Genetic Diversity and Population Structure of Topeka Shiners (Notropis Topeka) Alexander Peter Bybel Iowa State University

Iowa State University Capstones, Theses and Graduate Theses and Dissertations Dissertations

2018 Genetic diversity and population structure of Topeka shiners (Notropis topeka) Alexander Peter Bybel Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/etd Part of the Biology Commons

Recommended Citation Bybel, Alexander Peter, "Genetic diversity and population structure of Topeka shiners (Notropis topeka)" (2018). Graduate Theses and Dissertations. 16792. https://lib.dr.iastate.edu/etd/16792

This Thesis is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. Genetic diversity and population structure of Topeka shiners (Notropis topeka)

Alexander Peter Bybel

A thesis submitted to the graduate faculty

in partial fulfillment of the requirements for the degree of

MASTER OF SCIENCE

Major: Fisheries Biology

Program of Study Committee: Kevin J. Roe, Co-major Professor Michael J. Weber, Co-major Professor Clay L. Pierce

The student author, whose presentation of the scholarship herein was approved by the program of study committee, is solely responsible for the content of this thesis. The Graduate College will ensure this thesis is globally accessible and will not permit alterations after a degree is conferred.

Iowa State University

Ames, Iowa

2018

TABLE OF CONTENTS

Page

ACKNOWLEDGMENTS ...... iii

ABSTRACT ...... iv

CHAPTER 1. GENERAL INTRODUCTION ...... 1 Background ...... 1 Thesis Organization ...... 3 Literature Cited ...... 3

CHAPTER 2. RANGEWIDE GENETIC DIVERSITY AND POPULATION STRUCTURE OF THE ENDAGERED TOPEKA SHINER ...... 6 Abstract ...... 6 Introduction ...... 6 Methods ...... 10 Study Area ...... 10 Fish Sampling ...... 11 DNA Extraction and Genotyping ...... 12 Genetic Analysis...... 13 Results ...... 15 Discussion ...... 19 Literature Cited ...... 25 Tables ...... 33 Figures ...... 42

CHAPTER 3. METAPOPULATION STRUTURE OF TOPEKA SHINERS IN IOWA AND MINNESOTA ...... 49 Abstract ...... 49 Introduction ...... 49 Methods ...... 53 Fish Sampling ...... 53 DNA Extraction and Genotyping ...... 54 Genetic Analysis...... 55 Results ...... 57 Discussion ...... 59 Literature Cited ...... 66 Tables ...... 74 Figures ...... 82

CHAPTER 4. GENERAL CONCLUSIONS ...... 88 Summary ...... 88 Literature Cited ...... 91

APPENDIX. SUMMARY INFORMATION FOR MICROSATELLITE LOCI ...... 93 iii

ACKNOWLEDGMENTS

I’d like to thank my committee chairs, Dr. Kevin Roe and Dr. Michael Weber as well as committee member Dr. Clay Pierce for support and guidance throughout the duration of this project. I also want to thank Aleshia Kenny, Kim Emerson, Scott Ralston, and Gibran Suleiman of the United States Fish and Wildlife Service, Chelsea Pasbrig of

South Dakota Game, Fish, and Parks, Scott Campbell of the Kansas Biological Survey and Chelsea Titus and Leah Berkman of Missouri Department of Conservation as well as the many other people within state and federal agencies who provided me with sampling sites, helped me in the field, or collected genetic samples for me. I also need to thank my technicians as their help in the field and in the lab made this project possible.

Additionally, I would like to thank Sara Schwarz for her help in the lab and with data analysis. Lastly, I want to thank Courtney Zambory and Nick Simpson, fellow Topeka shiner graduate students, for their collaboration on this project and all the other things we worked on together throughout my time here at Iowa State University. Partial funding for this project was provided through the U.S. Fish and Wildlife Service's State Wildlife

Grants Competitive Program (F15AP010369) and the Iowa DNR. iv

ABSTRACT

Species and populations with low genetic diversity are at a higher risk of extinction because of an inability to adapt to threats such as environmental change.

Human activities have drastically reduced the amount of habitat, the number of populations, and population connectivity of many species throughout the world.

Fragmented populations can exist as isolated populations with no gene flow or as metapopulations - groups of discrete demes that maintain some level of gene flow. The

Topeka shiner (Notropis topeka) is an endangered cyprinid native to the Midwestern

United States. Topeka shiners have experienced drastic reductions in distribution due to stream alterations that have eliminated both instream and off-channel habitats, leaving smaller, more isolated populations throughout the range. Species with small isolated populations can suffer from low genetic diversity and lack of gene flow. A total of nine polymorphic microsatellite loci were used to asses genetic diversity, analyze population structure, compare migration rates across the range, and characterize metapopulation structure across the range of the Topeka shiner. Rangewide analysis of population structure revealed eight distinct populations with moderate levels of genetic diversity that can serve as management units for conservation. Estimates of historical and contemporary migration indicated that populations were more connected thousands of years ago whereas current populations have almost no gene flow between them. These results suggest habitat destruction has led to a reduction in genetic connectivity across the range of the Topeka shiner leaving the remaining populations geographically isolated.

Analysis of genetic structure within the Rock River and Boone River basins indicated deviations from Hardy-Weinberg Equilibrium, significant genetic isolation by distance, v and low but significant genetic differences between sites. These results suggest Topeka shiners in these basins are acting as metapopulations composed of discrete demes.

Analysis of migration patterns indicate larger demes may be acting as potential populations sources that export individuals to other demes. Although Topeka shiner populations exhibit moderate levels of genetic diversity throughout the range, failure to correct the factors that led to a decline of this species will result to reduced diversity, leaving populations susceptible to stochastic events. Given the vast geographic distance between many populations, natural reestablishment is unlikely between these populations. Managers, therefore, should focus on preserving or improving genetic diversity within these populations. Improving the amount and distribution of quality habitat to will allow Topeka shiners metapopulations to expand and persist in within basins.

CHAPTER 1. GENERAL INTRODUCTION

Background

Throughout North America, many freshwater species are in need of conservation as a result of human activities (Jansen 1998; Abell 2002; Wiens 2007). Approximately 39% of fishes and 47% of cyprinids in North America are imperiled (Jelks et al. 2008). One major threat to these species is habitat destruction through land conversion for agriculture (McGill et al. 2015). Loss of habitat can lead to a reduction and extirpation of local populations and fragmentation distributions. Species that experience long-term isolation both between and within populations are often left vulnerable to the effects of low genetic diversity and drift and therefore less able to adapt to environmental change (Quattro and Vrijenhoek 1989;

Saccheri et al. 1998; Zhou et al. 2016). Therefore, determining the genetic variability in populations of threatened and endangered species is critical to conservation efforts. Genetic data can provide information on diversity, population structure, and trends in population connectivity (Abdul-Muneer 2014). Isolation due to habitat loss can result in population fragmentation (Rybicki and Hanski 2013). Fragmented populations exhibit reduced connectivity and often become more inbred overtime, leading to an erosion of genetic diversity (Jackson and Fahrig 2016).

Before human settlement, the United States was covered by natural forests, prairies, and wetlands (Whitney 1994). However, there has been a substantial shift in land use towards agriculture post human settlement. Agriculture now covers over 915 million acres (40%) of the United States, but the Midwestern United States has had a particularly high amount of land conversion (USDA 2014). Row crop agricultural in the Midwest is now widespread because of nutrient rich soil, with 70-90% of land in several states currently in agricultural 2 production (USDA 2014). This agricultural conversion has resulted in immense modification to streams through channelization. Stream channelization provides many benefits to agriculture through irrigation and drainage of fields as well as flood control (Simpson et al.

1982). However, there are also substantial negative effects of channelization on aquatic ecosystems. Stream channelization creates a homogenous run habitat that reduces the stream’s connectivity to its flood plain and eliminates riffles, pool, and off-channel habitats that are preferred habitat of many aquatic species.

One species affected by these agricultural practices is the Topeka shiner (Notropis topeka, Gilbert 1884). The Topeka shiner is a small cyprinid distinguished by its silver color with a dark stripe along its side and a black chevron shaped wedge at the base of the caudal fin, with males developing an orange fin coloring during the breeding season (Pflieger 1997).

In 1998, the Topeka shiner was federally listed as endangered under the Endangered Species

Act (USFWS 1998) because their distribution had declined to 20% of their historical range in

Kansas, Missouri, Nebraska, South Dakota, Minnesota, and Iowa (Tabor 1998).

Throughout their current range, Topeka shiners inhabit stream segments that have slow moving pools, side channels, and back channels with little flow that have not been eliminated by agricultural conversion (Schrank et al. 2001; Kuitunen 2001; Ceas and

Anderson 2004; Bakevitch et al. 2015). Topeka shiners have also been found in off-channel habitat such as oxbows in Iowa and Minnesota and dugout ponds in South Dakota (Thomson and Berry 2009; Bakevich et al. 2013). Oxbows are natural meanders of streams that have become disconnected over time from the main channel, only reconnecting during flooding events (Clark 2000; Thomson and Berry 2009; Bakevich et al. 2013). Starting in 2002, the

United States Fish and Wildlife Service (USWFS) initiated a program to restore oxbow 3 habitat in Iowa (Kenney 2013), and to date, there have been more than 80 oxbows restored in

Iowa and 60 in Minnesota (USFWS 2018). Sampling for Topeka shiner in oxbows has produced evidence of natural reproduction and in some instances, more than 1,000 individuals, indicating oxbows represent a potentially important habitat for Topeka shiners

(Kenney 2013).

The reduction of habitat due to stream channelization has left Topeka shiners currently existing in a very fragmented and reduced distribution across the range leaving populations at higher risk of future extirpation. Therefore, the goals of this study were to characterize the current genetic structure, diversity and connectivity both range-wide and within populations. Results from this study will help managers indicate low diversity areas in greatest need of conservation, delineate genetically distinct populations that could serve as management units, and characterize migration that can be used to guide restoration efforts to improve population stability and connectivity.

Thesis Organization

Chapter 1 is a general introduction to this thesis. Chapter 2 focuses on genetic diversity, structure and connectivity of Topeka shiners across the current range. Chapter 3 concentrates on within basin metapopulation dynamics of Topeka shiners. Chapters 2 and 3 each include an abstract, introduction, methods, results, and discussion followed by tables and figures. Chapter 4 provides general conclusions on the overall project. Appendices provide further summary of data and statistical analyses.

Literature Cited

Abell, R. 2002. Conservation biology for the biodiversity crisis: a freshwater follow‐up. Conservation Biology 16:1435-1437.

Abdul-Muneer, P. M. 2014. Application of microsatellite markers in conservation genetics and fisheries management: recent advances in population structure analysis and conservation strategies. Genetics Research International 11p.

Bakevich, B. D., C. L. Pierce, and M. C. Quist. 2013. Habitat, fish species, and fish assemblage associations of the Topeka shiner in West-Central Iowa. North American Journal of Fisheries Management 33:1258-1268.

Bakevich, B. D., C. L. Pierce, and M. C. Quist. 2015. Status of the Topeka shiner in west- central Iowa. American Midland Naturalist 174:350-358.

Ceas, P. A., and Y. C. Anderson. 2004. Results of a pilot monitoring project for Topeka shiners in Southwestern Minnesota. Minnesota Department of Natural Resources St. Paul, MN, USA.

Clark, S. J. 2000. Relationship of Topeka shiner distribution to geographic features of the Des Moines Lobe in Iowa. M.S. Thesis, Iowa State University, Ames, IA, USA.

Gilbert, C. H. 1884. Notes on the fishes of Kansas. Bulletin of the Washburn College Laboratory of Natural History 1:1-16.

Jackson, N. D., and L. Fahrig. 2016. Habitat amount, not habitat configuration, best predicts population genetic structure in fragmented landscapes. Landscape Ecology 31:951- 968.

Janzen, D. 1998. Essays on science and society: gardenification of wildland nature and the human footprint. Science 279:1312-1313.

Jelks, H. L., S. J. Walsh, N. M. Burkhead, S. Contreras-Balderas, E. Diaz-Pardo, D. A. Hendrickson, J. Lyons, N. E. Mandrak, F. McCormick, J. S. Nelson, S. P. Platania, B. A. Porter, C. B. Renaud, J. J. Schmitter-Soto, E. B. Taylor, and M. L. Warren Jr. 2008. Conservation status of imperiled North American freshwater and diadromous fishes. Fisheries 33:372-407.

Kenny, A. 2013. The Topeka shiner: shining a spotlight on an Iowa success story. Endangered species program. US Fish and Wildlife Service.

Kuitunen, A. 2001. Microhabitat and instream flow needs of the Topeka shiner in the Rock River watershed, MN. Minnesota Department of Natural Resources, St. Paul, MN, USA.

McGill, B. J., M. Domelas, N. J. Gotelli, and A. E. Magurran. 2015. Fifteen forms of biodiversity trend in the Anthropocene. Trends Ecological Evolution 104-113.

Pflieger, W. L. 1997. The fishes of Missouri, Columbia, Missouri, Missouri Department of Conservation, Jefferson City, MO, USA. 372p. 5

Quattro, J. M., and R. C. Vrijenhoek. 1989. Fitness differences among remnant populations of the endangered Sonoran topminnow. Science 245:976-978.

Rybicki, J., and I. Hanski. 2013. Species–area relationships and extinctions caused by habitat loss and fragmentation Ecology Letters 16:27-38.

Saccheri, I., M. Kuussaari, M. Kankare, P. Vikman, W. Fortelius, and I. Hanski. 1998. Inbreeding and extinction in a butterfly metapopulation. Nature 392:491-494.

Schrank, S. J., C. S. Guy, M. R. Whiles, and B. L. Brock. 2001. Influence of instream and landscape-level factors on the distribution of Topeka shiners (Notropis topeka) in Kansas streams. Copeia 413-421.

Simpson, P. W., J. R. Newman, M. A. Keirn, R. M. Matter, and P. A. Guthrie. 1982. Manual of stream channelization impacts on fish and wildlife. U.S. Fish and Wildlife Service Technical Report FWS/OBS-82/24.

Tabor, V. M. 1998. Final rule to list the Topeka shiner as endangered. Federal Register 63(240):69008-69021.

Thomson, S. K. and C. R. Berry. 2009. Stream Fishes inhabit livestock watering ponds (dugouts) near six mile creek, Brookings County, South Dakota. North American Journal of Fisheries Management 88:127-138.

U.S. Department of Agriculture (USDA). 2014. Farms and farmland numbers, acreage, ownership, and use. Available https://www.agcensus.usda.gov/Publications/2012/ Online_Resources/Highlights/Farms_and_Farmland/Highlights_Farms_and_Farmlan d.pdf.

U.S. Fish and Wildlife Service (USFWS). 1998. Endangered and threatened wildlife and plants; final rule to list the Topeka shiner as endangered. Federal Register 65:57114- 57119.

U.S. Fish and Wildlife Service (USFWS). 2018. Final Topeka shiner Species Status Assessment. Species Status Assessment Reports. South Dakota Ecological Services Office.

Whitney, G. G. 1994. From coastal wilderness to fruited plain: a history of environmental change in temperate North America, 1500 to the present. Cambridge University Press.

Wiens, J. A. 2007. The demise of wildness? Bulletin of the British Ecological Society 38:78- 79.

Zhou, X., X. Meng, Z. Liu, J. Chang, B. Wang, and M. Li. 2016. Population genomics reveals low genetic diversity and adaptation to hypoxia in Snub‐Nosed Monkeys. Molecular Biology and Evolution 33:2670. 6

CHAPTER 2. RANGEWIDE GENETIC DIVERSITY AND POPULATION STRUCTURE OF THE ENDAGERED TOPEKA SHINER

Abstract

Species and populations with low genetic diversity are at a higher risk of extinction because of an inability to adapt to threats such as environmental change. Human activities have drastically reduced the amount of habitat and the number of populations of many species throughout the world. The Topeka shiner, (Notropis topeka), is an endangered cyprinid native to the Midwestern United States that has experienced drastic reductions in its distribution due to stream alterations, leaving fragmented populations throughout the range.

Nine polymorphic microsatellite loci were used to assess genetic diversity, analyze population structure, and compare historical and contemporary migration rates across the range of the Topeka shiner. Analysis of population structure revealed seven distinct populations with moderate levels of genetic diversity. Estimates of historical and contemporary migration indicated that populations were significantly more connected historically whereas current populations exhibit little gene flow. Overall, this study suggests there has been a reduction of genetic connectivity across the range of the Topeka shiner, leaving the remaining populations geographically isolated. Continual decline of this species will lead to reduced genetic diversity, leaving populations susceptible to stochastic events.

Improving habitat connectivity among Topeka shiner populations will help ensure their survival.

Introduction

Analysis of genetic data has become very important for informing the long-term conservation of threatened and endangered species (Hedrick 2012). Genetic data can provide information on diversity, population structure, and trends in population connectivity (Abdul- 7

Muneer 2014). Species that have lower or reduced genetic diversity are less able to adapt to threats, such as disease and environmental change, and are therefore at a higher risk of extinction (Quattro and Vrijenhoek 1989; Saccheri et al. 1998; Zhou et al. 2016). Therefore, determining the genetic variability in populations of threatened and endangered species is crucial to conservation efforts. Another threat to species is a reduction in their distributions due to habitat loss, resulting in population fragmentation (Rybicki and Hanski 2013).

Fragmented populations exhibit reduced connectivity between fragments that often become smaller and more inbred overtime, leading to lowered fitness and erosion of genetic diversity

(Jackson and Fahrig 2016). Over time reproductive isolation between populations will cause populations to become genetically distinct and genetic data can be used to estimate the number of distinct populations that can serve as management units for recovery (Storfer

1999) and estimate the amount of gene flow within and between them.

The Topeka shiner, (Notropis topeka, Gilbert 1884), is a cyprinid fish that was historically distributed in headwater streams across Kansas, Missouri, Nebraska, South

Dakota, Minnesota, and Iowa, USA (USFWS 2018). In 1998, the Topeka shiner was listed as endangered under the Endangered Species Act (USFWS 1998) because its distribution had declined to 20% of the historical range (Tabor 1998). Before human settlement, Midwestern streams were naturally meandering with numerous oxbows, back-channels, and side-channels

(Berg et al. 2004). These areas of slow-moving water in headwater streams were a preferred habitat of Topeka shiners across their range (Minckley and Cross 1959; Ceas and Anderson

2004; USFWS 2018). However, widespread stream channelization in support of intensive agricultural practices has resulted in a reduction of natural meanders and preferred pool habitat in streams (Simpson et al. 1982). 8

Because of the reduction of preferred habitat and subsequent extirpation of populations, the current distribution of the Topeka shiner consists of a relatively small number of populations scattered across the Midwest (USWFS 2018). When comparing the historical and contemporary distribution patterns, the United States Fish and Wildlife Service

(USWFS) divided the range of the Topeka shiners into a Northern portion (South Dakota,

Minnesota, and Iowa) including 70% of current distribution but only 20% of the historic distribution and a Southern portion (Kansas, Missouri, and Nebraska) portion that historically contains 30% in the present range but 80% of the historic range (USFWS 2009).

There have been reductions in the distribution of Topeka shiners in some areas of the

Northern portion of the range, but the overall distribution in this region has remained stable.

The increase in the number of known Topeka shiner locations in the Northern portion since it was listed has been due to an increase in sampling efforts that have discovered new populations (Wall and Thompson 2007). Populations in South Dakota and Minnesota are considered to be stable in the James, Vermilion, Big Sioux, and Rock river basins (Pasbrig and Lucchesi 2012; Cunningham 2017) whereas some portions of the Rock River Basin are thought to be declining (Nagle and Larson 2014; Bakevich et al. 2015). Iowa populations of

Topeka shiners were estimated to have experienced a 73% decline in historical distribution and only occur in three geographically isolated basins, the Boone, North Raccoon, and Rock river basins (Clark 2000; Bakevich et al. 2015). Recent sampling has found an increase in

Topeka shiners in the Boone River Basin, possibly due to increased sampling efforts

(Simpson et al. in review). However, there has been a decrease in distribution in the North

Raccoon River Basin (Clark 2000; Bakevich 2015; Simpson et al. 2017; Zambory et al.

2017). 9

The southern portion of the Topeka shiner range has experienced drastic reductions in distribution, with many historical populations now completely extirpated. Although Topeka shiners were once widespread throughout Kansas (Cross and Collins 1975), there are now only two basins located in the Flint Hills, the Kansas River and Cottonwood River basins, where Topeka shiners still naturally occur (USFWS 2018). Both basins are unsuitable for row crop agriculture, which has allowed Topeka shiners to continue to occupy these streams

(Gerkin and Paukert 2013). A third basin of historical Topeka shiner occupancy, Willow

Creek, presently exists only under captive conditions in Kansas as the wild stock has become extirpated due to the introduction of sportfish (Campbell et al. 2016). Topeka shiner distribution in Missouri has also experienced great declines. Historically, the largest population of Topeka shiners was centered around the Missouri River in the middle of the state (Pflieger 1997). Other isolated populations existed in tributaries of the Missouri River in

Northwest Missouri and the Grand and Chariton River drainages (Pflieger 1997) whereas only the Sugar Creek and Moniteau Creek basins still contain Topeka shiners (MDC 2010).

Nebraska has experienced the most severe decline in distribution. Topeka shiners were historically distributed throughout Eastern Nebraska (Meek 1894). There are currently only two areas that possibly harbor Topeka shiners, although neither has been recently confirmed

(USFWS 2018). Topeka shiners were last seen in the Elkhorn River in 2016 while the last record from the North Loup River is from 1989 (USFWS 2018).

Previous genetic research on Topeka shiners has indicated varying levels of population structure across its range. The only range-wide characterization of Topeka shiner genetic variation used mitochondrial DNA (mtDNA) sequences from eleven locations across the range and identified three distinct groups based on a lack of shared haplotypes: the Upper 10

Missouri, Lower Missouri, and Arkansas rivers (Michels 2000). More recent studies based on microsatellite data suggest Minnesota, South Dakota, and Missouri may be distinct populations (Anderson and Sarver 2008). Additionally, fine scale genetic variation has also been identified in Topeka shiners, suggesting possible segregation of populations between the Grand River Basin (Sugar Creek) and Lower Missouri Basin (Moniteau Creek) in

Missouri (Bergstrom et al. 1999). Some genetic variation was also recently documented between three sampled drainages in South Dakota (Blank et al. 2011). Although microsatellite studies suggest the possibility of finer structuring of Topeka shiner populations than the previous mtDNA study indicated, the scope of these studies have been small and covered only portions of the range. Additionally, there have been no studies that have assessed historical and contemporary genetic connectivity of populations. High historical and low contemporary genetic connectivity may suggest that human activities have affected gene flow throughout the range, whereas low historical and contemporary rates would suggest

Topeka shiner populations would have existed in isolation before human colonization.

The objectives of this study are to assess genetic diversity and population structure as well as historical and contemporary migration of Topeka shiners using nine microsatellite loci across the entire range. Results from this study will determine potential management units of Topeka shiners and indicate areas at risk, helping conservation managers to preserve genetic viability across the range of this species.

Methods

Study Area

The Des Moines River Basin, the Missouri River Basin, and the Arkansas River

Basin are the three major basins that are occupied by Topeka shiners in Minnesota, Iowa,

Missouri, Kansas, Nebraska, and South Dakota, USA (Figure 2.1). These basins drain into 11 the Mississippi River and have variable land use within their respected watersheds, but

Topeka shiners only occupy low gradient, prairie, or agriculturally impacted headwater streams in these basins (Dodds et al. 2004). Forty-three sites, spanning eleven smaller Eight-

Digit Hydrologic Unit Code (HUC8) basins currently containing Topeka shiners, were included in this study (Table 2.1, Figure 2.1). In the Des Moines River Basin, the two smaller basins sampled were the North Raccoon and Boone River basins. There were multiple smaller basins sampled in the Missouri River Basin. In the Northern part of the Missouri

River Basin, the James, Vermilion, Big Sioux and Rock River basins were sampled. The lower Missouri sampling locations consisted of the Sugar Creek, Moniteau Creek, and

Kansas River basins. The one site in the Arkansas Basin came from the smaller Cottonwood

Basin. There is one sample site from a population in Willow Creek that has become extirpated in the wild and is currently maintained in artificial ponds.

Fish Sampling

Topeka shiners were collected from oxbows with a bag seine (10.7m X 1.8m,

6.35mm mesh or 16.8 X 1.8m, 6.35mm mesh) (Thompson and Berry 2009; Bakevich et al.

2013) whereas streams were sampled using a single upstream DC (150-200 volts, 4-5 amps) electrofishing pass with a barge electrofishing unit (Bakevich et al. 2013). If the stream was unable to be sampled with a barge shocker, sampling was carried out with a Smith-Root LR-

20 backpack electrofishing unit. Electrofishing was followed by a variable number of seine pulls with either a seine (4.5m X 1.8m, 6.35mm mesh) or a bag seine (10.7m X 1.8m,

6.35mm mesh or 16.8m X 1.8m, 6.35mm mesh) as conditions permitted. Topeka shiners were collected from research ponds located at the Kansas Biological Station in Lawrence,

Kansas using unbaited minnow traps. Up to 20 adult or juvenile Topeka shiners were 12 randomly selected at each site and a portion of the anal fin of those fish was non- destructively clipped and preserved in 95% ethanol for DNA extraction.

DNA Extraction and Genotyping

Topeka shiner DNA was extracted from fin samples using the Qiagen DNeasy Blood and Tissue Kit following the manufacturer’s protocol. A total of 13 microsatellite loci were tested for successful polymerase chain reaction (PCR) amplification and polymorphism across sites (Bessert and Orti 2003; Burridge and Gold 2003; Turner et al. 2004; Anderson and Sarver 2008; Blank et al. 2011). Nine of thirteen loci were polymorphic and retained for analysis (Appendix A). PCR amplification of the microsatellite loci was performed in 10 µL reactions. Approximately 2 ng of genomic DNA was used as the template for each reaction.

Microsatellite forward primers were modified with an M13 tag

(5’AGGGTTTTCCCAGTCACGACGTT 3’) attached to their 5’ ends (Schuelke 2000). A 10

µL reaction for each sample included: 6.6 µL of sterile H20, 1 µL Biolase Buffer (10x), 0.3

µL MgCl2, 0.8 µL dNTP, 0.1 µL modified forward primer, 0.1 µL reverse primer, 0.05 µL

M13 fluorescently labeled oligo, 0.05 µL Biolase Taq polymerase, and 1 µL template DNA.

Eppendorf Master Cycler thermal cyclers under consistent conditions (95°C/5min; [94°C/30 sec.; 62°C/60 sec.; 72°C/30 sec.] X 10 cycles; [94°C/30 sec.; *annealing temperature

(Appendix A) °C/60 sec.; 72°C/30 sec.; 72°C/20 min] X 25 cycles; 72°C/4 min). Negative controls were included with every reaction to detect potential contamination. PCR products were visualized on 1% agarose gels before being sent to the Iowa State DNA Facility where samples were genotyped using an Applied Biosystems 3730 DNA Analyzer. The raw allele calls were evaluated and binned using the software GeneMarker version 1.95 (Hulce et al.

2011). 13

Genetic Analysis

Microsatellite loci were evaluated for genotyping errors from null alleles, stuttering, and large allele dropout using MicroChecker 2.2.3 (van Oosterhout et al. 2004). Deviations from Hardy-Weinberg Equilibrium (HWE) were tested using the Markov Chain Monte Carlo approximation of Fisher’s exact test in the adgenet package in Program R (Jombart 2008).

Linkage disequilibrium was tested via GenePop 4.2 (Raymond and Rousset 1995).

Genetic metrics such as the number of alleles per locus (NA) and expected heterozygosity (HE), inbreeding coefficient (FIS), and private alleles (AP) were calculated for sites using GenAlEx 6.5 (Peakall and Smouse 2006, 2012). Allelic richness (AR) was calculated and rarefacted to the smallest sample size using the program HP-Rare 1.1

(Kalinowski 2005). Sites were tested for evidence of bottlenecks using program Bottleneck version 1.2.02 (Piry et al. 1999; Chiucchi and Gibbs 2010). The two-phase model of the

Wilcoxon signed rank test was performed with 1,000 iterations at Psum = 0.78 and variance of

3.1 based on values estimated from other studies (Peery et al. 2012). Because some areas of the Topeka shiner range are thought to have experienced fragmentation and isolation in recent years, sites were tested for mode shifts in allele frequency distribution, which would indicate populations have undergone a recent bottleneck (Luikart et al. 1998).

To estimate the number of genetic populations (K), STRUCTURE v2.3.4 was used

(Pritchard et al. 2000) employing the admixture model. A burn-in period of 100,000 was used for simulation of K = 1-43 with 20 replicates. The optimal value of K was determined using the ∆K method (Evanno et al. 2005). Values for ∆K were generated by Structure

Harvester (Earl 2012). Since STRUCTURE mainly detects the uppermost distinct structuring for the given dataset, additional runs of STRUCTURE were performed using the populations identified in the original run to determine if additional structure occurred within those areas. 14

The same burn-in period and length of 100,000 was used for simulation of K = 1-5 with 20 replicates for each additional run. Bar plots of STRUCTURE results were constructed using

POPHELPER (Francis 2016). To further analyze population structure, a neighbor joining tree based on Nei’s distance (DA) with 1,000 boot straps was constructed using Poptree2

(Takezaki et al. 2014). Pairwise FST values were generated to test genetic differentiation between sites and populations using the Program R package poppr (Kamvar et al. 2014;

Kamvar et al. 2015). An Analysis of Molecular Variation (AMOVA) was performed using

GenAlEx to determine the significance of the genetic differences between the populations determined from previous analyses.

Historical and contemporary rates of migration between populations were compared using the programs BAYESASS (Wilson and Rannala 2003) and MIGRATE (Beerli 2008).

Topeka shiners from the Moniteau Creek site were considered a separate population for this analysis due to large geographic distances between it and other sites that it had been grouped with in STRUCTURE. These programs both estimate a migration rate based on gene flow, but over a different timescale. BAYESASS estimates asymmetrical pairwise migration value

(m), which represents the fraction of migrants per generation in one population that is derived from a source population over the past few (< 5) generations. Because Topeka shiners have a generation time of 1-2 year (Kerns and Bonneau 2002), BAYESASS measures the migration rate over the past 5-10 years. Alternatively, MIGRATE uses a coalescent approach to estimate asymmetrical gene flow M (m/µ) between populations over the past 4Ne generations

(1,000s of years). To obtain estimates of m from MIGRATE to compare to BAYEASS, values of M were multiplied by the mutation rate µ = 5x10-4 (Garza and Williamson 2001).

These two estimates of migration were tested for a correlation using a mantel test in Program

R with 10,000 permutations.

In BAYESASS, migration values with a 95% confidence interval that contained values that were smaller than or equal to 0.02 (1 migrant per ~50+ years) were considered not to be biologically significant. A Bayesian deviance metric (Spiegelhalter 2002) and the program TRACER v1.7 were used to select the optimal parameters and determine convergence. Delta parameters were modified to achieve 40-60% acceptance. Run length and parameters were adjusted to ensure convergence. BAYSESASS reached convergence after five runs with a different initial seed using a run length of 1x108 iterations with sampling every 100 iterations and a burn-in period of 5x107.

Migration rates m derived from MIGRATE (Beerli and Felsenstein 2001; Beerli

2006) with a 95% confidence interval that encompasses values equal to or lower than 0.02 were not considered biologically significant. The starting parameters of θNE and M were estimated by an FST calculation and uniform priors (θNE: minimum = 0, maximum = 10, delta

= 1; M: minimum = 0, maximum = 100, delta = 2) with a variable mutation rate at each locus. Convergence was determined by assessing smoothness and distribution of θNE and M estimation histograms. The final parameters were set as a single long chain of 1x105 iterations with sampling every 300 iterations. A burn-in period of 3x104 with a static heating chain of 1.0, 1.5, 3.0, and 10,000.0 were used (Beerli 2009).

Results

There was some evidence for null alleles and also deviation from HWE across a few sites at different loci, but no evidence for linkage disequilibrium for any loci across the range. Out of a total of 388 loci by population comparisons, MICROCHECKER only 16 identified eight that were out of HWE, and only one locus (NTB42) was out of HWE at more than one site. Therefore, all loci and sites were retained for analysis.

Low to moderate levels of genetic diversity were estimated within sites, with metrics indicating that this level of diversity is similar across most sites (Table.2.2). The number of alleles per locus was variable between loci. Locus NTA22 had the lowest number of alleles with three and locus NTB8 had the most with 30 alleles. The other 10 loci had 7-10 alleles.

The average number of alleles per locus ranged from 2.111 to 4.444 (mean = 3.560) whereas allelic richness ranged from 1.800 to 2.870 (mean = 2.560; Table 2.2). The Cottonwood

River Basin site had the lowest values for all metrics of genetic diversity and the highest values came from sites on Prairie Creek in the Boone River Basin. However, number of alleles per locus and allelic richness were fairly homogenous across the other basins (Table

2.2.). Values of FIS ranged from -0.199 indicating outbreeding, to 0.132 indicating inbreeding

(Table 2.2). Although most sites had no private alleles, four sites had one private allele

(Mound1, Mound2, Eagle1, and Prairie7), two sites had three private alleles (Sugar1 and

Moniteau1), and one site (SDRock1) had eight private alleles (Table 2.2).

There was evidence of genetic bottlenecks in five of the 44 sites (Table 2.3). In the

Boone River Basin, only one site (Prairie10) showed evidence for a bottleneck while all the other 18 sites in that basin, including nearby stream and oxbow sites, showed no evidence of a bottleneck. In the Raccoon River Basin, one site on East Buttrick Creek was significant for a mode shift, indicating a more recent bottleneck. The three other sites with evidence for a bottleneck all came from Kansas. Willow1, a wild population that had been transplanted into a research pond, showed evidence of a bottleneck under the two-phased model. The other research pond site was Deep2 that showed evidence for a more recent bottleneck as indicated 17 by a under the two-phased model. Finally, a pond in the Cottonwood River Basin showed evidence of a bottleneck through both the two-phase model and a mode shift.

Analysis of population structure across the range of Topeka shiners revealed multiple populations were present. The initial STRUCTURE analysis indicated three populations

(Figures 2.2, 2.3). Sites clustered into larger basins of the Des Moines River and the Missouri

River basins, with a third grouping in Kansas that encompassed the Kansas River and

Cottonwood River basins as well as Sugar Creek in Missouri (Figures 2.2, 2.3). Given large geographic distances between sites, further genetic structuring was possible within these large basins. Therefore, these three populations were analyzed in STRUCTURE separately.

The analysis of the Kansas and Sugar Creek group indicated it consisted of two populations, one group consisting of three sites in the Kansas River and a second consisting of the Cottonwood River Basin and Sugar Creek Basin (Figures 2.2, 2.3). Because of the geographic and genetic differences between the sites in these grouping, STRUCTURE analysis was repeated again separately on the newly identified groups. The Kansas River group split again with Willow Creek Basin forming one population and the two sites from

Deep Creek in the Kansas River Basin forming the other (Figures 2.2, 2.3). The Cottonwood

River Drainage and the Sugar Creek Drainage were also identified as separate groups

(Figures 2.2, 2.3). In contrast to the sites in Kansas, sites in the Missouri River Basin were grouped by STRUCTURE in a single population. This population includes basins from South

Dakota and Minnesota, all of which flow into the Missouri River and are located in a cluster in the northern part of the range. An additional site located in the state of Missouri, Moniteau

Creek that flows into the Missouri River, also groups into this population. Topeka shiners in 18 the Des Moines River Basin were separated into the North Raccoon and the Boone river basins (Figures 2.2, 2.3).

In total, STRUCTURE revealed seven genetic populations (Figure 2.4). Phylogenetic clustering through the construction of a neighbor joining tree revealed similar population structuring (Figure 2.5) to the STRUCTURE analysis. The neighbor-joining tree is in agreement with the STRUCTURE results, except for a lack of structuring in the Kansas and

Willow Creek Basins (Figures 2.3, 2.4). The test for genetic variation between populations indicated that the seven proposed populations were genetically distinct (AMOVA: P ≤

0.001). Approximately 14% of the variance in allele frequencies was attributed to differences between populations, 17% was attributed to difference within the populations, and 69% of the differences was attributed within individuals.

There was a wide range in pairwise FST values between sites, indicating varying levels of population structure across the study area. Values ranged from 0.008 between adjacent sites in the Boone River Basin to 0.309 between the Willow Creek research pond site in

Kansas and a site from an oxbow on the Rock River in Minnesota. Within basins, most sites showed little differentiation <0.050, while differences between sites in separate basins generally had FST values of 0.100 - 0.200, indicating moderate to great differentiation, with some paired sites having FST values >0.300. Pairwise FST indicated significant genetic differentiation among all populations (Table 2.4), corroborating the populations from both the STRUCTURE and neighbor joining trees. The only exception was the Boone River Basin and North Raccoon River Basin having a relatively low but significant FST of 0.031 (Table

2.4). 19

Historical and contemporary migration were compared to test for a change in population connectivity after human settlement. Historical migration was not correlated with contemporary migration (r = 0.26, P = 0.24), indicating a difference between historical and contemporary gene flow. There were 12 instances of biologically significant (one migrant or more per 50 years) historical migration rates determined by MIGRATE (Table 2.5, Figure

2.6). A few populations had bidirectional migration (Figure 2.6). In contrast, estimates of contemporary gene flow indicated only two instances, Moniteau Creek Basin to the Missouri

River Basin and the Willow Creek Basin into the Kansas River Basin (Table 2.6, Figure 2.7).

In these instances, migration between populations was biologically significant at the level of one or more migrant per 50 years.

Discussion

Species that are considered threatened or endangered often exhibit low genetic diversity, especially when populations are small or disconnected (DeHaan et al. 2012).

Higher levels of genetic diversity allow species to adapt to environmental changes and reduces their susceptibility to catastrophic extinction events (Quattro and Vrijenhoek 1989;

Hedrick and Hurt 2012). In this study, the number of alleles per locus (NA = 3.585) and allelic richness (mean AR = 2.560) in the Topeka shiner were low compared to other cyprinids. For instance, the Central stoneroller (Campostoma anomalum), a widespread stream fish (NA = 18.4), and the Oregon chub (Oregonichthys crameri), a threatened species that also inhabit similar off-channel habitats as Topeka shiners (NA = 3.44-12.22), both have high genetic diversity (Blum et al. 2012; DeHaan et al. 2012). However, the endangered

Cape Fear shiner (Notropis altipinnis) had similar levels of genetic diversity (NA = 5.09 -

5.27) to Topeka shiners (Burridge and Gold 2003). Expected heterozygosity, however, indicated moderate Topeka shiner genetic diversity (average HE = 0.413) in almost all sites. It 20 is important to note that historical levels of Topeka shiner genetic diversity are unknown, therefore it is not known whether current genetic diversity is similar to genetic diversity before distribution declines.

The site that had the lowest Topeka shiner genetic diversity was in the Cottonwood

River Basin, with the lowest values for average alleles per locus, allelic richness, expected heterozygosity, and private alleles but did not suffer from inbreeding. However, this was also the only site where a bottleneck was detected through both the two-phase model and a mode shift, suggesting that this population was founded from a few individuals. Bottlenecks occur when populations experience a rapid decline in numbers and are usually caused by some stochastic or founding event (Hundertmark and Van Daele 2010). This population was located in a constructed pond in Tallgrass National Prairie Preserve, KS that drains into a small tributary to Fox Creek. How Topeka shiners colonized this pond is unknown, but

Topeka shiners are known to be present downstream in Fox Creek (Portofee et al. 2018), however there is no upstream passage to the pond. Colonizing populations can have reduced genetic variation and fitness (Szucs et al. 2017). Unfortunately, this was the only sample site in the Cottonwood River. Thus, it remains unknown if this low diversity is restricted to

Topeka shiners in this site or characteristic of the larger basin. Either way, low diversity in this site puts it at greater risk of extirpation than other genetic populations in our analysis.

Although the diversity of all other sites was higher than Cottonwood Basin, there was other incidents of bottlenecks across the range. One site that experienced a bottleneck occurred in an oxbow (Eastbutt1). Although oxbows can be ephemeral habitats with high rates of recolonization (Fischer et al. 2018), bottlenecks did not occur in any of the other twenty-one oxbow sites, making this a unique case. Another bottleneck occurred in a stream 21 site (Prairie10). It is unclear why the stream site in Prairie Creek showed evidence for a bottleneck when no nearby oxbows or stream sites, which had great genetic similarity to this site, showed any signs of a bottleneck. The other two bottlenecked sites both came from the research ponds in Kansas. A total of 93 Topeka shiners were taken from Willow Creek and were distributed through six ponds in 2002 and these populations have grown to hundreds of individuals per pond (Campbell et al. 2016). The other captive population in these ponds are from Deep Creek in the Kansas River Basin and were introduced in 2000. Captive populations of endangered species often start with a few individuals from one genetic lineage, which can often lead to bottlenecks (Leberg and Firmin 2008).

There was significant genetic variation across the range and seven distinct genetic populations were identified. The Cottonwood River, Willow Creek, Kansas River, Sugar

Creek, North Raccoon River, and Boone River populations exist as isolated basins with no neighboring populations of Topeka shiners. This lack of gene flow led to population differentiation in the majority of the range. A similar pattern of genetic isolation was found in

Oregon chub populations, which have also become fragmented due to human activities

(DeHaan et al. 2012).

The Missouri River population is the most extensive and is comprised of five HUC8 basins.

The Missouri River population contrasts with the small isolated populations observed elsewhere across the range as four of the five HUC8s comprising this population are neighboring basins in Minnesota and South Dakota and drain into the Missouri River. This northern part of the distribution is the only area of the range where Topeka shiners have not experienced drastic declines and large numbers of Topeka shiners still exist (USFWS 2018).

Sites within this population are genetically homogeneous indicating there is gene flow 22 between them. This larger population appears to be the most demographically stable (Pasbrig and Lucchesi 2012; Nagle and Larson 2014). Moniteau Creek in Missouri, located 1,091 river kilometers downstream from the Missouri River sites, also grouped into this population.

There are other genetic populations closer to Moniteau Creek and current migration of individuals is unlikely given that estimates of cyprinid movements generally are only a few kilometers (Johnston 2003; Goforth and Foltz 2006). It is possible that this basin was once part of a larger, more contiguous population encompassing tributaries of the Missouri River.

Populations formerly inhabiting intermediate tributaries of the Missouri River in Iowa,

Nebraska, and Missouri that may have historically connected the Missouri River and

Moniteau Creek Sites are now extirpated (USWFS 2009). Cyprinids are often transported to new areas as baitfish (Drake and Madrak 2014), and it is possible, that this happened to

Topeka shiners in this case. Genetic similarity between Moniteau Creek and the Missouri

River population is unexpected and worth further investigation. The Missouri River population is located within the Northern Glaciated Plains and Western Cornbelt Plains and the Moniteau Creek population is within the Ozark Highlands (EPA 2013). Thus, Topeka shiners likely are locally adapted to habitats in these different ecoregions (Lean et al. 2017).

Moniteau Creek is likely a distinct population from the Missouri River due to geographic isolation, resulting in a total of eight Topeka shiner populations across the range.

An earlier range-wide analysis of Topeka shiner populations by using mtDNA identified three populations (Michels 2000). Population structuring was found to be finer in the current study, resulting in seven genetic populations. Differences between the markers used can explain some of the differences between the two studies. Microsatellite markers evolve more quickly than mtDNA sequences (Zhang and Hewitt 2003; Wan et al. 2004), 23 resulting in finer resolution in population structure. However, there are two areas where this study conflicts with the previous one. Mitochondrial DNA analysis grouped all sites from

Iowa, Minnesota, and South Dakota together into the Upper Missouri River population.

There was a historical connection between the headwaters of these basins (Pflieger 1971) that may be the source of genetic similarity in the mtDNA sequences. However, measures of genetic structure and gene flow indicate that these populations are currently genetically distinct and isolated from each other. Another difference between the two studies is the placement of Moniteau Creek, that mtDNA grouped into the Lower Missouri River with sites from Kansas and Missouri. This study indicates that Moniteau Creek has genetic similarity with the Upper Missouri River sites. Individual genetic markers can have different histories, and therefore mtDNA does not always agree with nuclear data (Wan et al. 2004) However, this study does generally corroborate the genetic structure from the previous mtDNA study with finer resolution.

Results presented here indicated a significant loss of gene flow across the range of the

Topeka shiner. Topeka shiners are thought to have originated in the ancestral Iowa River prior to the Pleistocene and spread to other basins as historical river connections shifted due to receding glaciers (Bailey and Allum 1962; Pflieger 1971). Estimates of historical migration (~1000s years ago) between the Des Moines Basins (North Raccoon and Boone

River) and sites located in the Missouri River tributaries support these suspected connections.

Historical levels of connectivity between these populations was generally high, suggesting that historical distribution of Topeka shiners may have been much more widespread, as indicated by previous historical distribution estimates (Bailey and Allum 1962; Pflieger

1971). However, current estimates of migration revealed almost no gene flow between these 24 populations. There was only evidence for contemporary migration between two sets of populations, which contrasts with the general pattern of isolation across the range. The first set is from Willow Creek to the Kansas River Basin and is likely not the result of gene flow in wild populations. These two captive populations both exist in research ponds located in the same facility (Campbell et al. 2016) where migration is likely a result of human assisted movement of fishes between ponds. The second instance is from Moniteau Creek to the

Missouri River. Given there are no known intermediate sources of Topeka shiners between those areas separated by 1,091 river km, migration is extremely unlikely (Johnston 2003;

Goforth and Foltz 2006). Other basins, such as the North Raccoon, Boone, and Arkansas river basins were separated from the Missouri River due to post-glacial changes to river connections (Pflieger 1971). However, many intermediate Topeka shiners populations have become extirpated since agricultural conversion leaving Topeka shiner populations isolated throughout the range (USFWS 2018). Therefore, the majority of the reduction of gene flow likely stems from this fragmentation of distribution.

This study indicates that Topeka shiners generally exist as small genetically distinct populations with no genetic connectivity between them. Management units (MUs) are defined as demographically independent populations (Palsboll et al. 2007; Ralls et al. 2018).

Therefore, the eight genetically distinct populations established in this study could be considered separate MUs. Low genetic diversity is a concern in small or declining populations because it reduces the ability of a population to adapt to changing environments and exposes them to negative effects of genetic drift (Allendorf 1986; Reed and Frankham

2003). Genetic diversity was low to moderate in almost all sites and if populations remain isolated and continue to decline in size, genetic diversity would be expected to decrease. 25

Given the level of isolation and genetic differences across the range, a loss of any population may result in a significant loss of overall genetic diversity. Therefore, MUs that are currently declining are candidates for priority consideration for restoration.

Once widely distributed and well interconnected throughout the Midwest, landscape, floodplain, and stream alteration has reduced the distribution of Topeka shiners to multiple fragmented and genetically isolated populations with moderate genetic diversity. Although full reestablishment of historical connectivity may not be feasible due to large geographic distances and anthropogenic barriers between populations, increasing the amount and distribution of suitable habitat to maintain current populations and allow natural reestablishment of Topeka shiners is a worthy goal. The maintenance of genetic diversity and therefore adaptive potential is crucial to the survival of Topeka shiners in a changing environment.

Literature Cited

Allendorf, F. W. 1986. Genetic drift and the loss of alleles versus heterozygosity. Zoo Biology 5:181-190.

Anderson, C. M., and S. K. Sarver. 2008. Development of polymorphic microsatellite loci for the endangered Topeka shiner, Notropis topeka. Molecular Ecology 8:311-313.

Bakevich, B. D., C. L. Pierce, and M. C. Quist. 2015. Status of the Topeka shiner in west- central Iowa. American Midland Naturalist 174:350-358.

Bailey, R. M., and M. O. Allum. 1962. Fishes of South Dakota Ann Arbor: Museum of Zoology, University of Michigan. Ann Arbor, MI, USA. 140p.

Beerli, P., J. Felsenstein. 2001. Maximum likelihood estimation of a migration matrix and effective population sizes in n subpopulations by using a coalescent approach. Proceedings of the National Academy of Sciences of the United States of America 98:4563-4568.

Beerli, P. 2006. Comparison of Bayesian and maximum likelihood inference of population genetic parameters. Bioinformatics 22:341-345.

Beerli, P. 2008. Migrate version 3.0: a maximum likelihood and Bayesian estimator of gene flow using the coalescent. Distributed over the internet at http://popgen.scs. edu/migrate. Html.

Beerli, P. 2009. How to use migrate or why are Markov Chain Monte Carlo programs difficult to use? Conservation Biology 17:42-79.

Berg, J. A., T. A. Petersen, Y. Anderson, and R. Baker. 2004. Hydrogeology of the Rock River Watershed, Minnesota, and associated off-channel habitats of the Topeka shiner. Minnesota Department of Natural Resources, Divisions of Waters and Ecological Services, St Paul, MN, USA. 14 p.

Bessert, M. L., and G. Orti. 2003. Microsatellite loci for paternity analysis in the Fathead minnow, Pimephales promelas (Teleostei : Cyprinidae). Molecular Ecology Notes 3:532-534.

Bergstrom, D., T. Holtsford, and J. Koppelman. 1999. Genetic structure of Notropis topeka in Missouri. Appendix III in “An action plan for the Topeka shiner (Notropis topeka) in Missouri.” Missouri Department of Conservation, MO, USA. 39 p.

Blank, M., B. Bramblett, S. Kalinowski, J. Cahoon, and K. Nixon. 2011. Impacts of barriers on Topeka shiner populations. Report SD2006-07-F to South Dakota Department of Transportation, Pierre SD, USA. 88 p.

Blum, M. J., M. J. Bagley, D. M. Walters, S. A. Jackson, F. B. Daniel, and D. J. Chaloud. 2012. Genetic diversity and species diversity of stream fishes covary across a land‐ use gradient. Oecologia 168:83-95.

Burridge, C. P., and J. R. Gold. 2003. Conservation genetic studies of the endangered Cape Fear shiner Notropis mekistocholas (Teleostei: Cyprinidae). Conservation Genetics 4:219-225.

Campbell, S. W., C. S. Szuwalski, V. M. Tabor, and F. DeNoyelles. 2016. Challenges to reintroduction of a captive population of Topeka shiner (Notropis topeka) into former habitats in Kansas. Transactions of the Kansas Academy of Science 119:83-92.

Ceas, P. A., and Y. C. Anderson. 2004. Results of a pilot monitoring project for Topeka shiners in Southwestern Minnesota. Minnesota Department of Natural Resources, St. Paul, MN, USA. 10 p.

Chiucchi, J. E., and L. L. Gibbs. 2010. Similarity of contemporary and historical gene flow among highly fragmented populations of an endangered rattlesnake. Molecular Ecology 19:5345-5358.

Clark, S. J. 2000. Relationship of Topeka shiner distribution to geographic features of the Des Moines Lobe in Iowa. Master’s Thesis, Iowa State University, Ames IA, USA. 54 p.

Cross, F. B., and J. T. Collins. 1975. Fishes in Kansas. University of Kansas Natural History Museum, Public Education Series Number 3:1-189.

Cunningham, G. 2017. Topeka shiner monitoring in Minnesota: 2017. Minnesota Department of Natural Resources, St. Paul, MN, USA. 13p.

DeHaan, P. W., P. D. Scheerer, R. Rhew, and W. R. Ardren. 2012. Analyses of genetic variation in populations of Oregon chub, a threatened floodplain minnow in a highly altered environment, Transactions of the American Fisheries Society 141:533-549.

Dodds, W. K., K. Gido, M. R. Whiles, K. M. Fritz, and W. J. Matthews. 2004. Life on the edge: The ecology of Great Plains prairie streams, BioScience 54:205-216.

Drake, D. A., and N. E. Mandrak. 2014. Bycatch, bait, anglers, and roads: quantifying vector activity and propagule introduction risk across lake ecosystems. Ecological Applications 24: 877-894.

Earl, D. A. 2012. STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conservation Genetics Resources 4:359-361.

Evanno, G., S. Regnaut, and J. Goudet. 2005. Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology 14:2611- 2620.

Fischer, J. R., B. D. Bakevich, C. P., Shea, C. L. Pierce, and M. C. Quist. 2018. Floods, drying, habitat connectivity, and fish occupancy dynamics in restored and unrestored oxbows of West Central Iowa, USA. Aquatic Conservation: Marine and Freshwater Ecosystems 28:630-640.

Francis, R. M. 2016. POPHELPER: An R package and web app to analyse and visualise population structure. Molecular Ecology Resources 17:27-32.

Garza, J. C., and E. G. Williamson. 2001. Detection of reduction in population size using data from microsatellite loci. Molecular Ecology 10:305-318.

Gerkin, J. E., and C. P. Paukert. 2013. Fish assemblage and habitat factors associated with the distribution of Topeka shiner (Notropis topeka) in Kansas streams. Journal of Freshwater Ecology 28:503-516.

Gilbert, C. H. 1884. Notes on the fishes of Kansas. Bulletin of the Washburn College Laboratory of Natural History 1:1-16.

Goforth, R. R., and J. W. Foltz. 1998. Movements of the Yellowfin shiner, Notropis lutipinnis. Ecology of Freshwater Fish 7:49-55.

Hedrick, P. W., and C. R. Hurt. 2012. Conservation genetics and evolution in an endangered species: research in Sonoran Topminnows. Ecological Applications 5:806-819.

Hulce, D., X. Li, T. Snyder-Leiby, and C. S. Liu. 2011. GeneMarker® genotyping software: tools to increase the statistical power of DNA fragment analysis. Journal of Biomolecular Techniques S35-S36.

Hundertmark, K. J., and L. J. Van Daele. Founder effect and bottleneck signatures in an introduced, insular population of elk. Conservation Genetics 11:139-147.

Jackson, N. D., and L. Fahrig. 2016. Habitat amount, not habitat configuration, best predicts population genetic structure in fragmented landscapes. Landscape Ecology 31:951- 968.

Johnston, C. E. 2000. Movement patterns of imperiled Blue shiners (Pisces: Cyprinidae) among habitat patches. Ecology of Freshwater Fish 9:170-176.

Jombart, T. 2008. adegenet: a R package for the multivariate analysis of genetic markers. Bioinformatics 24:1403-1405.

Kalinowski, S. T. 2005. HP-Rare: a computer program for performing rarefaction on measures of allelic diversity. Molecular Ecology Notes 5:187-189.

Kamvar, Z. N., J. F. Tabima and N. J. Grünwald. 2014. Poppr: an R package for genetic analysis of populations with clonal, partially clonal, and/or sexual reproduction. PeerJ 2:e281.

Kamvar, Z. N., J. C. Brooks, and N. J. Grünwald. 2015. Novel R tools for analysis of genome-wide population genetic data with emphasis on clonality. Frontiers in Genetics 6:208.

Kerns, H. A., and J. L. Bonneau. 2002. Aspects of the life history and feeding habits of the Topeka shiner (Notropis topeka) in Kansas Transactions of the Kansas Academy of Science 105:125-142.

Leberg, P. L., and B. D. Firmin. 2008. Role of inbreeding depression and purging in captive breeding and restoration programmes. Molecular Ecology 17:334-343.

Lean, J., M. P. Hammer, P. J. Unmack, M. Adams, and L. B. Beheregaray. 2017. Landscape genetics informs mesohabitat preference and conservation priorities for a surrogate indicator species in a highly fragmented river system. Heredity 118:374-384.

Luikart, G., F. W. Allendorf, B. Sherwin, and J. M. Cornuet. 1998. Distortion of allele frequency distributions provides a test for recent population bottlenecks. Journal of Heredity 12:238-247.

Meek, S. E. 1894. Notes in the fishes of western Iowa and eastern Nebraska. Bulletin of the US Fish Commission 14:133-138.

Michels, A. 2000. Population genetic structure and phylogeography of the endangered Topeka shiner (Notropis topeka) and the abundant Sand shiner (Notropis ludibundus) using mitochondrial DNA sequence. Ph.D. Dissertation, University of Kansas. Lawrence KS, USA. 68 p.

Minckley, W. L., and F. B. Cross. 1959. Distribution, habitat, and abundance of the Topeka shiner, Notropis topeka (Gilbert), in Kansas. American Midland Naturalist 61:210- 217.

Missouri Department of Conservation (MDC). 2010. A ten year strategic plan for the recovery of the Topeka shiner in Missouri. MO, USA. 63p.

Nagle, B. C., and K. A. Larson. 2014. Topeka shiner monitoring in Minnesota: 2014. Minnesota Department of Natural Resources, St. Paul, MN, USA. 17p.

Palsboll, P. J., M. Bérubé, and F. W. Allendorf. 2007. Identification of management units using population genetic data. Trends in Ecology & Evolution 22:11-16.

Pasbrig, C. A., and D. O. Lucchesi. 2012. Topeka shiner (Notropis topeka) monitoring in eastern South Dakota streams (2010-2012). South Dakota Department of Game, Fish and Parks, Pierre, SD, USA. 87p.

Peakall, R. D., and P. E. Smouse. 2006. GENALEX 6: genetic analysis in Excel. Population genetic software for teaching and research. Molecular Ecology Notes 6:288-295.

Peakall, R., and P. E. Smouse 2012. GenAlEx 6.5: genetic analysis in Excel. Population genetic software for teaching and research-an update. Bioinformatics 28:2537-2539.

Peery, M. Z., R. Kirby, B. N. Reid, R. Stoelting E. Doucet-Beer, S. Robinson, C. Vasquez- Carrillo, J. N. Pauli, and P. J. Palsboll. 2012. Reliability of genetic bottleneck tests for detecting recent population declines. Molecular Ecology 21:3403-3418.

Pflieger, W. L. 1971. A distributional study of Missouri fishes. University of Kansas Publication, Museum of Natural History 20:225-570.

Pflieger, W. L. 1997. The fishes of Missouri, Missouri Department of Conservation, Jefferson City, MO, USA. 372p.

Piry, S., G. Luikart, and J. M. Cornuet. 1999. BOTTLENECK: a computer program for detecting recent reductions in the effective population size using allele frequency data. Journal of Heredity 90:502-503.

Pritchard, J. K., M. Stephens, and P. Donnelly. 2000. Inference of population structure using multilocus genotype data. Genetics 155:945-959.

Portofee, M. C., D. R. Edds, K. J. Hase, and D. L. McCullough 2018. Fishes of the Tallgrass Prairie National Preserve, Chase County, Kansas. Transactions of the Kansas Academy of Science 121:1-2.

Quattro, J. M., and R. C. Vrijenhoek. 1989. Fitness differences among remnant populations of the endangered Sonoran Topminnow. Science 245:976-978.

Ralls, K., J. D. Ballou, M. R. Dudash, M. D. Eldridge, C. B. Fenster, R. C. Lacy, P. Sunnucks, and R. Frankham. 2018. Call for a paradigm shift in the genetic management of fragmented populations. Conservation Letters 11p.

Raymond, M., and F. Rousset. 1995. GENEPOP (version 1.2): a population genetic software for exact tests and ecumenicism. Journal of Heredity 86:248-249.

Reed, D. H., and R. Frankham. 2003. Correlation between fitness and genetic diversity. Conservation Biology 17:230-237.

Rybicki, J., and I. Hanski. 2013. Species–area relationships and extinctions caused by habitat loss and fragmentation. Ecology Letters 16:27-38.

Saccheri, I., M. Kuussaari, M. Kankare, P. Vikman, W. Fortelius, and I. Hanski. 1998. Inbreeding and extinction in a butterfly metapopulation. Nature 392:491-494.

Schuelke, M. 2000. An economic method for the fluorescent labeling of PCR fragments: a poor man’s approach to genotyping for research and high-throughput diagnostics. Nature Biotechnology 18:233-234.

Spiegelhalter, D. J. 2002. Bayesian measures of model complexity and fit. Journal of the Royal Statistical Society. Series B, Statistical methodology 64:583-639. 31

Simpson, N. T., C. L. Pierce, K. J., Roe, and M. J. Weber. 2017. Boone River watershed stream fish and habitat monitoring, IA. 2017 Annual Progress Report. Iowa State University. Ames, IA, USA.

Simpson, N. T., A. P. Bybel, C. L. Pierce, K. J. Roe and M. J. Weber. (In Review). Abiotic and fish assemblage associations of the endangered Topeka shiner and fish species of greatest conservation need in streams of Iowa and Minnesota, USA. Aquatic Conservation: Marine and Freshwater Ecosystems.

Storfer, A. 1999. Gene flow and endangered species translocations: a topic revisited. Biological Conservation 87:173-180.

Szűcs, M., B. A. Melbourne, T. Tuff, C. Weiss‐Lehman, R. A. Hufbauer, and A. Angert. 2017. Genetic and demographic founder effects have long‐term fitness consequences for colonising populations. Ecology Letters 20:436-444.

Tabor, V. M. 1998. Final rule to list the Topeka shiner as endangered. Federal Register 63:69008-69021.

Takezaki, N., M. Nei, and K. Tamura. 2014. Web version of POPTREE for constructing population trees from allele frequency data and computing other population statistics Molecular Biology and Evolution 31:1622-1624.

Thomson, S. K. and C. R. Berry. 2009. Stream fishes inhabit livestock watering ponds (dugouts) near six mile creek, Brookings County, South Dakota. North American Journal of Fisheries Management 88:127-138.

U.S. Environmental Protection Agency (EPA) 2013. Level III ecoregions of the continental United States: Corvallis, Oregon, U.S. EPA – National Health and Environmental Effects Research Laboratory, map scale 1:7,500,000, https://www.epa.gov/eco- research/level-iii-and-iv-ecoregions-continental-united-states

U.S. Fish and Wildlife Service (USFWS). 1998. Endangered and threatened wildlife and plants; final rule to list the Topeka shiner as endangered. Federal Register 65:57114- 57119.

U.S. Fish and Wildlife Service (USWFS). 2009. Topeka shiner 5-year review, summary and evaluation. Kansas Ecological Services, Manhattan, KS, USA. 43p. 32

U.S. Fish and Wildlife Service (USFWS). 2018. Final Topeka shiner Species Status Assessment. Species Status Assessment Reports. South Dakota Ecological Services Office. van Oosterhout, C., W. F. Hutchinson, D. P. Wills, and P. Shipley. 2004. Micro-Checker: software for identifying and correcting genotyping errors in microsatellite data. Molecular Ecology Notes 4:535-538.

Wall, S. S., and S. K. Thomson. 2007. Topeka shiner (Notropis topeka) monitoring in eastern South Dakota streams (2004-2006). Report submitted to South Dakota Game, Fish and Parks, Pierre, SD, USA.

Wan, Q., H. Wu, T. Fujihara, and S. Fang. 2004. Which genetic marker for which conservation genetics issue? Electrophoresis 25:2165-2176.

Wilson, G. A., and B. Rannala. 2003. Bayesian inference of recent migration rates using multilocus genotypes. Genetics 163:1177-119.

Zambory, C. L, A. P. Bybel, C. L. Pierce, K. J. Roe and M. J. Weber. 2017. Habitat improvement projects for stream and oxbow fish of greatest conservation need. 2017 annual progress report. Iowa State University, Ames IA, USA. 41p.

Zhang, D., and G. M. Hewitt. 2003. Nuclear DNA analyses in genetic studies of populations: practice, problems and prospects. Molecular Ecology 12:563-584.

Tables

Table 2.1. Site abbreviation, stream, basin, state, sample size (number of individual samples collected), and habitat type for all sites sampled for Topeka shiners.

Site Abbreviation Stream Basin State Sample Size Habitat Type Willow1 Willow Creek Willow Kansas 18 Research Pond Deep1 Deep Creek Kansas Kansas 12 In-stream Deep2 Deep Creek Kansas Kansas 18 Research Pond Cottonwood1 Trib. to Fox Creek Cottonwood Kansas 20 Pond Sugar1 Sugar Creek Sugar Missouri 20 In-stream Moniteau1 Moniteau Creek Moniteau Missouri 20 In-stream Camp1 Camp Creek Vermillion South Dakota 20 In-stream

SDRock1 Rock Creek James South Dakota 20 In-stream Beaver1 Beaver Creek Big Sioux Minnesota 20 Restored Oxbow Elk1 Elk Creek Rock Minnesota 18 Restored Oxbow Rock2 Rock River Rock Iowa 19 Oxbow Rocktrib1 Trib. To Rock River Rock Minnesota 13 In-stream Mound1 Mound Creek Rock Minnesota 25 Restored Oxbow Mound2 Mound Creek Rock Minnesota 15 In-stream Poplar1 Poplar Creek Rock Minnesota 19 Restored Oxbow Rocktrib2 Trib. to Rock River Rock Minnesota 20 In-stream Rocktrib3 Trib. to Rock River Rock Minnesota 18 Restored Oxbow Rocktrib4 Trib. to Rock River Rock Minnesota 19 Restored Oxbow Rocktrib5 Trib. to Rock River Rock Minnesota 15 In-stream Rocktrib6 Trib. to Rock River Rock Minnesota 20 Restored Oxbow Rocktrib7 Trib. to Rock River Rock Minnesota 19 In-stream Cedar1 Cedar Creek North Raccoon Iowa 15 Restored Oxbow Cedar2 Cedar Creek North Raccoon Iowa 19 In-stream

Table 2.1 (continued)

Cedar3 Cedar Creek North Raccoon Iowa 20 Restored Oxbow Eastbutt1 East Buttrick Creek North Raccoon Iowa 9 Oxbow Westbutt1 West Buttrick Creek North Raccoon Iowa 18 Restored Oxbow Eagle1 Eagle Creek Boone Iowa 20 Oxbow Eagle2 Eagle Creek Boone Iowa 13 Oxbow Boone1 Boone River Boone Iowa 18 Oxbow Boone2 Boone River Boone Iowa 16 In-stream Prairie1 Prairie Creek Boone Iowa 20 Oxbow Prairie2 Prairie Creek Boone Iowa 20 In-stream Prairie3 Prairie Creek Boone Iowa 19 In-stream Prairie4 Prairie Creek Boone Iowa 17 Oxbow Prairie5 Prairie Creek Boone Iowa 20 In-stream

Prairie6 Prairie Creek Boone Iowa 20 In-stream

Prairie7 Prairie Creek Boone Iowa 17 Oxbow Prairie8 Prairie Creek Boone Iowa 16 Restored Oxbow Prairie9 Prairie Creek Boone Iowa 18 In-stream Praire10 Prairie Creek Boone Iowa 16 In-stream Praire11 Prairie Creek Boone Iowa 20 Oxbow Prairie12 Prairie Creek Boone Iowa 17 Restored Oxbow Prairie13 Prairie Creek Boone Iowa 16 In-stream Prairie14 Prairie Creek Boone Iowa 19 Restored Oxbow Prairie15 Prairie Creek Boone Iowa 18 In-stream

Table 2.2 Site abbreviation, basin, number of alleles per locus (NA), allelic richness (AR), expected heterozygosity (HE), inbreeding coefficient (FIS), and private alleles (AP) across all

Topeka shiner loci.

Site Basin NA AR HE FIS AP

Willow1 Willow 3 2.4 0.419 0.023 0 Deep1 Kansas 3.333 2.62 0.43 0.11 0 Deep2 Kansas 3 2.42 0.385 0.005 0 Cottonwood1 Cottonwood 2.111 1.8 0.268 -0.004 0 Sugar1 Sugar 3.333 2.57 0.431 0.091 3 Moniteau1 Moniteau 3.333 2.5 0.399 -0.044 3 Camp1 Vermillion 3 2.27 0.335 -0.099 8 SDRock1 James 4.333 2.69 0.434 0.069 0 Beaver1 Big Sioux 3.222 2.36 0.378 -0.014 0 Elk1 Rock 3.444 2.45 0.353 0.005 0 Rock2 Rock 3.333 2.3 0.337 -0.095 0 Rocktrib1 Rock 3.556 2.52 0.343 -0.008 0 Mound1 Rock 3.444 2.3 0.352 -0.199 1 Mound2 Rock 2.889 2.24 0.319 -0.075 1 Poplar1 Rock 4.000 2.39 0.405 0.113 0 Rocktrib2 Rock 3.778 2.53 0.382 0.083 0 Rocktrib3 Rock 3.556 2.39 0.37 0.063 0 Rocktrib4 Rock 3.444 2.58 0.407 0.139 0 Rocktrib5 Rock 3.333 2.41 0.352 -0.086 0 Rocktrib6 Rock 4 2.68 0.407 0 0 Rocktrib7 Rock 3.333 2.21 0.314 -0.023 0 Cedar1 North Raccoon 3.556 2.61 0.428 0.018 0 Cedar2 North Raccoon 4 2.66 0.422 0.044 0 Cedar3 North Raccoon 3.778 2.68 0.443 -0.033 0 Eastbutt1 North Raccoon 3.333 2.7 0.407 -0.077 0 Westbutt1 North Raccoon 3.333 2.31 0.347 -0.068 0 Eagle1 Boone 3.556 2.56 0.416 -0.053 1 Eagle2 Boone 3 2.28 0.398 -0.075 0 Boone1 Boone 3.778 2.67 0.44 0.132 0 Boone2 Boone 3.667 2.69 0.429 -0.067 0 Prairie1 Boone 4.444 2.85 0.479 -0.051 0 Prairie2 Boone 3.667 2.68 0.482 -0.054 0 Prairie3 Boone 4.111 2.85 0.488 0.025 0 Prairie4 Boone 3.778 2.79 0.482 0.067 0 36

Table 2.2 (continued)

Prairie5 Boone 4.111 2.86 0.503 0.064 0 Prairie6 Boone 4.222 2.78 0.456 0.106 0 Prairie7 Boone 3.556 2.63 0.428 -0.031 1 Prairie8 Boone 3.889 2.77 0.46 -0.116 0 Prairie9 Boone 3.778 2.77 0.465 0.015 0 Praire10 Boone 3.444 2.73 0.482 0.019 0 Praire11 Boone 4.222 2.87 0.484 0.044 0 Prairie12 Boone 3.778 2.73 0.451 -0.017 0 Prairie13 Boone 4.000 2.72 0.444 0.101 0 Prairie14 Boone 3.778 2.67 0.484 0.067 0 Prairie15 Boone 3.778 2.69 0.451 0.087 0 Total 3.585 2.56 0.413 0.006 0.4

Cd1

Table 2.3. BOTTLENECK results of the Wilcoxon's sign test using two-phase model (TPM) assessed at α=0.05. Distribution patterns of the mode shift test are also represented L-shaped represents a normal distribution of allele frequencies and Mode shift indicates sites with a bottleneck.

Site Basin Wilcoxon's Sign Rank Test Mode Shift (TPM) Willow1 Willow 0.343 Shifted Deep1 Kansas 0.234 L-shape Deep2 Kansas 0.004* L-shape Cottonwood1 Cottonwood 0.031* Shifted Sugar1 Sugar 0.289 L-shape Moniteau1 Moniteau 0.371 L-shape Camp1 Vermillion 0.344 L-shape SDRock1 James 0.986 L-shape Beaver1 Big Sioux 0.0781 L-shape Elk1 Rock 0.719 L-shape Rock2 Rock 0.594 L-shape Rocktrib1 Rock 0.973 L-shape Mound1 Rock 0.422 L-shape Mound2 Rock 0.531 L-shape Poplar1 Rock 0.289 L-shape Rocktrib2 Rock 0.852 L-shape Rocktrib3 Rock 0.500 L-shape Rocktrib4 Rock 0.188 L-shape Rocktrib5 Rock 0.719 L-shape Rocktrib6 Rock 0.527 L-shape Rocktrib7 Rock 0.406 L-shape Cedar1 North Raccoon 0.125 L-shape Cedar2 North Raccoon 0.156 L-shape Cedar3 North Raccoon 0.098 L-shape Eastbutt1 North Raccoon 0.422 Shifted Westbutt1 North Raccoon 0.973 L-shape Eagle1 Boone 0.230 L-shape Eagle2 Boone 0.527 L-shape Boone1 Boone 0.320 L-shape Boone2 Boone 0.844 L-shape Prairie1 Boone 0.629 L-shape Prairie2 Boone 0.180 L-shape Prairie3 Boone 0.098 L-shape Prairie4 Boone 0.098 L-shape

Table 2.3 (continued)

Prairie5 Boone 0.500 L-shape Prairie6 Boone 0.473 L-shape Prairie7 Boone 0.406 L-shape Prairie8 Boone 0.273 L-shape Prairie9 Boone 0.273 L-shape Praire10 Boone 0.020* L-shape Praire11 Boone 0.230 L-shape Prairie12 Boone 0.629 L-shape Prairie13 Boone 0.344 L-shape Prairie14 Boone 0.367 L-shape Prairie15 Boone 0.422 L-shape

Table 2.4. Pairwise genetic distance between populations, measured using FST for seven genetic Topeka shiner populations. All differences were significant at α = 0.001.

North Cottonwood Willow Kansas Sugar Missouri Raccoon River Creek River Creek River River Willow Creek 0.264 Kansas River 0.346 0.130 Sugar Creek 0.237 0.193 0.123 Missouri River 0.212 0.209 0.117 0.073 North Raccoon River 0.238 0.235 0.162 0.088 0.081 Boone River 0.195 0.172 0.099 0.069 0.052 0.031

Table 2.5. Asymmetrical pairwise historical migration rates m for Topeka shiners (± 95% confidence intervals) generated by

MIGRATE. Sources sites are listed in the left column and receiving sites are listed in the top row. * indicates significant values at m =

.0200 (1 migrant per 50 years).

North Raccoon Willow Creek Kansas River Cottonwood River Sugar Creek Moniteau Creek Missouri River River Boone River

Willow Creek Basin 0.24715* 0.247165* 0.0098325 0.0046445 0.004978 0.008235 0.001165 C. I. (95%) (0.2100, 0.2830) (0.2100, 0.2830) (0.0003, 0.0190) (0.0000, 0.0137) (0.0000, 0.0137) (0.0000, 0.0170) (0.0000, 0.0093) Kansas River Basin 0.28985* 0.002165 0.0175 0.006165 0.009165 0.016165 0.01115 C. I. (95%) (0.1486, 0.3237) (0.0000, 0.0160) (0.0070, 0.0277) (0.0000, 0.0157) (0.0000, 0.0183) (0.0047, 0.0277) (0.0001, 0.0021) Cottonwood River Basin 0.0104915 0.0075 0.0098335 0.005165 0.0075 0.006165 0.014835 C. I. (95%) (0.045, 0.0167) (0.0000, 0.0160) (0.0004, 0.0187) (0.0000, 0.0150) (0.0000, 0.0160) (0.0000, 0.0253) (0.0013, 0.0025) Sugar Creek Basin 0.009563 0.0228335 0.0075 0.33385* 0.0305* 0.05385* 0.0038335

C. I. (95%) (0.0005, 0.0187) (0.0100, 0.0360) (0.0000, 0.0160) (0.2937, 0.4367) (0.0040, 0.0613) (0.0384, 0.0700) (0.0000, 0.0019)

Moniteau Creek Basin 0.005835 0.007835 0.003165 0.1835* 0.20785* 0.01215 0.0125 C. I. (95%) (0.0000, 0.0143) (0.0000, 0.0170) (0.0000, 0.0113) (0.0085, 0.0211) (0.029, 0.3870) (0.0000, 0.0025) (0.0000, 0.0025) Missouri River Basin 0.005305 0.009835 0.006835 0.011835 0.016165 0.025165 0.019165 C. I. (95%) (0.0000, 0.0140) (0.0003, 0.0193) (0.0000, 0.0160) (0.0000, 0.0417) (0.0033, 0.0290) (0.0005, 0.0497) (0.0077, 0.0307) North Raccoon River Basin 0.005015 0.017835 0.0075 0.13065* 0.0151665 0.014835 0.3715* C. I. (95%) (0.0000, 0.0130) (0.0067, 0.0233) (0.0000, 0.0227) (0.0747, 0.1823) (0.0050, 0.0253) (0.0000, 0.0300) (0.2515, 0.4715) Boone River Basin 0.007165 0.013165 0.0045 0.0225 0.3605* 0.13215* 0.112* C. I. (95%) (0.0000, 0.0157) (0.0020, 0.0237) (0.0000, 0.0130) (0.0000, 0.0570) (0.2907, 0.4040) (0.0070, 0.1944) (0.0727, 0.1677)

Table 2.6. Asymmetrical pairwise contemporary migration rates m for Topeka shiners (± 95% confidence intervals) generated by

BAYESASS. Sources sites are listed in the left column and receiving sites are listed in the top row. * indicates significant values at m

= .0200 (1 migrant per 50 years).

North Raccoon Willow Creek Kansas River Cottonwood River Sugar Creek Moniteau Creek Missouri River River Boone River Willow Creek Basin 0.2307* 0.0128 0.0129 0.0133 0.0128 0.0129 0.0129 C. I. (95%) (0.2011, 0.2603) (0.0004, 0.0252) (.0004, 0.0256) (0.0006, 0.0160) (0.0004, 0.0252) (0.0005, 0.0253) (0.0005, 0.0253) Kansas River Basin 0.0088 0.0088 0.0105 0.0088 0.0173 0.0116 0.0184 C. I. (95%) (0.0001, 0.0165) (0.0002, 0.0164) (0.0003, 0.0212) (0.0002, 0.0164) (.0036, 0.0300) (0.0007, 0.0225) (0.0039, 0.0331) Cottonwood River Basin 0.0119 0.012 0.0123 0.0121 0.0128 0.0121 0.0122 C. I. (95%) (0.0004, 0.0234) (0.0005, 0.0235) (0.0004, 0.0142) (0.0005, 0.0237) (0.0005, 0.0251) (0.0004, 0.0248) (0.0004, 0.0240) Sugar Creek Basin 0.0119 0.0131 0.012 0.0121 0.0032 0.0037 0.0179

C. I. (95%) (0.0004, 0.0134) (0.0005, 0.0255) (0.0005, 0.0135) (0.0004, 0.0238) (0.0006, 0.0580) (0.0011, 0.0638) (0.0010, 0.0266)

Moniteau Creek Basin 0.0119 0.0122 0.0126 0.0326 0.1952* 0.0201 0.0212 C. I. (95%) (0.0004, 0.0234) (0.0004, 0.0240) (0.0004, 0.0248) (-0.0046, 0.0698) (0.1478, 0.2426) (0.0018, 0.0388) (0.0002, 0.0407) Missouri River Basin 0.0012 0.0012 0.0013 0.0014 0.0061 0.0019 0.0058 C. I. (95%) (0.000, 0.0024) (0.000, 0.0024) (0.000, 0.0026) (0.000, 0.0028) (0.0033, 0.0089) (0.0001, 0.0037) (0.0015, 0.0073) North Raccoon River Basin 0.0037 0.0038 0.0051 0.0048 0.0039 0.0112 0.0478 C. I. (95%) (.0.000, 0.0074) (.0.000, 0.0076) (0.0003, 0.0098) (0.0002, 0.0094) (0.0001, 0.0077) (0.0019, 0.0201) (0.0195, 0.0761 Boone River Basin 0.001 0.001 0.001 0.0012 0.0011 0.0107 0.0042 C. I. (95%) (0.000, 0.0020) (0.000, 0.0020) (0.000, 0.0020) (0.000, 0.0024) (0.0001, 0.0021) (0.0024, .0180) (0.0006, 0.0078) 42

Figures

Figure 2.1. Current distribution (dark grey) and approximate historical range (light grey) of

Topeka shiners. Black dots represent sites that were included in the study. White dots represent source populations for Topeka shiners in Kansas Biological Station Research ponds.

Figure 2.2. Number of genetic populations (K) as determined by the second order rate of change in the distribution of likelihoods of K using the Evanno method for all samples (A),

Kansas River, Arkansas River, and Sugar Creek Basins (B), Missouri River Basin (C), Des

Moines River Basin (D), Kansas River Basin (E), Arkansas River and Sugar Creek

Basins (F).

Figure 2.3. STRUCTURE clustering for all samples, K=3 (A), Kansas River, Arkansas River and Sugar Creek Basins, K=2. (B),

Missouri River Basin, K=1 (C), Des Moines River Basin, K=2 (D), Kansas River Basin, K=2 (E) Arkansas River and Sugar Creek

Basins, K=2 (F)

Figure 2.4 Topeka shiner populations indicated by microsatellite analysis. Circle color represent genetically distinct pop 46

Figure 2.5 Unrooted neighbor joining tree for Topeka shiners collected at 44 sites based on

Nei’s genetic distance. River basin of origin listed on right.

Figure 2.6. Historical Topeka shiner gene flow patterns estimated by MIGRATE. The arrow colors and arrow heads indicate the direction of migration. The black dotted lines represent bidirectional migration between populations.

Figure 2.7 Contemporary Topeka shiner gene flow patterns estimated by BAYESASS. The arrow colors and arrow heads indicate the direction of migration. 49

CHAPTER 3. METAPOPULATION STRUTURE OF TOPEKA SHINERS IN IOWA AND MINNESOTA

Abstract

Threatened and endangered species with fragmented distributions often exist as metapopulations, groups of discrete demes within a population that maintain some level of gene flow. The Topeka shiner (Notropis topeka) is an endangered species that has experienced population declines across its range due to habitat destruction. Within basins

Topeka shiners occur in pool and off-channel habitat which has been greatly reduced in number due to human activity. The objectives of this study were to evaluate the genetic structure of Topeka shiners to determine if they are panmictic, isolated, or a metapopulation and also to determine the level of connectivity within the Rock River and Boone River

Basins in Minnesota and Iowa. Nine microsatellites loci were used to assess genetic diversity and isolation by distance within each basin whereas genetic variation was used to assign site distinct demes and determine first generation migrants between demes. Results indicate deviations from Hardy-Weinberg Equilibrium, genetic isolation by distance, and low but significant genetic differences between sites that are consistent with the definition of a metapopulation. Each basin was composed of four discrete demes and analysis of migration patterns indicate demes that cover larger areas may be acting as potential population sources, exporting individuals to other demes. Overall this study found Topeka shiners exhibited characteristics of highly connected metapopulations. Continued efforts to restore habitat distribution will help to maintain and improve Topeka shiner connectivity within basins.

Introduction

One of the basic objectives of population genetics is to determine population structure, genetic variation within and among populations of interbreeding individuals 50

(Templeton et al. 1995). Population structure occurs when there is some level of reproductive isolation that occurs as a result of geographic distance or barriers between groups of individuals that limits gene flow, in turn resulting in differences in allele frequencies (Wright

1951). Groups of individuals that are isolated for long periods of time can form separate populations (Hoelzel et al. 2002). Understanding population structure can help define management units for imperiled species, identify areas of low genetic diversity, and determine genetic bottlenecks, but also provide information on the interaction between and within these populations (Abdul-Muneer 2014).

Genetic populations can be classified in three different ways (Rowe et al. 2000). First, individuals can exist as a panmictic population where there is no population structure and no restrictions to mating between individuals (Wirth and Bernatchez 2001; Als et al. 2011). This occurs when groups of individuals (demes) are highly interconnected and there are no barriers to gene flow (Bergmann et al. 2006). Alternatively, populations can act as completely isolated demes, where groups of individuals are completely reproductively isolated from each other and there is no gene flow between demes (Ferreiria et al. 2009).

Finally, populations can act as a metapopulation, in which demes experience gene flow from occasional migrants (Levins 1969).

The classical metapopulation model is often described as a population of populations, consisting of a series of demes that have high extinction rates but also have a high probability of recolonization (Levins 1969). These demes may exhibit low genetic differentiation as a result of high dispersal (Planes et al. 1996). Metapopulations can also exhibit source-sink dynamics. In this model, some demes act as sources, consistently producing migrants that 51 are able to colonize other areas, whereas other demes act as sinks, which rely on immigration to sustain the demes over time (Pulliam 1988).

These abstract models have been adapted to characterize metapopulations to improve conservation efforts (Akcakaya et al. 2007). Although some species form metapopulations due to naturally patchy distributions, other species form metapopulations as a result of human activities that have fragmented formerly contiguous distributions (Fischer and Lindenmayer

2007). Throughout North America, many freshwater species have experienced population declines and fragmentation, leaving them vulnerable to extinction (Abell 2002).

Approximately 39% of fishes and 47% of cyprinids in North America are imperiled (Jelks et al. 2008), largely a result of habitat destruction (McGill et al. 2015). Post-European settlement, Midwestern states experienced a drastic increase (70-90%) in land conversion to agriculture (USDA 2014). Agricultural practices often involve channelization and straightening streams to improve irrigation, field drainage, and flood control (Simpson et al.

1982). However, these alterations have had negative effects on aquatic habitat (Gallant et al.

2011). Channelization results in streams with homogenized habitats and reduced floodplain connectivity (Simpson et al. 1982), leading to geographic fragmentation, which causes declines in many fish species’ distributions (Allan and Flecker 1993; Richter et al. 1997;

Jelks et al. 2008).

The Topeka shiner (Notropis topeka, Gilbert 1884), is a fish species that has been greatly affected by habitat loss due to stream channelization. The Topeka shiner is a small cyprinid found in headwater streams and associated off-channel habitats in parts of Kansas,

Missouri, Nebraska, South Dakota, Minnesota, and Iowa, USA (USWFS 2018). Topeka shiners were once widely distributed in streams that were naturally meandering with many 52 areas of slow to no flow (Dahl 2001; Hatch 2001) as well as off-channel habitats, such as oxbows, that connect with the stream during periods of high flow (Clark 2000; Thomson and

Berry 2009; Ceas and Larson 2010; Bakevich et al. 2013). These off-channel habitats are now less abundant than before agricultural land use changes (Wall et al. 2004; Panella 2012).

The alteration of this original habitat through stream channelization coincided with populations declining to approximately 20% of historical distribution (Tabor 1998), resulting in Topeka shiners being federally listed as endangered under the Endangered Species Act in

1998 (USFWS 1998).

Agricultural streams and their floodplain are also subject to periods of flooding and drought that vary in intensity (Matthews et al. 2013). Drought can cause streams to dry up, trapping and killing fish or forcing them to move to refugia (Lake 2003; Magoulick and

Kobza 2003) whereas high flow allows fish to migrate back into these habitats (Matthews et al. 2013). Oxbows commonly experience this extinction-recolonization phenomena due to their position in the flood plain (Fischer et al. 2018). Oxbows are separated from the stream channel the majority of time, but connect to streams during high-flow events, allowing fish to move between streams and oxbows (Ishii and Hori 2016; Fischer et al. 2018). Altered hydrology of prairie streams has disrupted oxbow formation, and reduced water retention

(Russell 1947; Miller et al. 2009), not only reducing the number of oxbows present in the landscape, but also the frequency of their connection to the adjacent stream (Russell 1947;

Miller et al. 2009). Shallow oxbows are susceptible to low dissolved oxygen and can dry up in the summer, as well as completely freeze in the winter, subsequently eliminating fish

(Escaravage 1990; Townsend et al. 1992). Therefore, starting in 2002, the United States Fish and Wildlife Service (USWFS) initiated a program to restore oxbow habitat (Kenney 2013). 53

Currently there have been over 80 oxbows restored in Iowa and 60 in Minnesota (USFWS

2018). Given the fragmentation of habitat as well as the propensity for extirpation and subsequent recolonization of streams and oxbows due to the drought-flood cycles in

Midwestern streams, Topeka shiners may exhibit some form of metapopulation structure in these altered basins. Successful restoration of species of conservation concern often depends on establishing or re-establishing population connectivity (Rudnick et al. 2012).

This study will provide conservation managers with information on the connectivity of Topeka shiners within basins to aid restoration efforts. The first objective of this study was to evaluate the genetic structure of Topeka shiners to determine if they are panmictic, isolated, or a metapopulation. The second objective was to characterize the connectivity between demes and determine potential population sources and sinks within these areas through the analysis of population genetic data.

Methods

Fish Sampling

Topeka shiners were sampled in two 8-digit Hydrologic unit code (HUC 8) basins in

Iowa and Minnesota, USA (Bakevich et al. 2013). Topeka shiners were collected from 12 sites in the Rock River Basin in southwest Minnesota and northwest Iowa, along the Rock

River and its tributaries, including Elk Creek, Mound Creek, and Poplar Creek (Table 3.1,

Figure 3.1. Topeka shiners were also collected from 19 sites in the Boone River Basin located in central Iowa, including Prairie Creek, Eagle Creek, and Boone River (Table 3.1;

Figure 3.1). Topeka shiners were collected between May and August in 2016 and 2017.

Given differences in habitat, oxbows and stream sites were sampled using different methods.

Oxbows lacked flow, were generally shallow with a silt bottom, and free of snags such as trees. These sites were sampled for fish using one to three passes with a bag seine (10.7m X 54

1.8m, 6.35mm mesh or 16.8 X 1.8m, 6.35mm mesh) following methods from previous studies (Thompson and Berry 2009; Bakevich et al. 2013; Simpson et al. in review A). Fishes were identified to species, counted, held in pens, and released alive after all passes were complete. Stream site conditions consisted of varying flows, substrates ranging from sand to cobble, and often contained large woody debris. Therefore, multiples gear types were used to increase the likelihood of capturing Topeka shiners (Bakevich et al. 2013; Simpson et al. in review B). Fish sampling started with a single upstream electrofishing pass using a barge electrofishing unit, or if the stream was impassable with a barge electrofishing unit, sampling was conducted with a Smith-Root LR-20 backpack electrofishing unit. Electrofishing was followed by a variable number of seine pulls with either a seine (4.5m X 1.8m, 6.35mm mesh) or a bag seine (10.7m X 1.8m, 6.35mm mesh or 16.8m X 1.8m, 6.35mm mesh) as conditions permitted. Up to 20 adult and juvenile Topeka shiners were randomly selected from the collection at each site for non-destructive genetic sampling. A portion of the anal fin was clipped from these fish and preserved in 95% ethanol for genetic analysis.

DNA Extraction and Genotyping

Genomic DNA was extracted using the Qiagen DNeasy Blood and Tissue Kit following the manufacturer’s protocol. Polymerase chain reaction (PCR) amplification of the microsatellite loci were performed using 10 µl reactions. Approximately 2-ng of genomic

DNA was used as the template for the reactions using the M13 protocol of Schuelke (2000).

A 10 µL reaction for each sample included: 6.6-µL H20, 1-µL-Biolase buffer (10x), 0.3-µL

MgCl2, 0.8-µL dNTP, 0.1-µL forward primer, 0.1-µL reverse primer, 0.05-µL M13 labeled oligo, 0.05-µL Biolase Taq polymerase, and 1-µL template DNA. The PCR reactions were then performed in an Eppendorf Master Cycler thermal cycler with the following conditions

(95°C/5min; [94°C/30 sec.; 62°C/60 sec.; 72°C/30 sec.] X 10 cycles; [94°C/30 sec.; 55 annealing temperature (Appendix A)/60 sec.; 72°C/30 sec.; 72°C/20 min] X 25 cycles;

72°C/4 min). Negative controls were included with every reaction to detect potential contamination. PCR products were checked on 1% agarose gels before being sent to the Iowa

State DNA Facility where samples were visualized using capillary electrophoresis. Raw data was evaluated and edited using the software GeneMarker version 1.95 (Hulce et al. 2011). A total of 13 microsatellite loci were successfully PCR amplified (Bessert and Orti 2003;

Burridge and Gold 2003; Turner et al. 2004; Anderson and Sarver 2008; Blank et al. 2011), nine of which were polymorphic and used for subsequent genotyping (Appendix A).

Genetic Analysis

Topeka shiner genotypes were tested for deviations from Hardy Weinberg

Equilibrium (HWE) due to possible errors in genotyping, large allele dropout, stuttering, and null alleles using MicroChecker 2.2.3 (van Oosterhout et al. 2004). Deviations from HWE were then tested using the Markov Chain Monte Carlo (MCMC) approximation of Fisher’s exact test using the adgenet package in Program R (Jombart 2008). Tests for linkage disequilibrium were performed using GenePop 4.2 (Raymond and Rousset 1995).

Measures of genetic diversity, such as number of alleles per locus, expected heterozygosity, and unbiased heterozygosity, were estimated for each site using GenAlEx 6.5

(Peakall and Smouse 2006, 2012). Allelic richness was calculated and rarefacted to the smallest sample size using the program HP-Rare 1.1 (Kalinowski 2005). Measures of genetic diversity from oxbows and stream sites were compared across all basins and within each basin using a Mann Whitney U test in GenAlEx 6.5.

Metapopulation demes that experience occasional gene flow will have genetic differentiation correlated with the geographic distance between demes, known as isolation by distance (IBD). To determine IBD, a standardized genetic distance (FST/(1-FST), where FST is 56 genetic difference between sites, was calculated between sites and compared to a geographic distance using a Mantel test in GenAlEx 6.5 (Peakall and Smouse 2006, 2012). As fish movement is likely to be restricted to streams and their respective floodplains, geographic distance between sample locations for IBD analysis was calculated as stream distance between sites using the stream centerline data created from 2m Digital Elevation Model

(DEM; Zambory 2018).

Pairwise genetic difference between sites or populations can give an indication of gene flow, and genetically similar sites can be considered as single demes. FST was used as a measure of genetic difference and was calculated using GenAlEx 6.5 (Peakall and Smouse

2006, 2012). Sites were grouped into demes based on genetic similarity between sites. Sites that did not have significant genetic difference at P ≤ 0.05 were grouped into the same deme.

An important characteristic of metapopulations is migration of individuals from existing demes and to unoccupied habitats (Levins 1969; Pulliam 1988). First generation migration between demes was determined using two Baysian methods to determine the amount of immigration and emigration. First, the program GeneClass2 was used to calculate first generation migrants in populations (Piry et al. 2004). The likelihood ratio

(L=Lhome/Lmax), where L is the likelihood of migration, Lhome is likelihood that an individual belongs to the deme it was sampled from, and Lmax is the highest likelihood value among all demes for an individual, was used as the criterion for the assignment likelihood (Paetkau et al. 2004) and a partial Bayesian method was used to assign individuals to a population

(Rannala and Mountain 1997). The MCMC resampling algorithm (Paetkau et al. 2004) was used and 10,000 individuals were simulated with α = 0.01. The second Bayesian method for determining first generation migrants was implemented in BIMr (Faubet et al. 2007) that uses 57

MCMC and reversible jump MCMC methods to estimate first generation migrants and directional migration rates between demes (Faubet and Gaggiotti 2008). Demes that have a higher rate of migration into other demes than from other demes are considered net emigrators and demes that have a lower rate of migration into other demes than from other demes are considered net immigrators. Multiple burnin in period were tested from 10,000 to

1x10-7, but it did not have an impact on the pattern of migration rates. Therefore, a burn-in period of 100,000, with 100,000 runs and 10 replicates were used for both basins.

Results

Seven oxbow and five stream sites were sampled in tributaries along the main stem of the Rock River Basin, with one site located in the headwaters of Elk Creek (Table 3.1, Figure

3.1). Distance was variable between sites, with some oxbow and stream sites only a few meters apart but with a maximum of 119 river km between the Elk Creek site and the headwater sites in the Rock River (Table 3.2) The Boone River Basin included ten oxbow and nine stream sites (Table 3.1, Figure 3.1). Distance between sites range from 0-135 river km, but the majority of sites were located within a 20 river km stretch of Prairie Creek (Table

3.3), with more isolated sites on the Boone River and in Eagle Creek.

The genotypes for individuals were tested for PCR or genotyping errors at each locus.

Null alleles were identified at a few sites, but no site had more than one locus with a null allele. There was no evidence for large allele dropout, stuttering, or linkage disequilibrium at individual sites. A few locus by population comparisons were out of HWE (Table 3.4) but all loci and sites were retained for further analyses. When all sites were pooled by basin, the

Rock River Basin was out of HWE at 4 out of 9 loci and the Boone River Basin was out of

HWE at 5 out of 9 loci. 58

Isolation by distance produced different results in the two basins (Figure 3.2).

Distance between sites in the Rock River Basin ranged from 0.02 km to 120 km (Table 3.2) whereas distance between sites in the Boone River Basin ranged from 0 km to 135 km, with a large cluster of 13 sites concentrated in a 20 km section of Prairie Creek (Table 3.3). Genetic distances between sites in the Rock River basin was generally low (Table 3.2) and genetic distance was not related to river distance (Mantel R = 0.159, P ≤ 0.19; Figure 3.2). In contrast, sites in the Boone River Basin exhibited low to moderate genetic differentiation that was positively related to river distance (Mantel R = 0.562, P ≤ 0.01; Figure 3.2).

Pairwise FST values generally showed low differentiation between sites within both basins, indicating gene flow was occurring between sites (Table 3.2, 3.3). The lowest difference in the Rock River Basin was 0.011 between an adjacent in-channel and off- channel site and the highest differentiation was 0.052 observed between the southernmost and northernmost sites on the Rock River (Table 3.2). The Boone River Basin exhibited low

FST values between sites, but a few sites showed moderate differences. The smallest difference was 0.009 between an adjacent oxbow and stream site on Prairie Creek and the largest genetic difference was 0.093 between a site in Eagle Creek and a site in headwaters of

Prairie Creek 128 river km away (Table 3.3).

Since Topeka shiners occupy two distinct habitat types, genetic diversity streams and oxbows, sites were compared using Mann-Whitney U tests, to determine if habitat had an effect on genetic diversity. Number of alleles, allelic richness, expected heterozygosity, and inbreeding coefficient FIS values were moderate with little difference between sites overall

(Table 3.5). Number of alleles per locus ranged from 2.89 to 4.44 and allelic richness ranged from 2.3 to 2.87. Values of FIS were low and variable ranging from -0.199 to 0.132. No 59 comparison of genetic diversity between oxbows and streams was significantly different

(Mann-Whitney U tests, P ≤ 0.05). The one exception was expected heterozygosity in the

Rock River Basin where oxbows had higher diversity than streams (Table 3.6).

Four distinct demes were identified in both the Rock River (Tables 3.2, 3.5) and

Boone River based on FST values (Table 3.3, 3.5). First generation migrants between demes were also detected in both basins. In the Rock River Basin, a single first generation migrant from the Headwater Rock deme to Elk Creek deme was identified using Geneclass2 (Figure

3.3). Analysis of migration in the Rock River Basin using BIMr indicated five first generation migrants: two migrants from the Middle Rock deme to the Elk Creek deme, two migrants from the Headwater Rock deme to the Lower Rock deme, and one migrant from the

Headwater Rock deme to the Middle Rock deme (Figure 3.3). Directional migration analysis in BIMr showed the Headwater Rock River and the Middle Rock River as net emigrators

(Probability of migration > 0.05 while the Lower Rock River and Elk Creek were net immigrators (Probability of migration > 0.05; Figure 3.4). In the Boone River Basin, there was only one first generation migrant from Lower Prairie deme to the Headwaters Prairie deme detected using Geneclass2 (Figure 3.5). There were six first generation migrants identified in the Boone River Basin using BIMr, all from Lower Prairie into the other demes

(Figure 3.6). Directional migration analysis of the Boone River Basin showed that only the

Lower Prairie deme had outward migration (Probability of migration > 0.05; Figure 3.6).

Discussion

Within population connectivity can greatly affect the choice of strategies employed by conservation managers (Hodgson et al. 2009). This can be particularly important for species in great conservation need like the Topeka shiner. Populations can be generally characterized as behaving in three different ways with respect to gene flow: single panmictic 60 populations, completely isolated sub-populations, or as interconnected meta-populations

(Rowe et al. 2000). Because of the dynamic and fragmented nature of habitat in these basins, population connectivity is variable.

Populations in panmixia have no breeding restrictions, therefore populations will be in HWE and no population structure will be present. Topeka shiners in both the Rock River and Boone River basins can be considered individual genetic populations within this study region (Chapter 2) but distribution-wide analyses often fail to determine fine scale structure when gene flow is present (Falush et al. 2016). The Rock River Basin was out of HWE for 4 of 9 loci and the Boone River Basin was out of HWE for 5 of 9 loci due largely to a deficit of heterozygotes, indicating a lack of panmixia. Populations not in panmixia will be out of

HWE due to population structuring which often produces a deficit of heterozygotes (Rowe et al. 2000).

Isolated demes will exhibit high genetic differentiation between demes and little to no genetic differentiation within demes (Harrison and Hastions 1996). Metapopulations display characteristics in between those two extremes and consist of distinct demes with occasional migration to other demes or uninhabited areas. Analysis of IBD and FST shows a presence of structure in the Boone River Basin that is consistent with a metapopulation structure. If migration of individuals is occurring between sites, as expected in a metapopulation, then genetic differentiation between sites will be positively correlated with distance and IBD will occur (Rousset 1997). The Boone River Basin was significant for IBD, suggesting that these Boone River Basin populations are experiencing some level of gene flow. Sites in close proximity may experience a greater amount of migration while sites farther apart experience a lesser amount of migration (Rowe et al. 2000). Values of FST 61 generally showed low differentiation, as values were below 0.05. Similarly, metapopulations of Natterjack toads (Epidalea calamita) showed low but significant differentiation in breeding ponds with FST values lower than 0.05, indicating high amounts of gene flow (Frei et al. 2016). Additionally, Atlantic salmon (Salmo salar) metapopulation demes have low but significant differentiation as a result of migration between streams (Palstra and Ruzzante

2010). Therefore, given the low, but significant gene flow between sites, Topeka shiners in the Boone River Basin appear to exhibit metapopulation structure.

IBD was not significant in the Rock River Basin, suggesting high gene flow over great distances or very low gene flow between sites even at short distances (Gold and Turner

1999; Kark et al. 1999). Estimates of genetic differentiation between sites in the Rock River

Basin did not fall into either of these two categories, as there was significant differentiation between some sites, whereas other sites had no differentiation. Populations that have uneven distribution of ideal habitat could experience differential rates of migration that would affect the strength of IBD (Sexton et al. 2014). Genetic differentiation, although significant between between Topeka shiners in some sites was relatively low overall, suggesting that there is gene flow with the basin. Therefore, the Rock River Basin is also likely acting as a metapopulation.

Oxbows have the potential to be drivers of metapopulation dynamics for Topeka shiners due to the high rate of colonization and extinction of oxbows (Thomson and Berry

2009; Fischer et al. 2017), but oxbows did not appear to have a direct influence on genetic diversity. The tests for differences in genetic diversity between oxbows and stream sites were not significant overall or in either basin, likely due to the high amount of gene flow between stream and adjacent oxbows that can lead to low differentiation and high diversity (Whitlock 62

1992; Richards 2000). Oxbows generally reconnect to the adjacent streams multiple times a year that may allow Topeka shiners to retain genetic similarity between adjacent sites.

Metapopulation structure is dependent upon the formation of genetically distinct but connected groups of individuals referred to as demes. Demes are composed of genetically similar individuals that occur in close geographic proximity. Within the Rock River Basin,

Topeka shiner demes were comprised of small stream clusters. The Boone River Basin exhibited similar metapopulation structure, with the same number of demes, which encompassed a single stream or stream segment. Other studies found that an increase in habitat will result in increased turnover and an increase in genetic similarity betetween demes. (Hokit et al. 1999; Cosentino et al. 2010). Topeka shiners may exist in more areas between sample sites than documented in this study, but given the distribution of habitat within basins, demes are still likely spatially and genetically fragmented. Geographic distance and genetic distance was greater between demes (~30 - 120 river km, average FST =

0.047) than within demes (~0 – 20 river km, average FST = 0.017). Increased distance between habitat patches will increase genetic isolation (Hanski and Gaggiotti 2004).

Therefore, due to the distances between habitats, there is likely differential migration between these habitats.

Interaction between demes is a driving factor in metapopulation dynamics (Hanski

2001). The movement of individuals either to other established demes or to founding demes determines metapopulation connectivity. The Rock River and Boone River basins exhibited different metapopulation structure and therefore have varying interaction between demes.

Metapopulation structure is also likely to change over time. Some demes will persist over time and others may go extinct only to be recolonized at a later time (Stacey and Taper 1992; 63

Reigada et al. 2015). Thus, exact characterization of these metapopulations only applies to the current study period, as the composition and connectivity of demes is likely to change due to the dynamic nature of these systems.

Directional migration analysis between demes in the Rock River Basin indicated that migrants from the two demes that consisted of more sites and covered a larger area

(Headwater Rock and Middle Rock), were moving into the two smaller downstream demes

(Elk Creek and Lower Rock). Larger demes are less susceptible to extinction and are more likely to export colonizers to other areas (Hull et al. 2011). In contrast, smaller demes are more susceptible to extinction and recolonization events (Hanski 1999). The two demes receiving migrants (Lower Rock and Elk Creek) both consisted of one site and were located on the edge of the known Topeka shiner distribution within the basin (Ceas and Anderson

2004; Ceas and Larson 2010; Cunningham 2017). Although the persistence of these specific demes is unknown, it is possible they could fluctuate over time as a result of extinction and recolonization by migrants from other demes.

The Boone River Basin had a different metapopulation dynamic compared to the

Rock River Basin where migrants originated from only one deme (Lower Prairie) that was centrally located between the other demes in the basin. Additionally, there was a migrant from Lower Prairie detected in all other demes. Sampling throughout the basin was widespread, but Topeka shiners appeared most frequently in areas located within the Lower

Prairie deme. Well-connected metapopulations often have more suitable habitat available, allowing neighboring demes to colonize uninhabited areas (Furrer and Pasinelli 2016).

Individual migrants appear to be moving from this larger deme into areas where preferred habitat may be less occupied. Conversely, the Headwater Prairie, Headwater Boone River, 64 and Eagle Creek demes appear to be smaller in area and only consisted of single or adjacent sites. The smaller demes may have been recently founded or are experiencing population decline. Knowledge of fine scale historical Topeka shiner distribution in this basin is incomplete, as sampling in the past was minimal and infrequent (Bakevich et al. 2015). Thus, it is unclear how persistent these smaller demes may be. Analysis of genetic metrics did not indicate lower diversity in these demes which would be expected in colonizing populations that are formed by only a few individuals, but are not always detectable due to gene flow from multiple demes (Greenbaum et al. 2014). It is possible these are persistent demes that may exhibit a source sink pattern.

Both the Rock River and Boone River basins showed some characteristics of source- sink dynamics. Source population are demographically stable and will produce migrants

(Pulliam 1988). The Upper Rock, Middle Rock, and Lower Prairie demes appear to be stable and exporting fish to other areas and are behaving as sources. Natterjack toads displayed similar migration patterns in ponds designated as sources (Frei et al. 2016). When metapopulations have greater connectivity, sources are often easier to detect than sinks

(Furrer and Pasinelli 2016). This appears to be the case in this study, as Topeka shiner population sinks were not easily discernable. There were a few demes that were receiving migrants, a major characteristic of sinks, but sinks are also unable to persist without migration from other areas (Pulliam 1988). Although sinks are likely to occur in Topeka shiner metapopulations, identifying these locations would require greater temporal sampling to measure migration rates and population stability over time. It is also possible that these smaller demes are producing migrants into areas not sampled in this study. 65

Metapopulations are dependent upon connectivity of habitat that allows individuals to migrate between demes and colonize new areas (Hanski and Gaggiotti 2004). Although the current metapopulation structure appears stable, further isolation of quality habitat may result in a reduction of colonization events and the loss of connectivity within Topeka shiner metapopulations. If Topeka shiners demes are too distant to recolonize habitats after extinction events, those habitats may remain unoccupied. In contrast, an increase in preferred instream pool and off-channel oxbow habitat would improve connectivity and the stability of these metapopulations (Furrer and Pasinelli 2016). Current restoration efforts in these basins focus on restoring off-channel oxbows to increase the amount of preferred habitat for Topeka shiners and other species (USWFS 2018). The current metapopulation structure in the two study basins had high levels of connectivity with potential for colonization. Topeka shiners use oxbows immediately after construction (Zambory et al. 2017) and these newly restored habitats likely support natural reproduction (Kenney 2013). Therefore, oxbows placed on the edges of the Topeka shiner distribution have the potential to experience rapid colonization and integration into the metapopulation. In addition, demes with larger population sizes maintain higher genetic diversity (Ellegren and Galtier 2016) and are more likely to act as sources. Therefore, oxbows placed near streams with established demes may bolster abundance of Topeka shiners that could help maintain or improve the current levels of diversity and promote migration to other areas within the metapopulation.

Despite large historical changes to hydrology, this study indicated a high level of gene flow within basins in Iowa and Minnesota. Although Topeka shiners have a fragmented distribution in these areas, the flood-drought cycle facilitates connectivity of Topeka shiners between headwater streams, giving these populations characteristics of highly connected 66 metapopulations. Future reduction of connectivity of habitat will cause a decline in distribution, abundance, and genetic diversity of Topeka shiners in these basins. Therefore, sustained monitoring of Topeka shiner persistence will be important. Increasing the quality and abundance of habitat has the potential to bolster population size and distribution. Larger populations with improved connectivity would insulate these metapopulations from loss of genetic diversity and connectivity, thereby decreasing the potential for extirpation of Topeka shiners in these basins.

Literature Cited

Abell, R. 2002. Conservation biology for the biodiversity crisis: a freshwater follow‐up. Conservation Biology 16:1435-1437.

Akçakaya, H. R., G. Mills, and C. P. Doncaster. 2007. The role of metapopulations in conservation. Key Topics in Conservation Biology. Blackwell Publishing 64-84.

Allan, J. D. and A. S. Flecker. 1993. Biodiversity conservation in running waters. Bioscience 43:32-43.

Als, T. D., M. M. Hansen, G. E. Maes, M. Castonguay, L. Riemann, K. Aarestrup, P. Munk H. Sparholt, R. Hanel, and L. Bernatchez. 2011. All roads lead to home: panmixia of European eel in the Sargasso Sea. Molecular Ecology 20:1333-1346.

Anderson, C. M. and S. K. Sarver. 2008. Development of polymorphic microsatellite loci for the endangered Topeka shiner, Notropis topeka. Molecular Ecology 8:311-313.

Bakevich, B. D., C. L. Pierce and M. C. Quist. 2015. Status of the Topeka shiner in west- central Iowa. American Midland Naturalist 174:350-358.

Bergemann, S. E., G. W. Douhan, M. Garbelotto, and S. L. Miller, 2006. No evidence of population structure across three isolated subpopulations of Russula brevipes in an oak/pine woodland. New Phytologist 170:177-184.

Bessert, M. L., and G. Orti. 2003. Microsatellite loci for paternity analysis in the Fathead minnow, Pimephales promelas (Teleostei : Cyprinidae). Molecular Ecology Notes 3:532-534.

Blank, M, B., S. Bramblett, J. Kalinowski, K. Cahoon, K. Nixon. 2011. Impacts of barriers on Topeka shiner populations. Report SD2006-07-F to South Dakota Department of Transportation, Office of Research Pierre, SD, USA 88p.

Burridge, C. P. and J. R. Gold. 2003. Conservation genetic studies of the endangered Cape Fear shiner Notropis mekistocholas (Teleostei: Cyprinidae). Conservation Genetics 4:219-225.

Ceas, P. A., and Y. C. Anderson. 2004. Results of a pilot monitoring project for Topeka shiners in Southwestern Minnesota. Department of Natural Resources, St Paul, MN, USA.

Ceas, P. A., and K. A. Larson. 2010. Topeka shiner Monitoring in Minnesota: Year Seven. Minnesota Department of Natural Resources, St. Paul, MN, USA.

Clark, S. J. 2000. Relationship of Topeka shiner distribution to geographic features of the Des Moines Lobe in Iowa. M.S. Thesis, Iowa State University Ames, IA, USA.

Cosentino, B. J., R. L. Schooley, and C. A. Phillips. 2010. Wetland hydrology, area, and isolation influence occupancy and spatial turnover of the Painted turtle, Chrysemys picta. Landscape Ecology 25:1589-1600.

Cunningham, G. 2017. Topeka shiner monitoring in Minnesota: 2017. Minnesota Department of Natural Resources St. Paul, MN, USA.

Dahl, S. P. 2001. Studies of the Topeka shiner (Notropis topeka) life history and distribution in Minnesota. M.S. Thesis, University of Minnesota Minneapolis, MN, USA.

Escaravage, V. 1990. Daily cycles of dissolved oxygen and nutrient content in a shallow fishpond: the impact of water renewal. Hydrobiologia 207:131-136.

Ellegren, H., and N. Galtier. 2016. Determinants of genetic diversity. Nature Reviews Genetics 17:422-433.

Faubet, P., R. S. Waples, and O. E. Gaggiotti. 2007. Evaluating the performance of a multilocus Bayesian method for the estimation of migration rates. Molecular Ecology 16:1149-1166.

Faubet, P., and O. E. Gaggiotti. 2008. A new Bayesian method to identify the environmental factors that influence recent migration Genetics 178:1491-1504.

Falush, D. L. van Dorp, and D. Lawson. 2016. A tutorial on how (not) to over-interpret STRUCTURE/ADMIXTURE bar plots. bioRxiv 066431.

Ferreira, E. L., A. M. Suoto, V. M. Soares, and C. Fonseca. 2009. Genetic structure of the Wild Boar population in Portugal: evidence of a recent bottleneck Mammalian Biology 74:274-285.

Fischer, J, and D. B. Lindenmayer. 2007. Landscape modiﬁcation and habitat fragmentation: a synthesis. Global Ecology and Biogeography 16:265-280.

Frei, M., D., S. Csencsics, E. Brodbeck, C. Schweizer, F. Bühler, J. Gugerli, and J. Bolliger. 2016. Combining landscape genetics, radio-tracking and long-term monitoring to derive management implications for Natterjack toads (Epidalea calamita) in agricultural landscapes. Journal for Nature Conservation 32:22-34.

Furrer, R. D., and G. Pasinelli. 2016. Empirical evidence for source–sink populations: a review on occurrence, assessments and implications. Biological Reviews 91:782-795.

Gallant, A. L., W. Sadinski, M. F. Roth, and C. A. Rewa. 2011. Changes in historical Iowa land cover as context for assessing the environmental benefits of current and future conservation efforts on agricultural lands. Journal of Soil and Water Conservation 66: 67A-77A.

Gilbert, C. H. 1884. Notes on the fishes of Kansas. Bulletin of the Washburn College Laboratory of Natural History 1:1-16.

Gold, J. R., and T. F. Turner. 2002. Population structure of Red Drum (Sciaenops ocellatus) in the northern Gulf of Mexico, as inferred from variation in nuclear-encoded microsatellites. Marine Biology 140:249-265.

Greenbaum, G., A. R. Templeton, Y. Zarmi, and S. Bar-David. 2014. Allelic Richness following Population Founding Events – A Stochastic modeling framework incorporating gene flow and genetic drift. PLoS ONE 9:115-203.

Hanski, I. 1999. Metapopulation ecology. Oxford University Press.

Hanski, I. 2001. Spatially realistic theory of metapopulation ecology. Naturwissenschaften 88:372.

Hanski, I., and O. E. Gaggiotti. 2004. Ecology, genetics, and evolution of metapopulations. Academic Press, San Diego, CA, USA. 69

Harrison, S., A. Hastions. 1996. Genetic and evolutionary consequences of metapopulation structure. Trends in Ecological Evolution 11:180-183

Hatch, J. T. 2001. What we know about Minnesota’s first endangered fish species: the Topeka shiner. Journal of the Minnesota Academy of Science 65:31-38.

Hodgson, J. A., C. D. Thomas, B. A. Wintle, and A. Moilanen. 2009. Climate change, connectivity and conservation decision making: back to basics. Journal of Applied Ecology 46:964-969.

Hoelzel, A. R., S. D. Goldsworthy, and R. C. Fleischer. 2002. Population genetics. In: marine mammal biology: an evolutionary approach. Blackwell Science 325-352.

Hokit, D. G., B. M. Stith, and L. C. Branch. 1999. Effects of landscape structure in Florida scrub: a population perspective. Ecological Applications 9:124-134.

Hulce, D., X. Li, T. Snyder-Leiby, C. Johathan, and S. Liu. 2011. GeneMarker® genotyping software: tools to increase the statistical power of DNA fragment analysis. Journal of Biomolecular Techniques 22:S35-S36.

Hull, V., A. T. Morzillo, and J. Liu. 2011. Impact of a classic paper by H. Ronald Pulliam: the first 20 years. In Sources, Sinks and Sustainability. Cambridge University Press 3- 22.

Ishii, Y. and K. Hori. 2016. Formation and infilling of oxbow lakes in the Ishikari lowland, northern Japan. Quaternary International 397:136-146.

Jombart, T. 2008. adegenet: a R package for the multivariate analysis of genetic markers. Bioinformatics 24:1403-1405. Kalinowski, S. T. 2005. HP-Rare: a computer program for performing rarefaction on measures of allelic diversity. Molecular Ecology Notes 5:187-189.

Kark, S., P. U. Alkon, U. N. Safrie, E. Randi. 1999. Conservation priorities for Chukar partridge in Israel based on genetic diversity across an ecological gradient. Conservation Biology 13:542-552.

Kenny, A. 2013. The Topeka shiner: shining a spotlight on an Iowa success story. endangered species program. Endangered species bulletins and technical reports, United States Fish and Wildlife Service. 70

Lake, P. S. 2003. Ecological effects of perturbation by drought in flowing waters. Freshwater Biology 48:1161-1172.

Levins, R. 1969. Some demographic and genetic consequences of environmental heterogeneity for biological control. Bulletin of the Entomological Society of America 15:237-240.

Magoulick, D. D., R. M. Kobza. 2003. The role of refugia for fishes during droughts: a review and synthesis. Freshwater Biology 48:1186-1198.

Matthews, W. J., E. Marsh-Matthews, R. C. Cashner. 2013. Disturbance and trajectory of change in a stream fish community over four decades. Oecologia 173:955-969.

McGill, B. J., M. Domelas, N. J. Gotelli, and A. E. Magurran. 2015. Fifteen forms of biodiversity trend in the Anthropocene. Trends Ecological Evolution. 104-113.

Miller, B. A., W. G. Crumpton, and A. van der Valk. 2009. Spatial distribution of historical wetland classes on the Des Moines Lobe, Iowa. Wetlands 29:1146-1152.

Paetkau D., R. Slade, M. Burden, and A. Estoup, 2004. Genetic assignment methods for the direct, real-time estimation of migration rate: a simulation-based exploration of accuracy and power. Molecular Ecology 13:55-65. Palstra, F. P., and D. E. Ruzzante. 2010. A temporal perspective on population structure and gene flow in Atlantic Salmon (Salmo salar) in Newfoundland, Canada. Canadian Journal of Fisheries and Aquatic Science 67:225-242.

Panella, M. J. 2012. Topeka shiner (Notropis topeka) a species conservation assessment for the Nebraska natural legacy project. Nebraska Game and Parks Commission, Lincoln, NE, USA.

Peakall, R. D., and P. E. Smouse 2006. GENALEX 6: genetic analysis in Excel. Population genetic software for teaching and research. Molecular Ecology Notes 6:288-295.

Peakall, R. and P. E. Smouse. 2012. GenAlEx 6.5: genetic analysis in Excel. Population genetic software for teaching and research-an update. Bioinformatics 28:2537-2539.

Planes, S., R. Galzin, and F. Bonhomme. 1996. A genetic metapopulation model for reef fishes in oceanic islands: the case of the Surgeonfish, Acanthurus triostegus. Journal of Evolutionary Biology 9:103-117. Piry, S., A. Alapetite, J. M. Cornuet, D. Paetkau, L. Baudouin, and A. Estoup. 2004. GeneClass2: A software for genetic assignment and first-generation migrant Detection. Journal of Heredity 95:536-539.

Pulliam, H. R. 1988. Sources, sinks, and population regulation. American Naturalist 32:652- 661.

Rannala, B., and J. L. Mountain. 1997. Detecting immigration by using multilocus genotypes. Proceedings of the National Academy of Sciences of the United States of America 94:9197-9201.

Raymond, M., and F. Rousset, 1995. GENEPOP (version 1.2): a population genetic software for exact tests and ecumenicism. Journal of Heredity 86:248-249.

Reigada, C., S. J. Schreiber, F. Altermatt, and M. Holyoak. 2015. Metapopulation dynamics on ephemeral patches. The American Naturalist 185:183-195.

Richards, C. M. 2000. Inbreeding depression and genetic rescue in a plant metapopulation. American Naturalist 155:383-394.

Richter, B. D., D. P. Braun, M. A. Mendelson, and L. L. Master. 1997. Threats to imperiled freshwater fauna. Conservation Biology 11:1081-1093.

Rousset, F. 1997. Genetic differentiation and estimation of gene flow from F-statistics under isolation by distance. Genetics 145:1219-1228.

Rowe, G., T. J. Beebee, T. Burke. 2000 A microsatellite analysis of Natterjack toad, Bufo calamita, metapopulations. Oikos 88:641-651.

Rudnick, D., S. Ryan, P. Beier, S. Cushman, F. Dieffenbach, and C. Epps. 2012. The role of landscape connectivity in planning and implementing conservation and restoration priorities. Issues in Ecology 16:1-20.

Russell, R. J. 1947. Stream meanders. Annals of the Association of American Geographers 37:65-67.

Sexton, J. P., S. B. Hangartner, and A. A. Hoffmann. 2014. Genetic isolation by environment or distance: which pattern of gene flow is most common? Evolution 68:1-15.

Simpson, N. T., A. P. Bybel, C. L. Pierce, K. J. Roe and M. J. Weber. (In Review A). Abiotic and fish assemblage associations of the endangered Topeka shiner and fish species of greatest conservation need in streams of Iowa and Minnesota, USA. Aquatic Conservation: Marine and Freshwater Ecosystems.

Simpson, N. T., A. P. Bybel, C. L. Pierce, K. J. Roe and M. J. Weber. (In Review B). Occurrence, abundance, and associations of Topeka shiners (Notropis topeka) in restored and unrestored oxbows in Iowa and Minnesota, USA. Aquatic Conservation: Marine and Freshwater Ecosystems.

Stacey, P. B., and M. Taper. 1992. Environmental variation and the persistence of small populations. Ecological Applications 2:18-29.

Tabor, V. M. 1998. Final rule to list the Topeka shiner as endangered. Federal Register 63(240):69008– 69021. U.S. Department of the Interior, U.S. Fish and Wildlife Service, Manhattan, KS, USA. 26 p.

Templeton, A. R., E. Routman, and C. A. Phillips. 1995. Separating population structure from population history—a cladistic analysis of the geographical distribution of mitochondrial-DNA haplotypes in the Tiger Salamander, Ambystoma tigrinum. Genetics 140:767-782.

Thomson, S. K., and C. R. Berry. 2009. Stream fishes inhabit livestock watering ponds (dugouts) near Six Mile Creek, Brookings County, South Dakota. Proceedings of the South Dakota Academy of Science 88:127-138.

Townsend, S. A., K. T. Boland, and T. J. Wrigley. 1992. Factors contribution to a fish kill in the Australian wet/dry tropics. Water Research 26:1039-1044.

Turner, T. F., T. E. Dowling, R. E. Broughton, and J. R. Gold. 2004. Variable microsatellite markers amplify across divergent lineages of cyprinid fishes (subfamily Leuciscinae). Conservation Genetics 5:279-281. van Oosterhout, C., W. F. Hutchinson, D. P. Wills, and P. Shipley. 2004. Micro-Checker: software for identifying and correcting genotyping errors in microsatellite data. Molecular Ecology Notes 4:535-538.

U.S. Fish and Wildlife Service (USFWS). 1998. Endangered and threatened wildlife and plants; final rule to list the Topeka shiner as endangered. Federal Register 65:57114- 57119.

U.S. Fish and Wildlife Service (USFWS). 2018. Final Topeka shiner species status assessment. species status assessment reports. South Dakota Ecological Services Office.

Wall, S. S., C. R. Berry Jr., C. M. Blausey, J. A. Jenks, and C. J. Kopplin. 2004. Fish-habitat modeling for gap analysis to conserve the endangered Topeka shiner (Notropis topeka). Canadian Journal of Fisheries and Aquatic Sciences 61:954-973.

Whitlock, M. C. 1992. Temporal fluctuations in demographic parameters and the genetic variance among populations. Evolution 46:608-615.

Wirth, T., and L. Bernatchez, 2001. Genetic evidence against panmixia in the European Eel. Nature 409:1037-1040.

Wright, S. 1951. The genetical structure of populations. Annual Eugenics 15:323-354.

Zambory, C. L. 2018. Geospatial methods for aquatic conservation: Topeka shiner Restoration site selection and the development of an Iowa watershed health assessment M. S. Thesis, Iowa State University Ames, IA, USA.

Tables

Table 3.1. Site abbreviation, stream, sample size (number of individual samples collected), and habitat type for sites sampled for Topeka shiners in the Rock River and Boone River basins.

Site Abbreviations Stream Sample Size Habitat Type Rock River Basin Elk1 Elk Creek 18 Restored Oxbow Rock1 Rock River 19 Oxbow Rocktrib1 Trib To Rock River 13 In-stream Mound1 Mound Creek 25 Restored Oxbow Mound2 Mound Creek 15 In-stream Poplar1 Poplar Creek 19 Restored Oxbow Rocktrib2 Trib to Rock River 20 In-stream Rocktrib3 Trib to Rock River 18 Restored Oxbow Rocktrib4 Trib to Rock River 19 Restored Oxbow Rocktrib5 Trib to Rock River 15 In-stream Rocktrib6 Trib to Rock River 20 Restored Oxbow Rocktrib7 Trib to Rock River 19 In-stream Boone River Basin Eagle1 Eagle Creek 20 Oxbow Eagle2 Eagle Creek 13 Oxbow Boone1 Boone River 18 Oxbow Boone2 Boone River 16 In-stream Prairie1 Prairie Creek 20 Oxbow Prairie2 Prairie Creek 20 In-stream Prairie3 Prairie Creek 19 In-stream Prairie4 Prairie Creek 17 Oxbow Prairie5 Prairie Creek 20 In-stream Prairie6 Prairie Creek 20 In-stream Prairie7 Prairie Creek 17 Oxbow Prairie8 Prairie Creek 16 Restored Oxbow Prairie9 Prairie Creek 18 In-stream Praire10 Prairie Creek 16 In-stream Praire11 Prairie Creek 20 Oxbow Prairie12 Prairie Creek 17 Restored Oxbow Prairie13 Prairie Creek 16 In-stream Prairie14 Prairie Creek 19 Restored Oxbow Prairie15 Prairie Creek 18 In-stream

Table 3.2. Estimates of pairwise FST (below diagonal) and river distances (km; above diagonal) between sites in the Rock River Basin.

* indicates significant values (P= 0.05).

Elk1 Rock2 Rocktrib1 Mound1 Mound2 Poplar1 Rocktrib2 Rocktrib3 Rocktrib4 Rocktrib5 Rocktrib6 Rocktrib7 Elk1 55 46 62 63 108 107 109 118 118 119 119 Rock2 0.044* 31 47 47 92 92 93 102 102 103 103 Rocktrib1 0.03 0.028 15 17 61 61 62 71 71 73 73 Mound1 0.042* 0.019 0.030* 1 52 51 53 62 61 63 63 Mound2 0.038* 0.038* 0.033 0.041* 53 52 53 63 63 64 64 Poplar1 0.022 0.033* 0.03 0.018 0.041 26 27 36 36 38 38 Rocktrib2 0.031 0.052* 0.047* 0.040* 0.04 0.03 0 10 10 12 12 Rocktrib3 0.022 0.038* 0.039* 0.03 0.027 0.019 0.019 9 9 10 10 Rocktrib4 0.031* 0.043* 0.049* 0.037* 0.045* 0.023 0.028 0.014 0 3 3 75 Rocktrib5 0.034* 0.021 0.034* 0.017 0.027 0.028 0.033* 0.02 0.035* 3 3

Rocktrib6 0.023 0.031* 0.039* 0.018 0.028 0.013 0.02 0.011 0.017 0.015 0 Rocktrib7 0.029* 0.02 0.018 0.017 0.029 0.028* 0.049* 0.032* 0.046* 0.018 0.025*

Table 3.3. Pairwise FST (below diagonal) and river distances (km; above diagonal) between sites in the Boone River Basin. * indicates significant values (P ≤ 0.05).

Eagle1 Eagle2

Prarie1 Praire5

Boone1 Boone2

Prairie2 Prairie3 Prairie4 Prairie6 Prairie7 Prairie8 Prairie9

Prarie13 Prairie10 Prairie11 Prairie12 Prairie14 Prairie15 Eagle1 7 60 99 84 84 89 89 91 91 91 96 96 96 96 96 106 106 135 Eagle2 0.025 53 92 77 77 82 82 84 84 84 89 89 89 89 88 99 98 128

Boone1 0.017 0.046* 40 25 25 30 30 31 31 31 36 36 36 36 36 46 46 75 Boone2 0.033* 0.071* 0.025 20 20 25 25 27 27 27 31 31 31 31 31 41 41 71 Prairie1 0.035* 0.058* 0.026* 0.031* 0 5 5 7 7 7 11 11 11 11 11 21 21 51 Prairie2 0.025* 0.044* 0.021 0.041* 0.018 5 5 7 7 7 12 11 11 11 11 21 21 51 Prairie3 0.0328* 0.056* 0.017 0.038* 0.024* 0.028* 0 2 2 2 6 6 6 6 6 16 16 46 Prairie4 0.029* 0.050* 0.015 0.038* 0.027* 0.030* 0.009 2 2 2 6 6 6 6 6 16 16 46

Prairie5 0.030* 0.058* 0.017 0.041* 0.036* 0.033* 0.013 0.013 0 0 5 5 5 5 5 15 15 44 Prairie6 0.031* 0.056* 0.026 0.029* 0.019 0.027* 0.02 0.024 0.034* 0 5 5 5 5 5 15 15 44 Prairie7 0.045* 0.072* 0.045* 0.044* 0.018 0.041* 0.042* 0.044* 0.053* 0.031* 4 4 4 4 4 14 14 44 Prairie8 0.033* 0.069* 0.021 0.023 0.009 0.027* 0.013 0.021 0.027* 0.016 0.024 0 0 0 0 10 10 39 Prairie9 0.022* 0.055* 0.015 0.026 0.029 0.030* 0.017 0.015 0.012 0.023 0.044* 0.02 0 0 0 10 10 39 Praire10 0.030* 0.043* 0.017 0.035* 0.018 0.019 0.023 0.024 0.026 0.028 0.032* 0.026 0.028 0 0 10 10 39 Praire11 0.041* 0.059* 0.028 0.042* 0.007 0.02 0.022 0.023 0.036* 0.021 0.025* 0.016 0.035* 0.019 0 10 10 39 Prairie12 0.037* 0.066* 0.019 0.037* 0.046* 0.043* 0.019 0.019 0.016 0.041 0.071* 0.032* 0.014 0.037* 0.048* 10 10 39 Prairie13 0.038* 0.074* 0.031 0.026 0.021 0.032* 0.027 0.036* 0.035* 0.014 0.033* 0.015 0.022 0.036 0.031 0.044* 0 29 Prairie14 0.043* 0.072* 0.028 0.036* 0.015 0.028* 0.018 0.021 0.028* 0.027 0.030* 0.012 0.025 0.028 0.018 0.029* 0.028 29 Prairie15 0.055* 0.093* 0.041* 0.040* 0.025 0.045* 0.022 0.031* 0.039* 0.033 0.036* 0.014 0.036* 0.045* 0.026* 0.045* 0.036* 0.018

Table 3.4. Deviations from Hardy-Weinberg equilibrium for all loci across sites generated in GenAlEx. Loci that were monomorphic at a site are denoted as mono. * indicates significant values (P ≤0.05).

NTA22 NTB42 NTC15 NTC81 NME178 NME208 Ppro48 NTB8 LCO4 Elk1 mono 0.562 mono 0.518 0.611 0.803 0.846 0.106 mono Rock1 mono 0.198 mono 0.889 0.670 0.906 0.137 0.682 0.906 Rocktrib1 mono 0.714 mono 0.914 0.137 0.885 0.503 0.188 0.875 Mound1 mono 0.001* mono 0.160 0.390 0.750 0.870 0.219 mono Mound2 mono 0.439 mono 0.221 0.127 0.782 0.816 0.210 0.885 Poplar1 mono 0.342 mono 0.556 0.698 0.063 0.436 0.081 0.996 Rocktrib2 mono 0.445 mono 0.149 0.069 0.717 0.160 0.433 0.890

Rocktrib3 mono 0.057 mono 0.358 0.075 0.063 0.442 0.310 mono 77

Rocktrib4 mono 0.990 0.000* 0.691 0.077 0.568 0.857 0.083 mono Rocktrib5 mono 0.533 mono 0.820 0.453 0.977 0.920 0.222 mono Rocktrib6 0.909 0.758 0.814 0.752 0.135 0.619 0.244 0.600 mono Rocktrib7 mono 0.198 mono 0.158 0.709 0.906 0.060 0.487 0.906 Eagle1 mono 0.154 0.814 0.160 0.783 0.257 0.128 0.614 0.649 Eagle2 mono 0.675 0.725 0.753 0.920 0.391 0.505 0.396 0.171 Boone1 mono 0.432 0.803 0.993 0.138 0.396 0.185 0.067 0.262 Boone2 mono 0.624 0.790 0.767 0.939 0.473 0.830 0.241 0.885 Prairie1 mono 0.382 0.619 0.995 0.695 0.808 0.929 0.814 0.235 Prairie2 0.717 0.114 0.939 0.841 0.525 0.809 0.009* 0.165 0.795 Prairie3 mono 0.450 0.967 0.984 0.096 0.245 0.745 0.392 0.240 Prairie4 mono 0.827 0.790 0.862 0.830 0.467 0.690 0.963 0.074 Prairie5 0.909 0.459 0.879 0.557 0.552 0.264 0.866 0.074 0.058 Prairie6 mono 0.360 0.901 0.468 0.450 0.476 0.076 0.105 0.341

Table 3.4 (continued)

Prairie7 mono 0.101 mono 0.657 0.232 0.118 0.733 0.670 0.913 Prairie8 mono 0.953 0.790 0.118 0.831 0.182 0.966 0.122 0.568 Prairie9 mono 0.523 0.985 0.450 0.270 0.648 0.243 0.583 0.143 Praire10 mono 0.194 0.354 0.262 0.216 0.985 0.248 0.153 0.170 Praire11 mono 0.443 0.536 0.402 0.457 0.929 0.670 0.062 0.858 Prairie12 mono 0.224 0.998 0.377 0.811 0.388 0.930 0.163 0.221 Prairie13 mono 0.710 mono 0.141 0.369 0.094 0.125 0.064 0.187 Prairie14 0.906 0.236 0.709 0.903 0.833 0.396 0.933 0.964 0.109 Prairie15 mono 0.771 0.904 0.717 0.472 0.536 0.919 0.176 0.447

Table 3.5 Deme membership for each site, and number of alleles per locus (NA), allelic richness (AR), observed heterozygosity (HO), expected heterozygosity (HE), inbreeding coefficient (FIS), and private alleles (AP) averaged across all loci in the Rock River and

Boone River basins.

Site Deme NA AR HO HE FIS AP Rock River Basin Elk1 Elk Creek 3.444 2.45 0.333 0.353 0.005 0 Rock1 Lower Rock 3.333 2.3 0.37 0.337 -0.095 0 Rocktrib1 Middle Rock 3.556 2.52 0.331 0.343 -0.008 0 Mound1 Middle Rock 3.444 2.3 0.416 0.352 -0.199 1 Mound2 Middle Rock 2.889 2.24 0.325 0.319 -0.075 1

Poplar1 Middle Rock 4 2.39 0.359 0.405 0.113 0 79

Rocktrib2 Headwater Rock 3.778 2.53 0.327 0.382 0.083 0 Rocktrib3 Headwater Rock 3.556 2.39 0.357 0.37 0.063 0 Rocktrib4 Headwater Rock 3.444 2.58 0.389 0.407 0.139 0 Rocktrib5 Headwater Rock 3.333 2.41 0.378 0.352 -0.086 0 Rocktrib6 Headwater Rock 4 2.68 0.394 0.407 0 0 Rocktrib7 Headwater Rock 3.333 2.21 0.316 0.314 -0.023 0 Boone River Basin Eagle1 Eagle Creek 3.556 2.56 0.446 0.416 -0.053 1 Eagle2 Eagle Creek 3 2.28 0.431 0.398 -0.075 0 Boone1 Headwater Boone 3.667 2.69 0.454 0.429 -0.067 0 Boone2 Lower Prairie 3.778 2.67 0.374 0.44 0.132 0 Prairie1 Lower Prairie 4.444 2.85 0.496 0.479 -0.051 0 Prairie2 Lower Prairie 3.667 2.68 0.502 0.482 -0.054 0 Prairie3 Lower Prairie 4.111 2.85 0.462 0.488 0.025 0 Prairie4 Lower Prairie 3.778 2.79 0.458 0.482 0.067 0

Table. 3.5 (continued)

Prairie5 Lower Prairie 4.111 2.86 0.468 0.503 0.064 0 Prairie6 Lower Prairie 4.222 2.78 0.394 0.456 0.106 0 Prairie7 Lower Prairie 3.556 2.63 0.433 0.428 -0.031 1 Prairie8 Lower Prairie 3.889 2.77 0.503 0.46 -0.116 0 Prairie9 Lower Prairie 3.778 2.77 0.451 0.465 0.015 0 Praire10 Lower Prairie 3.444 2.73 0.492 0.482 0.019 0 Praire11 Lower Prairie 4.222 2.87 0.458 0.484 0.044 0 Prairie12 Lower Prairie 3.778 2.73 0.46 0.451 -0.017 0 Prairie13 Lower Prairie 4.000 2.72 0.401 0.444 0.101 0 Prairie14 Lower Prairie 3.778 2.67 0.44 0.484 0.067 0 Prairie15 Headwater Prairie 3.778 2.69 0.401 0.451 0.087 0

Average 3.700 2.60 0.413 0.421 0.006 0.13

Table 3.6.measures of genetic diversity averaged between all oxbows and between all stream sites in both basins as well as the Rock and Boone river basins separately. * indicates significant differences between oxbows and streams (p ≤ 0.05).

Measure of Diversity Both Basins Rock Boone Oxbow Stream Oxbow Stream Oxbow Stream

Average Alleles per Locus (NA) 3.67 3.74 3.51 3.51 3.78 3.91

Allelic Richness (AR) 2.58 2.63 2.42 2.42 2.68 2.76

Expected Heterozygosity (HE) 0.421 0.424 0.386* 0.352* 0.452 0.469

Figures

Figure. 3.1 Topeka shiner sample site locations in the Rock River Basin (A), and Boone

River Basin (B). Black triangle represent stream sites and grey circles represent oxbow sites.

Figure 3.2. Isolation by Distance (IBD) using a standardized FST value (FST/(1-FST) and river distance between sites for the Rock River and Boone River basins.

Figure 3.3. First generation migrations detected by Geneclass2 (left) and BIMr (right) in the Rock River Basin. Migration direction indicated by arrow and arrow color.

Figure 3.4. Directional migration rates of deme in the Rock River Basin. EC = Elk Creek, LR = Lower Rock River, MR = Middle

Rock River, HWR = Headwater Rock River.

Figure 3.5. First generation Topeka shiner migrants from Geneclass2 (left) and BIMr (right) in the Boone River Basin. Migration direction indicated by arrow and arrow color.

Figure 3.6. Directional migration rates of Topeka shiner demes in the Boone River Basin. EC = Eagle Creek, HWB = Headwater

Boone River, LPC = Lower Prairie Creek, HPC = Headwater Prairie Creek.

CHAPTER 4. GENERAL CONCLUSIONS

Summary

The Topeka shiner has experienced an overall decline in distribution and abundance due to stream alteration and loss of preferred habitat, leaving this species scattered across its once more widespread historical range. Given these drastic changes to distribution, Topeka shiner connectivity has been reduced on multiple geographic scales leading to is listing as endangered. This study contributed to Topeka shiner conservation by characterizing genetic diversity, population structure, and connectivity across the range as well as the metapopulation dynamics of these populations to inform management decisions.

Sites across of the range Topeka shiners generally showed low to moderate diversity.

Only one site showed low diversity along with a bottleneck. Cottonwood1 in the Arkansas

River basin may have also experienced a founder effect due to a small founding stock, but also appears to be genetically isolated leading to low diversity. Further analyses of diversity in the surrounding population is needed to assess its viability. Overall, the genetic diversity of these sites is not currently low, but a further reduction of population sizes and distribution could result of future losses of diversity.

Analysis of population structure revealed eight distinct populations. Topeka shiners currently exist in geographically isolated and small (≤ HUC8 basin) populations across most of the range, including the Boone, Norther Raccoon, Sugar Creek, Kansas, Willow and

Arkansas River basins. There was one population along the Missouri River that was much larger than the others. This population covers multiple neighboring basins in South Dakota and Minnesota that appear to have remained stable over time and show genetic similarity due to current or recent gene flow. Moniteau Creek in Missouri also showed great genetic

89 similarity with this population, but given the great river distance between these sites, it is likely that this is also a separate population. A further analysis using multiple markers may provide more information on the nature of this connection. This range-wide study revealed that there was much greater genetic connectivity of Topeka shiner populations before human settlement, which may have existed in a more consistent distribution than current populations. Estimates of historical migration suggest high amount of gene flow between populations throughout the range whereas contemporarily migration between these populations is virtually nonexistent. Consequently, Topeka shiners now exist in isolated populations. Therefore, losses to any of these populations will result in a loss in overall genetic diversity.

Often when a species has lost a substantial amount of its range, reintroduction is used to expand the range and bolster connectivity (Hayes and Banish 2017). Topeka shiners have been reintroduced in Northern Missouri into watersheds with suitable habitats (Weichman

2014). A brood stock was obtained from the only stream known to contain Topeka shiners in

Sugar Creek in northern Missouri and use to produce Topeka shiners and initial stocking began in streams in 2013 (Wiechman 2014). Of the three basins stocked, all showed evidence of reproduction and expansion (Thornhill 2015; Wiechman 2015) but further genetic analysis could determine future viability of these populations. Although reintroduction can be an effective technique to promote genetic connectivity (Malone et al. 2018), the genetic fitness of the source population will be an important consideration (Schwartz et al. 2007; Weeks et al. 2011). Translocation between populations can affect localized adaption (Fitzpatrick et al.

2015; Lean et al. 2017) and is not recommended.

Although there is isolation between populations on a range-wide level, in the basins examined, Topeka shiners exist as highly connected metapopulations. Sites in the Boone

River and Rock River basins were not in panmixia, but instead formed multiple genetically and geographically distinct demes. Topeka shiners prefer slow moving instream as well as off channel habitat. Oxbow did not appear to have a great influence on the formation of these off-channel demes. Given the dynamic existence of Topeka shiner populations in oxbows, there is potential for further research. Mark-recapture and tracking studies paired with genetic analysis have shown the potential to identify source-sink dynamics of Natterjack toads in wetland ponds (Lowe and Allendorf 2010) and a similar approach may be applicable to determine the direct effect of oxbows on metapopulation structure in Topeka shiners.

There has been reduction and fragmentation of this habitat, leaving these demes fragmented.

Theses demes were not completely isolated, however, as there was evidence of first generation migration between demes which is characteristic of a metapopulation. In both basins, there were some demes which appear to act as sources that were exporting individuals that then would colonize new areas or migrate to other demes. The migration between demes keeps Topeka shiners within these populations genetically connected, allowing Topeka shiners to persist despite frequent drought in some areas of these basins. Sinks, which are demes that rely on migration inputs to persist, are likely to exist in these dynamic systems.

Sinks were unable to be identified using limited temporal sampling, but a study encompassing multi generations would address persistence in these demes and determine sinks.

The Topeka shiner was once widely distributed and well connected throughout the

Midwest, but stream alteration has reduced the Topeka shiners to multiple fragmented and

91 genetically isolated populations with low to moderate genetic diversity. Within these isolated basins, Topeka shiners act as metapopulations, forming demes around fragmented habitat that experience some migration. Although full reestablishment of historical connectivity is unlikely due to large geographic distances and anthropogenic barriers between populations, increasing the amount and distribution of suitable habitat will allow colonization and persistence of Topeka shiners demes within basins. The maintenance of these populations preserves overall genetic diversity and therefore adaptive potential that is crucial to the survival of Topeka shiners in a changing environment.

Literature Cited

Fitzpatrick, S. W., J. C. Gerberich, J. A. Kronenberger, L. M. Angeloni, and W. C. Funk. 2015. Locally adapted traits maintained in the face of high gene flow. Ecology Letters 18:37-47.

Hayes, M. F., and N. P. Banish. 2017. Translocation and reintroduction of native fishes: a review of Bull trout Salvelinus confluentus with applications for future reintroductions. Endangered Species Research 34:191-209.

Lowe, W. H., and F. W. Allendorf. 2010. What can genetics tell us about population connectivity? Molecular Ecology 19:3038-3051.

Malone, E. W., J. S. Perkin, B. M. Leckie, M. A. Kulp, C. R. Hurt, D. M. Walker. 2018. Which species, how many, and from where: Integrating habitat suitability, population genomics, and abundance estimates into species reintroduction planning. Global Change Biology 24:3729-3748.

Schwartz, M. K., G. Luikart, and R. S. Waples. 2007. Genetic monitoring as a promising tool for conservation and management. Trends in Ecology & Evolution. 22:25-33.

Thornhill, D. R. 2015. Monitoring of Non-essential Experimental Population Topeka shiners in the Spring Creek Watershed. Missouri Department of Conservation. 3p.

Weeks, A. R., C. M. Sgro, A. G. Young, R. Frankham, N. J. Mitchell, K. A. Miller, and M. F. Breed. 2011. Assessing the benefits and risks of translocations in changing environments: A genetic perspective. Evolutionary Applications 4:709-725.

Wiechman, J. 2014. Monitoring Experimental Populations of Topeka shiners in Little Creek and East Fork Big Muddy Creek, Northwest Missouri: 2014 Update. Missouri Department of Conservation 15p.

Wiechman, J. 2015. Monitoring Experimental Populations of Topeka shiners in the Little Creek and East Fork Big Muddy Creek. Missouri Department of Conservation.

APPENDIX. SUMMARY INFORMATION FOR MICROSATELLITE LOCI

APPENDIX. Table A1. GenBank accession number, primers, annealing temperature (A T), number of alleles across all study sites,

Original species that loci was developed for, and citation for loci used. * monomorphic loci were exclude from analysis.

GenBank Accession A T Number of Locus Number Primers (5'-3') (°C) Alleles Original species Source Anderson and Sarver NTA22 DQ659300 F: GCAACGCTGGAATAACATCT 49 3 Topeka shiner 2008 R: CTGTTCCCCTTGTCTTGTCG F: Anderson and Sarver NTB8 DQ659301 ACGGTAAGTGAAAATAAATC 49 30 Topeka shiner 2008 R: GTCTGCTCGCTATTTATAAA Anderson and Sarver

NTB42 DQ659302 F: ACAAGGTGTGGTTACTCC 49 7 Topeka shiner 2008 93

R: CTATTCCTATGGCAGCAA Anderson and Sarver NTC15 DQ659303 F: AGTGACAGATGATTGACAGC 49 6 Topeka shiner 2008 R: CTCGTTTAAGGTGCGTTTT Anderson and Sarver NTC39* DQ659304 F: GGGTTGAATGTGGATGTGTT 49 1 Topeka shiner 2008 R: TTAGCAAGGGGTTTCAAACT Anderson and Sarver NTC81 DQ659305 F: GTTTGCATCTTTTCTTCATT 55 7 Topeka shiner 2008 R: ATTTTACAGCAGACACAGAA F: Anderson and Sarver NTD10* DQ659306 AGCAAGGGCAGAGGATTAGT 49 1 Topeka shiner 2008 R: CCCCTCTGTCTTTAGTAT Anderson and Sarver NTF43* DQ659307 F: TGCTCCTCATCCTCTCTGT 49 1 Topeka shiner 2008 R: ACTTTATCTCCACCGGCAC

Table A1 (continued)

Ppro48 AY254350 F: TGCTCTGCTCTCCTGCGTGTCATT 55 7 Fathead minnow Bessert and Orti 2003 R: CAGCCTCGGCGGTGTTGTTGC Ppro118* AY254352 F: CCGGATGCACTGGTGGAGAAAA 55 1 Fathead minnow Bessert and Orti 2003 R: CCAGCAATCATAGCAGGCAGGAAC Lco4 AY318780 F: ATCAGGTCAGGGGTGTCACG 55 8 Common shiner Turner et al. 2004 R: TGTTTATTTGGGGTCTGTGT NME178 AF532582 F: TCAAACCCTACAGACAGCAAGACT 55 10 Cape Fear shiner Burridge and Gold 2003 R: TTCTCAGGGGGCTCCAACAAG NME208 AF532584 F: AAAAAGGCCTCCCAGTGC 55 4 Cape Fear shiner Burridge and Gold 2003 R: AATTATATGTCGGTGACCAGATTG 94