium
(HWE) expectations at each locus, with significance estimated using the Markov Chain
Monte Carlo method (MCMC) via 1000 randomization procedures (Guo and Thompson,
1992) in GENEPOP v4.0 (Rousset, 2008; http://mbb.univmontp2.fr/MBB/subsection/downloads.php?section=2). Any deviations were analyzed
for excess or deficiency of homozygotes, and loci were tested for linkage disequilibrium
(LD). Levels of significance for both tests were adjusted using sequential Bonferroni
14
corrections (Rice, 1989) to minimize Type 1 error. Possible presence of null
(nonamplified) alleles was tested with MICRO-CHECKER v2.3.3 (van Oosterhout et al.,
2004, 2006; http://www.microchecker.hull.ac.uk). To test hypothesis 1, whether genetic
diversity differs among spawning populations, expected and observed heterozygosity
values (HE and HO) were calculated in GENEPOP v4.0, and number of alleles (NA) and
allelic richness (AR; number of alleles per locus independent of sample size, adjusted by
rarefaction per El Mousadik and Petit (1996)) were determined in FSTAT v2.9.3.2
(Goudet, 2002; http://www2.unil.ch/ popgen/softwares/fstat.htm). We tested for
significant differences in observed heterozygosity or allelic richness (hypothesis 1)
among samples with Friedman sum rank tests in the R statistical software suite v2.14.0
(R Development Core Team, 2011; http://www.r-project.org/), with loci treated as blocks
per spawning population. The number of private alleles (NPA; those occurring only in a
single spawning population) was computed in CONVERT v1.31 (Glaubitz, 2004;
http://www.agriculture.purdue.edu/fnr /html/faculty/rhodes/students %20and%20
staff/glaubitz/software.htm).
To investigate hypothesis 2a, whether genetic compositions significantly differ
among yellow perch spawning populations, unbiased θ estimates of F statistics (Weir and
Cockerman, 1984) and their levels of significance were evaluated in FSTAT. Models
using θST (the FST estimate of Weir and Cockerman, 1984) have been shown to better
resolve relationships among such recently diverged populations (Balloux and LugonMoulin, 2002). Pairwise comparisons between samples also were conducted using a
nonparametric (exact G) procedure (Raymond and Rousset, 2005), with probability
estimated from MCMC in GENEPOP v4.0; this approach does not assume a normal
15
distribution and is not influenced by sample size, but may have less statistical power
(Goudet et al., 1996). In all pairwise comparisons, sequential Bonferroni corrections were
used to minimize the potential for Type 1 statistical error (Rice, 1989). Lastly, numbers
of migrants (NM) among spawning populations were estimated in ARLEQUIN v3.5.1.3
(Excoffier and Lischer, 2010), following Slatkin (1991).
To further evaluate connectivity and divergence patterns (hypothesis 2a), we used
BARRIER v2.2 (Manni et al., 2004 a, b; http://www.mnhn.fr/mnhn/ ecoanthropologie
/software/barrier.html) to identify discontinuous groups of sampling sites independent
from an a priori knowledge of their relationships. Pairwise θST estimates were mapped
onto a matrix of sample site geographic coordinates (latitude and longitude). The
resulting “barriers” denote populations whose genetic distances are greater than predicted
from spatial proximity. Relative support for each barrier was evaluated by the number of
loci that supported it, and by bootstrap analysis of the multilocus θST matrix with 2000
iterations in GENELAND v3.1.4 (Guillot et al., 2005a, b, 2008;
http://www2.imm.dtu.dk/~gigu/Geneland/) through R. Genetic barriers supported by a
majority of loci and bootstrap values ≥ 50% are reported.
Patterns of genetic connectivity and stock structure (hypothesis 2a) were further
evaluated using a Bayesian approach in STRUCTURE v2.3.3 (Pritchard et al., 2000;
Pritchard and Wen, 2004; http://pritch.bsd.uchicago.edu/structure.html) to identify
similar groups of individuals, regardless of their true sample origin. Membership to
groups was analyzed by 10 independent runs at K=1 (a single spawning group, i.e.,
panmixia) to K=14 (double the N of spawning sites sampled), with burn-ins of 100000
and 500000 replicates. We analyzed consistency among runs, compared the probabilities
16
of individual assignments to groups, and calculated log-likelihood values. Optimal K
scenarios were determined from the ∆K likelihood evaluations of Evanno et al. (2005).
Partitioning of genetic variation among Lake Huron, the HEC, and western Lake
Erie, and among individual spawning populations (hypothesis 2a) was analyzed with
Analysis of MOlecular VAriance (AMOVA; Excoffier et al., 1992) in ARLEQUIN
v3.5.12 (Excoffier et al., 2005; Excoffier and Lischer, 2010; http://cmpg.unibe.ch/
software/ arlequin35/). We additionally tested for correspondence between genetic
distances (θST/1-θST) and geographic distances (hypothesis 2b), measured as the shortest
waterway distances between each pair of samples (both natural log transformed and
nontransformed), using ISOLDE in GENEPOP (Rousset, 1997). The regression line fit
and significance were calculated using Mantel’s (1967) procedure with 1000
permutations.
2.4 RESULTS
2.4.1 Genetic diversity of yellow perch spawning populations along the HEC
(Hypothesis 1)
The 15 loci analyzed in this study for 248 spawning individual yellow perch from
seven sampling sites did not show null alleles and all samples conformed to HWE and
LD expectations after sequential Bonferroni correction (Table 1). Loci Svi2, YP13, and
Mpf 5 were the most informative for discerning divergence among spawning populations,
17
as indicated by their higher FST values (Table 1). In contrast, loci Svi17 and Mpf1 and 2
had relatively low FST values and showed more modest differentiation (Table 1).
Numbers of alleles per locus ranged from 5 (Svi2) to 41 (Mpf2), with the
population spawning in the Detroit River having the most (Tables 1 and 2). Spawning
populations from Saginaw Bay in Lake Huron (site A in Fig. 1) and Anchor Bay (C) in
Lake St. Clair also had high numbers of alleles (180 and 176 respectively; Table 2).
Allelic richness significantly differed among yellow perch spawning in Lake Huron, the
HEC, and western Lake Erie (χ2=8.9, df=2, p=0.01), with the HEC having the highest
values (14.04±2.34). Allelic richness likewise significantly varied among spawning
populations (χ2=30.9, df=6, p<0.0001), and was greatest (10.51±1.63) at Belle Isle (E) in
the Detroit River, followed by Algonac (B) on the St. Clair River (9.25±1.61); both are
within the HEC. All spawning populations had private alleles, which numbered from 1 at
L’anse Creuse Bay (D) in Lake St. Clair to 28 at Belle Isle (E) in the Detroit River.
Yellow perch spawning in the HEC overall had more private alleles (58;
proportion=0.19), than were found in Lake Huron (13; 0.07) and western Lake Erie (8;
0.05). Observed heterozygosity (HO) differed among spawning populations in Lake
Huron, the HEC, and western Lake Erie (χ2=13.7, df=2, p=0.001), with higher values in
Lake Huron (0.678) and the HEC (0.683) than in western Lake Erie (0.587). Observed
heterozygosity also significantly differed among individual spawning sites (χ2=30.5,
df=6, p<0.0001), ranging from 0.699 (E; Belle Isle, MI; Detroit River) to 0.568 (F;
Monroe, MI; western Lake Erie). These results show that levels of genetic diversity differ
among yellow perch spawning populations, supporting hypothesis 1.
18
2.4.2 Genetic divergence and connectivity (Hypotheses 2a and 2b)
Pairwise θST and exact G tests discerned that all yellow perch spawning
populations significantly differed in allelic composition (Table 3). Migration values (NM)
were low among all populations, indicating spawning group specificity and little gene
flow (Table 4). These results support hypothesis 2a, that spawning populations of this
area have distinct genetic compositions and low genetic exchange.
BARRIER analyses revealed four major genetic discontinuities among the
spawning populations tested (Fig. 1), with the largest genetic barrier (Barrier I; 56%, 13
loci) separating those in Lake Erie from all others (Fig. 1). Barrier II (59%, 11 loci)
isolated the population spawning at Belle Isle (E) in the Detroit River from the others.
Barrier III (58%, 12 loci) separated the spawning populations at Monroe, MI (F) and
Sturgeon Creek (G) in western Lake Erie from one another. Barrier IV (53%, 10 loci)
divided yellow perch from Saginaw Bay (A) in Lake Huron from those spawning at
Algonac (B) in the St Clair River and at Anchor Bay (C) and L’anse Creuse Bay (D) in
Lake St. Clair. Overall, high levels of genetic structure indicated little gene flow among
spawning populations, with Belle Isle (E) in the Detroit River showing pronounced
genetic divergence. These patterns support hypothesis 2a and suggest that divergence
patterns vary spatially.
Bayesian STRUCTURE analyses (Fig. 2a) identified two likely scenarios (K=3
and K=7; Fig. 2b). The first scenario, K=3, highlighted the unique allelic compositions of
populations spawning at Saginaw Bay (A; red color) in Lake Huron and Belle Isle (E) in
the Detroit River (green color; Fig. 2a). The remaining sites showed a mixed genetic
19
signature, which appeared consistent with AMOVA results that found genetic variance
was not partitioned among the three hypothesized regions of Lake Huron, the HEC, and
western Lake Erie (1.3% variation explained; p=0.2, N.S.), but rather among individual
spawning populations (4.7% variation explained; p<0.0001), congruent with θST and
exact G test results. The STRUCTURE analysis (K=7) supported seven population
groups along the HEC, consistent with pairwise test results, which mirrored findings from
BARRIER analysis that showed genetic differences among yellow perch spawning
populations from Saginaw Bay in Lake Huron (A), Belle Isle in the Detroit River (E), and
the two in western Lake Erie (F and G). Yellow perch spawning at Algonac in the St.
Clair River (B), and in Lake St. Clair at Anchor Bay (C) and L’anse Creuse Bay (D)
revealed a mixed genetic signature, indicating closer relationships to one another
although all significantly differed. The mixed signature of individuals at Algonac in the
St. Clair River (B), and in Lake St. Clair at Anchor Bay (C) and L’anse Creuse Bay (D)
may be due in part to reduced assignment success that comes with lower FST values less
than 0.03 (Latch et al., 2006). Overall, our results indicated that all sites sampled from
Lake Huron, the HEC, and western Lake Erie contain distinct and divergent spawning
populations.
Genetic relationships among the spawning populations did not follow a pattern of
isolation by geographic distance (nontransformed; y=0.0001x+0.04, R2 = 0.24, p=0.16;
Fig. 3), with some nearby groups being very different. For example, those spawning at
Belle Isle (E) in the Detroit River were more genetically distinct than predicted by
geographic distance. The natural log transformed regression discerned congruent results
(y=0.012x-0.002, R2=0.21, p=0.16; not shown). Our results thus did not reject null
20
hypothesis 2b, indicating that factors other than geographic distance regulate gene flow
and divergence among spawning populations in the HEC.
2.5 DISCUSSION
2.5.1 Genetic diversity of yellow perch (Hypothesis 1)
Our results showed that overall genetic diversity levels of yellow perch spawning
populations across Lake Huron, the HEC, and western Lake Erie were higher (mean
HO=0.637±0.020; range=0.568-0.699) than those across Lake Erie (0.533±0.010; 0.4790.593; Sepulveda-Villet and Stepien, 2011), the entire Great Lakes (0.551±0.013; 0.4780.635; Sepulveda-Villet and Stepien, 2012, accepted), or their native North American
range (0.533±0.016; 0.333-0.670; Sepulveda-Villet and Stepien, 2012, accepted). Those
studies were done in our laboratory with the same 15 microsatellite loci; thus these values
are directly comparable. Our study and the others focused on variation at neutral markers,
which does not directly address whether habitat alterations resulted in changes that
influenced overall fitness. However, several recent studies have linked such variation at
neutral loci as predictive of trends at adaptive loci in a variety of fishes (Allendorf et al.,
2010; Tymchuk et al., 2010).
An early study of yellow perch genetics using allozymes (HO range=0.000-0.039,
Todd and Hatcher, 1993) recovered modest overall levels of genetic diversity, attributed
to the lower resolution power of those genetic markers compared to the microsatellite loci
used here. More recent studies of genetic diversity using mtDNA control region
sequences (mean HD=0.395±0.026, range=0.000-0.822, Sepulveda-Villet et al., 2009) and
21
microsatellite loci (mean HO=0.533±0.016; range=0.333-0.670; Sepulveda-Villet and
Stepien, 2012, accepted) recovered greater diversity levels across the North American
range of yellow perch. Similar values also have been recovered from populations of their
close percid relatives: the European perch (mtDNA control region sequences; mean
HD=0.340±0.330; range=0.000-0.870; Nesbo et al., 1998, 1999) and the ruffe
Gymnocephalus cernua (mtDNA control region sequences; 0.732±0.025; Stepien et al.,
1998, 2005). These levels of genetic diversity are consistent with those predicted for
freshwater fishes based on results summarized by DeWoody and Avise (2000). In
contrast, genetic variability levels are somewhat higher in walleye Sander vitreus than in
yellow perch using mtDNA control region sequences (mean HD=0.690±0.001,
range=0.360-0.790; Faber and Stepien, 1998) and microsatellite loci (mean
HO=0.698±0.013; range=0.512-0.783; Stepien et al., 2009). These differences in average
levels of genetic variability of populations may reflect a life history trend for yellow
perch remaining in kin-related groups as has been shown for the European perch
(Behrmann-Godel and Gerlach, 2008). This merits experimental investigation.
Along the HEC, levels of yellow perch genetic diversity appear lower than those
of walleye spawning populations (mean HO=0.722±0.009, range=0.680-0.760; Haponski
and Stepien, 2013), which may reflect lower gene flow among yellow perch spawning
populations. Like yellow perch, walleye diversity in the HEC was somewhat higher than
its values across Lake Erie (0.704±0.011; 0.660-0.780; Strange and Stepien, 2007), the
Great Lakes (0.711±0.011; 0.650-0.780; Stepien et al., 2009, 2010), and the native North
American range (0.684±0.02; 0.512-0.783; Stepien et al., 2009). Thus, genetic diversity
levels of both percid species are higher in the HEC than in other spawning populations
22
range-wide. These relatively high diversity levels in Lake Huron, the HEC, and western
Lake Erie are not consistent with reduced population levels that might be predicted from
a history of severe habitat loss, degradation, and population fragmentation (Mills and
Smouse, 1994). In contrast, smaller populations of stream-dwelling char Salvelinus
leucomaenis (Morita and Yamamoto, 2002) from Japan and the killifish Aphanius
fasciatus (Angeletti et al., 2010) from wetlands in central Italy, linked low genetic
diversity to habitat degradation and decreased population connectivity, which improved
after restoration. The abundance of percid spawning habitat and their large population
sizes in the Great Lakes likely have maintained the genetic diversity of spawning
populations despite habitat changes along the HEC. Our results indicate that the HEC
houses diverse spawning populations of both yellow perch and walleye (Haponski and
Stepien, 2013), which should be monitored as restoration efforts continue to ensure
retention of their respective unique genetic signatures.
2.5.2 Genetic composition, divergence, and connectivity of yellow perch stocks
(Hypotheses 2a)
Marked genetic differences between upper and lower Great Lakes spawning
populations indicate low genetic exchange that likely stemmed from former drainage
isolation and post-glacial colonization pathways (see Sepulveda-Villet and Stepien, 2012
accepted). Our results indicate a large genetic break separated yellow perch spawning
population groups between the upper and lower Great Lakes (also see Sepulveda-Villet
and Stepien, 2012 accepted), along with spawning populations of smallmouth bass
(Stepien et al., 2007) and walleye (Stepien et al., 2009, 2010; Haponski and Stepien
23
2013). Differentiation among these fish reproductive populations likely became
pronounced when the early Lake Huron and Lake St. Clair/Erie systems had separate
drainages following the glaciations, dating to ~14 kya (Lewis et al., 1994, 2008). Similar
genetic patterns among these species reflect congruent biogeographic histories. These
patterns likely did not result from recent habitat loss or fragmentation (Laroche and
Durant, 2004; Bessert and Orti, 2008), as their appreciably high genetic diversity levels
exclude recent genetic bottlenecks.
Kin group recognition and fidelity may result in high genetic divergence among
yellow perch spawning populations located in close proximity due to behavioral isolation
(Gerlach et al., 2001; Behrmann-Godel and Gerlach, 2008). Although spawning fidelity
of kin-groups has not been investigated for yellow perch, the closely-related European
perch has been shown to use olfactory cues to discriminate kin from unrelated individuals
(Gerlach et al., 2001; Behrmann-Godel et al., 2006; Behrmann-Godel and Gerlach,
2008). Those studies indicated that European perch clustered in long-term population
groups composed of full and half siblings (Gerlach et al., 2001; Behrmann-Godel and
Gerlach, 2008). Reproductive success was significantly lower in non-kin groups,
reducing pre-zygotic and post-zygotic fitness. Notably, kin-groups exhibited higher
fertilization rates and higher hatching success (Behrmann-Godel and Gerlach, 2008). A
similar reproductive strategy by yellow perch may explain the high genetic divergence
we find among spawning populations across the HEC (mean θST=0.051±0.005;
range=0.011-0.099; here), Lake Erie (mean θST=0.233±0.020; range=0.000-0.665;
Sepulveda-Villet and Stepien, 2011), and their Great Lakes range (mean
24
θST=0.127±0.007; range=0.008-0.282Sepulveda-Villet and Stepien, 2012 accepted). This
merits further investigation.
2.5.3 Genetic isolation is not explained by geographic distance (Hypothesis 2b)
Our findings showed no significant relationship between genetic distance and
geographic distance among yellow perch spawning populations along the HEC, similar to
patterns found in Lake Erie (Sepulveda-Villet and Stepien, 2011). Instead, some
proximate groups were distinguished by high genetic divergence, and some of those
separated by larger geographic distances appeared genetically more similar (here and
Sepulveda-Villet and Stepien, 2011). Similarly, genetic isolation by geographic distance
did not explain relationships among walleye spawning populations along the HEC
(Haponski and Stepien, 2013). Some walleye spawning populations showed greater
similarity among more distant groups, while some closely spaced ones were more
divergent (Stepien et al., 2009, 2010; Haponski and Stepien, 2013). Thus relationships
among reproductive populations of both percids likely are regulated by homing behavior
to natal sites and possible fidelity of kin groups.
2.5.4 Significance of the Belle Isle restoration site
Our results show that the yellow perch spawning population at Belle Isle (E) is
genetically diverse (HO=0.699±0.063; PPA=0.126) and significantly differs from the
populations sampled. Similarly, Haponski and Stepien (2013) discerned high genetic
diversity of walleye spawning at Belle Isle (HO=0.730±0.030; PPA=0.068). Restoration
efforts at the Belle Isle reefs likely have provided increased spawning habitat for many
25
lithophilic spawners including several sucker species (Moxostoma and Hypentelium),
sturgeon (Ascipenser fulvescens), lake whitefish (Coregonus clupeaformis), and walleye
(Manny et al., 2007; Roseman et al., 2007; HTG 2010; Haponski and Stepien, 2013).
Further restoration efforts targeting nearshore phytophilic spawners (i.e., vegetation and
woody debris) likely will increase spawning habitat areas for additional species,
including the yellow perch. Such habitat enhancement may conserve and augment
population genetic diversity, leading to long-term adaptive potential and stability of
fishery stocks along the HEC. Future monitoring efforts and genetic assessments are
important to track this trend.
2.5.5 Conclusions
Our landscape genetic analysis of yellow perch spawning populations discerned
appreciable genetic diversity and distinctiveness in the HEC. These genetic patterns
likely have been maintained despite habitat loss, degradation, and fragmentation.
Pronounced yellow perch population structure likely denotes fidelity to specific spawning
groups (Aalto and Newsome, 1990; Sepulveda-Villet and Stepien, 2011), little exchange
of individuals among them (Haas et al., 1985), and potential kin-recognition, as
implicated for European perch (Behrmann-Godel et al., 2006). As spawning habitat is
restored, this genetic diversity will underlie the long-term success of yellow perch and
other populations along the HEC.
26
2.6 ACKNOWLEDGEMENTS
This research was funded by NOAA Ohio Sea Grant R/LR-13“Temporal and
spatial analyses of walleye and yellow perch genetic stock structure: A high-resolution
database for fisheries management”, USEPA CR-83281401-0“High-resolution
delineation of Lake Erie fish populations: DNA databases for fishery management”, and
the Great Lakes Restoration Initiative Project #70 “Fish habitat enhancement strategies
for the Huron-Erie Corridor”. TJS was supported by an NSF GK-12 DGE#0742395
fellowship “Graduate fellows in STEM high school education: An environmental science
learning community at the land-lake ecosystem interface” and a summer research
assistantship from NOAA project #NA09OAR4170182 “Effects of Bayshore power plant
on ecosystem function in Maumee Bay, western Lake Erie”. Collections were aided by M.
Bagley J. Boase, A. Bowen, A. Cook, J. Chiotti, S. DeWitt, J. Diemond, D. Fielder, K.
Glomski, T. Hartman, P. Kocovsky, R. Kuhaneck, E. Roseman, M. Thomas and M.
Werda, the Ontario Ministry of Natural Resources, Michigan Department of Natural
Resources, Ohio Department of Natural Resources, U.S. Geological Survey, and U.S.
Fish and Wildlife Service. We thank past and present members of the Great Lakes
Genetics Laboratory: D. Murphy, A. Haponski, H. Dean, S. Karsiotis, O.J. SepulvedaVillet, and L. Pierce for field/ laboratory help, and suggestions. J. Pritt, M. Dufour and R.
Kuhaneck also provided input, and P. Uzmann and M. Gray gave logistical support. This
is publication #2013-04 from the Lake Erie Research Center.
27
Table 2.1
Summary statistics for 15 microsatellite loci across Huron-Erie Corridor yellow perch
spawning populations, including: PCR annealing temperature (TA), number of alleles per
locus (NA), allelic size range in base pairs (bp), mean deviation from Hardy-Weinberg
equilibrium within subpopulations (FIS), and among subpopulations (FST).
Locus
TA (ºC)
NA
Svi2
54
5
Svi3
54
Svi4
Size range (bp)
FIS
FST
206-216
0.130
0.088
10
130-152
0.102
0.032
62
33
114-188
-0.060
0.030
Svi7
53
11
158-212
-0.106
0.013
Svi17
60
16
148-184
0.012
0.013
Svi33
60
40
100-190
-0.024
0.021
YP13
54
8
235-271
0.187
0.312
YP17
56
6
208-223
0.066
0.077
Mpf1
56
35
227-321
0.020
0.009
Mpf2
56
41
215-321
0.024
0.005
Mpf3
54
20
107-149
-0.095
0.049
Mpf4
58
31
159-239
0.167
0.047
Mpf5
54
17
127-163
0.021
0.124
Mpf6
54
9
124-160
-0.027
0.038
Mpf7
53
20
142-194
0.029
0.066
Total
---
309
---
0.024
0.057
28
Table 2.2
Geographic and genetic parameters of yellow perch spawning populations, including: location, sample size (N), observed
heterozygosity (HO) ± standard error (s.e.), deviation from Hardy-Weinberg equilibrium within subpopulations (FIS) ± s.e., total
number of alleles across all loci (NA), allelic richness (AR) ± s.e., number of private alleles (NPA), and proportion of private alleles
(PPA). Total values are calculated across all sites as a single unit. Mean values are averages of the seven spawning sites.
Water body
Locality
Year
Lat. (ºN)
Long. (ºW)
N
HO
FIS
NA
L. Huron
A) Saginaw Bay, MI
2004 (32)
43.4292
-83.7536
56
0.678±0.055
0.092±0.039
180
AR
NPA
PPA
8.58±1.27
13
0.07
2007 (24)
St. Clair R.
B)
Algonac, MI
2011
42.6524
-82.5139
23
0.690±0.065
0.075±0.035
148
9.25±1.61
11
0.07
L. St. Clair
C)
Anchor Bay, MI
2009
42.6319
-82.7764
47
0.619±0.072
-0.046±0.023
176
8.51±1.53
12
0.07
D) L’anse Creuse Bay, MI
2003
42.2457
-83.1198
23
0.633±0.071
0.026±0.034
140
8.71±1.50
1
0.01
Detroit R.
E)
2011
42.3469
-82.9533
48
0.699±0.063
0.098±0.037
222
10.51±1.63
28
0.13
HEC
(B-E)
---
---
---
141
0.683±0.055
0.075±0.034
275
14.04±2.34
52
0.19
F)
2009
41.8683
-83.3178
30
0.568±0.071
-0.123±0.036
127
7.10±1.11
3
0.02
G) Sturgeon Creek, ON
2010
42.0083
-82.5875
21
0.576±0.078
0.012±0.038
121
7.77±1.48
5
0.04
(F-G)
---
---
---
51
0.587±0.074
-0.041±0.027
171
11.26±2.13
8
0.05
Total
Across all samples (A-G)
---
---
---
248
0.678±0.062
0.084±0.031
302
20.02±3.26
---
---
Mean
7 spawning populations (A-G)
---
---
---
35
0.637±0.020
0.019±0.031
159
8.62±0.41
10
0.06
Western L.
Belle Isle
Monroe, MI
Erie
29
Table 2.3
Pairwise genetic divergences between yellow perch spawning populations, with θST values below the diagonal and χ2 values from
exact (G) tests of differentiation above the diagonal. Inf. = infinite value, as indicated by GENEPOP. All comparisons remained
statistically significant after sequential Bonferroni correction (Rice, 1989); p values are in italics below each metric. Mean values are
the average θST divergence from all other spawning populations ± standard error.
Location
A
B
C
D
E
F
G
A)
Saginaw Bay, MI
---
Inf.
Inf.
Inf.
Inf.
Inf.
Inf.
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
B)
Algonac, MI
0.039
Inf.
166.7
138.5
<0.0001
<0.0001
<0.0001
C)
Anchor Bay, MI
Inf.
Inf.
Inf.
<0.0001
<0.0001
<0.0001
<0.0001
D)
L’anse Creuse Bay, MI
---
Inf.
126.2
Inf.
<0.0001
<0.0001
<0.0001
E)
Belle Isle, MI
---
Inf.
Inf.
<0.0001
<0.0001
F)
Monroe, MI
---
Inf.
---
<0.0001
G)
Sturgeon Creek, ON
Mean θST
89.0
86.1
<0.0001
---
<0.0001
0.055
0.011
92.3
<0.0001
0.0007
0.040
0.019
0.022
<0.0001
0.0003
<0.0001
0.086
0.046
0.068
0.077
<0.0001
<0.0001
<0.0001
<0.0001
0.074
0.038
0.030
0.043
0.072
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
0.099
0.037
0.030
0.048
0.078
0.052
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
0.066±0.010
0.032±0.006
0.036±0.009
0.042±0.009
0.071±0.006
0.052±0.007
30
<0.0001
--0.057±0.011
Table 2.4
Estimated effective number of migrants per generation among spawning populations
from the equation FST=1/(4NM-1) (Slatkin, 1991). Mean values are the average NM ±
standard error from all other spawning populations.
Location
A
A) Saginaw Bay, MI
---
B) Algonac, MI
25
---
C) Anchor Bay, MI
16
84
---
D) L’anse Creuse Bay, MI
22
50
44
---
E) Belle Isle, MI
10
22
14
12
---
F) Monroe, MI
12
26
34
22
12
---
9
26
34
20
12
18
---
39±9.9
38±10
28±6.1
14±1.7
21±3.5
20±3.7
G) Sturgeon Creek, ON
Mean NM
16±2.7
B
C
31
D
E
F
G
Fig. 2.1. Map of yellow perch spawning populations sampled (lettered A-G according to
Table 1) (Hatch marks and X indicate spawning habitat identified by Goodyear et al.
1982). Lines = primary barriers to gene flow (ranked I–V, in order of decreasing
magnitude) from BARRIER v2.2 (Manni et al., 2004b). Barriers support is indicated by
percent bootstrap support and number of supporting loci (Barrier I: 56%, 13 loci; Barrier
II: 59%, 11 loci; Barrier III: 58%, 12 loci; Barrier IV: 73%, 11 loci).
Fig. 2.2. a) Estimated yellow perch population structure from Bayesian STRUCTURE
analysis (Pritchard et al., 2000; Pritchard and Wen, 2004) for K=3 and 7 groups
determined from ∆K evaluations (Evanno et al., 2005). Black lines separate different
spawning populations, with each individual fish as a thin vertical line colored according
to its estimated group membership. b) Results of ΔK computation (Evanno et al., 2005),
for each scenario tested (K=1-14) showing support for K=3 and 7.
Fig. 2.3. Mantel (1967) regression of the pairwise relationship between genetic distance
(θST/1-θST) and geographic distance (km) for yellow perch spawning populations
(y=0.0001x+0.04; R2 =0.24, p=0.16, N.S.).
32
Fig. 2.1
33
Fig 2.2 a)
b)
34
Fig. 2.3
35
Chapter 3
Temporal population genetic structure of the yellow perch Perca flavescens
(Percidae: Teleostei) within a complex lakescape
Currently in review for publication in Conservation Genetics as:
Sullivan, T.J., Stepien, C.A. (2013) Temporal population genetic structure of the
yellow perch Perca flavescens within a complex lakescape.
3.1 ABSTRACT: The yellow perch Perca flavescens comprises one of the most
important Great Lakes' fisheries whose population fluctuations have been attributed to
habitat alterations, exploitation, and environmental variability. These factors may impact
its genetic diversity, divergence patterns, and overall adaptive potential. This
investigation evaluates the temporal population genetic patterns of yellow perch
spawning groups across locations, collection years, and age cohorts, providing insights
for adaptive fishery management. Patterns and consistency of genetic composition over
space and time are analyzed for eight spawning groups from Lakes St. Clair, Erie, and
Ontario, using 15 nuclear DNA microsatellite loci. We compare genetic consistency of
spawning groups from two time periods (2001-5 and 2009). At two Lake Erie spawning
36
sites, we sampled in greater depth, encompassing six collection years (1985-2010) and 15
age cohorts (1980-2008) spawning at Dunkirk NY and seven age cohorts (1997-2004)
from Monroe MI. Results showed that relative levels of genetic diversity remained
consistent among spawning groups and across time (mean HO 2001-5=0.54±0.07,
2009=0.57±0.08). However, the genetic composition of spawning groups significantly
differed among locations, sampling years, and age cohorts. The 2003 cohort was the most
distinctive and temporal genetic divergence was not explained by genetic isolation over
time. Spatial patterns may result from limited migration and natal homing, whereas
temporal patterns may reflect kin group structuring and differential reproductive success.
Both merit further exploration.
3.2 INTRODUCTION
Understanding population variation and its spatial structure provides important
insight for conservation priorities, including maintaining genetic diversity, population
sizes, and adaptive potential of local populations across a meta-system. Genetic data may
be used to evaluate population responses to exploitation and habitat change (Schwartz et
al. 2007), aiding adaptive management practices (Schwartz et al. 2007, Allendorf et al.
2008). Halbert (1993) defined "adaptive management" as the feedback process of
developing and continually updating predictive tools to inform site-specific conservation
management. For example, population stressors may increase during periods of rapid
environmental change (Di Pippo et al. 2006) leading either to local extirpation (Quinn et
al. 2001, Olsen et al. 2004), or alternatively, population adaptation (Nacci et al. 2002).
37
Identification of population genetic structure and evaluation of genetic responses to
stressors can provide important insights for the conservation of genetic diversity, unique
variability, and adaptive capability.
Local populations may constitute important genetic reservoirs in the face of
stressors such as exploitation, habitat loss, competition with invasive species, and climate
change. In fisheries management, distinctive local populations are termed "stocks", which
are defined by Hallerman et al. (2003) as, “population subunits that freely interbreed,
share a common gene pool, and significantly differ from other such subunits”.
Knowledge of effective population size (NE) stocks, defined as the number of
reproducing individuals in a hypothetical population that would lose genetic variation by
genetic drift or inbreeding at the same rate as the number of real adults in the population
under investigation (Hallerman et al. 2003), further aids evaluation of impacts of
environmental and human interactions on population genetic patterns. This value is a
metric designed to indicate the genetic health of a spawning group. The smaller the
estimated effective population size the greater the chance that a group will experience
loss of genetic diversity due to random genetic drift (Allendorf and Luikart 2007).
Understanding spatial and temporal changes in genetic stock structure and effective
population size can inform adaptive management decisions with regard to stock
delineation and harvest management.
The Great Lakes, specifically Lake Erie, support the largest stocks of yellow
perch Perca flavescens range-wide averaging 180 million fish since 1990 (Minimum=43
Million, Maximum=501 million; Yellow Perch Task Group, YPTG 2012) for which
38
greater knowledge of the distribution of genetic variation may aid future management
decisions. The yellow perch has comprised an important Lake Erie fishery for over 100
years (Leach and Nepsy 1976). Its population size declined over the 1960s through 1985,
attributed to a combination of exploitation, poor water quality, and changes in prey
species composition (Henderson and Nepsy 1989, Tyson and Knight 2001). From 1985 to
the present, recruitment entered a period of shallow recovery (Munawar et al. 2005,
YPTG 2012), with some stocks showing improvement (~160 million Lake Erie yellow
perch estimated for 2012; YPTG 2012). Notably, 4,364 metric tons were harvested from
Lake Erie in 2011, with an estimated value of several hundred million US dollars
(Kocovsky et al. 2013 in press). The fishery of late has been dominated by a very large
2003 year class, followed by poor recruitment in subsequent cohorts (YPTG 2011). This
exemplifies the need for understanding its temporal and spatial genetic stock structure
(Ryan et al. 2003).
Management of Lake Erie's yellow perch is under the jurisdiction of the Great
Lake Fishery Commission’s Lake Erie Committee (with recommendations from the
YPTG), which employs a Management Unit (MU) framework to set annual quotas. This
binational process was enacted in 1955 to address historic differences in catches between
Canadian and United States waters and among the three bathymetric basins of Lake Erie
(Leach and Nepsy 1976, YPTG 2012). Kocovsky et al (2013 in review) indicated that
management has been successful under this framework, but finer-scale population
structure should be examined. Similarly, evaluation of the temporal consistency of stock
structure within the yellow perch fishery may also provide important insight that will
benefit the sustainability of these spawning groups.
39
Mechanisms that impact population structure of yellow perch have been
hypothesized to include limited reproductive movements (Kelso 1973, Rawson 1980,
MacGregor and Witzel 1987, Ontario Ministry of Natural Resources, OMNR 2011), natal
site fidelity (Aalto and Newsome 1990), and kin recognition/aggregation (Gerlach et al.
2001). Kelso (1973) found that most tagged perch were recovered 0-8.0 km from their
initial release area in Long Point Bay of northeastern Lake Erie. Similarly, Rawson
(1980) and OMNR (2011) found that most tagged perch were recovered proximal to their
initial location, suggesting that most stay relatively near their spawning locations.
MacGregor and Witzel (1987) reported that within Long Point bay of Eastern Lake Erie,
a majority of fish captured, tagged, transported outside the bay, and released were found
to return to capture locations, demonstrating some degree of homing in the yellow perch.
It is likely that if individuals stray during non-spawning times, they return to given
spawning locations during the spring (Aalto and Newsome 1990). Aalto and Newsome
(1990) found that following removal of perch egg masses from a spawning site in Nova
Scotia, Canada, fewer individuals spawned there in subsequent years. This indicated that
spawning site selection is not random, and that individuals return to given locations
(Aalto and Newsome 1990). The European perch P. fluviatilis, a close relative of the
yellow perch, use chemical and physical cues to identify related individuals (Gerlach et
al. 2001). Studies discerned that mating within kin groups of European perch may
increase overall fitness, manifested by increased fertilization success and greater hatching
rates (Behrmann-Godel et al. 2006, Behrmann-Godel and Gerlach 2008). Whether kin
recognition and aggregations of related individuals also characterize spawning groups of
yellow perch remains to be investigated.
40
Genetic investigations of yellow perch to date have found that spawning groups of
yellow perch possess lower genetic diversity than walleye Sander vitreus (SepulvedaVillet et al. 2009, Sepulveda-Villet and Stepien 2011, 2012) and are differentiated by
high divergence and very fine-scale population structure, markedly differing in
composition from site to site (Sepulveda-Villet and Stepien 2011, 2012, Sullivan and
Stepien 2013). These patterns were not related to geographic proximity (SepulvedaVillet and Stepien 2011, Sullivan and Stepien 2013), indicating that other factors shape
relationships among spawning groups. These divergences among spawning groups also
are manifested in morphological differences that mirror the genetic patterns (Kocovsky
and Knight. 2012, Kocovsky et al. 2013 in review). To date, it has been unknown
whether genetic variation of yellow perch is consistent among years or age cohorts.
Within Lake Erie, Sepulveda-Villet and Stepien (2011) analyzed 13 primary yellow perch
spawning groups using 15 nuclear microsatellite loci, finding appreciable genetic
diversity and that nearly all groups were distinctive. Although that research made an
important contribution to understanding the population structure of Lake Erie yellow
perch, its sampling design was limited since each spawning group was sampled just once
and different collection years were used, which may have confounded the spatial patterns.
Those issues thus are addressed here, along with an analysis of temporal variation among
collection years and age cohorts.
The present study analyzes the temporal and spatial genetic structure of yellow
perch spawning groups across collection years and age cohorts, aiming to increase
understanding of factors regulating population patterns. Specific questions are: (1) Does
genetic diversity differ among yellow perch spawning groups, sampling years, and/or age
41
cohorts? (2) Does their genetic composition vary significantly over time and/or space?
(3) Are there genetic differences among age cohorts at a given site? and (4) Does
effective population size of the spawning population at Dunkirk NY change among
sampling years?
3.3 MATERIALS AND METHODS
3.3.1 Sample collection, DNA extraction, and amplification
Yellow perch were sampled by state and federal agency biologists from eight
spawning sites (Fig. 1) at spring spawning times during 2001-5 and again in 2009. Sites
included: Anchor Bay MI in Lake St. Clair (site A; 42.6319, -82.7764), Monroe MI (B;
41.8683, -83.3178), Erieau ONT (C; 42.2515, -81.9167), Fairport OH (D; 41.8058, 81.4178), Perry OH (E; 41.8077, -81.1452), Erie PA (F; 41.1728, -80.2694), and Dunkirk
NY (G; 42.5047, -79.3339) in Lake Erie, and Rochester NY (H; 43.2880, -77.1414) in
Lake Ontario. The samples collected between 2001 and 2005 were used previously for a
larger geographic scale study by Sepulveda-Villet and Stepien (2011). Here, we
resampled spawning yellow perch from a subset of the same sites in 2009 to test for
consistency in genetic structure between the two sampling time periods. To further test
for temporal variability, we analyzed the Dunkirk NY reefs (G) site in depth, adding
samples from the 1985, 2008, and 2010 spawning events. Age data were provided for the
samples from Monroe MI (B) and Dunkirk NY (G), allowing for analysis of variation
among age cohorts. DNA from the 1985 samples was obtained from uncleaned dried fish
42
scales, whereas all other samples were fin clips preserved in 95% ethanol in the field that
then were stored at room temperature.
Genomic DNA from the ethanol preserved fin clips was extracted and purified
using DNeasy Qiagen kits (QIAGEN, Inc., Valencia Ca.), and aliquots were frozen,
labeled, and archived. Genomic DNA from the dried scale samples was extracted using
the technique of Hutchinson et al. (1999). Genetic variation was analyzed from 15
nuclear DNA microsatellite loci following Sepulveda-Villet and Stepien (2011, 2012)
using primers Svi2, 3, and 7 from Eldridge et al. (2002), Svi4, 17, and 33 from Borer et al.
(1999), YP13 and 17 from Li et al. (2007), and Mpf1-7 from Grzybowski et al. (2010).
For the scale samples, loci Svi7, 17 and Mpf1 failed to amplify, attributed to lower
template quality. Therefore, analyses with the 1985 samples were based on analyses of
the remaining 12 loci. Forward primers were synthesized with one of four 5' fluorescent
labels, allowing pool-plexing during analysis (grouped as follows: Svi2+7, Svi3+33,
Svi4+17, YP13+17, Mpf1+2+5+6, and Mpf3+4+7).
Polymerase chain reactions (PCR) followed the methodology used by SepulvedaVillet and Stepien (2011 and 2012) and Sullivan and Stepien (2013). Reactions contained
50 mM KCl, 1.5 mM MgCl2, 10 mM Tris-HCl, 50 μM of each deoxy-nucleotide, 0.5 μM
each of the forward and reverse primers, 2% dymethyl sulfoxide (DMSO), 5-30 ng DNA
template, and 0.6-1.2 μM of Taq polymerase per 10 μL reaction volume. Positive and
negative controls were included in each reaction. An initial cycle of 2 min at 94°C was
used for strand denaturation, followed by 40 cycles of denaturation (94°C, 30 s), primer
annealing (1 min) at a primer-specific temperature (TA; Table 1), and polymerase
43
extension (72°C, 30 s). A final extension at 72°C for 5 min was included to minimize
partial strands.
Amplification products were diluted 1:50 with ddH2O, then a 1 μL aliquot was
added to 13 μL of a formamide and ABI GeneScan-500 size standard solution, loaded
onto 96-well plates, and denatured for 2 min at 95°C. The denatured products were
analyzed on our ABI 3130XL Genetic Analyzer with GENEMAPPER 3.7 software
(Applied Biosystems Inc., Foster City, CA). Output profiles were reviewed manually to
confirm correct identification of the allelic size variants.
3.3.2 Data analyses
Population samples were tested for conformance to Hardy-Weinberg Equilibrium
(HWE) expectations at each locus, and significance was estimated using the Markov
Chain Monte Carlo method (MCMC) with 1000 randomization procedures (Guo and
Thompson 1992) in GENEPOP v4.0 (Rousset 2008, http://mbb.univmontp2.fr/MBB/subsection/downloads.php?section=2). Any deviations were analyzed
for excess or deficiency of homozygotes, and loci were tested for linkage disequilibrium
(LD). Levels of significance for both tests were adjusted using sequential Bonferroni
corrections (Rice 1989) to minimize Type I error. Possible presence of nonamplified or
null alleles was tested with MICRO-CHECKER v2.3.3 (van Oosterhout et al. 2004, 2006,
http://www.microchecker.hull.ac.uk).
To test whether genetic diversity differed among spawning populations, year
groups, collection years, and/or age cohorts, expected and observed heterozygosity values
(HE and HO) and the inbreeding statistic (FIS) were calculated in GENEPOP v4.0.
44
Number of alleles (NA) and allelic richness (AR; number of alleles per locus independent
of sample size per El Mousadik and Petit (1996)) were determined in FSTAT v2.9.3.2
(Goudet 2002, http://www2.unil.ch/ popgen/softwares/fstat.htm). Number (NPA) and
proportion (PPA) of private alleles were calculated using CONVERT v1.31 (Glaubitz
2004, http://www.agriculture.purdue.edu/fnr/html/faculty/rhodes/students%20and%20
staff/glaubitz/software.htm). We tested whether observed heterozygosity or allelic
richness values differed among samples using Friedman sum rank tests in the R statistical
software suite v2.15.2 (R Development Core Team 2012, http://www.r-project.org/), with
loci treated as blocks per spawning population.
To investigate whether genetic composition significantly differed among
spawning populations (for each sampling period), sampling years, collection years (at
Dunkirk, NY), and cohorts (at Monroe MI and Dunkirk NY), unbiased θ estimates of F
statistics (Weir and Cockerman 1984) and their levels of significance were evaluated in
FSTAT. Models using θST (the FST metric of Weir and Cockerman 1984) have been
shown to have greater resolution among recently diverged populations (Balloux and
Lugon-Moulin 2002), and thus were employed here. Pairwise comparisons also were
conducted using a nonparametric (exact G) procedure (Raymond and Rousset 2005), with
probability estimated from MCMC in GENEPOP v4.0. This approach does not assume a
normal distribution and is not influenced by sample size, but may have less statistical
power (Goudet et al. 1996). Sequential Bonferroni corrections (Rice 1989) were used to
minimize the potential for Type I error for all pairwise comparisons.
Changes in the spatial relationships among spawning groups were tested using a
computational geometric approach in BARRIER v.2.2 (Manni et al. 2004 a, b). Pairwise
45
divergence values (θST) were mapped onto a matrix of sample coordinates (latitude and
longitude) to identify genetically different spawning groups, irrespective of their de facto
relationships. Relative magnitudes of each genetic barrier were ranked based on their
genetic divergence relative to their spatial positioning. Support for barriers were
evaluated according to (1) their number of supporting loci and (2) bootstrap analysis of
2000 multilocus θST matrices created in GENELAND v.3.1.4 (Guillott et al. 2005 a, b,
2008) using R. Genetic barriers supported by a majority of loci and >50% bootstrap
support are reported here.
Analysis of Molecular Variance (AMOVA; Excoffier et al. 1992) in ARLEQUIN
v.3.5.12 (Excoffier et al. 2010) evaluated the distribution of genetic variation under a
series of possible scenarios: (a) among the eight spawning groups and between their two
sampling periods, (b) between the two sampling periods and among their eight spawning
groups, (c) between the sexes and among their spawning groups, and (d) between two
sites (Monroe MI (B) and Dunkirk NY (G)) and among their age cohorts (born in 1998,
1999, 2001, 2002, or 2003).
The program GENECLASS v2.2 (Piry et al. 2004, http://www1.montpellier.
inra.fr/ URLB/ index.html) tested the self-assignment of individuals to population groups
using a simulated population size of 10,000 individuals per site, with a 0.01 rejection
level (Cornuet et al. 1999). Population structure further was assessed through the
Bayesian-based program STRUCTURE v.2.3.4 (Pritchard et al. 2000, Pritchard and Wen
2004, http://pritch.bsd.uchicago.edu/structure.html). Results were compared from 10
independent runs at K=1 (a single spawning group) to K=24 (1.5x the number of
sampling sites), with burn-ins of 100,000 and 500,000 replicates. We evaluated
46
consistency among runs, the probabilities of individual assignments to groups, and loglikelihood values. The optimal K scenario was determined via the posterior probability
procedure of Pritchard et al. (2000) and ΔK likelihood from Evanno et al. (2005).
Correspondence between genetic distances (θST/1-θST) and geographic distances
(shortest waterway distances) or time (separating age cohorts) was tested using ISOLDE
in GENEPOP (Rousset 1997). The regression line fit and significance were calculated
using Mantel’s (1967) procedure with 1000 permutations.
We evaluated effective population size (NE) for the spawning group at Dunkirk
NY using temporal methods, for which we grouped together age cohorts that were
separated by at least nine years. This resulted in three cohort groups (1980-1982, 19951998, and 2007-2008), whose individuals likely represented independent generations. NE
and its 95% confidence intervals were calculated using the following three methods: (1)
MNE v1.0 (Wang and Whitlock 2003) estimated NE under the psuedo-likelihood method
of Wang (2001), assuming a single isolated group. An infinite value for the confidence
interval was produced when any simulation approached 5000; (2) CoNe v.1.01
(Anderson 2005) estimated NE between samples using the likelihood model of Berthier et
al (2002) with a Monte Carlo approximation; (3) NeEstimator v1.3 (Peel et al. 2004)
calculated NE with the moment based approach for each interval from Waples (1989).
Lastly, the long-standing effective population size was derived from the harmonic mean
of the individual NE estimates, following Franckowiak et al. (2009).
47
3.4 RESULTS
3.4.1 Genetic diversity of spawning groups, collection years, and cohorts (Question 1)
All loci were discerned to be unlinked and conformed to expectations of HWE
after sequential Bonferroni correction (Rice 1989). There were 368 alleles recovered for
the 15 loci (24.5 alleles/locus), with the number per locus ranging from 6 at Svi2 to 49 at
Mpf1 and 2. Based on FST values, Mpf5, YP13, and Svi33 showed the most divergence
among spawning groups.
All spawning groups possessed consistent levels of genetic diversity over the two
sampling periods (Table 1). Their mean observed heterozygosities (2001-5=0.54±0.07;
2009=0.57±0.07), mean allelic richnesses (6.8±1.2; 7.1±1.2), and proportions of private
alleles (0.03; 0.05) were similar between the sampling events (Table 1). Observed
heterozygosity values also were similar across the spatial distribution of the spawning
groups (2001-5:χ2=2.9, df=7, p=0.89, N.S.; 2009: χ2=7.2, df=7, p=0.41, N.S.). However,
their allelic richness values varied across the sampling design (2001-5: χ2=14.5, df=7,
p=0.04; 2009: χ2=16.6, df=7, p=0.02; Table 1). For the 2001-5 samples, allelic richness
appeared highest at Monroe MI (A; 7.8±1.3) and lowest at Fairport OH (D; 5.5±0.9); in
2009, it was highest at Erie PA (F; 8.1±1.4) and lowest at Monroe MI (B; 6.2±0.9). The
overall proportion of private alleles was similar among spawning groups and between the
sampling year periods, except at Erie PA (F) and Rochester NY (H), where more private
alleles were recovered in 2009. This likely was due to the larger sample sizes taken at
those sites during 2009.
48
Levels of genetic diversity (HO) of yellow perch spawning at Dunkirk NY (G) in
eastern Lake Erie were consistent across the six collection years (χ2=13.9, df=5, p=0.92,
N.S.; Table 2A). Allelic richness significantly varied (χ2=21.8, df=5, p=0.001; Table 2A),
being greatest for the 1985 samples (11.1) and lowest in 2004 (7.4). Age cohorts from
Dunkirk NY (G) and Monroe MI (B) displayed different patterns. Observed
heterozygosity of the Dunkirk NY cohorts was consistent from 1980-2008 (χ2=7.5,
df=14, p=0.92, N.S.; Table 2B), but those from Monroe MI in western Lake Erie showed
significant variation (1997-2004, χ2=20.1, df=6, p=0.003) being higher in the 2001 and
2002 cohorts (Table 3). Allelic richness significantly differed among the Dunkirk NY age
cohorts (χ2=32.3, df=14, p=0.004; Table 2B), being highest in 1980 (5.3), 1981 (5.1) and
1982 (5.2). In contrast, allelic richness was similar among all cohorts from Monroe MI
(χ2=9.3, df=6, p=0.16, N.S.; Table 2C). Private alleles were recovered in all collection
years (Table 2A) and characterized nearly all Dunkirk NY cohorts (G; Table 2B), with
proportions ranging from 0.00 in 2008 to 0.10 in 1980. Greater proportions of private
alleles were identified in the earlier cohorts (1980-1982). At Monroe MI (B), all cohorts
contained private alleles, whose proportions ranged from 0.05 (1999) to 0.16 (1998;
Table 2C). Numbers of private alleles appeared to decrease in the later samples from
Dunkirk NY and Monroe MI.
3.4.2 Spatial and temporal population genetic structure (Questions 2 and 3)
The genetic composition of all yellow perch spawning groups were significantly
distinct from all other groups (2001-5: mean θST=0.094, range=0.007-0.168; 2009: mean
θST=0.042, range=0.002-0.069; Table 3). Their differences did not correspond to a
49
genetic isolation by geographic distance pattern (2001-5" y=0.00007x+0.09, R2=0.036,
p=0.22, N.S., Fig 4A; 2009: y=0.00003x+0.04, R2=0.078, p=0.054, N.S., Fig. 4B). In
2009, spawning samples from the nearby locations of Fairport OH (D) and Perry OH (E;
separated by 23 km) did not significantly differ in genetic composition as measured by
mean θST (0.002), but differed using the nonparametric exact G test. The magnitude of
overall divergence among sampling groups was lower in 2009 (mean θST=0.042,
range=0.002-0.069) than in the 2001-5 samples (0.094, 0.007-0.168).
Barrier analysis identified four primary genetic discontinuities in each sampling
period (Fig. 1). For the 2001-5 period, barrier I (10/15 loci, 89%) separated Anchor Bay
MI (A) from all other sites; barrier II (10, 67%) isolated the group spawning at Erie PA
(F); barrier III (11, 53%) distinguished the spawning group at Rochester NY (H); and
barrier IV (12, 66%) isolated the spawning group at Fairport OH from all others. For the
2009 samples, barrier I (12, 65%) divided groups spawning at Anchor Bay MI (A),
Monroe MI (B), and Erieau ONT (C) from those along the southern shore of Lake Erie
and in Lake Ontario; barrier II (14, 53%) separated the spawning group at Erieau ONT
(C) from those in Anchor Bay MI (A) and Monroe MI (B); barrier III (12, 59%) divided
perch at Rochester NY (H) from spawning groups along the southern shore of Lake Erie;
and Barrier IV (12, 54%) distinguished Anchor Bay MI (A) from those in western Lake
Erie at Monroe MI (B). The genetic barriers among spawning groups within Lake Erie
changed over time; however, no genetic exchange occurred among the lakes during either
sampling period. Thus the gene pools of each appeared isolated and distinctive.
Individuals from all spawning group samples showed very high self-assignment in
the GENECLASS analyses (Fig 2.), with very few cross-assigning between the sampling
50
periods (i.e., 2001-5 to 2009, and vice-versa). Results indicate substantial spatial and
temporal differentiation of spawning groups. STRUCTURE analyses supported 8
population groups (K=8), showing very little cross-assignment between the two sampling
periods, except at Rochester NY (H; Fig. 3). High self-assignments for population groups
spawning at Anchor Bay MI (A), Fairport OH (D), Erie PA (F), and Dunkirk NY (G)
characterized the 2001-5 sampling period, but appeared much less in 2009. Overall,
several spawning groups displayed a mixed assignment signature in 2009, with proximate
samples in Fairport OH (D) and Perry OH (E) appearing very similar to each other. This
mirrored the lower pairwise divergence values in 2009 (see Table 3).
AMOVA tests revealed relatively low yet significant partitioning of genetic
variation between the two sampling periods (variation=1.71%, p<0.001), similar to the
significant pairwise divergence values recovered between sampling periods for given
spawning groups (Table 3). Annual variation (1985-2010) also was large in the pairwise
comparisons for perch spawning at Dunkirk NY (1985-2010; mean θST=0.073,
range=0.011-0.141; Table 4A). AMOVA tests showed variation among the spawning
groups across their geographic distribution was about three times greater (6.23%,
p<0.001; Table 5) than that discerned between the sampling periods. This scenario
provided the best explanation for the data patterns, accounting for 7.93% of the overall
variation. When the partitioning scenario was reversed (first combining the two sampling
regimes into 8 spawning groups), much less of the variation was explained among sites
(<0.01%, p=0.598), and more was partitioned between the two sampling periods (7.76%,
p<0.001; Table 5). However, the latter scenario explained less of the overall pattern,
accounting for a total of 7.77% of the overall variation.
51
AMOVA results also indicated that no differences occurred between males and
females (<0.01%, p=N.S.; Table 5), but the genetic composition at a given spawning site
significantly varied among age cohorts (4.56%, p<0.001; Table 5). Pairwise comparisons
at Monroe MI (B, 1997-2004, mean θST=0.010, range=0.000-0.030; Table 4C) and
Dunkirk NY (G, 1980-2008, 0.061, 0.000-0.172; Table 4B) showed significant genetic
divergence among many of their age cohorts. At both sampling sites, individuals from the
2003 cohort showed marked levels of genetic difference (Table 4B and C). The isolation
by time test indicated that genetic distance among cohorts at Dunkirk NY (G) was not
related to their separation time (y=0.0009x+0.078, R2=0.026, p=0.67, N.S.; Fig. 4).
3.4.3 Temporal effective population size of the Dunkirk NY spawning group (Question 4)
Relatively low NE values were estimated for the group spawning at Dunkirk NY
from all three models. The moment based approach of Waples (1989) discerned
consistently low values, 15 (-95%=10 / +95%= 24) for the 1980 to 1990 comparison and
15 (10 / 26) for the 1990 to 2000 comparison. The longstanding (harmonic) mean was
estimated at 15 (10 / 25) under the moment based approach. The pseudo-likelihood
method of Wang (2001) indicated slightly higher values, which also appeared temporally
consistent. That method found 38 (28 / 57) for the 1980 to 1990 comparison and 72 (42 /
228) for the 1990 to 2000 comparison, with a longstanding (harmonic) mean of 49.7
(33.6 / 91.2). Anderson’s (2005) coalescent likelihood method revealed similarly low
values of 16 (11 / 23) for the 1980 to 1990 comparison, and a higher value of 398 (241 /
892) for the 1990 to 2000 comparison. Its longstanding mean was 30.8 (21 / 44.8),
similar to the values recovered from the other models. Overall, these results indicate that
52
the effective population size of the spawning group at Dunkirk NY is low, yet has
remained consistent over the past 30 years.
3.5 DISCUSSION
3.5.1 Spatial and temporal patterns in genetic diversity (Question 1)
The overall levels of genetic diversity recovered in this study (HO=0.56±0.08)
match those found by Sepulveda-Villet and Stepien (2012) for yellow perch spawning
groups across the Great Lakes (0.55±0.02) and their North American range (0.53±0.02),
using the same loci, procedures, and in the same laboratory. Values here were slightly
lower than those recovered for perch spawning in Lake Huron, the Huron-Erie Corridor
(HEC), and western Lake Erie (0.64±0.02; Sullivan and Stepien 2013). Diversity values
of yellow perch were lower than those recovered for its sister species, the European perch
(5 loci; HO=0.79, Demandt 2010) and for walleye spawning groups using 9 microsatellite
loci across Lake Erie (0.70±0.01, Strange and Stepien 2007), the HEC (0.72±0.009,
Haponski and Stepien 2013), the Great Lakes (0.71±0.011, Stepien et al. 2009), and its
native range (0.68±0.02, Stepien et al. 2009). Although genetic diversity of yellow perch
is lower than some other percid species, the values recovered here appear similar to other
freshwater fishes based on the meta-analysis of DeWoody and Avise (2000), which
showed average heterozygosity levels of 0.54 for 13 surveyed freshwater fishes (using 75
microsatellite loci and 7755 individuals).
The present study discerns consistent levels of observed heterozygosity and
significantly variable levels of allelic richness among spawning groups of yellow perch in
53
both 2001-5 and 2009. In contrast, allelic richness values were similar among spawning
groups of walleye across Lake Erie (Strange and Stepien 2007, Stepien et al. 2012),
which differed in heterozygosity levels, being higher in the eastern basin than in the
western basin. These comparisons indicate that the distribution of genetic diversity
among spawning groups differs between the two species.
Populations of other fishes showed temporal genetic stability in their number of
alleles and heterozygosity values over generations, including Atlantic salmon Salmo
salar (Tessier and Bernatchez 2002, Lage and Kornfield 2006), rainbow trout
Oncorhynchus mykiss (Heath et al 2002), and walleye (Stepien et al. 2012). However,
populations of the North Sea cod Gadus morhua (Hutchinson et al. 2003) and New
Zealand snapper Pagrus auratus (Hauser et al. 2002) underwent temporal changes,
marked by declines in observed heterozygosity and numbers of alleles. Those temporal
reductions were linked to high levels of exploitation from commercial and recreational
fishing. In our study, allelic richness values fluctuate among the six sampling years and
the 15 age cohorts for the Dunkirk NY spawning group, but no significant decreases were
discerned. The age cohorts of yellow perch spawning at Monroe MI show stable values
for allelic richness, and their observed heterozygosity values increase during recent years.
These results indicate that exploitation levels in Lake Erie do not appear to have
decreased genetic diversity of yellow perch. This may be attributed to effective use of
management quotas protecting spawning groups from genetic impacts associated with
overharvest of individuals. It also is possible that this pattern reflects increased
population sizes, as lakewide estimates of abundance have risen slowly from 1985 to the
present (Munawar et al. 2005, YPTG 2012) following declines from the early 1960’s
54
(Henderson and Nepsy 1989). The variability recovered in the estimates of
heterozygosity and allelic richness of cohorts over time also may be influenced by lower
sample sizes.
3.5.2 Spatial and temporal patterns in genetic divergence and stock structure
(Questions 2 and 3)
These results show that spawning groups of yellow perch genetically diverge
across Lake Erie, at a much smaller scale and greater level than was known prior to
recent studies (see Sepulveda-Villet and Stepien 2011, 2012). Thus, spawning groups
have high genetic distinctiveness, significantly differing in composition from others.
Findings from this study match conclusions across the HEC (Sullivan and Stepien 2013),
the Great Lakes, and the North American range of yellow perch (Sepulveda-Villet and
Stepien 2012). The present investigation uniquely evaluates the stability of these patterns
over time. Results show that the magnitude of divergence among spawning groups was
lower in 2009 compared to the 2001-5 sampling period. This likely reflects the span of
the previous sampling design, since samples were not taken from all locations during the
same spawning year by Sepulveda-Villet and Stepien (2011).
Our investigation found that major genetic divisions or barriers distinguishing the
gene pools of spawning groups vary between the two sampling time periods. It is likely
that these patterns are influenced by a combination of factors. First, since samples for the
second sampling period were collected at spawning time in the same year, there is much
less sampling variance than in the 2001-5 span for the earlier samples (see SepulvedaVillet and Stepien 2011). The second may reflect the preponderance of the prolific 2003
55
year class of yellow perch, which was the largest in recent decades (YPTG 2012). This
large 2003 year class characterized both Lake Erie yellow perch and walleye populations.
Research by Strange and Stepien (2007) and Stepien et al. (2012) hypothesized an
increase in genetic connectivity of walleye spawning groups along Lake Erie’s southern
shore during reproduction for that year. Thus the yellow perch year class of 2003 may
have been the result of greater genetic mixing among spawning groups (see SepulvedaVillet and Stepien 2011, Kocovsky and Knight 2012, Kocovsky et al. 2013 in review). It
is important to note that the separation of Lake Erie gene pools from those of neighboring
lakes (Lakes St. Clair and Ontario) has remained consistent over time for yellow perch
(Sepulveda-Villet and Stepien 2011, 2012, Sullivan and Stepien 2013), as well as for
walleye (Stepien and Faber 1998, Strange and Stepien 2007, Stepien et al. 2009). This
divergence pattern appears to trace to post glacial recolonization history (Holcombe et al.
2003) and subsequent long time isolation of gene pools (Stepien and Faber 1998;
Sepulveda-Villet and Stepien 2012, Bernatchez and Wilson 2002).
Yellow perch collected in different years (at Dunkirk NY) and belonging to
different age classes (at Monroe MI and Dunkirk NY) vary genetically, but to a lesser
degree than characterizes the divergence among spawning groups. Similar temporal
fluctuations in genetic composition were identified in spawning groups of steelhead trout,
with numbers of alleles and heterozygosity being relatively consistent. (Heath et al.
2002). The temporal changes we observed among age cohorts of yellow perch spawning
at Dunkirk NY do not correspond to genetic isolation by time (i.e., do not show a
consistent pattern of genetic drift). In other words, the differences among cohorts are
unrelated to the time period separating them, and may reflect sampling fluctuations. It
56
might alternatively suggest that selective pressures may lead to temporal variability,
which remains to be investigated.
In comparison, age cohorts (from 1988 to 1996) of the European white sea bream
Diplodus sargassus varied in observed heterozygosity and were highly divergent (Planes
and Lefant 2002, Lefant and Planes 2002). Similar patterns were identified in populations
of kelp bass Paralabrax clathratus, with adult and larval gene pools varying across a
season of larval settlement (Selkoe et al. 2006). Those results showed that impacts of
increased kinship within cohorts and changes in delivery of larvae to nursery grounds led
to temporal divergence. The genetic composition of five collection years for a European
perch spawning group in Sweden diverged from 1977 to 2000 (Demandt 2010). AMOVA
results indicated a significant amount of that variation was explained by temporal
changes (Demandt 2010). That study also discerned a significant relationship between
genetic divergence and time (Demandt 2010), in contrast to our results in the present
analysis. Both kelp bass and European perch aggregate in kin-groups (as has been
hypothesized for yellow perch) and experience high fecundity and high mortality at very
early life history stages (type III survivorship). Those life history characters may lead to
differential reproductive output for some adults, whereby a few individuals produce most
of the progeny (Hedgecock 1994). Differential reproductive outputs and selection
likewise may shape temporal genetic patterns of yellow perch age cohorts, collection
years, and spawning groups. Further genetic investigations are needed to elucidate the
mechanisms behind the patterns observed in the present study to understand their impacts
on stock structure.
57
3.5.3 Temporal effective population size of the Dunkirk NY spawning group (Question 4)
The effective population size of yellow perch spawning at Dunkirk NY was
relatively low and temporally consistent. These values support the hypothesis of small
spawning groups that are genetically distinct from others. A single NE estimate showed
an increase over time. Larger values (harmonic mean range: 124.6-180.2) were estimated
for walleye of Lake Escanaba (a much smaller system) from 1950-1990 by Franckowiak
et al. (2009). This pattern may be related to larger spawning group sizes of walleye
compared to yellow perch, or to impacts of differences in kin group structuring between
the two species, which merits further evaluation. Our estimates of NE for yellow perch are
very similar to values found for European perch spawning groups in Sweden (11-62
individuals; Demandt 2010). Since both species are closely related, it is likely that similar
life history patterns regulate their population structures. Both species of Perca likely are
structured by kin groups that tend to spawn together, have relatively high fecundity, and
experience high mortality in early life stages, which characterize type III survivorship.
Perca appear to correspond to a sweepstake reproductive success model, with a limited
number of individuals often being over-represented in age cohorts (Hedgecock 1994,
Hedgecock and Pudovkin 2011). Further research is necessary to understand these
relationships and how they impact yellow perch population structure.
3.5.4 Conclusions and management implications
An understanding of the genetic composition, diversity, and spatial stock structure is
essential for effective adaptive management of yellow perch, given its economic, social,
and ecological value in the Great Lakes ecosystem. Our genetic analysis identifies
58
important insights into the dynamic nature of the spatial and temporal patterns underlying
the genetic stock structure of yellow perch over the past three decades. First, yellow
perch genetic diversity levels can be generalized as being consistent and stable over time.
The influences of habitat degradation, loss, and exploitation might have acted to reduce
genetic diversity through declines in population size and bottlenecks; however, this does
not appear to have been the case. A recent study by Sullivan and Stepien (2013) likewise
described comparable levels of genetic diversity characterizing yellow perch spawning
within the highly modified Huron-Erie Corridor. Second, all population groups were
genetically different from one another with high self-assignment. This indicates the
presence of many genetically distinct and unique groups of yellow perch and supports the
hypothesis that spawning populations are structured at a very fine-scale (see SepulvedaVillet and Stepien 2011, 2012, Kocovsky and Knight 2012, and Kocovsky et al. 2013 in
press). Major genetic breaks distinguish yellow perch gene pools from neighboring lakes,
with others in Lake Erie appearing to shift between the two sampling regimes. This may
reflect possible greater genetic mixing leading to the success of the 2003 year class, as
seen in walleye (Strange and Stepien 2007, Stepien et al. 2012). These types of changes
underlie the importance of regular genetic monitoring as a management tool and will aid
in understanding the dynamics of populations that display variable patterns in year class
strength. Lastly, this investigation identifies temporal fluctuations in the genetic
compositions of spawning groups and age cohorts that merit further investigation. Results
suggest that multiple factors shape population reproductive structure. Identifying these
other factors, with respect to their differential effects among life history stages, and
assessing their impacts may help to preserve genetic diversity and local adaptation.
59
3.6 ACKNOWLEDGEMENTS
This research was supported by National Oceanic and Atmospheric Administration Ohio
Sea Grant (R/LR-13 and NA09OAR4170182 to C.A.S.), United States Environmental
Protection Agency (CR-83281401-0 to C.A.S), and a National Science Foundation GK12 program fellowship that paid the stipend of T.J.S (DGE#0742395 to C.A.S.). We
thank B. Beckwith, D. Einhouse, A. Ford, P. Kocovsky, C. Knight, R. Knight, A. M.
Gorman, C. Murray, M. Sanderson, D. Sek, M. Thomas, J. Tyson, D. Zellar, R. Zimar,
the Michigan Department of Natural Resources, the New York State Department of
Environmental Conservation, the Ohio Department of Natural Resources, the Ontario
Ministry of Natural Resources, the Pennsylvania Fish and Boat Commission, and the
United States Geological Survey for providing samples. We are grateful for help and
suggestions from past and present members of the Great Lakes Genetics/Genomics
Laboratory: D. Murphy, A. Haponski, H. Dean, S. Karsiotis, O.J. Sepulveda-Villet, S.
Woolwine, C. Prichard, and L. Pierce. This is publication number #20XX-XX from the
Lake Erie Research Center.
60
Table 1
Sample data and genetic parameters for yellow perch spawning populations,
based on variation at 15 nuclear microsatellite loci, including: collection date,
sample size (N), observed (HO) and expected (HE) heterozygosity ± standard error
(S.E.), deviation from Hardy-Weinberg equilibrium within subpopulations (FIS) ±
S.E., the number of alleles across all loci (NA), allelic richness (AR) ± S.E.,
number of private alleles (NPA), and proportion of private alleles (PPA).
2001-2005
Date
N
HO±S.E.
HE±S.E.
FIS
NA
AR
5/26/2005
39
0.53±0.08
0.60±0.08
0.10±0.07
152
Locality
NPA
PPA
6.8±1.2
5
0.03
Lake St. Clair
H)
Anchor Bay MI
Lake Erie
I)
Monroe MI (MU1)
4/13/2004
48
0.56±0.07
0.63±0.07
0.10±0.03
184
7.8±1.3
4
0.02
J)
Erieau ON (MU2)
4/15/2003
30
0.55±0.08
0.54±0.08
-0.02±0.02
146
7.2±1.3
2
0.01
K)
Fairport OH (MU2)
4/19/2003
20
0.51±0.08
0.50±0.08
-0.03±0.03
94
5.5±0.9
5
0.05
L)
Perry OH (MU3)
4/28/2003
48
0.52±0.07
0.59±0.08
0.08±0.03
111
7.4±1.3
6
0.05
M)
Erie PA (MU3)
5/05/2001
20
0.51±0.08
0.55±0.07
0.08±0.07
109
6.3±1.2
0
0
N)
Dunkirk NY (MU4)
5/15/2001
37
0.54±0.06
0.60±0.06
0.09±0.05
143
6.6±1.0
4
0.03
7/16/2002
14
0.52±0.07
0.62±0.06
0.16±0.06
100
6.9±1.1
1
0.01
Lake Ontario
O)
Rochester NY
Total
---
256
0.54±0.06
0.63±0.07
0.14±0.04
279
18.6±2.8
---
---
Mean
---
32
0.54±0.07
0.58±0.07
0.08±0.05
130
6.8±1.2
4
0.03
61
2009
Date
N
HO
HE
FIS
NA
AR
NPA
PPA
6/03/2009
47
0.62±0.07
0.64±0.07
-0.05±0.02
176
7.5±1.3
9
0.05
4/21/2009
30
0.57±0.07
0.66±0.09
-0.12±0.04
127
6.2±0.9
1
0.01
5/05/2009
36
0.54±0.09
0.52±0.07
0.02±0.06
145
6.3±1.0
4
0.03
4/30/2009
48
0.57±0.08
0.56±0.08
-0.02±0.03
173
7.1±1.2
7
0.04
4/30/2009
48
0.56±0.08
0.56±0.08
-0.01±0.03
171
7.0±1.2
10
0.06
5/12/2009
34
0.57±0.07
0.61±0.08
0.08±0.02
180
8.1±1.4
13
0.07
5/13/2009
30
0.57±0.09
0.54±0.08
-0.06±0.04
133
6.6±1.2
5
0.04
4/24/2009
48
0.55±0.08
0.59±0.07
0.09±0.06
190
7.8±1.5
15
0.08
---
321
0.58±0.08
0.60±0.07
0.02±0.02
340
22.6±3.6
---
---
---
40
0.57±0.08
0.59±0.08
-0.02±0.04
162
7.1±1.2
8
0.05
62
Table 2
Sample data and genetic parameters for yellow perch spawning at (A) Dunkirk NY for
six collection years (1985-2010; 12 loci), (B) age cohorts from Dunkirk NY (born in
1980-2008; 12 loci), and (C) age cohorts from Monroe MI (1997-2004; 15 loci). Data
include collection date or cohort birth year, sample size (N), observed (HO) and expected
(HE) heterozygosity ± standard error (S.E.), deviation from Hardy-Weinberg equilibrium
within subpopulations (FIS) ± S.E., the number of alleles across all loci (NA), allelic
richness (AR) ± S.E., number of private alleles (NPA), and proportion of private alleles
(PPA).
A)
Date
N
5/16/1985
HO
HE
FIS
NA
34 0.54±0.09 0.64±0.07
0.20±0.07
144 11.1±2.1 23
0.16
5/15/2001
37 0.57±0.06 0.61±0.06
0.08±0.06
116
8.7±1.5 14
0.12
5/13/2004
48 0.59±0.08 0.55±0.06 -0.09±0.07
110
7.4±1.3 10
0.09
5/13/2008
30 0.55±0.08 0.56±0.09 -0.01±0.02
121
9.7±2.1 11
0.09
5/13/2009
30 0.57±0.09 0.53±0.08 -0.06±0.05
111
8.9±1.9 11
0.10
5/13/2010
36 0.58±0.09 0.59±0.09
0.01±0.03
141 10.6±2.0
Total
215 0.57±0.08 0.62±0.07
0.09±0.05
237 19.7±3.3 ---
---
Mean
35.8 0.57±0.08 0.58±0.08
0.02±0.05
124
0.10
63
AR
NPA PPA
9
9.4±1.8 12
0.06
B)
Cohort
N
HO±S.E.
HE±S.E.
FIS±S.E.
1980
9
0.53±0.10
0.63±0.08
0.23±0.09
1981
13
0.58±0.07
0.66±0.07
1982
12
0.51±0.11
1995
9
1996
AR±S.E.
NPA
PPA
79
5.3±0.9
8
0.10
0.12±0.06
90
5.1±0.7
6
0.07
0.65±0.07
0.27±0.10
91
5.2±0.8
7
0.08
0.56±0.07
0.60±0.05
0.07±0.08
55
3.9±0.4
1
0.02
6
0.51±0.08
0.65±0.07
0.21±0.11
49
4.1±0.5
2
0.04
1998
14
0.60±0.09
0.56±0.07
-0.06±0.07
75
4.1±0.6
6
0.08
1999
7
0.57±0.07
0.63±0.05
0.12±0.11
53
4.2±0.5
1
0.02
2000
10
0.58±0.10
0.55±0.08
-0.05±0.08
65
4.2±0.6
2
0.03
2001
16
0.61±0.07
0.58±0.06
-0.07±0.07
98
4.2±0.5
5
0.05
2002
17
0.57±0.08
0.54±0.07
-0.05±0.07
73
3.9±0.6
4
0.05
2003
10
0.58±0.11
0.55±0.09
0.01±0.08
68
4.4±0.8
1
0.01
2005
17
0.55±0.07
0.60±0.08
0.07±0.05
99
4.8±0.7
6
0.06
2006
6
0.57±0.11
0.61±0.10
0.08±0.07
60
5.0±0.9
2
0.03
2007
14
0.59±0.09
0.61±0.09
0.01±0.04
91
5.0±0.8
3
0.03
2008
13
0.57±0.10
0.55±0.09
-0.03±0.02
84
4.6±0.8
0
0
Total
183
0.57±0.07
0.63±0.07
0.10±0.05
226
18.8±3.3
---
---
Mean
12
0.57±0.09
0.60±0.10
0.06±0.07
75
4.5±0.7
4
0.05
64
NA
C)
Cohort
N
1997
HO±S.E.
HE±S.E.
FIS±S.E.
8
0.57±0.07 0.63±0.08
0.08±0.04
84 4.3±0.67
7
0.08
1998
18
0.52±0.08 0.61±0.08
0.15±0.06
130 4.5±0.62
21
0.16
1999
5
0.52±0.09 0.55±0.09
0.02±0.06
60 4.0±0.62
3
0.05
2001
11
0.70±0.09 0.59±0.07
-0.17±0.05
100 4.3±0.53
12
0.12
2002
7
0.67±0.09 0.56±0.07
-0.16±0.05
67 3.8±0.44
5
0.07
2003
12
0.61±0.09 0.56±0.08
-0.07±0.05
88 3.9±0.51
8
0.09
2004
7
0.61±0.07 0.61±0.07
0.00±0.03
68 3.8±0.51
4
0.06
Total
68
0.60±0.08 0.60±0.07
0.02±0.02
194 12.9±2.2
---
---
Mean
9.7
0.60±0.08 0.59±0.05
-0.01±0.02
85 4.1±0.56
8.6
0.09
65
NA
AR±S.E.
NPA PPA
Table 3
Pairwise genetic divergences of spawning yellow perch between year groups at a given site (2001-5 vs. 2009 in bold, along diagonal)
and among sampling locations (2001-5 above diagonal; 2009 below diagonal), including θST/ (exact G). * = significant following
sequential Bonferroni correction (Rice, 1989); all exact G comparisons were significantly different.
Location
A
H) Anchor Bay MI
B
C
D
E
F
0.138*/(Inf.) 0.030*/(Inf.)
0.044*/(Inf.)
0.065*/(Inf.)
0.052*/(Inf.)
0.044*/(Inf.)
I) Monroe MI
0.133*/(Inf.)
0.026*/(Inf.)
0.042*/(Inf.)
0.069*/(Inf.)
0.048*/(Inf.)
0.055*/(Inf.)
J) Erieau ON
0.159*/(Inf.)
0.015*/(73.9) 0.037*/(Inf.)
0.056*/(Inf.)
0.047*/(Inf.)
0.037*/(Inf.)
K) Fairport OH
0.156*/(Inf.)
0.072*/(Inf.)
0.114*/(Inf.) 0.002/(50.8)
0.030*/(Inf.)
L) Perry OH
0.151*/(Inf.)
0.007*/(71.4) 0.012*/(54.5) 0.073*/(Inf.)
0.041*/(Inf.) 0.028*/(Inf.)
M) Erie PA
0.083*/(Inf.)
0.110*/(Inf.)
0.121*/(Inf.)
0.145*/(Inf.)
0.114*/(Inf.)
0.135*/(Inf.)
N) Dunkirk NY
0.168*/(Inf.)
0.029*/(Inf.)
0.036*/(Inf.)
0.085*/(Inf.)
0.028*/(Inf.)
0.124*/(Inf.)
O) Rochester NY
0.122*/(Inf.)
0.074*/(Inf.)
0.101*/(Inf.)
0.139*/(Inf.)
0.071*/(Inf.)
0.149*/(Inf.)
Mean (2001-2005)
0.139±0.011
0.063±0.018
0.074±0.021
0.106±0.015
0.065±0.020
0.121±0.008
0.072*/(Inf.)
66
G
H
Mean (2009)
0.040*/(Inf.)
0.033*/(Inf.)
0.044±0.005
0.061*/(Inf.)
0.051*/(Inf.)
0.051±0.005
0.033*/(94.3) 0.045*/(Inf.)
0.043±0.003
0.038*/(Inf.)
0.056*/(Inf.)
0.045±0.009
0.037*/(Inf.)
0.049*/(Inf.)
0.038±0.007
0.014*/(97.1) 0.034*/(Inf.)
0.034±0.005
0.064*/(Inf.)
0.034*/(Inf.)
0.037±0.005
0.075*/(Inf.)
0.055*/(Inf.) 0.043±0.004
0.078±0.020
0.104±0.012
67
Table 4
Pairwise genetic divergences for (A) spawning yellow perch from Dunkirk NY, eastern Lake Erie sampled in six different collection
years (1985, 2001, 2004, 2008, 2009, 2010), (B) cohorts at Dunkirk NY (1980-2008), and (C) age cohorts at Monroe MI (1997-2004).
Inf. = “infinite”. * = remained significant following sequential Bonferroni correction (Rice 1989).
A)
Year (N)
1985
2001
2004
2008
2009
2010
1985 (34)
-----
Inf*
Inf.*
Inf.*
Inf.*
Inf.*
2001 (37)
0.037*
-----
Inf.*
Inf.*
Inf.*
Inf.*
2004 (48)
0.056*
0.055*
-----
Inf.*
Inf.*
Inf.*
2008 (30)
0.125*
0.138*
0.141*
-----
Inf.*
Inf.*
2009 (30)
0.041*
0.072*
0.088*
0.105*
-----
52.4*
2010 (36)
0.011*
0.041*
0.057*
0.113*
0.014*
-----
0.079±0.017
0.124±0.007
0.064±0.016
Mean±S.E. 0.054±0.019 0.069±0.018
68
0.047±0.018
B)
Cohort (N)
1980
1981
1982
1995
1996
1998
1980 (09)
-----
33.4
35.2
69.4*
55.4*
66.1*
48.3
60.8*
60.9*
74.3
1981 (13)
0.002
-----
14.9
73.1*
67.6*
89.8*
53.4*
69.7*
96.4*
105.2
1982 (12)
0.003
0.000
-----
64.9*
71.5*
95.2*
50.1*
64.2*
99.8*
Inf.
1995 (09)
0.052
0.048
0.059
-----
68.0*
41.9
37.3
78.7*
77.0*
77.0
1996 (06)
0.070
0.068*
0.073
0.077
-----
65.1*
34.0
104.9*
111.9*
Inf.
1998 (14)
0.041
0.042*
0.062*
0.030
0.066
-----
37.1
Inf.*
Inf.*
Inf.
1999 (07)
0.031
0.024
0.036
0.009
0.034
0.011
-----
73.4*
59.9*
67.3
2000 (10)
0.057*
0.057*
0.062*
0.077*
0.142*
0.088*
0.067*
-----
34.6
49.9
2001 (16)
0.044*
0.056*
0.065*
0.047*
0.117*
0.074*
0.042
0.006
-----
2002 (17)
0.040*
0.052*
0.048*
0.053*
0.093*
0.064*
0.046
0.033
0.014
-----
2003 (10)
0.117*
0.102*
0.120*
0.135*
0.172*
0.149*
0.079
0.115*
0.104*
0.125*
2005 (17)
0.092*
0.071*
0.101*
0.105*
0.150*
0.114*
0.062
0.082*
0.086*
0.106*
2006 (06)
0.000
0.018
0.008
0.071
0.098
0.071*
0.031
0.061
0.054*
0.054
2007 (14)
0.000
0.009
0.014
0.045*
0.094*
0.039*
0.031
0.057*
0.055*
0.043*
2008 (13)
0.011
0.016
0.027
0.063*
0.117*
0.054*
0.037
0.060*
0.067*
0.055*
Mean±S.E.
0.040±0.010
0.040±0.008
0.048±0.010
0.062±0.008
0.098±0.010
0.065±0.009
0.039±0.005
0.069±0.009
0.059±0.008
0.059±0.008
69
1999
2000
2001
2002
30.1
2003
2005
2006
2007
2008
Inf.*
Inf.*
24.0
35.7
29.3
Inf.*
Inf.*
46.2
49.1*
46.4
Inf.*
Inf.*
29.7
45.9
45.3
Inf.*
Inf.*
69.2*
65.5*
61.1*
102.0*
120.7*
65.2*
83.3*
89.1*
Inf.*
Inf.*
74.5*
69.4*
78.4*
70.0
70.2
39.0
50.8*
42.9
Inf.*
Inf.*
55.4*
Inf.*
59.4*
Inf.*
Inf.*
63.0*
96.4*
78.4*
Inf.*
Inf.*
69.9*
81.1*
79.0*
31.6
41.9
Inf.*
Inf.*
48.0
Inf.*
Inf.*
41.0
34.1
----0.009
-----
0.070
0.058
-----
0.100*
0.071*
0.009
-----
0.111*
0.073*
0.025
0.000
-----
0.108±0.010
0.084±0.009
0.045±0.008
0.041±0.009
0.051±0.009
17.6
70
C)
Cohort (N)
1997
1998
1999
2001
2002
2003
2004
1997 (8)
---
19.6
20.0
22.7
30.5
50.8*
30.3
1998 (18)
0.000
---
31.3
42.6
36.5
79.4*
53.4*
1999 (5)
0.000
0.019
---
22.4
30.5
40.2
35.5
2001 (11)
0.000
0.021*
0.006
---
29.1
52.0*
37.9
2002 (7)
0.001
0.014
0.004
0.010
---
28.4
20.5
2003 (12)
0.010
0.030*
0.015
0.024*
0.000
---
26.7
2004 (7)
0.000
0.022*
0.018
0.019*
0.000
0.000
---
0.010±0.003
0.013±0.004 0.004±0.002 0.013±0.005 0.010±0.004
Mean±S.E. 0.002±0.002 0.018±0.004
71
Table 5
Hierarchical Analysis of MOlecular VAriance (AMOVA; Excoffier et al. 1992) showing
the distribution of genetic variation under four possible scenarios: (a) among the
spawning groups and between the two sampling periods, (b) between two sampling years
and eight spawning groups, (c) between the sexes and spawning groups within them, (d)
between spawning groups at Monroe MI and Dunkirk NY and among their age cohorts
(‘98, ‘99, ‘01, ‘02, and ’03). *= Significant.
Source of Variation
(a)
(b)
(c)
(d)
(%)
Variation
<0.01
Among the eight spawning groups
Fixation
Index
<0.001
p value
0.598
Between the two years (2001-2005 and 2009)
7.76
0.077
<0.001*
Between the two periods (2001-2005 and 2009)
1.71
0.017
<0.001*
Among the eight spawning groups
6.23
0.063
<0.001*
<0.01
<0.001
0.952
Among sampling sites (Anchor Bay MI, Erieau
ON, Erie PA, and Dunkirk NY) for males and
females
8.57
0.076
<0.001*
Between spawning groups (Monroe MI and
Dunkirk NY)
Among age cohorts (‘98, ‘99, ‘01, ‘02, ‘03) within
groups
1.78
0.018
0.001*
4.56
0.046
<0.001*
Between sexes (male vs. female)
72
Fig. 1. Map showing sampling sites of spawning yellow perch. Lines show genetic
barriers among spawning groups from BARRIER v 2.2 (Manni et al. 2004b), with Roman
numerals indicating the relative strength of the genetic differences, with I being the
strongest. Solid lines = genetic barriers for 2001-5. Dashed lines = genetic barriers in
2009. Significance of barriers is given as percent bootstrap support and the number of
supporting loci. 2001-5: barrier I (10 loci, 89%), barrier II (10, 67%), barrier III (11,
53%), barrier IV (12, 66%); 2009 barrier I (12, 65%), barrier II (14, 53%), barrier III (12,
59%), barrier IV (12, 54%).
Fig. 2. Results of GENECLASS v.2.2 (Piry et al. 2004) assignment tests of individual
fish to spawning groups.
Fig. 3. Population relationships from Bayesian STRUCTURE analysis (Pritchard et al.
2000, Pritchard and Wen, 2004), showing K=8 population groups, with posterior
probability (pp)=1 from ΔK likelihood evaluations (Evanno et al. 2005). Black lines
separate spawning samples; individuals are represented by thin vertical lines, with colors
showing their estimated membership to a population group. .
Fig. 4. Mantel (1967) regression of the pairwise relationship between genetic distance
(θST/1-θST) and geographic distance (km) for spawning yellow perch sampled in A) 20015 (y=0.00007x+0.09, R2=0.036, p=N.S.) and B) 2009 (y=0.00003x+0.04, R2=0.078,
p=N.S.).
Fig. 5. Mantel (1967) regression of the pairwise relationship between genetic distance
(θST/1-θST) versus time in years separating age cohorts (1980-2008) of yellow perch from
73
Dunkirk NY (y=-0.0009x+0.0782, R2=0.026, p=0.67, N.S.). Open diamonds =
comparisons involving individuals from the 2003 cohort.
74
Fig. 1
75
Fig. 2
76
Fig. 3
77
Fig. 4
78
Fig 5.
79
Chapter 4
4.1 CONCLUSIONS
This thesis investigates the fine-scale spatial and temporal genetic structure of
yellow perch spawning groups. The first experimental part, Chapter 2, analyzes the
spatial patterns of diversity and divergence of yellow perch spawning groups in the
Huron-Erie Corridor in relation to habitat restoration and life-history patterns. The second
experimental section, Chapter 3, focuses on evaluating temporal and spatial variation in
genetic diversity, divergence, and effective population size of yellow perch spawning
groups in Lake St. Clair, Lake Erie, and Lake Ontario. Overall, the results discern high
spatial structure among spawning groups and the presence of temporal variability, which
likely result from life-history patterns, such as site fidelity, potential kin recognition, type
III survivorship, and differential reproductive success. Findings of this thesis highlight
the importance of understanding population genetic structure as it relates to stock
identification, as well as the need for continued genetic monitoring of temporal and
spatial trends. These data can provide valuable information to aid long term adaptive
management strategies, which match a biologically relevant scale and aim to preserve
genetic diversity, unique variability, and adaptive potential of fishery stocks (Rice et al.
2012, Pracheil et al. 2012).
80
This thesis discerns levels of genetic diversity for yellow perch that are similar to
those recovered for other yellow perch populations, using a variety of genetic methods
(Miller 2003, LeClerc et al. 2008, Sepulveda-Villet et al. 2009, Gryzbowski et al., 2010,
Sepulveda-Villet and Stepien 2011, 2012). These levels of genetic variation are roughly
consistent with those of other freshwater fishes, based on microsatellite data summarized
by DeWoody and Avise (2002). This thesis discerns that genetic compositions
significantly differ among population groups spawning across the HEC and in Lakes St.
Clair, Erie, and Ontario. Their levels of observed heterozygosity and allelic richness are
relatively consistent over time at given locations.
Spawning groups of yellow perch in the Huron-Erie Corridor and Lakes St. Clair,
Erie, and Ontario are defined by patterns of high divergence from one another, which are
unrelated to genetic isolation by geographic distance. It appears likely that these patterns
are related to life history characters, such as limited movement (Kelso 1973, Rawson
1980), natal site fidelity (Aalto and Newsome 1990), and kin group recognition, as have
been described for the related European perch P. fluviatilis (Gerlach et al. 2001,
Behrmann-Godel et al. 2004, 2006).
Although levels of genetic diversity and overall patterns of genetic divergence
among spawning groups remain similar over time, some temporal variations
characterized different sampling years. This temporal variability in genetic composition
at given spawning sites occurred among collection years and among different age
cohorts. The variation among age cohorts appears to be unrelated to isolation by time and
81
may be governed by life history patterns, such as kin grouping, type III survivorship, and
differential reproductive success.
4.2 FUTURE RESEARCH
This thesis identifies two areas for future research to develop further
understanding of the interplay between life history, environmental factors, population
genetic variability, and yellow perch fisheries conservation and management. These
factors are: natal homing and kin group recognition and functional variability and
signatures of selection.
Although the pelagic larval stage of yellow perch has been studied in depth
(Dettmers et al. 2005, Fulford et al. 2006a), environmental factors, such as currents, may
act on the larvae during the drift stage to intermix those originating from different
spawning stocks (Caley et al. 1996, Mora and Sale 2002). This thesis and other works on
the population structure of yellow perch have identified patterns of high spatial structure
among spawning adult stocks (e.g., Miller 2003, LeClerc et al. 2008, Gryzbowski et al.,
2010, Sepulveda-Villet and Stepien 2011 and 2012). These patterns likely result from two
major behavioral mechanisms: (1) natal site fidelity by returns of spawning adults to
given locations and/or (2) kin-recognition and group aggregation.
Natal site fidelity of yellow perch was explored by Aalto and Newsome (1990),
who examined the impact of egg mass removal from identified spawning sites in terms of
the likelihood that females would return to that site over time. They found a significant
82
reduction in the number of individuals spawning at that area over time in comparison
with control sites. Advances in individual fish tagging tracking systems may allow for
more accurate monitoring of individuals over a longer time period (Huuskonen et al.
2012, Chase et al. 2013). Comparisons of adult movements over multiple spawning
periods may provide information about: (1) Do the same individuals return to the same
spawning location every year? (2) Do yellow perch spawn on an annual basis? (3) Do
individuals spawn at multiple geographic locations in a given year? and (4) Are patterns
congruent between males and females? Answers will further understanding of the spatial
and temporal population structure observed in this thesis.
The European perch has been shown to readily identify and selectively associate
in groups of related individuals (Gerlach et al. 2001, Behrmann-Godel et al. 2004 and
2006), and mating with their kin may confer a fitness advantage through decreased egg
mortality and increased production of larvae. If yellow perch behave in a similar manner,
this may account for the high spatial structure found among their spawning groups.
Individuals may preferentially associate with their kin, which would limit gene flow
among non-related groups. Understanding whether yellow perch can identify kin in situ
and sampling of individuals in close proximity at various life history stages, followed by
genetic analysis of their relative levels of relatedness, may provide insight into the
maintenance and dynamics of kin group aggregation as it relates to spawning stock
structure.
The third chapter of this thesis identifies temporal variability among age cohorts
of the spawning group at Dunkirk NY. These variations in genetic composition are not
83
due to a genetic isolation by time pattern, and likely are related to sampling fluctuations
or perhaps to natural selection. In the Great Lakes, increased algal blooms and harmful
toxins, increased size and prevalence of hypoxic zones, loss of spawning and nursery
habitat, competition with invasive species, size-selective fishing pressure, and climate
change may impact the relative reproduction and survival of yellow perch stocks in the
process of natural selection. However, this study and the previous genetic investigations
of the yellow perch have been based on neutral genetic markers that confer no adaptive
advantage. In order to understand the interplay between spawning stocks and their
habitats, it would be useful to analyze the relationship between genetic variability and
individual gene expression using functional genetic markers that show a signature of
selection. Understanding such functional patterns may facilitate an understanding of the
responses of populations to ongoing and future stressors, and may identify which groups
of individuals are best suited to given habitats for use in genetic rehabilitation projects if
necessary. These markers might be useful to evaluate differential selection during larval
drift, juvenile, and adult life history stages, providing essential information for
understanding highly variable recruitment.
Lastly, it is important to realize that the ability to analyze such complex
mechanisms as selection, recruitment, and genetic composition are only possible with
long term data sets, as that used here for yellow perch spawning at Dunkirk NY. Without
the collection of long-term samples, many questions cannot be addressed. Continued
compilation of long-term data and the use of functional genetic markers will bring new
insight to the multifaceted questions surrounding genetic composition, selection,
recruitment, and population stability of yellow perch and other taxa.
84
References
Aalto SK, Newsome GE (1990) Additional evidence supporting demic behavior of a
yellow perch (Perca flavescens) population. Can J Fish Aquat Sci 47: 1959-1962
Allen MR, Thum RA, Caceres CE (2010) Does local adaptation to resources explain
genetic variation among Daphnia populations? Mol Ecol 19: 3076-3087
Allendorf FW, Hohlenlohe PA, Luikart G (2010) Genomics and the future of
conservation genetics. Nat Rev Genet 11: 697-709
Allendorf FW, England PR, Luikart G, Ritchie PA, Ryman R (2008) Genetic effects of
harvest on wild animal populations. Trends Ecol Evol 23: 327-337
Allendorf FW, Luikart G (2007) Conservation and the Genetics of Populations.
Blackwell Puplishing. Malden MA. USA
Anderson EC (2005) An efficient Monte Carlo method for estimating NE from temporally
spaced samples using a coalescent-based likelihood. Genetics 170: 955-967
Angeletti D, Cimmaruta R, Nascetti G (2010) Genetic diversity of the killifish Aphanius
fasciatus paralleling the environmental changes of Tarquina salterns habitat.
Genetica 138: 1011-1021
Balloux F, Lugon-Moulin N, (2002) The estimation of population differentiation with
microsatellite Markers. Mol Ecol 11: 155-165
85
Behrmann-Godel J, Gerlach G, Eckmann R (2006) Kin and population recognition in
sympatric Lake Constance perch (Perca fluviatilis L.): can assortative shoaling
drive population divergence? Behav Ecol Sociobiol 59: 461-468
Behrmann-Godel J, Gerlach G (2008) First evidence for postzygotic reproductive
isolation between two populations of Eurasian perch (Perca fluviatilis L.) within
Lake Constance. Front Zool 5: 1-7
Bennion D, Manny B (2011) Construction of shipping channels in the Detroit River:
history and environmental consequences. U.S. Geological Survey Scientific
Investigations Report 2011-5122, Reston, Virginia. 14p.
Berkeley SA, Hixon MA, Larson, RJ, Love, MS (2004) Fisheries sustainability via
protection of age structure and spatial distribution of fish populations. Fisheries
29:23-32
Bessert M, Orti G (2008) Genetic effects of habitat fragmentation on blue sucker
populations in the upper Missouri River (Cycleptus elongates Lesueur 1918).
Conserv Genet 9: 821-832
Billington N (1993) Genetic variation in Lake Erie yellow perch (Perca flavescens)
demonstrated by mitochondrial DNA analysis. J Fish Biol 43: 941-943
Billington N (1996) Geographical distribution of mitochondrial DNA (mtDNA) variation
in walleye, sauger, and yellow perch. Annales Zoologici Fennici 33:699-706
Bini LM, Velho LFM, Lansac-Toha FA (2003) The effect of connectivity on the
relationship between local and regional species richness of testate amoebae
(Protozoa, Ryzopoda) in floodplain lagoons of the upper Parana River, Brazil.
Acta Oecol 24: S145-S151
86
Borer S, Miller L, Kapuscinski A (1999) Microsatellites in walleye Stizostedion vitreum.
Mol Ecol 8: 336-338
Brown TG, Runciman B, Bradford MJ, Pollard S (2009) A Biological Synopsis of
Yellow Perch (Perca flavescens). Canadian manuscript report of fisheries and
aquatic sciences 2883. Fisheries and Oceans Canada
Caley MJ, Carr MH, Hixon MA, Hughes TP, Jones GP, Menge BA, Recruitment and the
local dynamics of open marine populations. Annu Rev Ecol Syst 27:477-500
Chase R, Hemphill N, Beeman J, Juhnke S, Hannon J, Jenkins AM (2013) Assessment of
juvenile coho salmon movement and behavior in relation to rehabilitation efforts
in the Trinity River, California, using PIT tags and radiotelemetry. Environ Biol
Fish 96: 303-314
Clapp DF, Dettmers JM (2004) Yellow perch research and management in Lake
Michigan. Fisheries 29: 11-19
Constable AJ, Nicol S (2002) Defining smaller-scale management units to further
develop the ecosystem approach managing large-scale pelagic krill fisheries in
Antarctica. CCAMLR Sci. 9: 117-131
Cornuet JM, Luikart G (1996) Description and power analysis of two tests for detecting
recent population bottlenecks from allele frequency data. Am J Bot 144: 20012014
Cornuet JM, Piry S, Luikart G, Estoup A, Solignac, M (1999) New methods employing
multilocus genotypes to select or exclude populations as origins of individuals.
Genetics 158: 1898-2000
87
Craig JF (2000) Percid Fishes: Systematics, Ecology, and Exploitation. Blackwell
Science, Ltd. Oxford, England
Craig J (1987) The Biology of Perch and Related Fishes. Portland, OR: Timber Press.
333p
Demandt M (2010) Temporal changes in genetic diversity of isolated populations of
perch and roach. Conserv Genet 11: 249-255
Dettmers JM, Janssen J, Pientka B, Fulford RS, Jude DJ (2005) Evidence across multiple
scales for offshore transport of yellow perch (Perca flavescens) larvae in Lake
Michigan. Can J Fish Aquat Sci 62: 2683-2693
Devine JA, Wright PJ, Pardoe HE, Heino M (2012) Comparing rates of contemporary
evolution in life-history traits for exploited fish stocks. Can J Fish Aquat Sci 69:
1105-1120
DeWoody JA, Avise JC (2000) Microsatellite variation in marine, freshwater and
anadromous fishes compared with other animals. J Fish Biol 56: 461-473
Di Pippo T, Donadio C, Guida M, Petrosino C (2006) The case of Sarno river (southern
Italy): effects of geomorphology on the environmental impacts. Environ Sci
Pollut Res Int 13:184-191
Eldridge W, Bacigalupi M, Adelman I, Miller L, Kapuscinski A (2002) Determination of
relative survival of two stocked walleye populations and resident natural-origin
fish by microsatellite DNA parentage assignment. Can J Fish Aquat Sci 59: 282290
El Mousadik A, Petit RJ (1996) Chloroplast DNA phylogeography of the aragan tree of
Morocco. Mol Ecol 5: 547-555.
88
Evanno G, Regnaut S, Goudet J (2005) Detecting the number of clusters of individuals
using the software STRUCTURE: a simulation study. Mol Ecol 14 : 2611-2620
Excoffier L, Smouse P, Quattro J (1992) Analysis of molecular variance inferred from
metric distances among DNA haplotypes: application to human mitochondrial
DNA restriction data. Genetics 131: 479-491
Excoffier L, Lischer HE L (2010) ARLEQUIN suite ver 3.5: A new series of programs to
perform population genetics analyses under Linux and Windows. Mol. Ecol. Res.
10, 564-567. Available from http://cmpg.unibe.ch/software/arlequin35/ [Accessed
5 March 2013]
Ford A, Stepien CA (2004) Genetic variation and spawning population structure in Lake
Erie yellow perch Perca flavescens: A comparison with a maine population.
Proceedings of Percis III 131-132
Franckowiak RP, Sloss BL, Bozen MA, Newman SP (2009) Temporal effective size
estimates of a managed walleye Sander vitreus population and implications for
genetic-based management. J Fish Biol 74: 1086-1103
Fulford RS, Rice JA, Binkowski FP (2006a) Examination of sampling bias for larval
yellow perch in Southern Lake Michigan. J Great Lakes Res 32: 434-441
Gerlach G, Schardt U, Eckmann R, Meyer A (2001) Kin-structured subpopulations in
Eurasian perch (Perca fluviatilis L.). Heredity 86: 213-221
Glaubitz JC (2004) CONVERT: a user friendly program to reformat diploid genotypic
data for commonly used population genetic software packages. Mol Ecol Notes 4:
309-310 Available from:http://www.agriculture.purdue.edu/fnr/html/faculty
89
/rhodes/ students %20and %20staff/glaubitz/software.htm. [Accessed 19 July
2012]
Goodyear C, Edsall TA, Dempsey DM, Moss GD, Polanski PE (1982) Atlas of Spawning
and Nursery Areas of Great Lakes Fishes. Vol. 9: Lake Erie. U.S. Fish and
Wildlife Service, Washington, D.C.
Goudet J, Raymond M, de Meeus T, Rousset F, (1996) Testing differentiation in diploid
populations. Genetics 144: 1933-1940
Goudet J (2002) FSTAT, a program to estimate and test gene diversities and fixation
indices, Ver. 2.9.3.2. Available from
http://www2.unil.ch/popgen/softwares/fstat.htm [Accessed 5 March 2013]
Grzybowski M, Sepulveda-Villet OJ, Stepien CA, Rosauer D, Binkowski F, Klaper R,
Shepherd BS, Goetz F (2010) Genetic variation of 17 wild yellow perch
populations from the Midwest and east coast analyzed via microsatellites. Trans
Am Fish Soc 139: 270-287
Guillot G, Estoup A, Mortier F, Cosson JF (2005a) A spatial statistical model for
landscape genetics. Genetics 170: 1261-1280
Guillot G, Mortier F, Estoup A (2005b) GENELAND: A computer package for landscape
genetics. Mol Ecol Notes 5: 708-711
Guillot G, Santos F, Estoup A (2008) Analyzing georeferenced population genetics data
with GENELAND: a new algorithm to deal with null alleles and a friendly
graphical user interface. Bioinformatics 24: 1406-1407. Available from
http://www2.imm.dtu.dk/ ~gigu/Geneland/ [Accessed 5 March 2013]
Guo SW, Thompson EA (1992) Performing the exact test of Hardy-Weinberg proportions
90
for multiple alleles. Biometrics 48: 361-372
Guzzo MW, Haffner GD, Legler ND, Rush SA, Fisk AT (2013) Fifty years later: trophic
ecology and niche overlap of a native and non-indigenous fish species in the
western basin of Lake Erie. Biol Invasions
Haas RC, Bryant WC, Smith KD, Nuhfer AJ (1985) Movement and harvest of fish in
Lake St. Clair, St. Clair River, and Detroit River. Final Report Winter Navigation
Study U.S. Army Corps of Engineers Detroit, Michigan
Halbert CL (1993) How adaptive is adaptive management? Implementing adaptive
management in Washington State and British Columbia. Rev Fish Sci 1: 261-283
Hallerman E, Brown B, Epifano J (2003) An overview of classical and molecular
genetics, In: Hallerman, E.M. (Ed.), Population Genetics: Principles and
Applications for Fisheries Scientists. American Fisheries Society, Bethesda Md.
pp. 3-20
Haponski A, Stepien CA (2008) Molecular, morphological, and biogeographic resolution
of cryptic taxa in the Greenside Darter Etheostoma bellioides complex. Mol
Phylogenet Evol 49: 69-83
Haponski A, Bollin T, Jeclicka M, Stepien CA (2009) Landscape genetic patterns of the
rainbow darter Etheostoma caeruleum: a catchment analysis of mitochondrial
DNA sequences and nuclear microsatellites. J Fish Biol 75: 2244-2268
Haponski AE, Stepien CA (2013). Genetic connectivity and diversity of Huron-Erie
Corridor walleye (Sander vitreus) spawning in the Huron-Erie Corridor: Early
findings after habitat augmentation. J Great Lakes Res
Hauser L, Adcock GJ, Smith PJ, Bernal Ramirez JH, Carvalho GR (2002) Loss of
91
microsatellite diversity and low effective population size in an overexploited
population of New Zealand snapper (Pagrus auratus) PNAS 99: 11742-11747
Heath DD, Busch C, Kelly J, Atagi DY (2002) Temporal change in genetic structure and
effective population size in steelhead trout (Oncorhynchus mykiss). Mol Ecol
11:197-214
Hedgecock D (1994) Does variance in reproductive success limit effective population
sizes of marine organisms. In: Beaufort, A.R. (Ed.), Genetics and evolution of
aquatic organisms. Chapman and Hall, London, UK. pp. 122-134
Hedgecock D, Pudovkin AI (2011) Sweepstake reproductive success in highly fecund
marine fish and shellfish: a review and commentary. Bull Mar Sci 87: 971-1002
Helfman G (1979) Twilight activities of yellow perch (Perca flavescens). J Fish Res
Board Can 36:173-179
Henderson BA, Nepszy SJ (1989) Yellow perch (Perca flavescens) growth and mortality
rates in Lake St. Clair and the three basins of Lake Erie, 1963-1986. J Great Lakes
Res 15: 317-326
Hergenrader G, Hasler A (1968) Influence of changing seasons on schooling behaviour
of yellow perch. J Fish Res Board Can 25:711-716.
Holbombe TL, Taylor LA, Reid DF, Warren JS, Vincent PA, Herdendorf CE (2003)
Revised Lake Erie postglacial lake level history based on new detailed
bathymetry. J Great Lakes Res 29: 681-704
HTG (2009) Habitat Task Group Summary. Great Lakes Fishery Commission, Windsor,
Ontario.http://www.glfc.org/lakecom/lec/HTG_docs/annual_reports/HTG%20Re
port%202009.pdf [Accessed 19 July 2012]
92
HTG (2010) Habitat Task Group Summary. Great Lakes Fishery Commission, Windsor,
Ontario.http://www.glfc.org/lakecom/lec/HTG_docs/annual_reports/HTG_Annua
lReport2010.pdf [Accessed 19 July 2012]
Hutchinson WF, Carvalho CG Rogers SI (1999) A nondestructive technique for the
recovery of DNA from dried fish otiliths for subsequent molecular analysis. Mol
Ecol 8: 893-894
Hutchinson WF, van Oosterhout C, Rogers SI, Carvalho GR (2003) Temporal analysis of
archived sampled indicating marked genetic changes in declining North Sea cod
(Gadus morhua). P Roy Soc B-Biol Sci 270: 2125-2132
Huuskonen H, Haakana H, Leskela A, Piironen J (2012) Seasonal movements and habitat
use of river whitefish (Coregonus lavaretus) in the Koitajoki River (Finland), as
determined by Carlin tagging and acounstic telemetry. Aquat Ecol 46:325-334
Isaak DL, Thurow RF, Reiman BE, Dunham JB, (2007) Chinook salmon use of spawning
patches: relative roles of habitat quality, size, and connectivity. Ecol Appl 17:
352-364
Jackson JBC, Kirby MX, Berger WH, Bjorndal KA, Botsford LW, Bourque BJ, Bradbury
RH, Cooke R, Erlandson J, Estes JA, Hughes TP, Kidwell S, Lange CB, Lenihan
HS, Pandolfi JM, Peterson CH, Steneck RS, Tegner MJ, Warner RR (2001)
Historical overfishing and the recent collapse of coastal ecosystems. Science 293:
629-637
Jansen AC, Grab BDS, Willis DW (2009) Effect of a simulated cold front on hatching
success of yellow perch eggs. J Freshwat Ecol 24: 651-655
93
Keller I, Taverna A, Seehausen O (2011) Evidence of natural and adaptive genetic
divergence between European trout populations sampled along altitudinal
gradients. Mol Ecol 20: 1888-1904
Kelsoe JRM (1973) Movement of yellow perch (Perca flavescens), smallmouth bass
(Micropterus dolomieu), and white bass (Morone chrysops) released in Long
Point Bay, Lake Erie, during 1971 and 1972. Fisheries Research Board of Canada.
Technical Report #386, 18p
Kitchell JF, Johnson MG, Minns CK, Loftus KH, Greig L, Oliver CM (1977) Percid
habitat: The river analogy. J Fish Res Board Can 34: 1936-1940
Kocovsky PM, Knight CT (2012) Morphological evidence for discrete stocks of yellow
perch in Lake Erie. J Great Lakes Res 38: 534-539
Kocovsky PM, Sullivan TJ, Knight CT, Stepien CA (2013 in review) Genetic and
morphometric differences demonstrate population substructure of yellow perch
Perca flavescens in central Lake Erie. J Fish Biol
Kopp D, Figuerola J, Compin A, Santoul F, Cereghino R (2012) Local extinction and
colonization in native and exotic fish in relation to changes in land use. Mar
Freshwater Sci 63:175-179
Kreiger DA, Terrell JW, Nelson PC (1983) Habitat suitability information: Yellow perch.
U.S. Fish and Wildlife Service FWS/OBS-83/10.55. Washington. DC.
Lage C, Kornfield I (2006) Reduced genetic diversity and effective population size in an
endangered Atlantic salmon (Salmo salar) population from Maine, USA. Cons
Genet 7: 91-104
94
Lande R (1998) Anthropogenic, ecological and genetic factors in extinction and
conservation. Res Popul Ecol 40: 259-269
Laroche J, Durand J (2004) Genetic structure of fragmented populations of a threatened
endemic percid of the Rhone River: Zingel asper. Heredity 92: 329-334
Latch EK, Dharmarajan G, Glaubitz JC, Rhodes Jr OE (2006) Relative performance of
Bayesian clustering software for inferring population substructure and individual
assignment at low levels of population differentiation. Cons Genet 7: 295-302
Leach JH, Nepszy SJ (1976) The fish community in Lake Erie. J Fish Res Board Can 33:
622-638
Leach JH (1991) Biota of Lake St. Clair: habitat evaluation and environmental
assessment. Hydrobiologia 219: 187-202
LeClerc E, Mailhot Y, Mingelbier M, Bernatchez L (2008) The landscape genetics of
yellow perch (Perca flavescens) in a large fluvial ecosystem. Mol Ecol 27: 27021717
Lefant P, Planes S (2002) Temporal genetic changes between cohorts in a natural
population of marine fish, Diplodus sargus. Biol J Linn Soc 76: 9-20
Leslie JK, Timmons CA (1991) Distribution and abundance of young fish in the St. Clair
River and associated waters, Ontario. Hydrobiologia 219: 125-134
Lewis CFM, Moore TC, Rea DK, Dettman DL, Smith AM, Mayer LA (1994) Lakes of
the Huron basin: Their record of runoff from the Laurentide Ice Sheet. Quaternary
Sci Rev 13: 891-922
95
Lewis CFM, Karrow PF, Blasco SM, McCarthy FMG, King JW, Moore TC, Rea DK
(2008) Evolution of lakes in the Huron basin: Deglaciation to present. Aquat
Ecosyst Health 11: 127-136
Li L, Wang HP, Givens CB, Czesny S, Brown BL (2007) Isolation and characterization
of microsatellites in the yellow perch (Perca flavescens). Mol Ecol Notes 7: 600603
Liermann CR, Nilsson C, Robertson J, Ng RY (2012) Implications of dam obstruction for
global freshwater diversity. BioScience 62:539-548
Nesbo CL, Magnhagen C, Jacobsen KS, (1998) Genetic differentiation among stationary
and anadromous perch (Perca fluviatilis) in the Baltic Sea. Hereditas 129: 241249
Nesbo CL, Fossheim T, Vollestad LA, Jakobsen KS (1999) Genetic divergence and
phylogeographic relationships among European perch (Perca fluviatilis)
populations reflect glacial refugia and postglacial colonization. Mol. Ecol. 8:
1387-1404
MacGregor RB, Witzel LD (1987) A twelve year study of the fish community in the
Nanticoke region of long point bay, Lake Erie. Lake Erie Fisheries Assessment
Unit 1987-3. 65p
Manel S, Schwartz M, Luikart G, Taberlet P (2003) Landscape genetics: combining
landscape ecology and population genetics. Trends Ecol Evol 18: 189-197
Mangan MT (2004) Yellow perch production and harvest strategies for semi- permanent
wetlands in Eastern South Dakota. MSc thesis, Wildlife and Fisheries Sciences,
South Dakota State University. 85p
96
Manni FE, Guerard E, Heyer E (2004a) Geographical patterns of (genetic, morphologic,
linguistic) variation: How barriers can be detected by "Monmonier's algorithm".
Hum Biol 76: 173-190
Manni FE, Guerard E, Heyer E (2004b) BARRIER v2.2. Museum of Mankind, Paris.
Available from http://www.mnhn.fr/mnhn/ecoanthropologie/software/barrier.html
[Accessed 5 March 2013]
Manny BA, Kenaga D (1991) The Detroit River: effects of contaminants and human
activities on aquatic plants and animals and their habitats. Hydrobiologia 219:
269-279
Manny BA, Carl LM, Morrison S, Nichols SJ, Roseman EF, Riley SC (2004) The U.S.
Geological Survey Huron-Erie Corridor Initiative. Available from
http://www.huronerie.org/files/mannyheci2004.pdf. [Accessed 19 July 2012]
Manny BA, Kennedy GW, Allen JD, French JRP (2007) First evidence of egg deposition
by walleye (Sander vitreus) in the Detroit River. J Great Lakes Res 33: 512-516
Mantel N (1967) The detection of disease clustering and a generalized regression
approach. Cancer Res 27: 209-220
Marsden, J. E., and S. R. Robillard. 2004. Decline of yellowperch in southwestern Lake
Michigan, 1987–1997. N Am J Fish Manage 24:952–966
Miller L (2003) Microsatellite DNA loci reveal genetic structure of yellow perch in Lake
Michigan. Trans Am Fish Soc 132:503-513
Miller MR, Brunelli JP, Wheeler PA, Liu S, Rexroad III CE, Palti Y, Doe CQ, Thorgaard
GH, (2012) A conserved haplotype controls parallel adaptation in geographically
distant salmonid populations. Mol Ecol 21: 237-249
97
Mills L, Smouse P (1994) Demographic consequences of inbreeding in remnant
populations. Am Nat 144: 412-431
Mitchell S (1814) Perca flavescens. Report on fishes, New York. 28pp
Mora C, Sale PF (2002) Are populations of coral reef fish open or closed? Trends Ecol
Evo 17: 422-428
Morita K, Yamamoto S (2002) Effects of habitat fragmentation by damming on the
persistence of stream-dwelling char populations. 16: 1318-1323
Moulton R, Theime S (2009) History of dredging and compensation – St. Clair and
Detroit rivers. International Upper Great Lakes Study, Ottawa, Ontario. Accessed
at http://pub.iugls.org/en/ St_Clair_Reports/Sediment/Sediment-05.pdf, 147p.
[Accessed 19 July 2012]
Munawar M, Numawar IF, Mandrak NE, Fitzpatrick M, Dermott R, Leach J (2005) An
overview of the impact of non-indeginous species on food web integrity of North
American Great Lakes: Lake Erie example. Aquat Ecosyst Health 8: 375-395
Nacci DE, Gleason TR, Munns WR (2002) Evolutionary and cological effects of multigenerational exposure ot anthropogenic stressors. Hum Ecol Rish Assess. 8: 91-97
Olsen EM, Heino M, Lilly GR, Morgan MJ, Brattley J, Ernande B, Dieckmann U (2004)
Maturation trends indicative of rapid evolution preceded the collapse of northern
cod. Nature 428: 932-935
Ontario Ministry of Natural Resources (2011) 2006-2009 Annual Report. Lake Erie MU
117p
98
Parker AD, Stepien CA, Sepulveda-Villet OJ, Ruehl CB, Uzarski DG (2009) The
interplay of morphology, habitat, resource use, and genetic relationships in young
yellow perch. Trans Am Fish Soc 138:899-914
Peel D, Ovenden JR, Peel SL (2004) NEESTIMATOR: Software for estimating effective
population size, Version 1.3. Queensland Government: Department of Primary
Industries and Fisheries, Brisbane, Queensland
Piry S, Alapetite A, Cornuet JM, Paetkau D, Baudouin L, Estoup E (2004) GeneClass2:
A Software for Genetic Assignment and First-Generation Migrant Detection. J
Hered 95: 536-539 Available from http://www1.montpellier.inra.fr/URLB/
[Accessed 5 March 2013]
Planes S, Lefant P (2002) Temporal change in the genetic structure between and within
cohorts of a marine fish, Diplodus sargus, induced by a large variance in
individual reproductive success. Mol Ecol 11: 1515-1524
Pracheil BM, Pegg MA, Powell LA, Mestl GE (2012) Swimways: protecting paddlefish
through movement-centered management. Fisheries 37: 449-457
Pritchard JK, Stephens M, Donnelly P (2000) Inference of population structure using
multilocus genotype data. Genetics 155: 945-959.
Pritchard JK, Wen W (2004) Documentation for STRUCTURE software: Version 2.2.
Department Human Genetics, University of Chicago, Chicago, Ill. Available
from http://pritch.bsd.uchicago.edu/structure.html [Accessed 5 March 2013]
Quinn TP, Kinnison MT, Unwin MJ (2001) Evolution of chinook salmon (Oncorhynchus
tshawytscha) populations in New Zealand: patterns, rate, and process. Genetica
112-113:493-513
99
R Development Core Team (2011) R: A language and environment for statistical
computing, R foundation for statistical computing, Vienna. Available from
http://www.r-project.org/. [Accessed 5 March 2013]
Rawson MR (1980) Lake Erie field investigations: yellow perch movements. Ohio
Department of Natural Resources, Division of Wildlife, Federal Aid in Fish
Restoration Project F-35-R-18, Columbus Ohio, 25p
Raymond M, Rousset F, (2005) An exact test of population differentiation. Evolution 49:
1280-1283
Rice WR (1989) Analyzing tables of statistical tests. Evolution 43: 223-225
Rice J, Moksness E, Attwood C, Brown SK, Dahle G, Gjerde KM, Grefsrud ES,
Kenchington R, Kleiven AR, McConney P, Ngoile MAK, Naesje TF, Olsen E,
Olsen EM, Sanders J, Sharma C, Vestergaard O, Westlund L (2012) The role of
MPAs in reconciling fisheries management with conservation of biological
diversity. Ocean Coast Manage 69: 217-230
Robillard SR, Marsden JE (2001) Spawning substrate preferences of yellow perch along
a sand-cobble shoreline in southwestern Lake Michigan. Trans Am Fish Soc 21:
208-215
Roseman EF, Kennedy GW, Boase J, Manny BA, Todd TN, Stott W (2007) Evidence of
lake whitefish spawning in the Detroit River: Implications for habitat and
population recovery. J Great Lakes Res 33: 397-406
Rousset F (1997) Genetic differentiation and estimation of gene flow from F-statistics
under isolation by distance. Genetics 145: 1219-1228
Rousset F (2008) GENEPOP ‘007: a complete re-implementation of the GENEPOP
100
software for Windows and Linux. Mol Ecol Resour 8: 103-106. Available from
http://mbb.univ-montp2.fr/MBB/ subsection/downloads.php?section=2
[Accessed 5 March 2013]
Rush SA, Paterson G, Johnson TB, Drouillard KG, Haffner GD, Hebert CE, Arts MT,
McGoldrick DJ, Backus SM, Lantry BF, Lantry JR, Schaner T, Fish AT (2012)
Long-term impacts of invasive species on a native top predator in a large lake
system. Freshwater Biol 57: 2342-2355
Ruzzante DE, Taggart CT, Doyle RW, Cook D (2001) Stability in the historical pattern
of genetic structure of newfoundland cod (Gadus morhua) despite catastrophic
decline in population size from 1964 to 1994. Cons Genet 2: 257-269
Ryan PA, Knight R, MacGregor R, Towns G, Hoopes R, Culligan W (2003) Fish
community goals and objectives for Lake Erie. Special Publication 03-02, Great
Lakes Fishery Commission, Windsor, Ontario
Saalfeld DT, Reutebuch EM, Dickey RJ, Seesock WC, Webber C, Bayne DR (2012)
Effects of landscape characterstics on water quality and fish assemblages in the
Tallapoosa river basin, Alabama Southeast Nat 11: 239-252
Sato T (2006) Occurrence of deformed fish and their fitness-related traits in Kirikuchi
charr, Salvelinus leucomaenis japonicas, the southernmost population of the
genus Salvelinus. Zool Sci 23: 593-599
Schwartz MK, Luikart G, Waples RS (2007) Genetic monitoring as a promising tool for
conservation management. Trends Ecol Evol 22: 25-33
Scott WB, Crossman EJ (1973) Freshwater fishes of Canada. Bulletin of the Fisheries
Research Board of Canada
101
Selkoe KA, Gaines SD, Caselle JE, Warner RR (2006) Current shifts and kin aggregation
explain genetic patchiness in fish recruits. Ecology 87: 3082-3094
Sepulveda-Villet OJ, Ford AM, Williams JD, Stepien CA (2009) Population genetic
diversity and phylogeographic divergence patterns of the yellow perch (Perca
flavescens) J Great Lakes Res 35: 107–119
Sepulveda-Villet OJ, Stepien CA (2011) Fine-scale population genetic structure of the
yellow perch Perca flavescens in Lake Erie. Can J Fish Aquat Sci 68: 1435-1453
Sepulveda-Villet OJ, Stepien CA (2012) Waterscape genetics of the yellow perch (Perca
flavescens) from two genomes: Patterns across large connected systems and
isolated relict populations. Mol Ecol 23: 5795-5826
Stephenson RL (1999) Stock complexity in fisheries management: a perspective of
emerging issues related to population sub-units. Fish Res 43: 247-249
Stepien CA, Faber JE (1998) Population genetic structure, phylogeography, and
spawning philopatry in walleye (Stizostedion vitreum) from mitochondrial DNA
control region sequences. Mol Ecol 7: 1757-1769
Stepien CA, Dillon AK, Chandler MD (1998) Genetic identity, phylogeography, and
systematics of the ruffe Gymnocephalus in the North American Great Lakes and
Eurasia. J Great Lakes Res 24: 361-378
Stepien CA, Brown JE, Neilson ME, Tumeo MA (2005) Genetic diversity of invasive
species in the Great Lakes versus their Eurasian source populations: insights for
risk analysis. Risk Anal 25: 1043-1060
Stepien CA, Murphy DM, Strange RM (2007) Broad- to fine-scale population genetic
patterning in the smallmouth bass Micropterus dolomieu across the Laurentian
102
Great Lakes and beyond: an interplay of behavior and geography. Mol Ecol 16:
1605-1624
Stepien CA, Murphy DM, Lohner RN, Sepulveda-Villet OJ, Haponski AE (2009)
Signatures of vicariance, postglacial dispersal, and spawning philopatry:
population genetics of the walleye Sander vitreus. Mol Ecol 18: 3411-3428
Stepien CA, Murphy DM, Lohner RN, Haponski AE, Sepulveda-Villet OJ (2010) Status
and delineation of walleye genetic stock structure across the Great Lakes. In
Status of Walleye in the Great Lakes: Proceedings of the 2006 Symposium. Great
Lakes Fishery Commission Technical Report. 69: 189-223
Stepien CA, Banda JA, Murphy DM, Haponski AE (2012) Temporal and spatial genetic
consistency of walleye spawning groups. Trans Am Fish Soc 141:660-672
Strange RM, Stepien CA (2007) Genetic divergence and connectivity among river and
reef spawning populations of walleye (Sander vitreus) in Lake Erie. Can J Fish
Aquat Sci 64: 437-448.
Storfer A, Murphy MA, Evans JS, Goldberg CS, Robinson S, Spear SF, Dezzani R,
Delmelle E, Vierling L, Waits LP (2007) Putting the ‘landscape’ in landscape
genetics. Heredity 98: 128-142.
Sullivan TJ, Stepien CA (2013) Genetic diversity and divergence of yellow perch
spawning populations across the Huron-Erie corridor, from Lake Huron through
western Lake Erie. J Great Lakes Res
Tessier N, Bernatchez L (2002) Stability of population structure and genetic diversity
assessed by microsatellites among sympatric population od landlocked Atlantic
salmon (Salmo salar L.) Mol Ecol 8: 169-179
103
Thomas MV, Haas RC (2000) Status of yellow perch and walleye populations in
Michigan waters of Lake Erie, 1994-1998. Fisheries Division Research Report
No. 2054. Michigan Department of Natural Resources. Harrison Township,
Michigan. 39p.
Todd TN, Hatcher CO (1993) Genetic variability and glacial origins of yellow perch
(Perca flavescens) in North America. Can J Fish Aquat Soc 50, 1828-1834
Trautman MB (1981) The Fishes of Ohio. The Ohio State University Press, Columbus
OH, pp. 309-311
Tymchuk WV, O’Reilly P, Bittman J, MacDonald D, Schulte P (2010) Conservation
genomics of Atlantic salmon: variation in gene expression between and within
regions of the Bay of Fundy. Mol Ecol 19: 1842-1859
Tyson JT, Knight RL (2001) Response of yellow perch to changes in the benthic
invertebrate community of western Lake Erie. Trans Am Fish Soc 130: 766-782
U.S. Army Corps of Engineers (2004) Waterborne commerce of the United States,
calendar year 2004. Part 3-Waterways and harbors, Great Lakes, U.S, Army
Corps of Engineers, Waterborne Commerce Statistics Center. New Orleans,
Louisiana
van Oosterhout C, Hutchinson WF, Wills DPM, Shipley P (2004) MICRO-CHECKER:
software for identifying and correcting genotyping errors in microsatellite data.
Mol Ecol Notes 4: 535-538.
van Oosterhout C, Hutchinson WF, Wills DPM, Shipley P (2006) MICRO-CHECKER
Available from http://www.microchecker.hull.ac.uk. [Accessed 5 March 2013].
104
Vitousek PM, Mooney HA, Lubchenco J, Melillo JM (1997) Human domination of
earth’s ecosystems. Science 277: 494-499
Wang J (2001) A pseudo-likelihood method for estimating effective population size from
temporally spaced samples. Genet Res 78: 243-257
Wang J, Whitlock MC (2003) Estimating effective population size and migration rates
from genetic samples over space and time. Genetics 163:429-446
Waples RS (1989) A generalized approach for estimating effective population sizes from
temporal changes in allele frequency. Genetics. 121: 379-391
Weir BS, Cockerham CC, (1984) Estimating F-statistics for the analysis of population
structure. Evolution 38: 1358-1370
Whiteside M, Swindoll S, Doolittle W (1985) Factors affecting the early life history of
yellow perch Perca flavescens. Environ Biol Fish 12: 47-56
Wolter C, Minow J, Vilcinskas A, Grosch UA (2000) Long-term effects of human
influence on the fish community structure and fisheries in berlin waters: an urban
water system. Fisheries Manag Ecol 7: 97-104
YPTG (2006) Lake Erie Committee. Yellow Perch Task Group. Executive Summary.
Lake Erie Commission. Great Lakes Fishery Commission. Windsor, Ontario.
http://www.glfc.org/lakecom/lec/YPTG_docs/ annual_reports/
YPTG_report_2006.pdf [Accessed 5 March 2013]
YPTG (2011) Lake Erie Committee. Yellow Perch Task Group. Executive Summary.
Lake Erie Commission. Great Lakes Fishery Commission, Windsor, Ontario.
http://www.glfc.org/lakecom/lec/YPTG_docs/ annual_reports/
YPTG_report_2011.pdf [Accessed 5 March 2013]
105
YPTG (2012) Lake Erie Committee. Yellow Perch Task Group. Executive Summary.
Lake Erie Commission. Great Lakes Fishery Commission, Windsor, Ontario.
http://www.glfc.org/ lakecom/lec/YPTG_docs/annual_reports/
YPTGreport2012.pdf [Accessed 5 March 2013]
106