The use of regional phylogenies in exploring the structure of plant assemblages

Tammy L. Elliott

Doctor of Philosophy

Department of Biology

McGill University

Montr´eal, Qu´ebec, Canada

2015-09-015

A thesis submitted to McGill University in partial fulfillment of the requirements of the degree of Doctor of Philosophy

c Copyright Tammy L. Elliott, 2015 All rights reserved Dedication

I dedicate this thesis to my parents, who sadly both left this world much to early. I like to dream that you are both enjoying your time together in a place with no worries, where you can enjoy all of the wonderful things in life.

Dad—Although you left us when we were so young, I daily cherish the special times the two of us spent together. The memories of exploring the countryside, visiting neighbours, caring for the pigs and skipping school to fish are always close to my heart.

Mom—I miss your strength, interesting perspective (albeit humorously pessimistic), no-nonsense attitude towards life and listening ear. I hope that you are finding ways to enjoy your grandchildren and tend your beautiful gardens. I would like to assure you that yes—one day I will have a full-time job.

If Roses grow in Heaven Lord, please pick a bunch for me. Place them in my Mother’s arms and tell her they’re from me.

Tell her that I love her and miss her, and when she turns to smile, place a kiss upon her cheek and hold her for awhile

Because remembering her is easy, I do it every day, but there’s an ache within my heart that will never go away.

Author Unknown Acknowledgements

This thesis is the result of the input of several different people to whom I am grateful. First, I would like to thank my supervisor, Jonathan Davies, for accepting me into his lab. I feel that he has offered the right amount of guidance over the last few years, allowing me to grow by making my own decisions. Second, I would like to thank my Supervisory Committee members—Sylvie de Blois and Catherine Potvin for providing advice on experimental design, analytical methods and overall thesis direction and structure. In addition, I thank Martin Lechowicz and Marcia Waterway for their help during the early stages of my doctoral studies, as well as Pierre Legendre and Guillaume Larocque for their assistance with statistical methods. Many thanks are extended to my labmates for the support and camaraderie they have offered in and outside of the office over the last few years. In particular, I thank Eraclis Araclides, Chelsea Chisholm, Ria Ghai, Jean-Phillipe Lessard, Ignacio Morales-Castilla, Will Pearse and Stephanie Shooner for reviewing and offering comments on my work. Additional thanks are offered to Eraclis Araclides, Frederic Boivin, Maxwell Farrell and Will Pearse for helping with my analyses, as well as Michael Becker for his awesome filmmaking abilities. I also thank Malie Lessard-Therrien for sharing her data. I acknowledge the perseverance of Liam Harris, Genevi`eve Lajoie and Kyle Martins for enduring the black flies and long work days in Schefferville. My gratitude is offered to Peter White for originally creating the Mnt. Irony sampling grid and to Oksana Choulik, Neil Ednie and the McGill Subarctic Research Station for providing logistical support while in the Schefferville area. I thank David Maneli and Stephanie Shooner for their help in the field at Mont St. Hilaire, as well as Robin Beaus´ejour, M´elanie Lapointe, Arold Lavoie and Catherine Polcz for helping to locate additional species on the ‘mountain’. This research would not be possible without the support of the McGill Herbarium and the Marie-Victorin Herbarium. Thanks are offered to Nadia Cavallin for her help and iii support in the field, herbarium and outside of work. I am very grateful for the advice of Geoffrey Hall, whose knowledge of the local flora improved my collection efforts at Mont St. Hilaire and led to many stimulating botanical conversations. In addition, I thank Kyle Martins and Marcia Waterway for their help at the McGill Herbarium. I also thank Kathryn Flinn, Rapha¨elle Fr´echon and countless volunteers for their help preparing my hundreds of herbarium specimens. My appreciation is also extended to George Argus, Paul Catling, Jim Phipps, Anton Reznicek and Jeff Saarela for kindly offering their plant identification services. I thank the African Centre for DNA Barcoding research for processing my DNA sam- ples. In particular, I thank Olivier Maurin and Michelle Van der Bank for responding to my countless e-mail messages and for ensuring that all samples were correctly processed. My gratitude is extended to the several funding agencies whose financial support allowed me to pursue my research interests. Specifically, I thank the Garfield Weston Foundation, the Natural Sciences and Engineering Research Council of Canada (NSERC), the Qu´ebec Centre for Biodiversity Science (QCBS) and McGill University for their generous scholarships. In addition, I thank the Canadian Polar Commission for logistical support in the field in the form of Northern Scientific Training Program (NSTP) grants. Additional thanks are offered to the many anonymous reviewers who have provided valuable comments on the manuscripts included within this thesis. I am also grateful for the help of other researchers for providing comments on my work. In particular, I thank Frieda Beaugrand, Heather Collins, Barnabas Daru, Vincent Fug`ere, Kiyoko Gotanda, Pascal Kropf, Patrick McCamphill, S´ebastien Portalier and Melissa Ward for their reviews. Finally, I thank my family for standing by my side as I have embarked on this multi- year challenging adventure.

iv Abstract

Unravelling the complexities of biological communities has long interested community ecologists. Molecular, computational and conceptual advances in the last fifteen years have led to the increasing incorporation of phylogenetic information into ecological anal- yses, creating a new field of ‘community phylogenetics’. Although the field of community phylogenetics has improved our understanding of the processes determining community composition, more recently the field has been challenged because of its reliance on several critical assumptions. In this thesis, I address some of these critiques by exploring novel empirical and analytical approaches. In Chapter 2, I describe the major challenges to sampling a comprehensive species pool, essential for generating regional phylogenies and testing community assembly mechanisms. I use a case study based on my work at Mont St. Hilaire, Qu´ebec to illustrate these challenges, and I make several recommendations for sampling that can be used by other researchers pursuing similar efforts in the future. I switch focus from regional species pools to plant communities in Chapters 3 and 4, where I combine co-occurrence and phylogenetic data on sedges (i.e. plants of the Cyperaceae family) near Schefferville, Qu´ebec to address assumptions related to species coexistence. I use an individual-based focal sampling approach in Chapter 3 to show that the phyloge- netic neighbourhood of a plant varies with the clade membership of the species, addressing the critique that it is individuals of a species and not communities that are ‘filtered’ by environment. I explore the assumption that competitive exclusion is most probable between closely related species in Chapter 4 by examining whether specialists are better competitors than generalists. My results suggest that niche width differences translate into differences in competitive abilities and that co-occurrence is more likely between distantly related species with differing niche widths, adding further insights into the relationship between phylogenetic relatedness and competitive abilities. Finally, in Chapter 5 I eval- uate the effectiveness of phylogenetic beta diversity methods for delineating ecological v boundaries. Previous work has focused on regional and global scales, whereas similar methods might not be as effective at differentiating local scale patches. As I demonstrate in this thesis, the field of community phylogenetics will remain relevant for unravelling the complexities of biological communities by using novel approaches.

vi R´esum´e

D´emˆeler la complexit´e des communaut´es biologiques a depuis longtemps int´eress´eles ´ecologistes des communaut´es. Les avanc´ees en biologie mol´eculaire, en informatique et les progr`es conceptuels durant les quinze derni`eres ann´ees ont conduita ` l’int´egration croissante de l’information phylog´en´etique dans les analyses ´ecologiques, cr´eant le nouveau champ de la phylog´enie des communaut´es. Bien que le domaine de la phylog´enie des communaut´es ait am´elior´e notre compr´ehension des processus qui d´eterminent la composition des communaut´es, plus r´ecemment, ce domaine a ´et´e contest´eenraisonde sa d´ependance `al’´egard de plusieurs hypoth`eses critiques. Dans cette th`ese, je r´eponds `a plusieurs de ces critiques en explorant de nouvelles approches empiriques et analytiques. Dans le chapitre 2, je d´ecris les principaux obstaclesal’´ ` echantillonnage d’un groupe com- plet d’esp`eces, essentiel pour g´en´erer des phylog´enies r´egionales et tester les m´ecanismes d’assemblage de la communaut´e. J’utilise une ´etude de cas bas´ee sur mon travail `a Mont-Saint-Hilaire, au Qu´ebec, pour illustrer ces probl`emes, et je fais plusieurs recom- mandations pour l’´echantillonnage qui peuvent ˆetre utilis´ees par d’autres chercheurs qui poursuivront des efforts similaires dans l’avenir. Je passe ensuite des groupes r´egionaux d’esp`eces aux communaut´es v´eg´etales dans les chapitres 3 et 4, o`u je combine des donn´ees de cooccurrence et de phylog´enie sur les plantes de la famille des Cyp´erac´ees pr`es de Schefferville, au Qu´ebec, pour r´epondre aux hypoth`eses relatives `a la coexistence des esp`eces. J’utilise une m´ethode d’´echantillonnage focal bas´e sur l’individu dans le chapitre 3 pour montrer que la proximit´ephylog´en´etique d’une plante varie avec la composition de clade de l’esp`ece, r´epondant `a la critique que ce sont les individus d’une esp`ece et non pas les communaut´es qui sont filtr´es dans les communaut´es. J’explore l’hypoth`ese que l’exclusion comp´etitive est plus probable entre les esp`eces ´etroitement li´ees dans le chapitre 4 en examinant si les sp´ecialistes sont de meilleurs comp´etiteurs que les g´en´eralistes. Mes r´esultats du chapitre 4 sugg`erent que les diff´erences de largeur de niche se traduisent par vii des diff´erences dans les capacit´es comp´etitives et que la cooccurrence est plus probable entre les esp`eces apparent´ees de loin avec diff´erentes largeurs de niche, ce qui jette un nou- vel ´eclairage sur la relation entre la parent´ephylog´en´etique et les capacit´es comp´etitives. Enfin, au chapitre 5, j’´evalue l’efficacit´edesm´ethodes phylog´en´etiques de diversit´ebˆeta pour d´elimiter les fronti`eres ´ecologiques. Des travaux ant´erieurs ont mis l’accent sur des ´echelles r´egionales et globales, alors que les m´ethodes similaires pourraient ne pas ˆetre aussi efficace `a diff´erencier les patches `al’´echelle locale. Comme je le d´emontre dans cette th`ese, l’utilisation de nouvelles approches du domaine de la phylog´enie des communaut´es restera pertinente pour d´emˆeler la complexit´e des communaut´es biologiques.

viii Table of Contents

Dedication ...... ii Acknowledgements ...... iii Abstract ...... v R´esum´e ...... vii

List of Tables ...... xiii

List of Figures ...... xiv Preface ...... xvi

Thesisformatandstyle...... xvi Contributions of co-authors ...... xvii Original contributions to knowledge ...... xviii

1 Introduction ...... 1

1.1 Community ecology ...... 2 1.1.1 Community assembly rules ...... 3 1.1.2Betadiversity...... 5 1.2 Phylogenetic community assembly ...... 6 1.3 Phylogeneticbetadiversity...... 9 1.4 Community phylogenetics: summary ...... 10 1.5 Challenges to community phylogenetics ...... 10 1.6 Thesisoutline...... 12

2 Challenges to barcoding an entire flora ...... 21

2.1 Abstract...... 22 2.2 Introduction...... 23 2.3 Barcoding a complete flora—Mont St. Hilaire as a case study ...... 26 2.4 Consequences of incomplete sampling ...... 32 2.5 Lessons learned: suggestions for future studies ...... 33 Linking Statement #1 ...... 41

3 Contrasting lineage-specific patterns conceal community phyloge- netic structure in larger clades ...... 42

3.1 Abstract...... 43 ix 3.2 Introduction...... 44 3.3 Methods...... 46 3.3.1Studysystem...... 46 3.3.2Phylogenyreconstruction...... 48 3.3.3Speciesco-occurrencemetrics...... 48 3.4 Results...... 50 3.4.1Aggregatedcommunity-levelmetrics...... 50 3.4.2 Phylogenetic clustering in Trichophorum and overdispersion in Eriophorum ...... 51 3.4.3 Phylogenetic clustering in the Core Carex and overdispersion in the Vignea clade...... 51 3.4.4 Reduced species pool enhances contrasting patterns of phyloge- netic clustering and overdisperion in Carex ...... 52 3.5 Discussion...... 53 3.6 Conclusion...... 56 Linking Statement #2 ...... 64

4 Competitive trade-offs and species co-occurrence: integrating co- existence theory and community phylogenetics ...... 65

4.1 Abstract...... 66 4.2 Introduction...... 67 4.3 Methods...... 69 4.3.1Studysystemandexperimentaldesign...... 69 4.3.2Analyses...... 69 4.4 Results...... 71 4.4.1Nichewidth...... 71 4.4.2Pairwisespeciescomparisons...... 71 4.5 Discussion...... 72 Linking Statement #3 ...... 78

5 Delineating ecological boundaries at local scales: a comparison of phylogenetic versus species beta diversity metrics ...... 79

5.1 Abstract...... 80 5.2 Introduction...... 81 5.3 Methods...... 84 5.3.1 Sampling ...... 84 5.3.2Phylogeneticreconstruction...... 85

x 5.3.3 Environmental and beta diversity distances between plots . . . . . 87 5.3.4ClusteringplotsbyfuzzyK-means ...... 87 5.4 Results...... 89 5.4.1Speciesdiversityandphylogeneticreconstruction...... 89 5.4.2Optimalnumberofclusters...... 89 5.4.3Comparisonofclusteringpatterns...... 89 5.5 Discussion...... 91 5.5.1 The possible role of biome shifting ...... 94 5.5.2Justificationofchosenmetrics...... 95 5.5.3Conclusion...... 96

6 General conclusion ...... 102

6.1 Conclusion...... 103 List of References ...... 108

Appendices ...... 137 Appendix A: Supporting Information — Chapter 2 ...... 138 Appendix A1: Contributors to the WG1.2 Land Plants iBOL Working Group based on voluntary survey conducted in March and April 2013 ...... 139 Appendix A2: Updated terrestrial vascular plant species list for the Gault Reserve at Mont St. Hilaire, Qu´ebec as of April 9, 2014 ..... 140 Appendix A3: Time and cost estimates for two regional barcoding efforts in Qu´ebec,Canada...... 170 Appendix B: Supporting Information — Chapter 3 ...... 171 Appendix B1: Distribution of 700 focal sedge plots across nine different habitat types near Schefferville, Qu´ebec in the Canadian subarc- tic...... 172 Appendix B2: Locating and marking focal sedge plants in the Schefferville area...... 174 Appendix B3: Detailed description of methods for phylogenetic recon- struction...... 177 Appendix B4: Regional species pool of Cyperaceae from the subarctic Qu´ebecandadjacentLabradorregionofCanada...... 180 Appendix B5: Sequence data used to create a regional phylogeny of sedges found near Schefferville, in subarctic Qu´ebec ...... 183

xi Appendix B6: Phylogeny of the Cyperaceae species of the subarctic Qu´ebec andadjacentLabradorregionofCanada ...... 188 Appendix B7: Phylogeny of the Cyperaceae species of the subarctic Qu´ebec and adjacent Labrador region of Canada based on bayesian anal- ysis with blue node bars indicating estimated divergence times . . 189 Appendix B8: Mean net relatedness index to focal plant (FNRI) for the 35 Cyperaceae and 29 Carex focalspecies...... 190 Appendix B9: Mean nearest taxon index to focal plant (FNTI) for the 35 Cyperaceae and 29 Carex focalspecies...... 192 Appendix B10: Average species richness per focal species and sampling frequency of the 35 focal sedge species included in the analyses . 194 Appendix B11: Mean net related index to focal (FNRI) and nearest taxon index to focal (FNTI) compared to average species richness . . . 196 Appendix B12: Net relatedness to focal (FNRI) comparisons among dif- ferenthabitattypes...... 201 Appendix C: Supporting Information — Chapter 4 ...... 209 Appendix C1: Descriptions of four different specialization-generalization indices considered in the study ...... 210 Appendix C2: Justification for exclusion of one focal species from the fi- nalanalyses ...... 213 Appendix C3: Results from ordinary least squares regressions among pair- wise phylogenetic distances and the co-occurrence of generalist andspecialistspecies ...... 219 Appendix D: Supporting Information — Chapter 5 ...... 222 Appendix D1: Taxonomic sampling and background information used in phylogeneticreconstruction...... 223 Appendix D2: Supplementary five cluster analyses ...... 246 Appendix D3: Supplementary three and seven cluster analyses ...... 254

xii List of Tables

Table page 1–1 The combinations of phylogenetic trait distribution and ecological processes that result in the expected distributions of sample taxa at a site ...... 15 3–1 Mean net relatedness index (NRI) and mean nearest taxon index (NTI) val- ues as well as net related index to focal (FNRI) and nearest taxon index to focal (FNTI) for all Cyperaceae plots combined ...... 57 3–2 Mean net relatedness index to focal (FNRI) and nearest taxon index to focal (FNTI)forCyperaceaefocalplantsdividedintomajorclades...... 58 3–3 Differences in mean net relatedness index to focal (FNRI) and nearest taxon index to focal (FNTI) for the different clades of Cyperaceae and Carex in- cluded in the study ...... 59 3–4 Mean net relatedness index to focal (FNRI) and nearest taxon index to focal (FNTI) for Carex focalplantsdividedintomajorclades...... 61 5–1 Comparison between clusters using Gower dissimilarity (SiteEnv), beta di- versity(BD)andphylogeneticbetadiversitydistances(PhBD)...... 97

xiii List of Figures

Figure page 1–1 A hypothesis illustrating community assembly rules with ‘filtering’ ...... 16 1–2 A diagrammatic example illustrating the calculation of phylogenetic beta di- versity with the PhyloSor index ...... 17 1–3Locationofthetwofieldstudyareasforthisthesis...... 18 1–4 Vegetation on Mont St. Hilaire, where sampling for Chapter 2 was conducted 19 1–5 Four different habitats common throughout the Schefferville region, where sampling for Chapters 3, 4 and 5 was conducted ...... 20 2–1 Web of Science search results of published papers on barcoding plants .... 37 2–2 Results of 2012 and 2013 sampling efforts of the vascular plant species on the Gault Nature Reserve, Qu´ebec, based on a pre-existing species list of the reserve ...... 38 2–3 Location of sampling points for plant specimens from Mont St. Hilaire, Qu´ebec, during the 2012 and 2013 field seasons ...... 39 2–4 Number of terrestrial plant species collected per month during the 2012 and 2013 field seasons at Mont St. Hilaire, Qu´ebec...... 40 3–1 Net relatedness index to focal plant (FNRI) based on presence/absence data forthe35Cyperaceaefocalspeciesdividedintomajorclades...... 62 3–2 Mean FNRI values mapped per focal species weighted by percent cover es- timates for (a) all co-occurring Cyperaceae in the study and (b) only co- occurring Carexeachco-occurringspecies...... 63 4–1 Predicted relationship between phylogenetic distance and difference in niche width...... 75 4–2 Relationship between species co-occurrences and differences in niche width andphylogeneticdistances...... 76 4–3 Relationship among phylogenetic distance, niche width and species co-occurrence fitted with a generalized additive model ...... 77 5–1 Clusters of the 176 plots on Mont Irony, Labrador, shaded by cluster mem- bership...... 98 5–2Spatialassociationsinplotassignmentstoclusters...... 99

xiv 5–3 The proportion of shared species and branch lengths within five Gower dis- similarity (SiteEnv) clusters ...... 100 5–4 Species contributions to beta diversity values (SCBD) ...... 101

xv Preface

Thesis format and style

This thesis is written in a manuscript-based format, and consists of four manuscripts of which I am the lead author. The data for the chapters were collected during field research that I conducted near Schefferville and on Mont St. Hilaire, Qu´ebec. The style of each chapter follows the scientific journal Ecology Letters.

Chapter 2: Elliott, T.L. & Davies, T.J. (2014). Challenges to barcoding an entire flora. Molecular Ecology Resources, 14, 883-891.

Chapter 3: Elliott, T.L., Waterway, M.J., & Davies, T.J. (accepted for publication in Journal of Vegetation Science). Contrasting lineage-specific patterns conceal community phylogenetic structure in larger clades.

Chapter 4: Elliott, T.L. & Davies, T.J. (submitted to Ecology). Competitive trade-offs and species co-occurrence: integrating coexistence theory and community phylogenetics.

Chapter 5: Elliott, T.L. & Davies, T.J. (submitted to Ecography). Delineating ecological boundaries at local scales: a comparison of phylogenetic versus species beta diversity metrics.

xvi Contributions of co-authors

This thesis is the result of my own research. I am the primary author for all of the chapters, in close collaboration with my supervisor – Jonathan Davies. I collected and analyzed the data, as well as wrote the first draft of each chapter. Jonathan Davies helped develop the ideas and contributed to each chapter’s revisions. Chapter 3 benefited from the help of an additional co-author, which is explained in more detail below:

Chapter 2: Authors: Tammy L. Elliott and T. Jonathan Davies Tammy Elliott and Jonathan Davies conceived the idea for this chapter. Tammy Elliott gathered and analyzed the data, as well as wrote the first draft. Tammy Elliott and Jonathan Davies contributed to the chapter’s revisions.

Chapter 3: Authors: Tammy L. Elliott, Marcia J. Waterway and T. Jonathan Davies Tammy Elliott collected and analyzed the data, as well as wrote the first draft of the chapter. Marcia Waterway helped develop the study design and create the phylogenetic reconstruction. Tammy Elliott and Jonathan Davies worked on the chapter’s revisions.

Chapter 4: Authors: Tammy L. Elliott and T. Jonathan Davies Tammy Elliott collected and analyzed the data, in addition to writing the first draft of the manuscript. Tammy Elliott and Jonathan Davies contributed to the chapter’s revisions.

Chapter 5: Authors: Tammy L. Elliott and T. Jonathan Davies Tammy Elliott and Jonathan Davies developed the design for this study. Tammy Elliott gathered and analyzed the data, as well as wrote the first draft of the chapter. Tammy Elliott and Jonathan Davies worked on the chapter’s revisions.

xvii Original contributions to knowledge

In recent years, there have been large efforts to create comprehensive DNA barcode libraries. Whereas much work has focused on their potential uses and the selection of specific DNA markers, few papers have discussed the practical limitations to generating comprehensive libraries. In Chapter 2, I discuss several challenges to the creation of a DNA library based on a case study of the vascular plants of Mont St. Hilaire, Qu´ebec. I also contribute to the discipline by making several recommendations that can be used to guide similar efforts in the future. In Chapter 3, I use a novel, individual-based approach to evaluate lineage-specific co- occurrence patterns in closely related plants. While previous studies have used community wide metrics, my unique approach reveals lineage-specific patterns in phylogenetic structure that are obscured when metrics are averaged across species. In addition, my results are some of the first to suggest that different plant lineages vary in their response to community level processes such as environmental filtering and competition. A long-standing ecological question is whether specialist or generalist species make better competitors. Although this question has been addressed using conceptual models in the past, it has been difficult to examine empirically in natural communities, where specialists and generalists co-occur. In Chapter 4, I introduce a novel approach based on co-occurrence and phylogenetic relatedness to evaluate whether species with different niche breadths (i.e. specialists and generalists) differ in their competitive abilities. I also provide suggestions for future research on this question. In Chapter 5, I provide one of the few studies comparing the effectiveness of beta diversity and phylogenetic beta diversity metrics in delineating the boundary separating the boreal forest and arctic tundra biomes. In contrast to recently published work that has suggested that phylogenetic beta diversity is a powerful metric for delineating biome boundaries at large spatial scales, my work shows that phylogenetic beta diversity is xviii not more informative than traditional beta diversity indices in local scale boundary delineation. My work suggests for the first time that spatial scale is an important factor to consider when choosing between traditional beta diversity and more novel phylogenetic beta diversity metrics.

xix CHAPTER 1

Introduction

1 1.1. Community ecology

1.1 Community ecology

Unravelling the complexity of biological communities presents a daunting task for commu- nity ecologists. The difficulties in understanding this complexity beg the question—what motivates researchers to pursue this goal? For some researchers, the goal is prediction, whether it be to predict the impacts of global change on biological communities (Sim- berloff 2004) or the number and abundances of species that can coexist in a given area (Keddy 1992; Lawton 1999; Sutherland et al. 2013; Roughgarden et al. 1986). For others, the goal is to establish links between the patterns observed in biological communities and the processes causing these patterns (Loreau 2009; Vellend 2010), and yet for other re- searchers, the goal is understanding how nature ‘works’ (Simberloff 2004). These dissimilar goals have often been pursued independently, resulting in differences in terminology and approaches within the field of community ecology. The precise definition of a biological community changed over the 20th century, with the conflicting views of two North American plant ecologists dominating the first half of the century. Clements (1916) first argued that a community was a super ‘organism’ where the activities of different species combined to create integrated functions, accompanied by a definite birth, maturity and death of the community. This view was contrasted by a more individualistic concept of the community proposed by Gleason (1926), who stated that individual species pursued private agendas. Throughout the latter half of the 20th century the concept of the community became more phenomenological: defined as a group of interacting organisms in the same location (Whittaker 1975). This definition persists into the current century, with the common view that a community is a group of organisms representing multiple species living in a specified place and time (Vellend 2010). In addition to changes in the definition of a ‘community’, through the 20th century community ecology also shifted from a largely descriptive to a more predictive discipline

2 1.1. Community ecology within a hypothesis-driven framework. In particular, earlier work often focused on detailed descriptions of the communities present across a region (e.g. Weaver & Fitzpatrick 1934; Tisdale 1947); although it is worth mentioning that the recognition of different levels of diversity (alpha, beta and gamma) by Whittaker (1960) helped propel the field from simple description and recognized the importance of spatial scale. The heavy focus on community ecology as a descriptive discipline and its struggle to become more predictive was captured well by Richard Lewontin who wrote (1974, pp. 8), “[w]e cannot go out and describe the world in any old way we please and then sit back and demand that an explanatory and predictive theory be built on that description. The description may be dynamically insufficient. Such is the agony of community ecology.” Increasing interest in community assembly rules, most famously by Diamond (1975), represented a major paradigm shift which helped push the discipline to become a more predictive science.

1.1.1 Community assembly rules

Community ecology underwent a major transition following observations of Diamond (1975), who proposed a set of ‘assembly rules’ explaining the co-occurrence of 147 species of land birds on the islands of the Bismarck Archipelago near New Guinea. Based on his detailed observations of the bird fauna on these islands, Diamond (1975) outlined seven rules that guided the co-occurrence of the bird species, assuming that interspecific competition was the dominant process across the islands. Diamond’s work suggested that community assembly was guided by non-random processes, such as competition, with certain rules that could be used in predictive modelling, although his work soon became a source of contention amongst community ecologists (e.g Connor & Simberloff 1979; Strong et al. 1979; Grant & Abbott 1980).

3 1.1. Community ecology

The debate arising from Diamond’s assembly rules was heated, with both strong opposition and support. The seven assembly rules were criticized as being either tautolo- gies, trivialities or patterns that would be expected if species were distributed randomly (Connor & Simberloff 1979). It was argued that null models should have been used by Diamond (1975) to determine if the patterns that he ascribed to competition could have been generated by models that were free of competition (Connor & Simberloff 1979; Strong et al. 1979; Simberloff & Connor 1981). However, null models were criticized for several reasons, such as ignoring biological knowledge to determine which species should be incorporated into null hypothesis tests and for placing less importance on Type II errors (Grant & Abbott 1980; Diamond & Gilpin 1982). The legacy of this debate was the emergence of community ecology as a more statistically rigorous science, using carefully constructed null models for hypothesis testing (Gotelli & Graves 1996). Community ecology embraced the application of null models that allowed the testing of explicit assembly-based models predicting the identity and abundances of different species within a community. In the search to create the most accurate model, assembly rules were viewed as providing constraints on composition independent of ecological process (Weiher & Keddy 2001). Typically, assembly-based models included abiotic and/or biotic ‘filters’ (Fig. 1–1; Cornwell & Ackerly 2009) and were tested against null expectations derived from resampling the local species pool. In this framework, the ‘species pool’ represents the set of species that can potentially colonize and establish within a community (Ricklefs 1987; Keddy 1992; Weiher & Keddy 2001; Lessard et al. 2012). The species pool is then passed through a conceptual ‘sieve’ or ‘filter’, filtering into the community the set of species that have traits adapted to the specific abiotic growing conditions at a site, while the set of species unable to form viable populations are filtered out of the community (Harper 1977; van der Valk 1981; Keddy 1992; Weiher et al.

4 1.1. Community ecology

1998; Cornwell & Ackerly 2009). Species must then pass through a second biotic filter, corresponding to the interspecific (most often assumed competitive) interactions with neighbouring species (Keddy 1992; Cornwell & Ackerly 2009). Despite these conceptual advances, with some exceptions (for an exception see Shipley et al. 2006), progress in predicting community composition was limited. This time it was Lawton who was to call the field to account writing (1999, pp. 278), “community ecology is a mess, with so much contingency that useful generalizations are hard to find.” Even though Lawton (1999) also criticized ecology more generally for not having universal laws, his comments directed towards community ecology were particularly harsh, calling for ecologists to set the discipline aside to concentrate on more worthwhile pursuits such as macroecology. Lawton’s comments led community ecology into a period of soul-searching (e.g. Simberloff 2004; Roughgarden 2009).

1.1.2 Beta diversity

While historically much of community ecology had focused on alpha diversity (Harrison et al. 1992), the diversity and composition of particular communities, there was growing in- terest in how communities differed from each other. Whittaker (1960) first introduced the concept of beta diversity in his study of vegetation patterns along topographic, climatic and parent material gradients in the Siskiyou Mountains located on the Oregon-California border. Whittaker (1960) presented several different definitions of beta diversity, leading to future confusion in terminology (Tuomisto 2010), but it was perhaps his multiplicative formulation of beta diversity—the proportion by which species richness in a region exceeds average richness of a single site within that region (Whittaker 1960; Whittaker 1972)— that provided the most significant conceptual advance. Following Whittaker’s concept of beta diversity, a number of studies went on to analyze species turnover along abiotic gradients (e.g. Cody 1975; Routledge 1977; Wilson & Shmida 1984), distance decay in

5 1.2. Phylogenetic community assembly

the similarity of species composition (Condit et al. 2002) and the additive partitioning of beta diversity into hierarchical components that could be compared within and among communities at different levels (Lande 1996; Crist & Veech 2006). More recent advances have reformulated beta diversity as the variance of the composition table of the various species within a community (Legendre et al. 2005), allowing for the quantification of local contributions to beta diversity, for example, among sites or species (e.g. Legendre & De C´aceres 2013). With several beta diversity formulas to choose from, the applicability of the different methods have been questioned (see Vellend 2001; Koleff et al. 2003; Jurasinski et al. 2009; Tuomisto 2010; Anderson et al. 2011). Heated debates arose on the appropriateness of additive versus multiplicative beta diversity partitioning and how statistical dependence on alpha, beta and gamma influenced hypothesis testing (Crist & Veech 2006; Jost 2007; Jost 2010; Baselga 2010; Veech & Crist 2010; Chao et al. 2010). Vellend (2001) and Anderson et al. (2011) attempted to distinguish between approaches that quantified turnover in species composition along gradients versus those that measured compositional differences without reference to gradients. The performance of various beta diversity metrics have been compared to determine the conceptual and sampling properties of different presence-absence (Koleff et al. 2003) and abundance-weighted (Barwell et al. 2015) indices. Tuomisto (2010) even advocated for one ‘true’ measure of beta diversity, although this was soon criticized by others (Anderson et al. 2011). Many of these debates on beta diversity remain far from resolution.

1.2 Phylogenetic community assembly

As species of the same genus have usually, though by no means invariably, some similarity in habitats and constitution, and always in structure, the struggle will

6 1.2. Phylogenetic community assembly

generally be more severe between species of the same genus, when they come into competition with each other, than between species of distinct genera. Darwin (1910, pp. 68)

Darwin recognized that species’ evolutionary histories were important in determining community composition. However, with only few exceptions, the majority of ecologists throughout the last century focused on the ecological processes of community assembly without reference to their histories. A notable exception was Elton (1946) who expanded on Darwin’s idea of ‘limiting similarity’ amongst closely related species by examining species to genus ratios (S/G ratios) to quantify competition in plant and animal commu- nities. Although subsequent work on S/G ratios was mixed, with some studies supporting a link between S/G ratios and competitive interaction strength (e.g. Grant 1966), other studies questioned such links (e.g. Simberloff 1970). Nonetheless the concept of ‘limiting similarity’ amongst co-occurring species was firmly embedded within community ecology, based on the idea that ecologically similar species would tend to competitively exclude each other (see Gause 1934; MacArthur & Levins 1967). Thus, since co-occurring species were limited in their similarity because of competitive exclusion, we would predict that they would also be overdispersed in their morphological (trait) space (Moulton & Pimm 1987; Weiher et al. 1998). However, testing such hypotheses was difficult in the absence of detailed data on species traits, which is often lacking. By the late 20th century, an increased availability of DNA sequences as well as improved computational power and statistical techniques allowed for the creation of robust phylogenetic hypotheses based on genetic sequences (e.g. Palmer et al. 1988; Donoghue 1994). Soon after, information derived from the phylogenetic relationships amongst species was used for conservation purposes (Faith 1992) and to interpret ecological patterns (Warwick & Clarke 1995). In 2000, Webb (2000) introduced a comprehensive framework

7 1.2. Phylogenetic community assembly

to analyze community structure using phylogenetic approaches that represented an extension of the S/G ratio while building upon existing theory on trait dispersion. In Webb’s framework, phylogenetically clustered communities would arise when co-occurring species were more closely related than expected by chance (i.e. estimated from a random draw from the species pool), whereas phylogenetically overdispersed assemblages would be composed of species less related than expected by chance. Webb’s framework focused on the composition of individual biological assemblages (i.e. communities), and it can thus be considered as a phylogenetic extension of alpha diversity. Webb et al. (2002) tied the patterns in phylogenetic clustering and overdispersion to the processes of environmental filtering and competitive exclusion, respectively (Table 1–1). Within this framework, habitat filtering (i.e. environmental filtering) caused com- munities to be phylogenetically clustered if traits were evolutionarily conserved; however, filtering would cause communities to be phylogenetically overdispersed if traits were evolu- tionarily convergent (Webb et al. 2002). That is, species filtered into similar environments likely shared phenotypic traits that arose through phylogenetic trait conservatism or con- vergent evolution (Webb et al. 2002; Cavender-Bares et al. 2004). In contrast, competition would prevent the coexistence of species sharing similar traits if its effects were strong (Webb et al. 2002; Cavender-Bares et al. 2004). Therefore, in a competitively structured community, species composition would be phylogenetically overdispersed if traits were evolutionarily conserved, while random community compositions would occur if species had convergent traits (Table 1–1). Studies focusing on the processes of environmental filtering and competition have since dominated the field of community phylogenetics (reviewed in Emerson & Gillespie 2008; Cavender-Bares et al. 2009; Vamosi et al. 2009). With some exceptions (e.g. Anderson et al. 2004; Mouillot et al. 2005; Horner-Devine & Bohannan 2006), the majority

8 1.3. Phylogenetic beta diversity

of studies have focused on land plants and have found phylogenetic clustering to be more common (Vamosi et al. 2009). However, relatively few studies have focused on clades of closely related species at small spatial scales, where competition is expected to be the more important process (Vamosi et al. 2009). In one of the few studies to explore this spatial and taxonomic scale, Cavender-Bares et al. (2006) found greater overdispersion of species at smaller scales, supporting suggestions that competition should be more important in structuring communities at this scale. In addition, Cavender-Bares et al. (2006) showed that communities might be phylogenetically overdispersed within one lineage but clustered when other lineages are included in the analysis. Perhaps unsurprisingly, the signal for phylogenetic dispersion can be affected by the choice of community, as well as the size of the species pool and choice of null model (Kembel & Hubbell 2006; Kraft et al. 2007; Swenson et al. 2007). More recently, concepts from phylogenetic community ecology have been incorporated into other ecological theories, such as trophic interactions (e.g. Morlon et al. 2014) and metacommunity ecology (e.g. Peres-Neto et al. 2012) to further improve our understanding of community assembly.

1.3 Phylogenetic beta diversity

Parallel to the later interest in beta diversity within community ecology, phylogenetic beta diversity originally received less attention than phylogenetic community assembly within the field of community phylogenetics. Eight years after the Webb (2000) framework de- scribing patterns of over– and underdispersion, Bryant et al. (2008) published a pioneering study comparing the phylogenetic similarity of communities of microorganisms and an- giosperms along an elevational gradient in the Rocky Mountains of Colorado in the United States. While this study was not the first to introduce the concept of phylogenetic beta

9 1.4. Community phylogenetics: summary

diversity (see Lozupone & Knight 2005; Ferrier et al. 2007), Bryant et al. (2008) proposed the PhyloSor index, a direct corollary to the Sørenson index of community similarity but based on shared phylogenetic branch lengths (Fig. 1–2). In addition, Bryant et al. (2008) connected their results to the community-level processes of environmental filtering and competition. Because phylogeny represents evolutionary history, this measure of phylo- genetic beta diversity has the potential to provide insight into regional scale processes, such as speciation and trait evolution (Graham & Fine 2008), shaping species composition and can help define biogeographical boundaries that capture historical processes (e.g. Holt et al. 2013; Qian et al. 2013).

1.4 Community phylogenetics: summary

Together, studies on phylogenetic community assembly and phylogenetic beta diversity are encompassed within the field of community phylogenetics. This relatively new discipline integrates information derived from the phylogenetic relationships among co-occurring species to reveal additional ecological insights into the processes determining patterns in species co-occurrence, such as environmental filtering, competition (Cavender-Bares et al. 2009; Vamosi et al. 2009) and species turnover (Graham & Fine 2008). Although community phylogenetic studies have provided new insight into ecological processes, the assumptions underlying these approaches have recently been criticized.

1.5 Challenges to community phylogenetics

With an increasing number of studies in community phylogenetics, the application of many such methods have come under increasing scrutiny. Ecologists often use poorly- resolved or incomplete phylogenies, which have been shown to bias estimates of several metrics used in community phylogenetics (Kress et al. 2009; Swenson 2009; Davies et

10 1.5. Challenges to community phylogenetics

al. 2012). For example, the online query tool ‘Phylomatic’ (Webb & Donoghue 2004) creates phylogenies resolved only to the level of plant family, thus biasing some metric estimates, but it is still commonly used in community ecology (e.g. Arroyo-Rodr´ıguez et al. 2012; Qian et al. 2013; Li et al. 2015). The development of high-throughput molec- ular techniques has increased the accessibility of DNA sequence data, and recent DNA barcoding initiatives have provided ecologists with an alternative approaches for creating regional and community phylogenies that are better resolved than ‘Phylomatic’ phylo- genies (Kress et al. 2009; Kress et al. 2010; Burgess et al. 2011; Maurin et al. 2014). To generate well-resolved, comprehensive, regional phylogenies based on DNA barcode data, all of the species in the regional species pool must first be known and have available DNA sequences. This creates two main challenges: 1) sampling and identifying all of the species in the regional species pool and 2) obtaining DNA barcode sequences for those species. Further, since the size of the species pool can affect signals in phylogenetic dispersion (Kraft et al. 2007; Swenson et al. 2007), another critical, but often underappreciated, challenge is the need to correctly define the limits to the species pool. Alongside these data challenges, assumptions implicit within community phylogenetics approaches have also been challenged (e.g. Mayfield & Levine 2010; HilleRisLambers et al. 2012). First, interpretations are often made at the community level, but it is at the individual level where species compete or are ‘filtered’ (Cornwell & Ackerly 2009). Second, although support exists for the assumption that phylogenetically distant species are more ecologically different, there are several counterexamples refuting this claim (HilleRisLambers et al. 2012; Mouquet et al. 2012; Gerhold et al. 2015). For example, character displacement between co-occurring, closely related species can lead to a loss of signal between phylogenetic distance and ecological similarity at the regional level (Gerhold et al. 2015). Third, the assumption that competition exclusion is most likely

11 1.6. Thesis outline

between closely related species was challenged by Mayfield & Levine (2010), who argued that competition could also result in phylogenetic clustering when species have similar resource requirements and are separated by small competitive differences. Viewed as a critique to the Webb phylogenetic community assembly framework, the ideas presented in Mayfield & Levine (2010) have received much attention in the last few years (e.g. Bennett et al. 2013; Godoy et al. 2014; Kraft et al. 2015b). To date, phylogenetic beta diversity approaches have not been as well scrutinized. These metrics have been used to delineate boundaries at the global and regional scales (e.g. Holt et al. 2013; Li et al. 2015), where processes such as speciation, extinction and dispersal are the dominant processes (Cavender-Bares et al. 2009). However, their performance at local scales, where processes such as competition and environmental filtering increase in importance, has not been explored. In addition, few studies have quantified the added benefit of including phylogenetic data within such analyses, for example by comparing their performance to more traditional beta diversity metrics derived from species data.

1.6 Thesis outline

In the following chapters, I address several challenges to the field of community phyloge- netics by exploring different empirical and analytical approaches. Specifically, I address a) the definition and delineation of the regional species pool, b) lineage differences in patterns of phylogenetic clustering and overdispersion, c) niche and fitness differences among closely related species, and d) the delineation of ecological boundaries at local spatial scales using phylogenetic beta diversity metrics. In Chapter 2, I describe the major challenges in sampling a comprehensive species pool so as to reconstruct the phylogeny for a regional flora. I make recommendations for

12 1.6. Thesis outline sampling that can be adapted for similar efforts in the future. I approach this question by presenting a case study based on my work on Mont St. Hilaire, Qu´ebec (Fig. 1–3 & Fig. 1–4). Chapters 3 and 4 focus on patterns of species coexistence at local scales using data on co-occurring sedges (i.e. plants of the Cyperaceae family) near Schefferville, Qu´ebec (Fig. 1–3 & Fig. 1–5). Specifically, in Chapter 3, I use an individual focal species approach to examine lineage-specific patterns in community assembly. Using a unique sampling design and analytical approach, I show that the phylogenetic composition of the immediate neighbours of individual plants varies with the clade membership of the species. In Chapter 4, I adapt ideas presented in Mayfield & Levine (2010) to address a long-standing question in community ecology: is the specialist the better competitor? I examine this question using theory that suggests coexistence among competing species is possible if they have similar competitive abilities or if their niche differences are large, which questions the assumption that competitive exclusion is most probable between closely related species. In Chapter 5, I use new phylogenetic beta diversity methods to delineate patches with similar vegetation types on a sharp gradient in elevation on Mnt Irony, Labrador (Fig. 1–3). Previous work using these methods has focused on the regional and global scales, whereas I explore patches at the local scale where different ecological processes may dominate. My work in this chapter shows that although phylogenetic beta diversity is effective at delineating boundaries between different biomes types at larger scales, it might not perform as well in differentiating between patches at local scales. In recent decades, the discipline of community ecology has faced many challenges, and community phylogenetics is currently facing its own unique list challenges. Instead of resting on methods and assumptions in community phylogenetics that have been recently

13 1.6. Thesis outline called into question, the work presented in this thesis uses novel approaches to advance the field, so that it can remain a relevant option for addressing future ecological questions.

14 Chapter 1 Tables & Figures

Table 1–1 The combinations of phylogenetic trait distribution and ecological processes that result in the expected distributions of sample taxa at a site. Table after Webb et al. (2002).

Ecological traits phylogenetically Conserved Convergent Dominant ecological force: Environmental filtering Clustered Overdispersed Competitive exclusion Overdispersed Random

15 Chapter 1 Tables & Figures

Figure 1–1 A hypothesis illustrating community assembly rules with ‘filtering’. Once those species from the species pool with the adaptations to survive the abiotic growing conditions have passed through an abiotic (environmental) filter, they must survive the competitive interactions of their neighbouring species (i.e. the biotic filter).

16 Chapter 1 Tables & Figures

Figure 1–2 A diagrammatic example illustrating the calculation of phylogenetic beta diversity with the PhyloSor index. The phylogenetic relationships among the species in the regional pool are shown in panel (a). In panel (b), the branch lengths represented by the species present within two separate plots are indicated, while panel (c) shows the branch length overlap between the two plots. Panel (d) shows the formula for PhyloSor, which is one of the many methods used to calculate phylogenetic beta diversity.

17 Chapter 1 Tables & Figures

Figure 1–3 Location of the two field study areas for this thesis. The work presented in Chapter 2 was conducted on Mont St. Hilaire, Qu´ebec (represented by black circle). Data for Chapters 3, 4 and 5 were gathered near Schefferville, Qu´ebec, which is represented by the black star.

18 Chapter 1 Tables & Figures

Figure 1–4 Vegetation on Mont St. Hilaire, where sampling for Chapter 2 was conducted. Vege- tation found along the edge of Lac Hertel is shown in panel (a), whereas panel (b) highlights the spring vegetation of one of the slopes on the ‘mountain’.

19 Chapter 1 Tables & Figures

Figure 1–5 Four different habitats common throughout the Schefferville region, where sampling for Chapters 3, 4 and 5 was conducted. The following habitats are shown: spruce–moss forest (a), shoreline (b), fen (c) and tundra (d).

20 CHAPTER 2

Challenges to barcoding an entire flora

Elliott, T.L.1 & Davies, T.J.1

A version of this chapter is published in Molecular Ecology Resources, 14, 883-891.

1 Department of Biology, McGill University, Montr´eal, Qu´ebec, Canada

21 2.1. Abstract

2.1 Abstract

DNA barcodes are species-specific genetic markers that allow taxonomic identification of biological samples. The promise of DNA barcoding as a rapid molecular tool for conducting biodiversity inventories has catalysed renewed efforts to document and catalogue the diversity of life, parallel to the large-scale sampling conducted by Victorian naturalists. The unique contribution of DNA barcode data is in its ability to identify biotic material that would be impossible to classify using traditional taxonomic keys. However, the utility of DNA barcoding relies upon the construction of accurate barcode libraries that provide a reference database to match to unidentified samples. Whilst there has been much debate in the literature over the choice and efficacy of barcode markers, there has been little consideration of the practicalities of generating comprehensive barcode reference libraries for species-rich floras. Here, we discuss several challenges to the generation of such libraries and present a case study from a regional biodiversity hotspot in southern Qu´ebec. We suggest that the key challenges include (i) collection of specimens for rare or ephemeral species, (ii) limited access to taxonomic expertise necessary for reliable identification of reference specimens and (iii) molecular challenges in amplifying and matching barcode data. To be most effective, we recommend that sampling must be both flexible and opportunistic and conducted across the entire growing season by expert taxonomists. We emphasize that the success of the global barcoding initiative will depend upon the close collaboration of taxonomists, plant collectors and molecular biologists.

22 2.2. Introduction

2.2 Introduction

For centuries, plant hunters have travelled the world in search of specimens. Early collectors, such as Henry Compton (1632-1713), Hans Sloane (1660-1753), Carl Linnaeus (1707-1778), and the wave of Victorian naturalists that succeeded them, systematically documented the diversity of the natural world and populated the world's herbaria with their collections. In recent decades, plant–hunting expeditions have increasingly shifted focus to collecting specimens for use in medicinal and other scientific studies (Hepper 1989). Following the launch of the Consortium for the Barcode of Life (CBOL) in May 2004, the International Barcode of Life Project (iBOL) was established to create a barcode library for all eukaryotic life (Schindel & Miller 2005; Frezal & Leblois 2008; Vernooy et al. 2010). This new effort more closely parallels the large-scale sampling conducted by the Victorian naturalists, but in addition to submitting plant species to herbaria, collectors now also submit genetic data generated using new molecular tools to digital repositories such as The Barcode of Life Data System (BOLD; www.barcodinglife.org) (Ratnasingham & Hebert 2007). The iBOL initiative aims to produce biological inventories in the form of DNA barcode libraries that will provide a reference database to aid in the identification of unidentified or hard to identify taxa (Ratnasingham & Hebert 2007; Jinbo et al. 2011). Although there were initial fears that DNA barcoding might usurp the role of traditional taxonomists (e.g. Ebach & Holdrege 2005), it is now realized that successful barcoding ef- forts rely on strong collaborations between 'barcoders' and taxonomists (Frezal & Leblois 2008). To create a barcode library, a site's flora must first be barcoded. These libraries can then be used as a cheap and quick method for biodiversity assessments (Krishna- murthy & Francis 2012). Comprehensive, accurate barcode libraries also allow for the identification of fragmentary samples or morphologically indistinct life stages (e.g. shoots, eggs or larvae), and the detection of species in environments when more comprehensive

23 2.2. Introduction

sampling is not practical, such as in biosecurity (Armstrong & Ball 2005) or for early detection of invasive species (Dejean et al. 2012). The wealth of molecular sequence data that is available in such libraries has provided researchers with the opportunity to generate large, well-resolved regional phylogenies, allowing for more accurate quantification of community structure and species interactions (e.g. Kress et al. 2009). In addition, barcode reference libraries have been used to identify ingredients in herbal medicines (e.g. Li et al. 2011b; Han et al. 2012; Kool et al. 2012) and teas (e.g. Stoeckle et al. 2011) as well as examine the belowground spatial structure of plant communities (e.g. Kesanakurti et al. 2011). The exact DNA sequences to be used as barcode markers in plants have been a source of heated debate. Key features for an effective DNA barcode include universality (loci that can be amplified across taxa with common primers), quality (loci that most likely produce bidirectional sequences with few or no ambiguous base calls), discrimination (loci that distinguish the highest number of species) and cost-effectiveness (Hollingsworth et al. 2009). The combination of rbcLa (c. 599 base pairs at 5' end of gene) and matK (c. 846 base pairs at centre of gene) is now widely considered as the DNA barcode for land plants (Hollingsworth et al. 2011; Fazekas et al. 2008), although other markers (see Hollingsworth et al. 2011), such as the internal transcribed spacer (ITS), have also been proposed (Li et al. 2011a), and others have been used to discriminate species in some plant clades (e.g. Roy et al. 2010). Recent work has explored the efficacy of these markers in discriminating different species in regional floras (e.g. Burgess et al. 2011) and the resolution they provide in reconstructing phylogenetic trees of regional floras (e.g. Kress et al. 2009). As barcoding efforts have expanded, a major challenge has been to generate DNA barcode libraries for complete floras. However, to date, there have been few published studies on such efforts. We surveyed ISI Web of Science (WoS; accessed 28 March 2013) based on the following criteria: barcoding; plants; NOT alga; NOT fungi; NOT insects.

24 2.2. Introduction

This search returned 353 matching papers, only five of which focused on barcoding regional floras (Fig. 2–1), including all the vascular plants of Mt. Valerio, Italy (Bruni et al. 2012), Churchill, Canada (Kuzmina et al. 2012), and the Koffler Scientific Reserve, Canada (Burgess et al. 2011), as well as the angiosperms and conifers of Wales, United Kingdom (de Vere et al. 2012). Other notable studies, from more diverse regions, have focused on certain plant groups, for example, the woody trees, shrubs and palms on Barro Colorado Island, Panama (Kress et al. 2009). In part, the paucity of published studies reflects the relative youthfulness of the field; we therefore additionally surveyed contributors to the WG1.2 Land Plants iBOL Working Group. Ten of the 17 respondents described projects focused on barcoding regional floras at taxonomic scales of angiosperms or higher (Appendix A1). This estimate likely represents only a fraction of ongoing projects as the release of information by BOLD contributors was voluntary, and it is possible that there are other barcoding studies that do not use BOLD. These large barcoding efforts pose many significant practical challenges, some of which have been largely ignored in the recent literature, including (i) difficulties in collecting specimens for rare or ephemeral species; (ii) paucity of or limited access to sufficient taxonomic expertise and resources to obtain reliable, up-to-date identifications of samples used to build the barcode reference library; and (iii) molecular challenges of barcoding (e.g. difficulties in amplification of samples spanning a wide taxonomic spectrum and distinguishing between closely related species). Even though many of these challenges are not novel, the first two have tended to be overlooked in barcoding literature in comparison with the emphasis that has been placed on the third challenge—molecular methods and the performance of barcodes for species identification (e.g. Lahaye et al. 2008; Burgess et al. 2011). Here, we attempt to redress this balance and focus on the first two challenges. Specifically, we discuss the practicalities of generating DNA barcode libraries for complete floras. We illustrate these challenges using a case study of the flora

25 2.3. Barcoding a complete flora—Mont St. Hilaire as a case study

of the Gault Nature Reserve on Mont Saint Hilaire (MSH), a biosphere reserve in the Mont´er´egie region of southern Qu´ebec.

2.3 Barcoding a complete flora—Mont St. Hilaire as a case study

In 2012, we established an initiative to DNA barcode the entire flora of the Gault Nature Reserve, a small reserve approximately 30 km from the major urban centre of Montreal, Qu´ebec, Canada. The flora of the Gault reserve is well-documented with a nearly complete list of vascular plants available for the site (see Maycock 1961, Appendix A2) as well as the presence of herbarium specimens for a majority of the reserves species in either the McGill University Herbarium (MTMG) or the Marie-Victorin Herbarium (MT). The reserve is situated on Mont St. Hilaire, one of the eight Monteregian Hills of the St. Laurence Lowlands in southern Qu´ebec (Maycock 1961). The old growth forest of the reserve is dominated by the deciduous tree species Acer saccharum, Fagus grandifolia and Quercus rubra (Gilbert & Lechowicz 2004). Protection of this 1000 ha reserve goes back until the 1600s, and it is now recognized as a UNESCO Biosphere Reserve (Maycock 1961; Francis 2004; White et al. 2011). The reserve contains approximately 650 vascular plant species (Maycock 1961, Appendix A2), several of which are considered rare. Barcoding this flora provides an illustration of the more general (and generally more extreme) challenges to barcoding complete floras. The reserve is relatively small and easily accessible, and although the area is a regional diversity hotspot, species richness is relatively low compared with, for example, tropical floras of similar areal extent (Myers 1988). Over approximately 384 and 235 sampling hours during the 2012 and 2013 respective field seasons, 582 different terrestrial vascular species were collected with sampling focused on collecting one individual per species, except in cases where there were uncertainties in field identifications for which multiple samples were gathered. We found 111 species

26 2.3. Barcoding a complete flora—Mont St. Hilaire as a case study

not previously reported in the reserve and suspect that changing distributions in response to climate change (Parmesan 2006; Kelly & Goulden 2008), exotic species moving in with disturbance (Gilbert & Lechowicz 2005) and differing taxonomic skill sets of the various surveyors (Oredsson 2000; Chen et al. 2009) might all contribute to explaining this increase in number. We also think that the previously documented species that we did not find might have been extirpated because of the impacts of climate change and increased deer browsing (Cˆot´e et al. 2004; Gilbert & Lechowicz 2005). In addition, we identified several discrepancies in the list used to guide collection efforts, leading to an inflated species total from past years (Fig. 2–2). Of the 787 vascular species previously recorded, 143 were removed from the list because they were synonyms and 21 species names were duplicated due to typographical errors. Thus, our sample represents an impressive 84% of the revised species list, contributing 133 rbcL and 126 matK new sequences for species not previously represented through the BOLD public data portal as of December 2013. Sampling in future years will target the remaining 16%, but it is impossible to verify how many of the species on this list were originally misidentified as only a handful have matching voucher specimens (e.g. Maycock 1961) that we can cross-check. The flora of Mont St. Hilaire is likely much smaller than that targeted by regional barcoding efforts in tropical or larger areas; nonetheless, our efforts identified a number of key challenges to creating a comprehensive DNA barcode library based on sampling one individual per vascular plant species that will apply more generally (and likely to a greater degree). Many of these challenges arise from the time constraints imposed by the short– term nature of most research projects, which are often determined by funding cycles and the demands of thoroughly sampling all of the species within a site's boundaries during short growing seasons. These challenges parallel all such efforts aimed at conducting comprehensive biodiversity surveys (Palmer et al. 2002; Chen et al. 2009), but generating a barcode library imposes additional demands.

27 2.3. Barcoding a complete flora—Mont St. Hilaire as a case study

Simply creating complete regional floristic inventories, even for relatively small floras such as Mont St. Hilaire, is challenging and will require considerable effort, which is a substantial impediment, given current costs and budgeting constraints (Pautasso 2012). In addition, rare species will take more time to find, whilst common species can be quick and easy to sample (Preston 1948). For example, we collected a high number of species in a relatively short period of time during the 2012 field season because sampling targeted accessible areas with high species numbers (Fig. 2–3). However, this opportunistic sampling strategy reduced the amount of time spent searching for rarer species, which can be more difficult to find, illustrating diminishing returns to the amount of effort invested. During the 2013 field season, more targeted sampling based on previous location reports for rarer species was generally aimed at more remote locations within the reserve; however, this strategy required much larger effort but yielded fewer new species as these sites were often species poor (Fig. 2–4). Low population numbers of rare species can also present a challenge as it is not ethical to collect when sampling might impact population viability; thus, workers must spend valuable time searching for alternative sources of plant material. Differences in phenology can also present major challenges and limit the utility of short, intensive sampling efforts or 'bioblitzes' (a term originally used by Susan Ruby of U.S. National Park Service in 1996; http://www.biodiversityonline.ca/BioBlitz/intro.htm), as the collection of some plant species must be conducted during a brief window of time during which traits important for species identification are expressed as fruits and/or flowers (Miller & Nyberg 1995; Leponce et al. 2010). For example, the number of plant species collected during the field season at Mont St. Hilaire fluctuated as high numbers of spring ephemerals were collected whilst flowering early in the growing season, which made them easiest to identify, but species numbers dropped off as the summer progressed (Fig. 2–4). Combining phenological differences with the logistics of getting to remote locations presents an even larger challenge to comprehensive sampling. Within Mont St.

28 2.3. Barcoding a complete flora—Mont St. Hilaire as a case study

Hilaire, several rare species grow in difficult-to-access areas (Maycock 1961), including several locations that were impossible to access due to dangerous physical obstacles such as steep cliffs. In a few cases, several time-consuming return visits to these remote areas were required because of differences in the phenology of rare species found at these sites. Assuming sufficient time and funds exist for adequate collection, there remains a major challenge in correctly identifying species. First, even experts might overlook species if the material that is present at a site is not in an identifiable form for at least part of the year (see above). Second, species within genera that are the product of rapid radiations can represent suites of morphologically similar taxa that are difficult to distinguish either in the field or herbarium. In cases of taxonomic uncertainty, more time must then be spent collecting extra specimens of these taxa so that species that closely resemble each other are not missed, in addition to the extra time required to collect, process and send off material to taxonomic experts. Collection of specimens at the Gault Nature Reserve was conducted by experienced botanists; nonetheless, several genera such as the Carex, Dichanthelium, Crataegus, Salix and Amelanchier were problematic to distinguish in the field because of the presence of hybrids, agamic complexes and difficult-to-distinguish species characteristics. Correct identification of such specimens required reference to multiple resources and referral to specialist taxonomists in some cases, as has also been the case in other regional barcoding studies (e.g. Kuzmina et al. 2012). In general, the decline in taxonomists and limited access to herbarium resources and updated taxonomic treatments present a major challenge to accurate species identification—the taxonomic impediment (Hoagland 1996; Rodman & Cody 2003). Although these challenges might be familiar to most biodiversity surveys, the unique approach of DNA barcoding presents additional challenges related to species identification. Well-constructed barcode libraries allow end-users to identify plant material to species or genus based on tissue fragments (e.g. a single leaf, root or stem) that would be impossible to identify using traditional

29 2.3. Barcoding a complete flora—Mont St. Hilaire as a case study

taxonomic keys. However, it is not possible for end-users to detect errors in the original species identification because barcode matches are based only on sequence data and not morphology, although digital images and herbarium specimens can be revisited in the future to verify identifications. The correct identification of reference specimens is therefore paramount (Collins & Cruickshank 2013). Once plants have been collected, voucher specimens must be accessioned into a herbarium as required by BOLD. Mounting, labelling and accessioning our 680 specimens were an additional challenge to barcoding. Perhaps most surprisingly, time and cost estimates based on our experiences at the Gault Nature Reserve (Appendix A3) suggest that over one-third of the time required to complete floristic barcoding studies should be allocated to data organization and processing voucher specimens. Further, our experience conducting botanical surveys near Schefferville, in northern Qu´ebec, suggests these same two steps might still require a similar proportional time commitment when preparing barcode libraries for more remote areas (Appendix A3). Paid and dedicated staff would be necessary to process samples for larger collection efforts. Sequencing success is not guaranteed once specimens have been collected and vouchered. The choice of DNA barcodes has been a trade-off between factors such as universality (i.e. ability to amplify all species) and resolution (i.e. the ability to differentiate between species) (Hollingsworth et al. 2009). In plants, rbcL has proven relatively straightforward to amplify for most clades but provides only limited resolution, particularly among closely related taxa (Hollingsworth et al. 2009). In contrast, matK offers greater resolution, but the proposed primer sets perform poorly in non-angiosperm plant groups such as the gymnosperms and ferns (Hollingsworth et al. 2011; Li et al. 2011c), although alternative primer combinations are now available (Li et al. 2011c). We do not discuss further the details of marker choice here as the selection of DNA barcode markers has already been debated extensively in

30 2.3. Barcoding a complete flora—Mont St. Hilaire as a case study

the literature (e.g. Kress et al. 2005; Rubinoff et al. 2006; Kress & Erickson 2007; Pennisi 2007; Hollingsworth et al. 2011); however, additional challenges can arise postsequencing. Once barcode sequences are available, additional complexities must be considered before using barcode reference libraries to assign identifications to unknown plant material (e.g. Kesanakurti et al. 2011; Li et al. 2011b). First, the genetic distance differentiating species must be decided upon if cryptic species are suspected (Meyer & Paulay 2005; Collins & Cruickshank 2013). The answer to this is likely taxon and gene specific, and there is probably no correct answer in some circumstances (Hollingsworth et al. 2011), although sequences can be simply grouped with the closest matching barcode if a library is assumed 'complete' (Hajibabaei et al. 2007; Kesanakurti et al. 2011). Whilst species resolution is generally higher in regionally focused studies compared with those focused on particular taxonomic groups (e.g. Le Clerc-Blain et al. 2010; Roy et al. 2010; Burgess et al. 2011), even within the flora of Mont St. Hilaire, we suspect the identification of several genera such as Salix and Carex might be problematic, as these genera have previously been reported to be difficult to distinguish using DNA barcodes (Starr et al. 2009; von Cr¨autlein et al. 2011), and multiple species within each genus are found in the regional flora. A second (and related) question is how to decide the appropriate sampling density to get estimates of within-species variation in sequences (Meyer & Paulay 2005; Bergsten et al. 2012). One approach would be to maximize the geographical distance between samples (Bergsten et al. 2012). However, as the number of individuals required would likely vary by species and perhaps barcode region, there again is probably no single answer. The effort required for the implementation of this strategy would in any case likely be prohibitive for sampling species-rich floras as many individuals per species would have to be sampled to get an appropriate estimate. Finally, users of barcode data must decide how to interpret results if different genes give conflicting answers (Li et al. 2011a). Common barcoding genes in plants are perhaps unlikely to give conflicting answers

31 2.4. Consequences of incomplete sampling because they are from the chloroplast (Soltis et al. 1998); nonetheless, neutral evolution might still result in apparent conflict and chloroplast vs. nuclear genes might still show incongruence (Rieseberg & Soltis 1991). Possible solutions include adding sequence data and/or individuals or to explore neutral vs. non-neutral mutations separately. We are not aware of any review of this issue as it applies to barcoding, although it has been widely discussed in the broader phylogenetics literature (Pamilo & Nei 1988; Degnan & Rosenberg 2009). These are all important questions that must be addressed in the barcoding literature.

2.4 Consequences of incomplete sampling

Future conservation and research programmes will have to take into consideration the reality that sampling efforts will probably not yield comprehensive floristic inventories, and thus, barcode libraries will often be incomplete. Of concern, the taxa most likely to be missed are rare species (e.g. orchids on MSH), which might lack barcodes and could be of highest conservation priority. Limited sampling or misidentification of closely related species could also impact efforts to generate phylogenetic trees using barcode data. It is possible to generate phylogenetic hypotheses from partial or gappy molecular matrices, at least for those taxa with some barcode data (Wiens 2003; Wiens & Morrill 2011); however, missing sequence data might increase phylogenetic uncertainty, resulting in unresolved nodes (i.e. polytomies). The consequences of this lack of phylogenetic resolution will depend on the questions to which the phylogenetic hypotheses are applied. For example, testing fine-scale patterns in species co-occurrence would require more well- resolved phylogenetic trees at the species level (e.g. Slingsby & Verboom 2006; Araya et al. 2012). In addition, metrics of phylogenetic community structure differ in their sensitivity to phylogenetic resolution (Swenson 2009). Species-level resolution might be more important for capturing interspecific interactions, such as competition and

32 2.5. Lessons learned: suggestions for future studies

facilitation. Considerations such as these are important for determining whether it is worth investing the extra time and resources to retrieve complete barcode sequences for all species.

2.5 Lessons learned: suggestions for future studies

Based on our experiences, we suggest several ways forward to more effectively plan and execute regional barcoding efforts. First, herbaria, species inventories, local experts and the Global Biodiversity Information Facility (GBIF; www.gbif.org) can all provide possible locations for species that are hard to find, taking into account possible errors in identification and synonyms. Other suggestions are to use online resources such as Project BudBurst (www.budburst.org) and review herbarium specimens to design field sampling around times when plants are in flowering or fruiting stage. The current year's climatic conditions will have to be considered when interpreting the data available through such sources, and the resolution of GBIF location data might not help in the location of rare species. A targeted sampling approach could focus sampling on areas where rare species have been observed in the past, but to be effective, this method requires precise GPS co-ordinates or detailed, accurate information on herbarium labels for previously observed species. This strategy is of course only applicable when targets are known and their populations are not transient. Second, methods such as accumulation and rarefaction curves can be used to provide estimates of potential gains in species numbers provided by additional sampling, although these approaches do not give an absolute answer to the number of species in the area (Gotelli & Colwell 2010). Third, we suggest to be prepared to sample more than one field season by having a sampling plan that focuses on collecting the most common species the first year and targets rare species during subsequent field seasons. Follow-up grants can be used to support continued sampling in subsequent years, but diminishing returns of species numbers relative to the amount of time invested suggest

33 2.5. Lessons learned: suggestions for future studies

that continued full-time sampling would yield only a few substantial new results. We therefore recommend to maintain continuous sampling at a site by establishing strong long-term collaborations with other researchers at the site who might also benefit from the publically available barcode data. One alternative to reduce the time and costs spent supporting field work is to sequence DNA material from herbarium specimens, although DNA from such material can be degraded, inhibiting amplification of barcode markers. Nonetheless, herbarium specimens are useful for sampling very small populations of rare species that can be documented but not sampled because of their conservation status. Whilst herbaria should be part of the global barcoding initiative, they cannot, however, replace field sampling, especially where the flora for a region is incompletely described. Critically, field collection allows for the discovery of new species and can document shifting species distributions with climate change that would be missed by reviewing herbarium specimens for a region. During our study, 12% of the species on our final list were new records for Mont St. Hilaire (Fig. 2–2) and would have been omitted if we had relied only on herbarium specimens. Limited access to taxonomic expertise and resources can also hamper regional barcod- ing efforts; therefore, it is essential to collaborate with taxonomists as well as work with parataxonomists and herbarium resources when possible. Ideally, strong collaborations with taxonomists should be made before gathering data. Taxonomists are a key compo- nent to producing regional barcode libraries and can train the field collectors and notify them of problematic species before the field season and as such should be compensated appropriately for their time and effort (Tripp & Hoagland 2013). Parataxonomists can provide an additional resource and may help encourage local participation in regional barcoding efforts (Janzen 2004; Abadie et al. 2008; Janzen & Hallwachs 2011), but they are not appropriate for all study systems. We suggest using herbaria as much as possible

34 2.5. Lessons learned: suggestions for future studies

as they provide many resources crucial in species identification; however, extra time and money might be required for workers in areas without local herbaria. Finally, further tax- onomic confusion could be avoided by having data repositories such as BOLD recommend common databases and identification resources for attributing names to species. Comprehensively sampling entire areas to create barcode libraries will require large- scale field work incurring major financial costs. Funding such scientific endeavours is a socio-political issue that society needs to address and is largely beyond the scope of this paper. One possibility is to incorporate regional barcoding efforts into current research projects; however, the short–term nature of grant funding cycles and the considerable time commitment required to compile barcode libraries may provide a disincentive. The unfortunate reality is that adequately sampling large areas, such as northern Qu´ebec, will be expensive and might not be possible under current funding structures. As recent funding cutbacks have already ended the era of free barcode sequencing funded by the Natural Sciences and Engineering Research Council of Canada (NSERC) and other major granting agencies through BOLD, we might thus predict a decrease in the volume of barcode data being produced in the near future. This decline will be a double blow as we will not only see a decrease in the rate of data acquisition, but also because the value of existing barcode data increases as the database grows. We have shown that even for a relatively small, easily accessible and well-sampled site with a flora of only ∼650 vascular plant species, comprehensive taxonomic sampling to create a barcode library is a formidable challenge. We have identified a number of specific challenges to barcoding a regional flora that are sometimes overlooked in the literature, and we suggest some possible ways forward. In addition, we emphasize some of the particular benefits that may be gained from efforts to barcode all of the plant species at a site, such as documenting changes in floristic composition and contributing valuable new specimens to herbarium collections. However, regardless of the effort put into

35 Chapter 2 Figures collecting specimens in the field and advances in sequencing performance, the information provided by DNA barcodes will not be valuable to science unless the original specimens are correctly identified, and/or voucher specimens are available such that identifications can be revised later. The iBOL initiative aims to provide comprehensive DNA barcode libraries that can be used in biodiversity research, conservation and forensics, by allowing endusers to identify biological material through matching barcode sequences. In many cases, this material might be fragmentary or in various developmental stages that are difficult or impossible to identify using traditional taxonomic keys (Jinbo et al. 2011). Correct identification of barcoded specimens relies on resources produced by expert taxonomists, such as updated taxonomic keys, identification services and correctly identified sequences on GenBank. Thus, the success of the iBOL initiative will rely upon professionally trained taxonomists working together with plant collectors and molecular biologists to generate accurate barcode libraries.

36 Chapter 2 Figures

Figure 2–1 Web of Science search results of published papers on barcoding plants (n = 353). Papers were classified as ‘Barcoding regional flora’ if they focused on the taxonomic scale of angiosperms or higher within a single region. ‘Species discrimination’ papers focused on deter- mining which gene sequences best differentiated plant species within various genera or families, whereas papers on ‘Sequence selection’ examined the broader question of selecting the best bar- code sequences for land plants. ‘Barcoding methods’ papers either introduced novel methods that could be used in future barcoding research or presented methods to improve existing proto- cols. ‘Barcoding utility’ papers provided ideas or examples of new plant barcoding applications. ‘Review/idea’ papers were literature reviews or viewpoints on plant barcoding research. In addi- tion, a request of information from contributors to the BOLD Land Plants campaign yielded 10 unpublished projects working on barcoding regional floras.

37 Chapter 2 Figures

Figure 2–2 Results of 2012 and 2013 sampling efforts of the vascular plant species on the Gault Nature Reserve, Qu´ebec, based on a pre-existing species list of the reserve (Gault Nature Reserve, unpublished data). The original species list contained several ‘typographical errors’ and ‘synonyms’. ‘New species to list’ were taxa not included in the original 2012 list. Of the remaining taxa on the list, ‘collected’ species were those that were collected during the 2012 and 2013 field seasons, whereas ‘not collected’ were species that were not found during the two field seasons.

38 Chapter 2 Figures

Figure 2–3 Location of sampling points for plant specimens from Mont St. Hilaire, Qu´ebec, during the 2012 and 2013 field seasons. The reserve boundaries have been extended to include the point in the top right-hand side of the figure. Sampling locations for plants collected dur- ing the 2012 field season are indicated by stars, whereas plants collected during the 2013 field season are indicated by circles.

39 Chapter 2 Figures

Figure 2–4 Number of terrestrial plant species collected per month during the 2012 and 2013 field seasons at Mont St. Hilaire, Qu´ebec (note: collectors were not available to go to the field in July 2013).

40 Linking Statement #1

In my introduction, I identified a number of challenges to the field of community phylo- genetics. These included i) correctly defining the appropriate regional species pool and ii) embracing lineage specific differences in patterns of phylogenetic clustering and overdis- persion. In Chapter 2, I identified several key challenges to barcoding a regional flora and provided suggestions on how to overcome these challenges. Once the species in a regional flora are identified and barcoded, a regional species pool can be created that defines the species to include in null models and regional phylogenies. In this next chapter, I recon- struct the regional phylogeny of the Cyperaceae species (sedges) found near Schefferville, Qu´ebec. I then use this regional phylogeny to examine lineage-specific patterns in the co-occurrence of sedge species. The novel sampling design and analysis I use in Chapter 3 reveals that lineage-specific patterns in sedge co-occurrence are evident and when averaged together these patterns mask each other, leading to apparent randomness in the overall phylogenetic structure of sedge communities. CHAPTER 3

Contrasting lineage-specific patterns conceal community phylogenetic structure in larger clades

Elliott, T.L.1, Waterway, M.J.2, & Davies, T.J.1

A version of this chapter has been accepted to be published in Journal of Vegetation Science.

1 Department of Biology, McGill University, Montr´eal, Qu´ebec, Canada 2 Department of Plant Science, McGill University, Ste-Anne-de-Bellevue, Qu´ebec, Canada 42 3.1. Abstract

3.1 Abstract

Community phylogenetic approaches can provide information on the ecological forces structuring plant community composition. For example, assuming evolutionary con- servatism of ecological traits, environmental filtering has been suggested to result in phylogenetic clustering, whereas competition has been suggested to lead to phylogenetic overdispersion. However, current approaches report aggregate community-wide metrics and typically assume that all lineages respond similarly. Here, we question this assump- tion and evaluate evidence for lineage-specific patterns of co-occurrence. We use a novel individual-based, neighbourhood approach to evaluate lineage-specific co-occurrence pat- terns among 35 focal species of Cyperaceae (sedges). Cyperaceae is a species-rich clade, with many species sharing similar niche preferences and environmental tolerances, making this a model clade for evaluating interspecific interaction strengths. We reveal striking differences in co-occurrence patterns between clades. Within the Cyperaceae, the two species of the genus Trichophorum tended to cluster with each other, whereas species within Eriophorum co-occurred more often with less closely related neighbours. These contrasting patterns repeated themselves within Carex, with species in the Core Carex clade more often co-occurring with close relatives, while phylogenetic overdispersion was more prominent in the Vignea clade. As a consequence of these opposing lineage-specific relationships, the overall phylogenetic structure of sedge communities appears random. We show opposing lineage-specific patterns in phylogenetic structure that are obscured when species are aggregated into larger clades. Our results reveal cryptic complexities in plant community assembly across taxonomic scales, and indicate that different plant lineages may vary in their response to community-level processes. We suggest that the processes underlying community composition can be better understood by using alternative sam- pling and analytical approaches combined with thoughtfully-created null models.

43 3.2. Introduction

3.2 Introduction

The ecological and evolutionary forces that determine the structure of plant communities at the local scale are multifold (Vellend 2010); however, environmental filtering and competition are often considered the dominant processes (Weiher et al. 1998; Cornwell & Ackerly 2009). Environmental filtering determines which plant species in a regional pool can persist under a given set of environmental conditions (van der Valk 1981; Keddy 1992; Weiher et al. 1998). Once species have passed through the environmental filter, community membership might be determined by the strength of interspecific competition (Elton 1946; Weiher et al. 1998); a process that is expected to be strongest amongst close relatives at fine spatial scales (Vamosi et al. 2009). One enduring hypothesis is that ecological differences permit coexistence by reducing the intensity of competition among co-occurring species (MacArthur & Levins 1967). It remains a challenge, however, to quantify the relative strength and importance of these two processes in determining plant community composition (Kikvidze et al. 2011), but taxonomic affinities can provide some insights. Over the past century, ecologists have progressed from using species/genus ratios (e.g. Elton 1946), through indices of taxonomic distinctiveness (e.g. Warwick & Clarke 1995; Clarke & Warwick 1998) to explicit phylogenetic methods to infer the importance of different ecological processes in community assembly (Webb 2000; Webb et al. 2002; Vamosi et al. 2009; Cavender-Bares et al. 2009). With the increasing availability of phylogenetic information, the field of community phylogenetics has expanded rapidly (Vamosi et al. 2009; Cavender-Bares et al. 2009). Underpinning much of this work is the assumption that the phylogenetic distances separating species in a community can provide insights into the historical and evolutionary processes of community assembly. Following the framework developed by Webb and colleagues (Webb 2000; Webb et al. 2002), environmental filtering is typically inferred when communities are phylogenetically clustered, such that co-occurring species are more

44 3.2. Introduction

closely related than would be expected by chance. In contrast, competition is thought to be the dominant force structuring communities when communities are phylogenetically overdispersed—co-occurring species are more distantly related than expected (Webb 2000; Webb et al. 2002). These interpretations rely on various assumptions, of which two are perhaps key: first, that traits relevant to filtering or competition are phylogenetically conserved (Webb 2000; Cavender-Bares et al. 2004) and second, that all lineages respond similarly when exposed to the same ecological pressures. While the former assumption has been the focus of particular attention in the literature, the second assumption has been less explored (but see Ndiribe et al. 2013) and is our focus here. A growing number of studies have provided evidence for non-random phylogenetic structure in ecological communities, variously suggesting environmental filtering or competition as the dominant structuring force (Johnson & Stinchcombe 2007; Vamosi et al. 2009; Cavender-Bares et al. 2009). However, patterns in phylogenetic structure can vary depending on the taxonomic and spatial scale considered; for instance, the dominant pattern can switch from overdispersion to clustering as we move from small to large taxonomic and spatial scales (e.g. Cavender-Bares et al. 2006). In addition, competition might not always result in overdispersion; for example, when competitive differences between species are small we might expect greater co-occurrence among species with similar resource requirements (Mayfield & Levine 2010; Bennett et al. 2013). Thus, it might also be important to consider how trait differences determine competitive outcomes, although we do not address this complex issue here. The relative strength of phylogenetic clustering and overdispersion might also change depending on the evolutionary age of co-occurring lineages. For example, in English meadow communities old lineages nested deep in phylogeny are more phylogenetically clustered, whereas the species at the tips of the phylogeny are more overdispersed (Silvertown et al. 2006). Importantly, apparent random structure might result if opposing patterns of clustering and overdispersion are found within the same community (Hardy & Senterre 2007; Mayfield & Levine 2010). 45 3.3. Methods

One alternative approach, therefore, is to explore lineage-specific patterns that allow for opposing mechanisms across the tree topology (Hardy & Senterre 2007; Ndiribe et al. 2013). Lineage-specific approaches can provide a more nuanced understanding of community assembly, especially when combined with innovative sampling designs aimed at capturing the strength of ecological interactions. Because it is individuals, and not species, that are filtered into a community and compete (Cornwell & Ackerly 2009), one particularly pow- erful sampling method is to evaluate the phylogenetic neighbourhood of focal individuals (e.g. Hill & Kotanen 2009; Uriarte et al. 2010; Yguel et al. 2011). By combining this focal sampling approach with lineage-specific phylogenetic methods, it is possible to evaluate fine-scale interactions between neighbouring individuals (e.g. Bennett et al. 2013) and then map these back onto the phylogeny to identify lineage-specific trends. Here, we use an individual-based neighbourhood approach in a closely related group of plants, sedges in the family Cyperaceae, to explore changes in patterns of species co-occurrence across the phylogeny. Both phylogenetic niche conservatism and niche differentiation have been reported in different sedge clades (Dabros & Waterway 2008; Waterway et al. 2009), thus we predict that patterns in co-occurrence could shift across the phylogeny. For example, the Vignea clade of Carex tends to be more closely associated with fen habitats in our study region, whereas the Core Carex is often more widespread, and perhaps less subject to strong abiotic filtering. We compare our lineage-specific approach to more traditional community phylogenetic methods that index community- wide patterns.

3.3 Methods

3.3.1 Study system

We focus our study on three genera in the Cyperaceae, a highly diversified, monophyletic and almost cosmopolitan group that consists of nearly 5,500 species (Muasya et al. 2009). 46 3.3. Methods

Sedges are grass-like herbs that have mostly 3-ranked leaves with parallel veins, closed leaf sheaths and reduced floral morphology (Ball et al. 2002). Carex is the largest genus within the Cyperaceae with more than 1,800 species worldwide and about 480 in North America (Ball & Reznicek 2002; Muasya et al. 2009; Govaerts et al. 2015). Carex belongs to the tribe Cariceae, along with Cymophyllus, Kobresia, Schoenoxiphium and Uncinia (Waterway & Starr 2007). Recent molecular phylogenetic studies (Waterway & Starr 2007; Waterway et al. 2009) have shown that Carex is only monophyletic when treated in the broad sense to include all of the segregate genera, as recently proposed (Global Carex Group 2015). Three of five major clades of Carex s.l. are found in North America, including the Core Carex (primarily multispicate with terminal staminate and lateral pistillate spikes), Vignea (sessile bisexual spikes with predominantly distigmatic flowers) and Caricoid clade (mostly unispicate and androgynous, including all segregate genera) (Ford et al. 2006; Waterway & Starr 2007; Starr et al. 2008; Waterway et al. 2009). The study was conducted in a 500 km2 area surrounding Schefferville, Qu´ebec (54.035◦–54.888◦N; 66.753◦–67.240◦W), where winters are long, and summers are cool and wet (Lechowicz & Adams 1978). Deglaciation of this area occurred between 5,000 and 6,000 years ago, making it possibly the last area in North America to be deglaciated after the Wisconsinan glaciation (Ives 1960). Shorelines, ponds, fens, spruce-moss forests, alder thickets and tundra are suitable environments for sedge growth in this region (Waterway et al. 1984). Within the spruce-feather moss forests, sedges tend to occur on relatively moist and nutrient-rich sites, whereas these plants can also be found on higher mountains and ridges in alpine tundra. In addition, a few sedge species are restricted to tundra ponds and seepage areas (Waterway et al. 1984). The distribution of the sedges included in this study across the different habitat types of the Schefferville region are given in Appendix B1. We sampled 700 focal plants (20 individuals of each of the 35 most common sedge species in the Schefferville area) during the 2011 field season using a combination of the 47 3.3. Methods

‘ignorant man’ method (Ward 1974) and random walks (see Appendix B2 for further details). This sampling design ensured that each focal species was sampled at least 20 times and that plants were randomly chosen within populations in accessible areas. Previous work on sedge distributions in the area was used to determine the general locations in which to sample each species (Dabros 2004, Bell, G., Lechowicz, M.J., & Waterway, M.J.; unpublished data) and to maximize the breadth of growing conditions sampled. The 20 individuals of each focal species were sampled across the Schefferville landscape. One focal species was randomly selected out of all of the sedge species growing within a 1.5 metre wide area after randomly arriving at each potential plot, and plots were located at a minimum of 5 metres apart to avoid trampling. Circular 1.0 m2 quadrats were centred over each focal plant, and the percent cover of each co-occurring sedge species was estimated by consensus of two or three independent observers to document all of the sedge species within the focal plant’s neighbourhood. The percent cover estimates included both the central culms and all other vegetation for each sedge species in the plot since clonality often made it impossible to delineate individual plants.

3.3.2 Phylogeny reconstruction

The phylogenetic reconstruction was conducted using Bayesian inference including 51 taxa from four genera of Cyperaceae and two species of Juncaceae as an outgroup, with gene sequences of four plastid regions (rbcL protein coding region, matK protein coding region, trnL intron, and the trnL–F spacer) and three nuclear ribosomal spacers (ITS1, ITS2, and ETS-1f). More detailed methods describing the species pool and phylogenetic reconstruction are given in Appendix B3.

3.3.3 Species co-occurrence metrics

We first calculated mean phylogenetic distances (MPD) and mean nearest taxon phyloge- netic distance (MNTD) for all study plots, where MPD is the mean pairwise phylogenetic distance between all species in a plot, and MNTD is the mean pairwise phylogenetic 48 3.3. Methods

distance between the closest non-conspecific relatives in a plot (Webb 2000). The net relatedness index (NRI) and nearest taxon index (NTI) were then calculated for each plot as the effect size of MPD and MNTD relative to a null model that maintained sample species richness multiplied by -1 (Webb 2000). Abundance-weighted metrics for both NRI

(NRIab) and NTI (NTIab) were estimated by weighting with percent cover. Positive NRI and NTI values indicate that species are phylogenetically clustered, whereas negative values represent phylogenetic overdispersion (Webb 2000). Each analysis was calculated on 100 randomly selected phylogenetic trees from the posterior distribution of the Bayesian analysis discarding 40% burn-in, and averaged to calculate a mean value for each focal plot. Plots containing only one sedge species were omitted from the analysis (Appendix B4). We evaluated the phylogenetic relatedness of focal sedges with co-occurring sedge species using two complementary metrics, the net relatedness index to focal (FNRI) and nearest taxon index to focal (FNTI), by modifying community structure metrics from the PICANTE R-library (Kembel et al. 2010). We derived FNRI as the standardized mean phylogenetic distance between the focal species and each of its co-occurring species as follows: (i) for each focal species, we first calculated the mean of the phylogenetic distances between the focal individual and its co-occurring species (FNRIobs) and (ii) we then generated a null distribution of expected phylogenetic distances (FNRIrand)and the standard deviation of these distances (FNRIstd) from 999 simulated communities generated by randomly sampling the regional species pool (excluding the focal species) while maintaining sample species richness. A standardized effect size was then calculated as (FNRIobs - FNRIrand) / FNRIstd and multiplied by -1 to return the final FNRI value for each plot. To determine FNTI, the phylogenetic distance between the focal species and its

most closely related co-occurring species was first calculated to get FNTIobs. Equivalent

FNTIrand and FNTIstd were generated in a similar manner to FNRIrand and FNRIstd, but calculations only considered the phylogenetic distance to the closest-related species per 49 3.4. Results

plot. Abundance weighted metrics for both FNRI (FNRIab) and FNTI (FNTIab)were also estimated and all analyses were conducted on 100 randomly selected phylogenetic trees. Observed MPD, MNTD, FNRIobs and FNTIobs values occurring in the lowest or highest 2.5% of the distribution of distances from their respective null communities were considered significant (Kembel & Hubbell 2006). We evaluated co-occurrence patterns at nested taxonomic scales. First, we calculated FNRI and FNTI values for all 35 focal sedges and co-occurring Cyperaceae species for all study plots. In a second analysis, we calculated FNRI and FNTI values only for the 29 Carex focal plants and all co-occurring Cyperaceae species in the dataset. Third, we examined only Carex focal plants and their co-occurring congeners. For each set of analyses, we estimated the mean FNRI and FNTI for each focal species. One-sample t tests were used to determine if mean NRI, NTI, FNRI, or FNTI values differed from zero at the different taxonomic scales. Differences in mean FNRI and FNTI among clades were evaluated using Analysis of Variance (ANOVA) and the Tukey-Kramer multiple comparisons of means test (Tukey 1949; Dunnett 1980) using the multcomp R package (version 1.3, http://multcomp.r-forge.r-project.org/).

3.4 Results

3.4.1 Aggregated community-level metrics

Mean NRI was significantly less than zero, suggesting overdispersion, whereas NRIab

and NTIab values were significantly greater than zero, suggesting phylogenetic cluster- ing (Table 3–1). However, mean NTI values did not differ significantly from zero (Table 3–1), and individual plots showed a mix of clustering and overdispersion. Overall, signif- icant phylogenetic clustering was found in ∼ 1% of plots, whereas there was significant phylogenetic overdispersion in just over 7% of the plots in all four analyses combined (Table 3–1). In addition, mean values for both FNRI and FNTI approached zero when aggregating across all Cyperaceae focal plants and co-occurring Cyperaceae, with and 50 3.4. Results

without weighting by percent cover (Table 3–1, Fig. 3–1a). In summary, we found that the aggregated community-level metrics approach zero, thus supporting neither clustering nor overdispersion.

3.4.2 Phylogenetic clustering in Trichophorum and overdispersion in Erio- phorum

All four focal species metrics were significantly greater than zero within Trichophorum (Table 3–2), indicating a significant trend towards phylogenetic clustering. Further, a trend towards phylogenetic overdispersion was evident within Eriophorum,withmean

FNRIab and FNTI significantly less than zero, although mean FNRI and FNTIab values did not differ significantly from zero (Table 3–2). In Carex, however, we found no sig- nificant trend for either clustering or overdispersion in any of the four metrics (Table 3–2). When directly comparing the distribution of co-occurrence statistics across genera, we observed significantly higher FNTI, FNRI and FNRIab values within Trichophorum

compared to within Eriophorum and Carex (Table 3–3, Fig. 3–1b), although mean FNTIab values did not differ significantly (Table 3–3). In addition, FNTI values were significantly higher for Carex compared to Eriophorum. Even though the other three neighbourhood metrics did not significantly differ between the two genera, mean values for the percent- cover metrics were greater in Carex compared to Eriophorum (Table 3–3).

3.4.3 Phylogenetic clustering in the Core Carex and overdispersion in the Vignea clade

When focusing the taxonomic scope of our analysis on the Carex focal plants but consider- ing all co-occurring Cyperaceae, we observed random phylogenetic structuring for all four metrics (Table 3–4). However, interesting clade-specific patterns emerged when we split

Carex into the Vignea,CoreCarex and Caricoid clades. Mean FNRI, FNRIab and FNTI were significantly less than zero for the Vignea clade (Table 3–4), indicating phylogenetic overdispersion in this clade. Percent cover-weighted FNTI values showed a similar trend, 51 3.4. Results

but not significantly so. In contrast, phylogenetic clustering was evident in the Core Carex

clade (Table 3–4), with mean FNRIab, FNTI, and FNTIab values all significantly greater than zero. Mean FNRI values did not differ from zero. Contrasting patterns of phyloge- netic overdispersion and clustering were evident in the Caricoid clade (Table 3–4), with mean FNRI and FNRIab significantly greater than zero, mean FNTIab significantly less than zero and mean FNTI not significantly different from zero. Furthermore, we observed strong clade level patterns in the co-occurrence metrics when comparing between clades, with significant differences between the Vignea and Caricoid clades for FNRI, FRNIab and

FNTI as well as between the Vignea and Core Carex clades for FNRI, FNRIab, FNTI and

FNTIab (Table 3–3; Fig. 3–1c). Overall, our analyses indicate that phylogenetic overdis- persion was more evident in the Vignea clade compared to the Core Carex and Caricoid clades, with clustering more apparent in Core Carex (Table 3–3; Fig. 3–1c).

3.4.4 Reduced species pool enhances contrasting patterns of phylogenetic clustering and overdisperion in Carex

The clade-specific patterns were more pronounced when we further restricted our compar- isons to Carex focals and co-occurring Carex congeners (Fig. 3–2b). Here, we observed mostly random phylogenetic structure when considering all Carex focal plots for three of the four metrics, with just mean FNRIab indicating some evidence for phylogenetic overdis- persion (Table 3–4). However, we again observed contrasting patterns of phylogenetic overdispersion and clustering among clades (Vignea,CoreCarex and Caricoid); with the differences in mean values greater than when considering all co-occurring Cyperaceae (Ta- ble 3–3). Phylogenetic overdispersion was again evident in the Vignea clade (Table 3–4), with all four focal metrics significantly less than zero. In contrast, Core Carex showed a trend towards phylogenetic clustering (Table 3–4). Only FNRIab indicated phylogenetic clustering in the Caricoid clade, whereas mean FNRI, FNTI, and FNTIab values did not

52 3.5. Discussion

differ from zero (Table 3–4). In summary, we found significant differences in neighbour- hood structure of focal species among Carex clades. Individual focal species values for each of the four metrics are included in Appendices B8 and B9.

3.5 Discussion

In this study, we explored the phylogenetic neighbourhoods of sedges from the Cyperaceae family across several different habitats in subarctic Qu´ebec and adjacent Labrador. We revealed that the phylogenetic composition of the immediate neighbours of an individual plant varies with the clade membership of the species. Specifically, we found that phyloge- netic clustering in Trichophorum was contrasted with a tendency towards overdispersion in Eriophorum and Carex, and similarly, phylogenetic clustering in the Core Carex clade was contrasted by a trend towards overdispersion in the Vignea clade. Thus, as we av- eraged patterns across clades, these opposing trends effectively resulted in an aggregate random pattern of co-occurrence among the three Cyperaceae genera and Carex clades. These results demonstrate that patterns in community structure can vary significantly among clades and, when aggregated across all community members, these differences can mistakenly lead us to infer that community assembly has been a more or less random process. Several studies have previously documented random phylogenetic structuring of plant communities (e.g. Kembel & Hubbell 2006; Swenson et al. 2007; Bryant et al. 2008; Ndiribe et al. 2013), and it is likely that many more studies demonstrating random or null results might never have been published (Fanelli 2012). Here, we suggest that opposing lineage effects might provide one possible explanation for the prevalence of apparently random community structuring. If we had characterized phylogenetic structure by aggregating across focal species, we would have found that patterns of co-occurrence amongst sedges for the most part did not depart from random expectations. From this result we could have inferred that neutral or other processes that leave less of an 53 3.5. Discussion

imprint on phylogenetic community structure might have been dominant in shaping these communities. Our focal species approach, however, allowed us to reveal very different patterns in phylogenetic co-occurrence. We were able to detect significant differences among the Cyperaceae lineages included in our study. Although we acknowledge the difficulties in inferring process from pattern, it is possible, nonetheless, to interpret these lineage-specific differences within the commonly accepted community phylogenetics framework sensu Webb et al. (2002). For example, Trichophorum alpinum, a species that favours high pH conditions and roots near or just above the water table in fens (Dabros & Waterway 2008), clustered with its congener, Trichophorum cespitosum. These Trichophorum species share similar niches along the rooting depth to water table gradient in fens, although they may differentiate along pH gradients in these habitats (Dabros & Waterway 2008). However, T. cespitosum is more generalist, and it is additionally found growing across a number of different habitats from which T. alpinum is absent, and as a consequence this species tended to have more phylogenetically distant (overdispersed) neighbours. Nonetheless, the stronger signal for phylogenetic clustering of the more specialist T. alpinum eclipsed the weaker signal for overdispersion among the more generalist T. cespitosum, such that phylogenetic clustering dominates co-occurrence patterns in the two species of this genus represented in our study area. In contrast, Eriophorum co-occurrence patterns were more uniformly overdispersed, with Eriophorum species known to segregate along pH and rooting depth relative to water table gradients (Dabros & Waterway 2008). Even though some species within the genus (e.g. E. chamissonis and E. angustifolium) prefer similar abiotic growing conditions (Dabros & Waterway 2008), they tended to co-occur infrequently. Even within Carex we found contrasting patterns between clades, with a greater tendency towards overdispersion in the Vignea clade and clustering within Core Carex. This pattern remained even when correcting for differences in species occurrence frequen- cies using the ‘trialswap’ null model (Mikl´os & Podani 2004) that randomizes community 54 3.5. Discussion

structure while maintaining species occurrence frequency and site richness (see Appendix B11). The majority of focal species within the Vignea clade were located in fen habitats with the exception of Carex brunnescens (see Appendix B1), which interestingly had the highest phylogenetic clustering values in the clade. In the fen habitat, species within this clade are generally indicators of high pH but segregate by rooting at different depths relative to the water table, with some species rooting at or below the surface of the water table (e.g. Carex chordorrhiza) and others rooting above the surface of the water table (e.g. Carex gynocrates) (Dabros & Waterway 2008). We would typically associate strong relationships between species occurrences and abiotic environment with phylogenetic clustering, but here we observe phylogenetic overdispersion, perhaps suggesting evidence for convergent evolution on this habitat type. Interestingly, species richness was high- est in Carex focal plots that were more phylogenetically overdispersed (see Appendices B10 and B11), such that focal species in the Vignea clade had more neighbouring sedge species compared to focal species in the Core Carex consistent with the hypothesis that phylogenetic relatedness might limit co-occurrence. The Core Carex clade, on the other hand, showed a tendency towards greater clustering, suggesting a greater importance of environmental filtering for the clade. It is notable that Vignea, which tended to demonstrate patterns of overdispersion are predominantly fen species, whereas Core Carex, which tended to demonstrate clustering, are the more common species in the tundra. It is possible therefore that clade differences in species co-occurrence patterns are driven by habitat preferences. However, we did not find any significant trend for dispersion patterns to vary by habitat type (Appendix B12). Further, even within fens, the habitat with the largest number of plots, we still observed significant lineage-specific differences (Appendix B12). Therefore, while broad habitat preferences obviously influence the phylogenetic neighbourhood, they are not sufficient to explain the divergent evolutionary co-occurrence patterns we observed among Carex clades. 55 3.6. Conclusion

Our ecological interpretations of phylogenetic structure remain untested, and further work is needed to evaluate hypotheses relating to fine scale niche partitioning. Adaptation to anoxic, waterlogged environments represents one potential niche axis that could be ex- plored further, since previous work in this system demonstrate that sedges segregate along the rooting depth to water table gradient (Dabros & Waterway 2008). More generally, we emphasize here that phylogenetic structure in patterns of species co-occurrence should be interpreted cautiously. For example, even under assumptions of phylogenetic conservatism, Mayfield & Levine (2010) suggested that competition might lead to clustering rather than overdispersion if coexistence is favoured by similar competitive abilities rather than niche differences.

3.6 Conclusion

Lineage-specific patterns in phylogenetic community structure can bring new insights into community assembly processes. Using a fine-scaled sampling design and a neighbourhood analytical approach, we revealed hidden complexities in community composition. We showed that species of Cyperaceae in the eastern subarctic may have either phyloge- netically overdispersed or clustered neighbours. However, opposing patterns in different subclades resulted in a seemingly random phylogenetic structuring when all species are combined, as is typical in most analyses of phylogenetic community structure. We suggest that this mixed pattern of lineage-specific co-occurrence might be a general feature of ecological communities, especially when communities are comprised of clades with different environmental tolerances. Future studies should allow for lineage-specific responses to opposing community level processes; these complexities can be simply addressed with appropriate study designs and null model selection.

56 Chapter 3 Tables & Figures

Table 3–1 Mean net relatedness index (NRI) and mean nearest taxon index (NTI) val- ues as well as net related index to focal (FNRI) and nearest taxon index to focal (FNTI) for all Cyperaceae plots combined. The difference between the means and zero is repre- sented with P values from one-sample t tests. Each analysis had 612 degrees of freedom. Generally, all aggregated NRI, FNRI, NTI and FNTI values approach zero.

NRI NRI NTI NTI FNRI FNRI FNTI FNTI Sample mean -0.086 0.089 0.006 0.129 -0.024 0.054 0.058 0.038 P value 0.013 0.023 0.885 0.001 0.524 0.165 0.153 0.301 No. of plots high† 129 97 9 200 185 42 27 88 No. of plots low‡ 115 44 76 43 88 26 10 46

Weighted by percent cover † Number of plots with observed MPD, MNTD, FNRIobs and FNTIobs values occurring in the highest 2.5% of the distribution of distances from their respective null communities. See main text for an explanation of the abbreviations ‡ Number of plots with observed MPD, MNTD, FNRIobs and FNTIobs values occurring in the lowest 2.5% of the distribution of distances from their respective null communities

57 Chapter 3 Tables & Figures

Table 3–2 Mean net relatedness index to focal (FNRI) and nearest taxon index to fo- cal (FNTI) for Cyperaceae focal plants divided into major clades. The species included in each analysis are listed in Appendix B4†. The four metrics were significantly greater than zero for Trichophorum. Two or the four metric were significantly less than zero for Eriophorum.

FRNI FNRI FNTI FNTI Clade TRI df 39 39 39 39 Sample mean 0.345 0.825 1.140 0.259 P-value 0.032 0.005 < 0.001 0.037 ERI df 76 76 76 76 Sample mean 0.009 -0.190 -0.334 0.005 P-value 0.269 < 0.001 < 0.001 0.956 CAR df 495 495 495 495 Sample mean -0.075 0.030 0.032 0.026 P-value 0.080 0.459 0.439 0.548

Weighted by percent cover †ERI—Eriophorum, CAR—Carex,TRI—Trichophorum

58 Chapter 3 Tables & Figures

Table 3–3 Differences in mean net relatedness index to focal (FNRI) and nearest taxon index to focal (FNTI) for the different clades of Cyperaceae and Carex included in the study. The difference in the mean values between clades is represented by ‘Diff’ with sig- nificance levels from the Tukey-Kramer multiple comparisons of means test as indicated in the table footnote. Cyperaceae1 analysis includes all Cyperaceae focal species and co- occurring Cyperaceae plants, whereas Cyperaceae2 includes only Carex focal species and co-occurring Cyperaceae plants. The Carex set of analysis is further restricted to Carex focal species and only co-occurring Carex. The species included in each analysis are listed in Appendix B4. Trichophorum was more phylogenetically clustered Eriophorum and Carex for three of the four metrics. Within the Carex,theCoreCarex clade was more phylogenetically clustered than the Vignea clade for all four metrics, whereas the Vignea clade was phylogenetically overdispersed compared to both the Caricoid clade and Core Carex clades.

FRNI FNRI FNTI FNTI Cyperaceae1 P-value ANOVA 0.002 < 0.001 < 0.001 0.755 Diff ERI–CAR† 0.041 -0.084 -0.350 -0.067 P-value 0.924 0.735 0.007 0.822 Diff TRI–CAR† 0.524 0.629 1.056 0.081 P-value 0.002 < 0.001 < 0.001 0.855 Diff TRI–ERI† 0.483 0.713 1.414 0.013 P-value 0.020 < 0.001 < 0.001 0.997

Cyperaceae2 P-value ANOVA < 0.001 < 0.001 < 0.001 0.005 Diff CC–CRC† -0.126 -0.256 0.120 0.285 P-value 0.686 0.217 0.708 0.196 Diff VIG–CRC† -0.525 -0.519 -0.431 0.001 P-value 0.003 0.003 0.070 0.999 Diff VIG–CC† -0.400 -0.262 -0.551 -0.284 P-value < 0.001 0.006 < 0.001 0.006 continued (Table 3–3)...

59 Chapter 3 Tables & Figures

. . . continued (Table 3–3) FRNI FNRI FNTI FNTI Carex P-value ANOVA < 0.001 < 0.001 < 0.001 < 0.001 Diff CC–CRC† 0.385 0.486 0.204 0.337 P-value 0.014 < 0.001 0.286 0.101 Diff VIG–CRC† -0.430 -0.530 -0.530 -0.136 P-value 0.007 < 0.001 < 0.001 0.696 Diff VIG–CC† -0.816 -1.016 -0.725 -0.472 P-value < 0.001 < 0.001 < 0.001 < 0.001

Weighted by percent cover †ERI—Eriophorum, CAR—Carex,TRI—Trichophorum, CC—Core Carex clade, CRC— Caricoid clade of Carex,VIG—Vignea clade of Carex.

60 Chapter 3 Tables & Figures

Table 3–4 Mean net relatedness index to focal (FNRI) and nearest taxon index to focal (FNTI) for Carex focal plants divided into major clades. One-sample t tests were used to determine if sample means differed significantly from zero. The Carex 1 analysis includes all Carex focal species and co-occurring Cyperaceae plants, whereas Carex 2 includes only Carex focal species and co-occurring Carex plants. Degrees of freedom (df) for each anal- ysis are included under the clade title. The species included in each analysis are listed in Appendix B4†. Carex species from the Vignea clade were phylogenetically overdispersed when considering their co-occurrence with all Cyperaceae and only co-occurring Carex. Carex species from the Core Carex clade were phylogenetically clustered when consider- ing their co-occurrence with all Cyperaceae and only co-occurring Carex. Patterns for the Caricoid clade varied across metrics and the taxonomic scope of co-occurring sedges.

Taxa included: Clade† (df) FNRI FNRI FNTI FNTI Carex 1 CAR (495) Diff. from 0 -0.075 0.030 0.032 0.026 P-value 0.080 0.459 0.439 0.548 VIG (184) Diff. from 0 -0.211 -0.277 -0.316 -0.111 P-value 0.002 < 0.001 < 0.001 0.162 CC (272) Diff. from 0 -0.032 0.176 0.272 0.170 P-value 0.594 0.003 < 0.001 0.001 CRC (37) Diff. from 0 0.275 0.483 -0.002 -0.345 P-value 0.039 < 0.001 0.984 0.022 Carex 2 CAR (478) Diff. from 0 -0.060 -0.097 -0.072 -0.046 P-value 0.141 0.012 0.066 0.318 VIG (183) Diff. from 0 -0.412 -0.487 -0.332 -0.206 P-value < 0.001 < 0.001 < 0.001 0.019 CC (256) Diff. from 0 0.405 0.486 0.379 0.233 P-value < 0.001 < 0.001 < 0.001 < 0.001 CRC (37) Diff. from 0 0.007 0.293 -0.043 0.002 P-value 0.927 0.001 0.601 0.990

Weighted by percent cover † CAR—Carex, CC—Core Carex clade, CRC—Caricoid clade of Carex,VIG—Vignea clade of Carex

61 Chapter 3 Tables & Figures

Figure 3–1 Net relatedness index to focal plant (FNRI) based on presence/absence data for the 35 Cyperaceae focal species divided into major clades. Results for all plots com- bined are given in (a). Analysis of variance (ANOVA) indicated significant variation

among clades ((b) F 2,610 = 5.074, P = 0.002; (c) F 2,493 = 10.260, P < 0.001; and (d)

F 2,476 = 108.080, P < 0.001). In addition, significant differences among clades (Tukey- Kramer multiple comparison of means, P < 0.05) are indicated by different letters. Medi- ans for each plot are represented by thick lines, the boundaries of each box show the 25th and 75th percentiles, and whiskers above and below each plot represent the 10th and 90th percentiles. Outlying data is indicated by hollow circles and abbreviations for each clade are the same as those in Appendix B4. The species included in each analysis are listed in Appendix B4. Trichophorum was more phylogenetically clustered than the other two genera. Within the Carex,theCoreCarex clade was more phylogenetically clustered than the Vignea clade, whereas the Vignea clade was phylogenetically overdispersed compared to both the Caricoid clade and Core Carex clades. Corresponding P-values are given in Table 2–3. 62 Chapter 3 Tables & Figures

Figure 3–2 Mean FNRI values mapped per focal species weighted by percent cover es- timates for (a) all co-occurring Cyperaceae in the study and (b) only co-occurring Carex. Mean values were mapped onto one of 100 randomly chosen phylogenies from the Bayesian posterior distribution using the contMap function available in the phytools package (Rev- ell 2012) of R and are for illustration only. Blue colouring represents focal species with higher FNRI, whereas red colouring represents focal species with lower FNRI values. The colour bar in the bottom left hand corner of each phylogeny gives the scale of FNRI values and corresponding colours for each phylogeny. Abbreviations for each clade are the same as those in Appendix B4. Phylogenetic overdispersion in the Vignea clade and phyloge- netic clustering in the Core Carex clade became more evident when the species pool was narrowed to only co-occurring Carex. 63 Linking Statement #2

In Chapter 3, I showed lineage-specific patterns in co-occurrence that are obscured when species are aggregated into larger clades, and I demonstrated that these patterns became more evident when the species pool was narrowed to Carex focal plants and co-occurring congeners. In Chapter 4 I also investigate patterns of co-occurrence in Cyperaceae, and examine whether the specialist species is the better competitor; a question that has been difficult to explore in field studies. According to modern coexistence theory, coexistence is a balance between competitive inequalities and niche differences, and coexistence should be promoted when niche differences are large and competitive imbalances are relatively small. In Chapter 4 I evaluate this theory, equating niche differences to phylogenetic distances and differences in competitive abilities to dissimilarities in niche width. The results from this chapter suggest that dissimilarities in niche widths equate to differences in competitive abilities and show that coexistence theory can predict co-occurrence between species with different niche widths and phylogenetic relatedness. CHAPTER 4

Competitive trade-offs and species co-occurrence: integrating coexistence theory and community phylogenetics

Elliott, T.L.1 & Davies, T.J.1

A version of this chapter has been submitted to Ecology.

1 Department of Biology, McGill University, Montr´eal, Qu´ebec, Canada

65 4.1. Abstract

4.1 Abstract

Theory predicts that ecological specialists should be better competitors than generalists; however, few studies have explored the relative competitive ability of specialists versus generalists in natural systems. Here, we evaluate evidence for a trade-off between niche width and competitive ability by extending ideas from coexistence theory which suggest that coexistence among competing species is possible if they have similar competitive abilities or if their niche differences are large. We use data on sedges across 500 km2 of the Canadian subarctic to examine this trade-off under the assumption that phylogenetic distances reflect niche differences and niche breadth correlates with competitive ability. We find higher co-occurrence between phylogenetically distant species with comparable niche widths, although we explain only a small amount of the total variation in co- occurrence. We suggest that differences in niche widths translate into differences in competitive abilities and provide support for a trade-off between competitive ability and ecological generalization.

66 4.2. Introduction

4.2 Introduction

A central theme in community ecology has long been the trade-offs that allow co- occurrence among competing species (MacArthur & Pianka 1966; MacArthur 1972). Following Hutchinson’s conceptual framework of the fundamental niche describing the resources required for a species to persist (Hutchinson 1957), theoretical (e.g. MacArthur & Levins 1964; Colwell & Futuyma 1971) and empirical studies (e.g. McNaughton & Wolf 1970; Thompson et al. 1999) showed that species also differ in their relative niche widths. Species can thus be categorized according to their niche widths; a specialist is defined as a species with a narrow niche width, whereas a generalist species survives across a wider range of environmental conditions (MacArthur & Pianka 1966; MacArthur & Levins 1967). Because a species cannot be a superior competitor across all niches, a potential trade-off between niche width and competitive ability was proposed, with the greater niche width of a generalist coming at the expense of the better competitive ability of specialists (MacArthur & Levins 1967; Wilson & Yoshimura 1994). In this paper, we present a framework that combines advances in coexistence theory and community phylo- genetics (Mayfield & Levine 2010) to evaluate the trade-offs in the competitive abilities of generalists versus specialists (Fig. 4–1). Recently, coexistence theory has been extended to predict the likelihood of compet- itive exclusion based on differences in competitive ability (Chesson 2000; Abrams 2007). In general, we expect co-occurrence to be promoted when species differ in their resource requirements (i.e. niche differences are large), leading to weaker competitive interactions between them (Chesson 2000; Mayfield & Levine 2010; HilleRisLambers et al. 2012). How- ever, when resource requirements are similar (i.e. niche differences are small), we expect greater co-occurrence when competitive differences between species are small (Chesson 2000; Mayfield & Levine 2010; HilleRisLambers et al. 2012). Thus, co-occurrence is a balance between relative competitive strength and niche differences.

67 4.2. Introduction

Independent from these advances in coexistence theory, the field of community phylogenetics has emerged and expanded rapidly over the last fifteen years, stimulated by the increasing availability of phylogenetic information (Vamosi et al. 2009; Cavender- Bares et al. 2009). Interpretations based on this framework assume phylogenetic distances capture niche differences (Webb et al. 2002; Cavender-Bares et al. 2009); evidence that co-occurring species are more distantly related than expected has thus been interpreted as evidence that competition is the more important force structuring community composition (Webb et al. 2002). However, Mayfield & Levine (2010) suggested that the community phylogenetic paradigm was out of step with Chesson’s coexistence theory, and that close relatives (i.e. those predicted to occupy similar niches) might be able to coexist when their competitive differences are small (Chesson 2000). Here again it is the balance between niche differences (phylogenetic distance) and competitive abilities that promotes co-occurrence, with competitive exclusion occurring when the difference in competitive abilities between two species is greater than their niche difference. Here, we extend the framework of Mayfield & Levine (2010) to examine the trade- offs between specialism and competitive ability. Specifically, we test whether differences in niche widths translate to differences in competitive abilities, assuming phylogenetic distance captures niche differences. Under the trade-off hypothesis (Fig. 4–1), we would predict co-occurrence to be more likely among species with large niche differences (phylo- genetically distant) and similar competitive abilities (similar niche widths). In contrast, competitive exclusion would be most likely among closely related species with large differences in niche width. We evaluate predictions using co-occurrence patterns and phylogenetic distance data for sedges (plants of the Cyperaceae family) collected at the southern edge of the forest tundra in the Canadian subarctic, an area an area rich in sedge diversity. This species-rich clade is composed of closely related life forms with similar environmental tolerances and diverse habitat preferences, and therefore provides a model group for studying the importance of interspecific interactions on community composition.

68 4.3. Methods

4.3 Methods

4.3.1 Study system and experimental design

We evaluated patterns of co-occurrence within a single, strongly supported clade of the Cyperaceae family (sedges) (Muasya et al. 2009) in a 500 km2 area surrounding Schefferville, Qu´ebec during the 2012 growing season. We sampled 680 1m2 plots (20 plots centered on each of the 35 most common sedge species in the Schefferville area) using a combination of the ‘ignorant man’ method (Ward 1974) and random walks. A more detailed description of the sampling design is provided in Chapter 3 (see Appendix B1 and B2).

4.3.2 Analyses

We scored each species on a gradient of specialist to generalist using an index of habitat

diversity (Hi) according to the habitat descriptions in Waterway et al. (1984). To correct for variation in sampling effort in our dataset, we took the inverse of the number of plots sampled in each habitat type (Pj) divided by the total number of plots (PT ), as follows: 1 H = (4.1) i Pj PT We then weighted the per species frequency within a habitat type using this cor- rection factor. Shannon’s diversity was calculated on the corrected habitat values in the vegan package (Oksanen et al. 2013) of R version 3.1.0 (R Core Team 2014) to provide an index of habitat specificity (niche width) for each species. To explore model sensitivity, we evaluated two additional metrics of specialization (Appendix C1). Each index was rescaled between zero and one, with specialists having lower values and generalists having higher values. Pearson’s product-moment correlations were calculated to quantify strength of covariation among metrics. Phylogenetic signal was evaluated for each of the metrics separately using the K statistic (Blomberg et al. 2003) in the picante package (Kembel et al. 2010) of R.

69 4.3. Methods

We explored the relationship between co-occurrence, phylogenetic distance and niche width using a co-occurrence matrix across all pairwise combinations of the 34 Cyperaceae species. We derived the observed co-occurrence values by transposing the species presence matrix against itself and extracting the upper triangle. Standard deviations and mean expected co-occurrences were then calculated by conducting 999 random permutations of the pairwise co-occurrence matrix while maintaining plot species richness and species frequencies across the matrix. Standardized co-occurrence scores for each species pair were calculated by subtracting the mean expected from the observed co-occurrence values for each species pair and dividing this by the standard deviations of the expected co- occurrences. We extracted the pairwise phylogenetic distance matrix using the cophenetic function in the APE package of R (Paradis et al. 2004) on the maximum clade credibility tree taken from the posterior distribution of 36,000 trees from Chapter 3 (see Appendix B6). Phylogenetic distances were then square root transformed to reduce the effects of increasing variance with greater distances (Letten & Cornwell 2015). Finally, we computed a matrix of the absolute differences in niche width (as estimated above) for each species pair. Statistical relationships between differences in niche width, phylogenetic distance and co-occurrence were described using generalized additive models (GAMs) with a tensor product smoothing parameter (Wood 2006). Significance values are not reported because of the non-independence of species pairs, although all models were highly significant (P << 0.001). We also conducted an additional set of matching analyses that included only species pairs with divergence times within the last 15 million years (i.e. < 30 million years cophenetic distance), to focus on co-occurrence patterns among close relatives among which we might predict biotic interactions would be stronger.

70 4.4. Results

4.4 Results

4.4.1 Niche width

All reported results exclude Carex deflexa Hornem., which was omitted from the analysis because it occurred in habitats (e.g. coniferous woodlands and rock outcrops) that other sedges in the study generally did not occur (Appendix C2: Table C2-1; Fig. C2-1), thus strongly skewing our index of specialization (i.e. all other species would appear to be generalists relative to C. deflexa). All three specialization indices were significantly correlated (r = 36–44), although subsequent analyses were sensitive to choice of metric (see Discussion; Appendix C3: Fig. C3-1 and C3-2). Blomberg’s K values for all indices indicated that there was little or no phylogenetic signal in specialization (K = 0.09–0.16, all P > 0.05; Appendix C3: Fig. C3-1). Here, we present results using our Shannon index of specialization because we think it is the better estimate of the realized niche width of the species in our study system, and it has been used similarly before (e.g. Thompson et al. 1999). Matching results for the other indices are presented in Appendix C3.

4.4.2 Pairwise species comparisons

We found that generalists were more likely to co-occur with generalists and specialists were more likely to co-occur with specialists (Fig. 4–2a), and species co-occurrence increased with phylogenetic distances (Fig. 4–2b). The GAM including both phyloge- netic distance and the difference in niche width as well as their interaction explained approximately 8% (deviance explained = 8.06%) of the variance in species co-occurrence. Co-occurrence was highest among very distant relatives (> 90 MYA) with large differ- ences in niche width, although this pattern was driven by only a few species pairs. High co-occurrence was also observed among species pairs of intermediate phylogenetic distance but with large differences in niche width (Fig. 4–3a). Co-occurrence was lowest between species with large differences in niche width separated by intermediate phylogenetic distances. 71 4.5. Discussion

When considering species pairs separated by less than 30 million years (Fig 4–2c and Fig. 4–2d), percent deviance explained marginally increased (deviance explained = 9.07%; Fig. 4–3b), and we observed the highest co-occurrence among more distantly related species with larger differences in niche width.

4.5 Discussion

Ecological theory suggests a trade-off between niche width and competitive ability such that more generalist species (those with wider niche width) tend to be weaker competitors. Separately, coexistence theory predicts that stable co-existence is only possible when species niche differences exceed their differences in competitive ability, whereas competitive exclusion occurs when the reverse is true (Chesson 2000; Mayfield & Levine 2010; HilleRisLambers et al. 2012). In this study, we introduce a novel framework that combines coexistence theory with the common assumption from the community phylogenetics literature that phylogenetic distances reflect niche differences between species (Fig. 4–1; Webb 2000). We then evaluate evidence for a trade-off between niche width and competitive ability using patterns of species pairwise co-occurrences and a phylogenetic tree of sedges from subarctic Canada. If greater niche width translates to lower competitive ability we would predict that co-occurrence should be more common among species separated by large phylogenetic distances with similar niche widths (Fig. 4–1). Our results provide evidence supporting the prediction that co-occurrence should be low for closely related species (where niche differences are low) with large differences in niche width (where competitive differences are high). Further, this trend is stronger when we consider only more recently diverged species pairs (species pairs with divergence times < 15 MYA). The strength of biotic interactions on community structure is expected to be greater within plots of relatively small size and for clades of fairly closely related species—a scale termed the ‘Darwin-Hutchinson zone’ by Vamosi et al. (2009). Our small

72 4.5. Discussion

plot sizes likely fall within this zone, but the larger evolutionary distances separating the major clades might fall outside. When we exclude co-occurrence patterns between more distant relatives, not only do we find that explanatory power increases, but observed patterns matched more closely to expectations (compare Fig. 4–3b with Fig. 4–1); that is, co-occurrence was highest among more distantly related species with similar niche widths and lowest among closely related species with large differences in niche width. We consider our index of habitat specialization to be most representative of the realized niche width of the species in our study system, but we explored several alternative indices of specialization and show results differed depending on the choice of index (see Appendix C3). Capturing all dimensions of Hutchinson’s n-dimensional hypervolume would be an almost impossible task (Colwell & Futuyma 1971; Futuyma & Moreno 1988), and thus empirical data is always likely to fall short of theory. Nonetheless, we found qualitatively similar patterns for two of the three indices examined, and only our metric of specialization indexed on global geographic extent did not result in a similar relationship. It is perhaps not surprising that this latter index demonstrated different properties, as global range extent reflects both a species’ intrinsic niche preferences and the geographical extent of preferred habitats. For example, it is possible for a species with a narrow niche preference to have a large global extent if the preferred habitat type is geographically widespread, as might not be uncommon for species within the tundra biome. Despite high significance, our models explain only a small proportion of the variation in species co-occurrences. There are several possible explanations for the low explanatory power of our models. We make various assumptions in our analysis, some of which may only hold weakly. For example, we assume that phylogenetic distance correlates with niche differences. Although this relationship has been subject to debate in the literature (see Godoy et al. 2014), there are strong theoretical grounds for expecting that phylogenetic distance provides a good proxy for niche differences, especially when the niche is defined by a complex combination of traits. However, perhaps more importantly, other ecological

73 4.5. Discussion processes are undoubtedly important in determining species co-occurrence in our study system. For example, stochastic or neutral processes could dominate locally (Wiegert 1974; Aarssen 1983; Hubbell 2001), facilitative interactions might occur among some neighboring plants, so that not all pairwise interactions are competitive (Bertness & Callaway 1994), and non-sedge species might also exert competitive or facilitative effects. Further, our model only predicts the potential for co-existence; hence, the relationship between both phylogenetic distance and niche width with co-occurrence is approximately triangular: distant relatives and species pairs with similar competitive abilities might be more able to coexist, but they are of course not obliged to coexist. We have shown that co-occurrence is more likely among species that are phylogenet- ically distant and have similar niche widths. Our results support Chesson’s coexistence framework, where co-occurrence is possible in the face of competitive asymmetries if niche differences are small (Chesson 2000), assuming phylogenetic distance is a reasonable proxy for niche difference. We suggest future work could extend this framework to consider additional metrics to quantify competitive outcomes (e.g. survival percentage, primary production and germination rate — Callaway 2007) and look to integrate the competitive effects of multiple interacting species, for example, within a spatially-explicit neighborhood model (e.g Uriarte et al. 2010). As species ranges shift with climate change and communities become shuffled, it is increasingly important that we better incorporate biotic interaction and coexistence theory into species distribution models so that we can more accurately predict the impact of future changes on the distribution of biodiversity (Wisz et al. 2013). Here we have introduced a novel framework that allows us to examine relative competitive differences in coexisting species with different niche widths. Although this question has been explored previously through theory and experiments, we have shown how it is possible to address this same question within natural communities using new tools that incorporate phylogenetic relationships among species.

74 Chapter 4 Figures

Figure 4–1 Predicted relationship between phylogenetic distance and difference in niche width (adapted from Mayfield & Levine 2010). Dark shading represents high probability of co-occurrence, and light gray represents areas of low expected co-occurrence.

75 Chapter 4 Figures

Figure 4–2 Relationship between species co-occurrences and differences in niche width (a) and phylogenetic distances (b). Figures (c) and (d) show the same relationships, but for only those species pairs separated by phylogenetic distances less than 30 million years. Species co-occurrences represent standardized z-scores (see main text), and phylogenetic distance is square root transformed. Lines are for illustration only and represent best fit slopes from OLS regression (see Appendix C3).

76 Chapter 4 Figures

Figure 4–3 Relationship among phylogenetic distance, niche width and species co-occurrence fitted with a generalized additive model (see main text for details). Contours are on the scale of standardized co-occurrences. Predicted co-occurrence varies from high (dark) to low (light). Panel (a) shows the comparisons among all possible species pairs, whereas (b) includes only those species pairs separated by less than 30 million years.

77 Linking Statement #3

In Chapters 3 and 4, I examined the phylogenetic relationships of co-occurring species within individual plots, that is, co-occurrence patterns at the alpha diversity level. Chapter 5 extends my work to consider beta diversity, exploring among-plot species co-occurrence patterns. In this chapter I examine whether phylogenetic beta diversity is an effective method of delineating ecological boundaries at the local scale. I show that phylogenetic beta diversity is not any better in detecting ecological boundaries than more traditional beta diversity methods based on species presence/absence data, questioning the added value in incorporating phylogenetic information within local-scale analyses of beta diversity. CHAPTER 5

Delineating ecological boundaries at local scales: a comparison of phylogenetic versus species beta diversity metrics

Elliott, T.L.1 & Davies, T.J.1

A version of this chapter has been submitted to Ecography.

1 Department of Biology, McGill University, Montr´eal, Qu´ebec, Canada 79 5.1. Abstract

5.1 Abstract

The boundary separating the boreal forest and arctic tundra biome is moving northwards in response to climate change, but delineating its edge is complex because of heterogeneity in local scale vegetation. Recently, phylogenetic beta diversity (PhBD) indices that cap- ture information on shared evolutionary history have been used to help delineate biome boundaries at global and regional scales, providing additional insights into the ecological and historical processes determining the geographical limits of biomes. To date, the appli- cation of PhBD metrics for delineating boundaries at local scales, where community-level processes are more important, have yet to be fully explored. In this paper, we combine vascular plant co-occurrence data collected from an elevation gradient in Labrador, Canada, with information on phylogenetic relatedness to compare PhBD methods for boundary delineation with more traditional beta diversity (BD) approaches that use only species occurrence data. Assuming phylogenetic conservatism in abiotic niche preferences, we predicted that PhBD metrics would more closely match to abiotic boundaries com- pared to traditional species BD indices. In addition, we predicted that species clustered by habitat preferences would be more closely related than expected by chance. However, contrary to our expectations, we found that neither BD nor PhBD aggregated sites into clusters that reflected their environmental differences. Further, we found no evidence that species within patches were any more closely related than a random draw from the local species pool. Our results indicate that more novel phylogenetic beta diversity metrics do not necessarily provide additional information on ecological boundaries than species beta diversity metrics. However, we suggest that phylogenetic metrics might still be advanta- geous in delineating vegetation patches at the local scale when alpha taxonomy for a group is poorly known.

80 5.2. Introduction

5.2 Introduction

The biogeographic boundary separating the tundra and forest biomes is moving north- wards as the northern climate warms, and shifts in individual species distributions and entire communities are expected to follow (Zoltai 1988; Timoney et al. 1992; Chapin et al. 2005; Bernstein et al. 2007; Elmendorf et al. 2012). These shifts will likely be accompanied by altered ecological processes, such as changes in nutrient cycling, energy budgets and species interactions, which together comprise the biological responses to climate change (Fagan et al. 2003; Chapin et al. 2005). It is thus important to find more effective methods to detect shifting ecological boundaries and the patches that these boundaries separate if we are to make predictions on the ecological impacts of climate change. However, delineat- ing these landscape elements is complex as they vary both spatially and temporally (Fagan et al. 2003), and patches often do not have easily identifiable edges (Post et al. 2007). Choosing a precise yet workable definition that encompasses both patches and boundaries is also a challenging task (Post et al. 2007; Yarrow & Mar´ın 2007). Among the several possible definitions that encompass both boundaries and patches, we define here ecological boundaries as contact zones that arise when heterogeneous areas are partitioned into patches (Fagan et al. 2003; Cadenasso et al. 2003; Strayer et al. 2003). Patches and the boundaries that separate them might or might not share similar characteristics, but some defining characteristic must distinguish patches from each other, and the gradient associated with that characteristic is steeper in the boundary than in the patches (Cadenasso et al. 2003). Generally, it is the specific research question at hand that guides the methods and scale chosen to delineate patches, but most often they are distinguished compositionally, structurally or functionally by floristic or abiotic criteria (Cadenasso et al. 2003; Yarrow & Mar´ın 2007). The boundaries separating patches can be simple or convoluted in shape, and they are often not homogenous (Cadenasso et al. 2003; Strayer et al. 2003; Yarrow & Mar´ın 2007). Examples of ecological boundaries differing in

81 5.2. Introduction

shape and size include the well-studied boundaries between wetland and upland patches (Fortin et al. 2000), and, most relevant here, those between tundra and forest patches (Timoney et al. 1992). The boreal forest gradually transitions into the arctic tundra biome towards high latitudes, with vegetation ranging from closed boreal forests to the south to the treeless tundra in the more recently deglaciated north (L¨ove 1970; Timoney et al. 1992; Gajewski et al. 1993; Payette et al. 2001). Vegetation patterns are often heterogeneous within these two large biomes, creating a mosaic of patches on the landscape (Gajewski et al. 1993; Payette et al. 2001). For example, in addition to forested and tundra sites, a subarctic region might also have patches of other vegetation types, such as shorelines and peatlands (Waterway et al. 1984). Furthermore, as the richness of plant species decreases northwards (Fischer 1960; Willig et al. 2003), there are also accompanying shifts in taxonomic composition, with declines in the richness and abundance of non-angiosperm taxa, such as gymnosperms and ferns (Porsild & Cody 1980; Timoney et al. 1992; Kessler et al. 2011). Delineating ecological boundaries between patches has been a major theme in the eco- logical literature for the past several decades and has given rise to a diversity of methods and approaches. Since Womble (1951) introduced a method to delineate ecological bound- aries based on rates of change from regularly or irregularly spaced data (Fortin & Drapeau 1995), others have built on his spatial methods by, for example, adapting them to other forms of data and applying more formal test statistics (Fortin & Drapeau 1995; Fortin & Dale 2005; Jacquez et al. 2008; Fitzpatrick et al. 2010; Legendre & Legendre 2012a). The analyses of beta diversity, calculated from species presence-absence or abundance data, has provided an alternative approach for delineating ecosystem boundaries and the grouping of sites into patches of similarity (Whittaker 1960; Fortin & Dale 2005). A number of beta diversity indices have been developed that allow us to measure species turnover along environmental gradients (Vellend 2001; Anderson et al. 2011), with areas of high turnover suggested to indicate ecological boundaries (Fortin & Dale 2005). 82 5.2. Introduction

Recently, advances in phylogenetic methods and the availability of well-resolved phylogenetic trees have provided new insights into ecological patterns, with approaches combining community ecology and phylogenetics gaining popularity over the past decade (Webb 2000; Vamosi et al. 2009; Cavender-Bares et al. 2009). One of the basic assump- tions underlying community phylogenetics is that closely related species should be more ecologically similar, and thus should be filtered into similar environments, although ev- idence for this assumption is mixed (Cavender-Bares et al. 2004; Cavender-Bares et al. 2009; Graham et al. 2009; Vamosi et al. 2009). Traditional species-based beta diversity in- dices have also now been extended to include phylogenetic information (Bryant et al. 2008; Nipperess et al. 2010; Chiu et al. 2013). Similar to traditional measures of beta diversity, phylogenetic beta diversity (PhBD) indices can also be used to examine turnover along environmental gradients (e.g. Bryant et al. 2008; Graham et al. 2009) and thus delineate ecological boundaries (e.g. Holt et al. 2013). We might predict that PhBD indices would be more sensitive to detecting environmental turnover and thus ecological boundaries (Graham & Fine 2008) if phylogeny captures niche preferences through evolutionarily conserved traits (i.e. phylogenetic niche conservatism; Wiens et al. 2010). In this study, we explore fine-scale vegetation boundaries in the Canadian north. We compare traditional beta diversity approaches (BD) for delineating patches to approaches that include phylogenetic information (PhBD). We predict that patches defined using PhBD will more closely resemble those based on environmental data, and that patches will be composed of species that are closely related because abiotic variables should ‘filter’ species with similar traits into similar vegetation patches. Further, we predict different clustering patterns for PhBD when patches are defined by all vascular plants compared to only angiosperms, as vegetation types in this region are frequently characterized by changes in the abundance and richness of non-angiosperm taxa. We examine these predictions using vascular plant data collected from a sampling grid of 176 plots in the Canadian subarctic boreal-tundra transition zone. Our study site was located along a 83 5.3. Methods

sharp altitudinal gradient at the latitudinal transition zone separating these vegetation types, resulting in rapid turnover of vascular species within relatively short distances. Thus, we capture turnover of species typical of different biomes within several hundred metres, making this an ideal system to examine questions related to species compositional turnover and ecological boundaries.

5.3 Methods

5.3.1 Sampling

This study was conducted on Mnt Irony, Labrador (54.89◦N; 67.17◦W) in the Canadian subarctic, where winters are long and summers are cool and wet (Lechowicz & Adams 1978). The most dominant plant communities in this region include spruce-lichen wood- lands, spruce-moss forests, subalpine heath, alpine tundra, fens and shoreline communities (Waterway et al. 1984). We established 176 plots across the south-facing slope of Mnt Irony and into the lower elevation spruce-moss forest and fens. The first 88 plots for this study were positioned along a sampling grid with eight elevation bands located at 25 metre height intervals down the slope of the mountain. The horizontal distance between plots along each elevation band was approximately 100 metres with a total of 11 transects going down the slope. The distance between elevation bands was less than 100 metres. An additional 88 plots were added to the lower, flatter section of the grid at intervals of 100 metres, so that the 11 transects each had 16 sampling bands. At each plot, percent cover for all vascular plants within a 1.0 m2 square quadrat was estimated by consensus of two independent observers during July 2013, and elevation, slope and ‘sky visible’ (canopy cover over breast height) recorded. Each of the 176 plots were categorized into one of the following classes as an estimate of soil moisture: 1) hydric—sites covered by water for part of the year, 2) mesic-hydric—generally wet sites that were not covered by water for part of the year, 3) mesic—sites with intermediate 84 5.3. Methods

moisture levels that were not inundated for part of the year, 4) xeric-mesic—intermediate sites between xeric and mesic, and 5) dry sites that could be wind-swept and/or had high proportions of rock cover. In addition, depth to impermeable layer was measured by taking four measurements approximately 25 cm from each of the corners of the plots with an iron rod.

5.3.2 Phylogenetic reconstruction

The current names and authorities of the species sampled within our plots are in accor- dance with VASCAN (Brouillet et al. 2013, see Appendix D1: Table D1–1 for further details). A representative herbarium voucher of each species was accessioned into the McGill University Herbarium (MTMG), with duplicates accessioned into the Marie- Victorin Herbarium (MT). To reconstruct phylogenetic relationships, we first created a backbone constraint tree resolved to the family level using the Phylomatic online software ver. 3 (Webb & Donoghue 2004) for the vascular plant species recorded across the site (Appendix D1: Table D1–1). Four lycophte taxa found in our plots (Diphasiastrum complanatum (L.) Holub, Huperzia appressa (Desvaux) A.´ L¨ove & D. L¨ove, Lycopodium annotinum L. and Selaginella selaginoides (L.) P. Beauv. ex Mart. & Schrank) were manually added to the backbone constraint tree as an outgroup. An additional six species from outside of the study area were included in the phylogenetic reconstruction to improve resolution and branch length estimations in the Santalales and Cornales orders, as these clades had few species represented in our plots and proved problematic during our preliminary phylogenetic analyses. In addition, we also included 11 other vascular species collected from outside of our plots but found on Mnt Irony to further improve phylogenetic resolution. Gene sequences were retrieved from GenBank (Benson et al. 2010) and BOLD (Ratnasingham & Hebert 2007) for two plastid (rbcL protein coding region and matK coding region) and two nuclear ribosomal spacers (ITS1 and ITS2). For species

85 5.3. Methods

lacking publically available sequences, the African Centre for DNA barcoding sequenced an additional 22 rbcL and 13 matK sequences from specimens collected on Mnt Irony (sequences available on GenBank and BOLD; Appendix D1: Table D1–2). The ITS1 and ITS2 sequences were further cleaned using relaxed block selection criteria with Gblocks version 0.9 1b (Castresana 2000; Talavera & Castresana 2007). Sequence alignment was conducted using MAFFT version 7 (Katoh & Standley 2013) and manually edited in BioEdit version 7.0.5.3 (Hall 1999). Sequences for the four regions were concatenated with SequenceMatrix 1.7.8 (Meier et al. 2006), producing a final matrix of 3227 base pairs in length, with 35 missing sequences from 29 different species (see Appendix D1: Table D1–2). The optimal model of sequence evolution for each gene region was assessed using jModelTest2 (Darriba et al. 2012), based on corrected Akaike information criterion values calculated using 11 substitution schemes and a fixed BIONJ-JC starting tree. A general time-reversible model estimating rate variation across sites using a gamma distribution (GTR + G) was selected for rbcL and ITS, while the TVM + G model was selected for matK. Maximum likelihood (ML) phylogenetic reconstruction was performed in GARLI 2.01 (Zwickl 2006) enforcing the Phylomatic backbone as a constraint. Node posterior probabilities were calculated from 100 ML bootstrap runs using the SumTrees package in the DendroPy 3.12.0 phylogenetic computing library (Sukumaran & Holder 2010, see Appendix D1, Fig. D1–1). Finally, phylogenetic branch lengths were made proportional to time using Bayesian inference in Beast v1.8.0 (Drummond et al. 2012) assuming an uncorrelated lognormal relaxed clock model, with nine independent calibration points determined by fossil evidence (Appendix D1: Table D1–3). Six independent analyses were run for 50,000,000 generations each and assessed for convergence by examining estimated sample sizes, marginal probability distributions and traces in Tracer (version 1.6, http://tree.bio.ed.ac.uk/software/tracer/). The phylogeny used in the following analyses is shown in Appendix D1: Fig. D1–2. 86 5.3. Methods

5.3.3 Environmental and beta diversity distances between plots

Environmental distances between plots were calculated using data on elevation, slope, sky visible, soil moisture and depth to impermeable layer. Soil moisture was quantified on a scale between 1 and 5, with hydric and xeric sites receiving the lowest and highest values, respectively. Gower dissimilarity for mixed variables (Gower 1971) was then calculated using the gowdis function in the FD library (Lalibert´e et al. 2014) of R ver. 3.1.0 (R Core Team 2014). Matching plot-level beta diversity and phylogenetic beta diversity distances were calculated using the pairPhylo function from code provided by A. Chao. This function allows the calculation of both species beta diversity and phylogenetic beta diversity using Hill numbers (Hill 1973; Chao et al. 2010; Chiu et al. 2013), which are considered the preferred measure for diversity partitioning (Jost 2006; Chao et al. 2012). We used q = 1 (equivalent to the exponential of Shannon entropy) for all analyses to incorporate variation in species abundances. We separately estimated beta (BD) and phylogenetic beta (PhBD) diversity for all 114 species of vascular species and only the 99 angiosperm species present in our plots.

5.3.4 Clustering plots by fuzzy K-means

Plot measures of Gowers dissimilarity (SiteEnv), BD and PhBD were clustered using fuzzy K-means (Kaufman & Rousseeuw 2009) for both the all vascular (Vasc) and angiosperm only (Angio) data using the fanny function from the cluster R library (Maechler et al. 2015). We thus generated five sets of clusters: SiteEnv, VascBD, VascPhBD, AngioBD and AngioPhBD. The optimal number of fuzzy K-means clusters was estimated using the Calinski and Harabasz index (Cali´nski & Harabasz 1974; Milligan & Cooper 1985) in the cascadeKM function from the vegan R library (Oksanen et al. 2013). Clusters were visualized using principal coordinate analysis (PCoA) to maintain as best as possible the distance relationships among plots (Legendre & Legendre 2012b). The assignment of the plots to clusters in each of the five analyses was then compared using the adjusted Rand

87 5.3. Methods

index (Rand 1971; Hubert & Arabie 1985), available in the clusterSim R library (Walesiak et al. 2008). In addition, we also obtained the highest membership coefficients for each plot per clustering method and compared clusters using Analysis of Variance (ANOVA) and Tukeys range test (Tukey 1949). The spatial association of cluster assignments for each of the five methods was assessed with the Morans I index (Moran 1950), using the lets.correl function in the letsR R library (Vilela & Villalobos 2015). To compare taxonomic versus phylogenetic compositional differences among plots defined by environmental variables (SiteEnv clusters), we partitioned BD and PhBD between and among plots within the SiteEnv clusters using the PhD2014 function from code provided by A. Chao. Within and among cluster partitioning shows the average

proportion of shared species within clusters (BDwithin) compared to the average proportion

of shared species among clusters (BDamong), and equivalently for branch lengths when considering phylogenetic beta diversity (PhBDwithin and PhBDamong). We then calculated the species contribution to beta diversity (SCBD) for each set of clusters, following Legendre & De C´aceres (2013). The SCBD represents the degree of variation of individual species across plots. Finally, to examine if SiteEnv clusters were composed of closely related species as would be expected if phylogenetically similar species were being filtered into the same cluster, the correlation indices of individual species for each clusters were calculated in the multipatt function of the indicspecies R library (De C´aceres & Legendre 2009). These values were then mapped onto the phylogenetic reconstruction to calculate phylogenetic signal (Blomberg et al. 2003) using the phylosignal function in the Picante R library, with significance assessed as the variance of phylogenetically independent contrasts compared to a random tip shuffling algorithm (Kembel et al. 2010).

88 5.4. Results

5.4 Results

5.4.1 Species diversity and phylogenetic reconstruction

We sampled 114 vascular and 99 angiosperm species within the 176 plots. The most common species across plots were: Cornus canadensis L., Linnaea borealis L., Vaccinium uliginosum L., V. vitis-idaea L. and Betula glandulosa Michx. (see Appendix D1: Table D1–1). An average of 10.28 vascular plant species were recorded per plot, with species richness increasing with decreasing elevation (Ordinary Least Squares regression, F 1,174 = 57.33, P < 0.001; see Appendix D2: Fig. D2–1). The phylogenetic reconstruction is presented in Appendix D1 (Fig. D1–1 and Fig. D1–2). Branch tip topology and uncertainty was similar to recently published phylogenies, with low bootstrap support within some families and genera, including the Pinaceae (e.g. Gernandt et al. 2008), Poaceae (e.g. Davis & Soreng 2007), Asteraceae (e.g. Panero & Funk 2008), Ericaceae (e.g. Kron & Luteyn 2005; Gillespie & Kron 2010), Carex (e.g. Ford et al. 2006; Waterway et al. 2009) and Salix (e.g. Chen et al. 2010).

5.4.2 Optimal number of clusters

For the angiosperm data, the optimal number of clusters was three for SiteEnv, five for AngioBD and seven for AngioPhBD (Calinski and Harabasz index, Fig. 5–1; Appendix D3). To allow easier comparison between clusters, we chose to show the results for the intermediate number of clusters (i.e. five) across all dissimilarity measures; similar results were found for three and seven clusters (Appendix D3). When all vascular species were considered, the optimal number of clusters diverged between BD (four clusters) and PhBD (14 clusters) (see Appendix D2: Fig. D2–2).

5.4.3 Comparison of clustering patterns

Clustering patterns of VascBD and AngioBD were most similar, whereas the greatest difference was between SiteEnv and AngioPhBD (adjusted Rand index; Table 5–1). In general, we observed weak correlations between BD and SiteEnv, as well as between PhBD 89 5.4. Results

and SiteEnv (Table 5–1). Support coefficients for cluster membership showing which metric produced the most well-defined clusters varied among distance types (ANOVA,

F 4,525 = 254.40, P < 0.001; Appendix D2: Fig. D2–4). The highest support coefficients were for SiteEnv, followed by VascPhBD and AngioPhBD, despite the fact that optimal number of clusters for SiteEnv was three and not five (Fig. 5–1). The cluster membership coefficient was lowest for VascBD and AngioBD (Appendix D2: Table D2–1), indicating that clusters were relatively less well-defined for these two metrics compared to the other indices. The five cluster types showed broadly similar patterns in the spatial associations of plots, with significant spatial autocorrelation in cluster assignments (i.e. spatially adjacent plots fell into similar clusters; Fig. 5–2). However, the spatial aggregation of plots within clusters was strongest for SiteEnv (Fig. 5–2a; Appendix D2: Table D2–2). In comparison, the spatial association of plots within clusters was lower for our other indices (VascBD, AngioBD, VascPhBD and AngioPhBD), although there were slight differences between patterns for angiosperms versus all vascular plants (Fig. 5–2; Appendix D2: Table D2–2). The average proportion of shared species (BD) was lower than the average proportion of shared lineages (PhBD) for both vascular plants and angiosperms when comparing beta diversity within and among the SiteEnv clusters (Fig. 5–3). The five clusters showed similar patterns in the proportion of shared species (BDwithin) for both the vascular

and angiosperm data, with BDwithin marginally higher than BDamong for Cluster 4 and

BDwithin marginally lower than BDamong for Clusters 1,3 and 5 (Fig. 5–3). Shared branch lengths among plots within clusters (PhBDwithin) was similar to that between clusters

(PhBDamong) for all five clusters, with slight differences between the vascular plant and angiosperm analyses (compare Fig. 5–3a and Fig. 5–3b). Species contributions to beta diversity (SCBD) differed among the different cluster types and between the vascular plants and angiosperms (Fig. 5–4). For example, Picea mariana (Mill.) Britton, Sterns & Poggenb. had the highest SCBD for the SiteEnv 90 5.5. Discussion

clusters and the second highest SCBD for VascBD and VascPhBD (Fig. 5–4); however, since this species is a gymnosperm, it did not contribute to SCBD across the angiosperm clusters. Betula glandulosa had the highest contribution to SCBD across all beta diversity clusters (Fig. 5–4), whereas Salix vestita Pursh had the second highest contribution across the angiosperm clusters and the third highest values across the vascular plant clusters (Fig. 5–4). In general, correlations between the SCBD for the different clustering methods were high, ranging from r = 0.729 (SiteEnv versus VascBD) to r = 0.939 (AngioBD versus AngioPhBD) (Fig. 5–4). Notably, the highest correlations in SCBD for both the vascular and angiosperm analyses was between the BD and PhBD clusters (Fig. 5–4). There was little evolutionary pattern in the positive and negative correlations of species to different SiteEnv clusters. Phylogenetic signal in species correlations to individual clusters was low and not significant for all five clusters s (K < 0.03 and P > 0.05; Appendix D2: Table D2–3 and Fig. D2–5); that is, individual clusters were not composed of more phylogenetically-related species than expected.

5.5 Discussion

Using vegetation plots distributed across a steep environmental gradient in the Canadian subarctic, we show that differences in phylogenetic branch lengths and species membership aggregate sites into distinct clusters, but that neither matches closely to the aggregation of sites based on environmental dissimilarity. We predicted that branch length differences (PhBD) would cluster sites similarly to environmental differences (using Gowers distance), because phylogeny might capture evolutionarily conserved abiotic preferences; however, we found that clusters informed by phylogeny were no more similar to sites clustered by environment than clusters determined by species similarities (BD). In addition, we found that species within sites sharing similar environments were no more closely related than expected by chance, indicating little phylogenetic niche conservatism for site preferences.

91 5.5. Discussion

Finally, we found that the inclusion of non-angiosperm lineages had little influence on our results, despite the strong gradients in the occurrence of gymnosperms across this biome. We predicted that sites clustered by shared phylogenetic branch lengths would most closely match to environmental clusters because of the possibility of phylogenetically conserved abiotic site preferences (Webb 2000; Cavender-Bares et al. 2009; Wiens et al. 2010). Environmental clusters were strongly supported, and we observed a high turnover in species richness with elevation, which matches previous vegetation studies at this site (Lessard-Therrien et al. 2014), and further supports strong abiotic sorting of plants. However, clusters based upon species or branch lengths matched only poorly to clusters based on environment indicating that abiotic site preferences were not reflected by either species or branch length dissimilarities. Nonetheless, clusters based on abiotic site preferences and, to a lesser extent, branch length differences were more clearly defined, with stronger membership support, than clusters defined by species differences, indicating that species difference was a poorer index for defining patches. We suggest the ubiquitous presence of several species and genera across our plots might help explain the weaker separation of clusters defined by species and branch length membership. Across the subarctic boreal-tundra transition zone, entire plant families and genera are lost towards higher latitudes without new families entering the flora (Porsild & Cody 1980; Qian 1999). We observed a similar pattern on Mnt. Irony where plant families such as the Santalaceae and Onagraceae were lost from the highest elevations. Such a pattern might be expected to emphasize phylogenetic dissimilarities among our clusters. However, other families (e.g. Ericaceae, Cyperaceae and Salicaceae) were generally well- represented across the site, with the same species being found at both high and low elevations in some families (e.g. Ericaceae). We propose, therefore, that it is these shared lineages, rather than the lineage differences, that tend to dominate cluster membership. We predicted that clustering plots by branch lengths would be different when includ- ing ferns and gymnosperms compared to considering only angiosperms because of the 92 5.5. Discussion

large evolutionary distance separating these taxonomic groups (Chaw et al. 2000; Pryer et al. 2001), which show strong geographical limits to their distribution. Gymnosperms tend to be limited to the more southern latitudes of the tundra biome (Porsild & Cody 1980; Timoney et al. 1992) and leptosporangiate ferns show a general decrease in richness with increasing latitude (Kessler et al. 2011). For example, as one proceeds northwards or upwards past the treeline in the eastern North American subarctic, there is a decrease in abundance and the eventual loss of gymnosperm taxa such as P. glauca (Moench) Voss, P. mariana and Abies balsamea (L.) Mill. (Porsild & Cody 1980; Timoney et al. 1992). These geographical trends might suggest evidence that the species within these clades are being filtered (Keddy 1992; Webb 2000; Cavender-Bares et al. 2009) from the flora at high latitudes because of their narrow temperature tolerances (Qian et al. 2013). In our analyses, the relative abundances of ferns, gymnosperms and lycophytes were lowest at the highest elevations as expected (see Appendix D2: Fig. D2–3); however, we found that excluding these lineages did not greatly alter results, possibly because their abundances were in any case generally low, and some gymnosperms were found at both high and low elevations. Perhaps most surprisingly, we found no strong signal of phylogenetic conservatism for membership in the SiteEnv clusters; species within clusters were no more closely related than expected by chance. As suggested above, we think that by this latitude many plant taxa have already been filtered because of their narrow temperature tolerances (e.g. magnoliids — Qian et al. 2013). Thus, as one proceeds to higher latitudes and elevations, the remaining species belong to a subset of often distantly related higher-taxa with wider temperature tolerances (Qian & Ricklefs 2007) where different species of the same genus often replace each other along abiotic gradients. Within our Mnt Irony plots, for example, several congeners, such as Pyrola asarifolia Michx. and P. grandiflora Radius, were sampled either in the windswept tundra at the top of the study site or the more protected forest/fen habitats at lower elevations, respectively, while co-occurrence in the same plots 93 5.5. Discussion was low. Clade changes across the phylogeny between environmental patches were thus relatively small as the same families and genera were often represented within several different clusters but by different species. Within the last few years, shared phylogenetic branch lengths have been used to delineate patches in a variety of biological systems, ranging from a global study of amphibians, birds and mammals (Holt et al. 2013) to more regional studies of angiosperms in the province of Yunnan, in southwestern China (Li et al. 2015). Our study, in contrast, was conducted at a local scale. We propose that the difference in scale between the focal organisms and the spatial extent might have contributed to the strong patterns observed in previous studies (Gaston 2000). In our study, which focused on small-scale differences in vegetation types and which was less sensitive to regional-level patterns, we found that phylogenetic information identified clearly defined clusters, but these did not capture clustering of sites with abiotic environment. It is likely that the importance of ecological processes vary across spatial scales (Graham & Fine 2008). We suggest that ecological processes prevalent at continental and regional scales (e.g. speciation, extinction, and trait evolution) might be better captured by phylogeny, whereas local-scale processes, such as dispersal and environmental filtering, have a much weaker phylogenetic signature.

5.5.1 The possible role of biome shifting

The low association between clusters defined by environment and those defined by species or phylogenetic branch lengths, as well as the weak phylogenetic signal in abiotic cluster membership might reflect a high incidence of biome shifting through evolutionary time (Donoghue & Edwards 2014) in the subarctic boreal-arctic tundra transition zone. Both the arctic tundra and boreal forest are large, broadly connected biomes (Udvardy & Udvardy 1975; Donoghue & Edwards 2014) that undergo periodic glaciations (Hewitt 2000). This shared boundary might have provided the opportunity for many species to shift between biomes over geological time (Donoghue & Edwards 2014), perhaps facilitated

94 5.5. Discussion

by orbitally forced range dynamics that link to glacial cycles and major shifts in the geographical distributions of the biomes themselves (Dynesius & Jansson 2000). Further, the arctic tundra might be a particularly welcoming recipient of species over time because it is relatively young (Donoghue & Edwards 2014), and competition among co-occurring species within this biome is suggested to be relatively weak compared to nearby forest biomes (Callaway et al. 2002). Therefore, many species from disparate clades may have had opportunity to disperse into the tundra (with a few perhaps making the reverse transition), such that phylogenetic relatedness tends to be only a poor predictor of habitat preference in this transition zone.

5.5.2 Justification of chosen metrics

Our study examined local patterns in species and branch length differences between two different taxonomic levels using fuzzy K-means and Hill numbers. The use of fuzzy K- means has advantages since boundaries between different biological systems are rarely cleanly delineated (Strayer et al. 2003). Alternative clustering methods are also available, for example, we could have spatially constrained our clusters (Legendre & Legendre 2012a). We decided against this option as we were more interested in seeing whether spatial aggregations emerged from clustering by species and branch length differences and how these spatial patterns varied. Interest in using phylogenetic beta diversity, or similar indices such as phylogenetic diversity resemblance (Nipperess et al. 2010; Root & Nelson 2011), to compare different assemblages has been growing, with various studies using metrics based on both species presence/absence (e.g. Bryant et al. 2008; Holt et al. 2013; Li et al. 2015) and abundance-weighted data (e.g. Nipperess et al. 2010; Root & Nelson 2011). We chose corresponding BD and PhBD metrics that incorporate species abundance information through Hill numbers, allowing for the possibility to change the importance of species abundances in our calculations (Jost 2007; Chiu et al. 2013). For this study, we used the first Hill number, which is equivalent to the exponential of Shannons entropy

95 5.5. Discussion

(Hill 1973; Chiu et al. 2013); however, further studies could explore other Hill numbers to examine more closely the influence of abundance distributions and the importance of common versus rare species.

5.5.3 Conclusion

We clustered plots by species membership and phylogenetic differences, but neither group reflects the abiotic aggregation of sites that we think most closely represents the transition between biomes. Our study indicates that there might be only limited additional value in including phylogenetic data in delineating patches at the local scale, although its use in delineating larger-scale biogeographic boundaries has been shown by others (e.g. Holt et al. 2013; Li et al. 2015). We found our results somewhat surprising, given that our study system represented a biome-level transition. We suggest that our conclusions might apply more generally wherever there is a relatively broad spatial distribution of evolutionary distinct lineages, as these shared deep branch lengths might tend to overshadow evidence for habitat segregation among more closely related congeners. We propose, however, that phylogenetic beta diversity metrics might still be preferable to traditional beta diversity metrics, especially when species identities are uncertain.

96 Chapter 5 Tables & Figures

Table 5–1 Comparison between clusters using Gower dissimilarity (SiteEnv), beta diversity (BD) and phylogenetic beta diversity distances (PhBD). Values range between 0 and 1 and are based on the adjusted Rand index, where 1.00 indicates that the clustering of plots is identical. Two sets of beta and phylogenetic beta diversity analyses are included, one with all vasculars (Vasc) and a second to angiosperms only (Angio).

SiteEnv VascBD VascPhBD AngioBD AngioPhBD SiteEnv 1.00 0.19 0.16 0.16 0.15 VascBD 0.19 1.00 0.27 0.40 0.22 VascPhBD 0.16 0.27 1.00 0.24 0.28 AngioBD 0.16 0.40 0.24 1.00 0.26 AngioPhBD 0.15 0.22 0.28 0.26 1.00

97 Chapter 5 Tables & Figures

Figure 5–1 Clusters of the 176 plots on Mont Irony, Labrador, shaded by cluster membership. The top three panels show the results for the Gower dissimilarity (SiteEnv) distances. The middle three panels show beta diversity distances. The bottom three panels show results for phylogenetic beta diversity distances. Panels (a), (d) and (g) show principal coordinate ordi- nations (PCoA) on five K-means clusters, grey lines indicate the location of plots in ordination space. The optimal number of K-means clusters estimated by the Calinski and Harabasz in- dex are shown in panels (b), (e) and (h), with black circles indicating the optimal and grey circles showing the second highest recommended number of clusters for each analysis. Pan- els (c), (f) and (i) show the location of the plots corresponding to each of the five clusters on the sampling grid, with greyscale shading indicating cluster membership. All figures are for angiosperm data.

98 Chapter 5 Tables & Figures

Figure 5–2 Spatial associations in plot assignments to clusters for vascular plants (a) and angiosperms (b) compared to Gower dissimilarity (SiteEnv) clusters, as estimated by Morans I. Values above zero indicate positive spatial associations in cluster membership across plots, whereas negative values show negative spatial associations. Pairwise distances (metres) were divided into 16 bins with between 1924 and 1926 observations per bin. VascBD = vascular beta diversity, VascPhBD = vascular phylogenetic beta diversity, AngioBD = angiosperm beta diversity, AngioPhBD = angiosperm phylogenetic beta diversity.

99 Chapter 5 Tables & Figures

Figure 5–3 The proportion of shared species and branch lengths within five Gower dissimi-

larity (SiteEnv) clusters (BDwithin and PhBDwithin, respectively) compared to among cluster

beta diversity (BDamong) and phylogenetic beta diversity(PhBDamong). The black dashed line indicates among-cluster beta diversity and among-cluster phylogenetic beta diversity is shown in light grey. Black circles indicate within-cluster beta diversity, and light grey circles show within cluster phylogenetic beta diversity. Panel (a) shows values for vascular plants, whereas angiosperm values are indicated in panel (b).

100 Chapter 5 Tables & Figures

Figure 5–4 Species contributions to beta diversity values (SCBD). The length of individual bars corresponds to each SCBD, with longer bars corresponding to higher SCBD values and short bars representing low SCBD. Bars (a) indicate SCBD among the Gower dissimilarity (SiteEnv) clusters, whereas bars (b) and (c) show SCBD values among the beta and phyloge- netic beta diversity clusters for vascular plants, respectively. Considering only angiosperms, bar (d) indicate SCBD among the SiteEnv clusters, while bars (e) and (f) indicate SCBD values among the angiosperm beta and phylogenetic beta diversity clusters.

101 CHAPTER 6

General conclusion

102 6.1. Conclusion

6.1 Conclusion

Community ecology has undergone a major transformation in the last fifteen years driven in part by recent advances in molecular, computational and analytical methods. These advances have allowed for the integration of phylogenetic and ecological approaches, resulting in the emergence of the relatively new field of ‘community phylogenetics’ (Webb 2000; Cavender- Bares et al. 2009; Vamosi et al. 2009). Many of the early studies in community phylogenetics focused on dispersion patterns among community types, with phylogenetic overdispersion— co-occurring species more distantly related than expected—suggesting competition as the dominant force structuring communities and environmental filtering typically inferred as the more important force when communities are phylogenetically clustered—co-occurring species are more closely related than would be expected by chance (Webb 2000; Webb et al. 2002; Cavender-Bares et al. 2009; Vamosi et al. 2009). However, there has been growing awareness that we are limited in our ability to infer underlying community assembly processes from such patterns because of their reliance on certain key assumptions (Gerhold et al. 2015). Thus, the discipline requires increased rigour combined with creative alternative approaches to remain relevant as theoretical and analytical approaches continue to develop. One of the first steps that must be taken when conducting studies in community phylo- genetics is to define a species pool that can be used to create null models and phylogenetic reconstructions for subsequent analyses. Previous work has shown how poor phylogenetic resolution can bias metrics commonly used in community phylogenetics (e.g. Swenson 2009), with little attention given to the difficulties in defining the species pool itself, although the im- portance of the pool is well recognized (Swenson et al. 2007; Lessard et al. 2012). In Chapter 2, I describe several challenges that can be encountered when sampling species to include in a regional species pool. I illustrate that it can be almost impossible to sample every species within even a relatively small regional pool. I give several suggestions that can be used by others conducting similar sampling efforts to help maximize sampling efficiency and improve taxonomic coverage.

103 6.1. Conclusion

Throughout this thesis, I have advanced traditional community phylogenetics methods by developing and evaluating novel approaches, showing that the discipline can remain relevant with the incorporation of creative questions, sampling designs and analyses. In contrast to much previous work that has focused on overall patterns in community dispersion where a particular community might be classified as either phylogenetically overdispersed, clustered or show random patterns of assembly (e.g. Webb et al. 2002; Cavender-Bares et al. 2004), my work in Chapter 3 focuses on lineage-specific patterns in community structure. I show that each species in a community can have neighbours that are either more or less related than expected and that these patterns can depend on their clade membership. For example, within co-occurring Carex in the Schefferville region, phylogenetic clustering was more evident in species in the Core Carex clade, while phylogenetic overdispersion was more prominent in the Vignea clade. In addition, I demonstrate that contrasting lineage-specific patterns in community structure can mask each other, resulting in overall patterns that appear random. Although my approach ultimately focuses on patterns in community structure and not the underlying processes determining them, I argue that it is the species-specific responses to community assembly processes that need to be understood, since it is the individuals of a species that are ‘filtered’ into a community. Knowledge of both species and lineage-specific patterns in community structure can then be examined to help infer the processes structuring communities and species coexistences. More recently, advances in the understanding of species co-occurrence within communities have come from combining modern coexistence theory with the knowledge of the evolutionary relationships between co-occurring species (Chesson 2000; Mayfield & Levine 2010; HilleRis- Lambers et al. 2012; Godoy et al. 2014). Within this framework, it is the balance between differing competitive abilities and niche differences that determine species coexistence, with coexistence most likely between species with similar competitive abilities but different niche preferences (Mayfield & Levine 2010; HilleRisLambers et al. 2012). In Chapter 4, I combine ideas from modern coexistence theory with community phylogenetics to test the hypotheses

104 6.1. Conclusion that niche width differences translate to competitive differences and that phylogenetic dis- tances reflect niche differences between co-occurring species pairs. This novel approach allows us to revisit a long-standing ecological question that has been difficult to examine using field studies—whether the specialist makes the better competitor than the generalist? My results show higher co-occurrence between phylogenetically distant species with comparable niche widths. These results provide support for a trade-off between ecological generalization and competitive ability and suggest that differences in niche widths translate into differences in competitive abilities, while phylogenetic distance might indeed provide a useful proxy for niche differences. Another recent development in the field of ecological phylogenetics has been the use of phylogenetic beta diversity measures, for example, to describe biome boundaries at both regional and global scales (e.g. Holt et al. 2013; Li et al. 2015). Using this approach, biomes can be delineated when the species represented within an area differ greatly from other areas in the phylogenetic branches they represent (see Holt et al. 2013). In Chapter 5, I show that novel approaches using phylogenetic beta diversity are not more effective than more traditional beta diversity approaches for boundary delineation at the local scale and poorly capture abiotic structure in the environment. My findings suggest that the scale at which a study is conducted should be a key consideration when deciding which dimension of biodiversity to focus on, as local level processes such as environmental filtering and competition can obscure the signals of phylogenetic processes operating at larger scales (e.g. speciation, extinction and long distance dispersal). It is highly likely that the field of community phylogenetics will continue to evolve in the near future because of technological advances, allowing for the development of new research questions and approaches. With recent advances in next-generation sequencing, phylogenetic resolution is quickly improving as researchers tailor the genomic data used in their analyses to the specific research question at hand (Straub et al. 2012; Chan & Ragan 2013; Weitemier et al. 2014). These new phylogenomic approaches are rapidly gaining popularity because of

105 6.1. Conclusion their ability to incorporate information from kilobases of genome-level data from nuclear, mito- chondrial, and in plants chloroplast DNA, while the use of traditional phylogenetic approaches based on a small number of loci retrieved using Sanger sequencing has been decreasing in comparison (Straub et al. 2012). However, the increasing availability of phylogenomic data presents additional challenges and questions for researchers interested in community phyloge- netics, for example, what is the appropriate phylogenenetic/phylogenomic resolution required for a particular research question, and how much time and effort is worth investing to achieve the desired resolution? At the same time as these advances in genomics tools, community phylogenetics is becoming a more predictive science with increases in computational power and the volumes of data that can be incorporated into models (Luo et al. 2011). In addition, the increasing availability of new metrics (see Vellend et al. 2011; Pearse et al. 2014), as well as the greater ability of researchers to develop their own novel indices through tools such as the open-source programming language—R, are allowing for the creation of analyses designed for specific research questions. Paradigm shifts have been common throughout the history of community ecology (Lortie et al. 2004), and recent theoretical advances indicate that key paradigms are also changing in community phylogenetics. In the past five years, the community phylogenetics framework first introduced by Webb (2000) and Webb et al. (2002) has been losing traction to approaches that combine modern coexistence theory with information derived from the phylogenetic relationships between co-occurring species (see Mayfield & Levine 2010; HilleRisLambers et al. 2012; Godoy et al. 2014; Kraft et al. 2015b). Alongside these advances in theory, powerful sampling designs have been combined with innovative questions to improve our overall ability to infer the relative importance of the key processes determining plant community composition (e.g. Godoy et al. 2014; Kraft et al. 2015b). At the same time, theory on key processes such as environmental filtering is being refined (see Kraft et al. 2015a), further improving our understanding of community assembly by challenging researchers to more carefully define their study questions and analyses to capture the effects of the process that they are investigating.

106 6.1. Conclusion

Although the currently-accepted paradigm in community phylogenetics has shifted and key concepts have been further refined, this should not negate earlier work that has provided the foundation upon which the field has advanced and still provides a reservoir for new ideas. Understanding the complex relationships among co-occurring plants is a daunting task. Community phylogenetics should be regarded as one tool among many that might help move us further towards this goal. For community phylogenetics to remain relevant in the future, it is important that researchers carefully consider the assumptions underlying such approaches and how they affect the inferences that can be drawn from study results. As shown in this thesis, these considerations combined with unique study designs and analyses will allow researchers to address both old and new questions in community ecology, thus unravelling the seemingly endless complexities of plant assemblages.

107 List of References

Aarssen, L.W. (1983). Ecological combining ability and competitive combining ability in plants: toward a general evolutionary theory of coexistence in systems of competition. American Naturalist, 122, 707–731. Abadie, J.-C., Andrade, C., Machon, N., & Porcher, E. (2008). On the use of paratax- onomy in biodiversity monitoring: a case study on wild flora. Biodiversity and Conservation, 17, 3485–3500. Abrams, P.A. (2007). Specialist-generalist competition in variable environments; the consequences of competition between resources. In: The Impact of Environmental Variability on Ecological Systems. (eds. Vasseur, D.A. & McCann, K.S.). Springer, The Netherlands, pp. 133–157. Anderson, M.J., Crist, T.O., Chase, J.M., Vellend, M., Inouye, B.D., Freestone, A.L., Sanders, N.J., et al. (2011). Navigating the multiple meanings of beta diversity: a roadmap for the practicing ecologist. Ecology Letters, 14, 19–28. Anderson, T.M., Lachance, M.A., & Starmer, W.T. (2004). The relationship of phylogeny to community structure: the cactus yeast community. The American Naturalist, 164, 709–721. Araya, Y.N., Silvertown, J., Gowing, D.J., McConway, K.J., Linder, H.P., & Midgley, G. (2012). Do niche-structured plant communities exhibit phylogenetic conservatism? A test case in an endemic clade. Journal of Ecology, 100, 1434–1439. Armstrong, K.F. & Ball, S.L. (2005). DNA barcodes for biosecurity: invasive species identification. Philosophical Transactions of the Royal Society B: Biological Sciences, 360, 1813–1823. Arroyo-Rodr´ıguez, V., Cavender-Bares, J., Escobar, F., Melo, F.P.L., Tabarelli, M., & Santos, B.A. (2012). Maintenance of tree phylogenetic diversity in a highly fragmented rain forest. Journal of Ecology, 100, 702–711. List of References

Ball, P.W. & Reznicek, A.A. (2002). Carex Linnaeus. In: Flora of North America; North of Mexico. (eds. Ball, P.W., Gandhi, K., Kiger, R.W., Murray, D., Zarucchi, J.L., Reznicek, A.A., & Strother, J.L.). Vol. 23: Magnoliophyta: Commelinidae (in part); Cyperaceae. Oxford University Press, Oxford, United Kingdom, pp. 254–572. Ball, P.W., Reznicek, A.A., & Murray, D.F. (2002). Cyperaceae Jussieu. In: Flora of North America; North of Mexico. (eds. Ball, P.W., Gandhi, K., Kiger, R.W., Murray, D., Zarucchi, J.L., Reznicek, A.A., & Strother, J.L.). Vol. 23: Magnoliophyta: Commelinidae (in part); Cyperaceae. Oxford University Press, Oxford, United Kingdom, pp. 3–8. Barwell, L.J., Isaac, N.J.B., & Kunin, W.E. (2015). Measuring β–diversity with species abundance data. Journal of Animal Ecology, 1112–1122. Baselga, A. (2010). Multiplicative partition of true diversity yields independent alpha and beta components; additive partition does not. Ecology, 91, 1974–1981. Bennett, J.A., Lamb, E.G., Hall, J.C., Cardinal-McTeague, W.M., & Cahill, J.F. (2013). Increased competition does not lead to increased phylogenetic overdispersion in a native grassland. Ecology Letters, 16, 1168–1176. Benson, D.A., Karsch-Mizrachi, I., Lipman, D.J., Ostell, J., & Sayers, E.W. (2010). GenBank. Nucleic Acids Research, 38, D46–D51. Bergsten, J., Bilton, D.T., Fujisawa, T., Elliott, M., Monaghan, M.T., Balke, M., Hen- drich, L., et al. (2012). The effect of geographical scale of sampling on DNA barcod- ing. Systematic Biology, 61, 851–869. Bernstein, L., Bosch, P., Canziani, O., Chen, Z., Christ, R., Davidson, O., Hare, W., et al. (2007). IPCC, 2007: Climate Change 2007: Synthesis Report. Contribution of Working Groups I. Intergovernmental Panel on Climate Change, Geneva, Switzerland. Bertness, M.D. & Callaway, R. (1994). Positive interactions in communities. Trends in Ecology & Evolution, 9, 191–193.

109 List of References

Blomberg, S.P., Garland, T., & Ives, A.R. (2003). Testing for phylogenetic signal in comparative data: behavioral traits are more labile. Evolution, 57, 717–745. Brouillet, L., Coursol, F., Meades, S.F, Favreau, M., Anions, M., Belisle, P., & Desmet, P. (2013). VASCAN, the Database of Vascular Plants of Canada. http://data.canadensys.net/vascan/. Bruni, I., De Mattia, F., Martellos, S., Galimberti, A., Savadori, P., Casiraghi, M., Nimis, P.L., et al. (2012). DNA barcoding as an effective tool in improving a digital plant identification system: a case study for the area of Mt. Valerio, Trieste (NE Italy). PLoS One, 7, e43256. Bryant, J.A., Lamanna, C., Morlon, H., Kerkhoff, A.J., Enquist, B.J., & Green, J.L. (2008). Microbes on mountainsides: contrasting elevational patterns of bacterial and plant diversity. Proceedings of the National Academy of Sciences of the United States of America, 105, 11505–11511. Burgess, K.S., Fazekas, A.J., Kesanakurti, P.R., Graham, S.W., Husband, B.C., Newmas- ter, S.G., Percy, D.M., et al. (2011). Discriminating plant species in a local temperate flora using the rbcL + matK DNA barcode. Methods in Ecology and Evolution,2, 333–340. Cadenasso, M.L., Pickett, S.T.A., Weathers, K.C., & Jones, C.G. (2003). A framework for a theory of ecological boundaries. BioScience, 53, 750–758. Cali´nski, T. & Harabasz, J. (1974). A dendrite method for cluster analysis. Communica- tions in Statistics-Theory and Methods, 3, 1–27. Callaway, R.M. (2007). Positive Interactions and Interdependence in Plant Communities. Springer, Berlin, Germany. Callaway, R.M., Brooker, R.W., Choler, P., Kikvidze, Z., Lortie, C.J., Michalet, R., Paolini, L., et al. (2002). Positive interactions among alpine plants increase with stress. Nature, 417, 844–848.

110 List of References

Castresana, J. (2000). Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Molecular Biology and Evolution, 17, 540–552. Cavender-Bares, J., Ackerly, D.D., Baum, D.A., & Bazzaz, F.A. (2004). Phylogenetic overdispersion in Floridian oak communities. The American Naturalist, 163, 823–843. Cavender-Bares, J., Keen, A., & Miles, B. (2006). Phylogenetic structure of Floridian plant communities depends on taxonomic and spatial scale. Ecology, 87, S109–S122. Cavender-Bares, J., Kozak, K.H., Fine, P.V.A., & Kembel, S.W. (2009). The merging of community ecology and phylogenetic biology. Ecology Letters, 12, 693–715. Chan, C.X. & Ragan, M.A. (2013). Next-generation phylogenomics. Biology Direct,8. Chao, A., Chiu, C.-H., & Jost, L. (2010). Phylogenetic diversity measures based on Hill numbers. Philosophical Transactions of the Royal Society B: Biological Sciences, 365, 3599–3609. Chao, A., Chiu, C.-H., & Hsieh, T.C. (2012). Proposing a resolution to debates on diversity partitioning. Ecology, 93, 2037–2051. Chapin, F.S., Sturm, M., Serreze, M.C., McFadden, J.P., Key, J.R., Lloyd, A.H., McGuire, A.D., et al. (2005). Role of land-surface changes in Arctic summer warming. Science, 310, 657–660. Chaw, S.-M., Parkinson, C.L., Cheng, Y., Vincent, T.M., & Palmer, J.D. (2000). Seed plant phylogeny inferred from all three plant genomes: monophyly of extant gym- nosperms and origin of Gnetales from conifers. Proceedings of the National Academy of Sciences, 97, 4086–4091. Chen, G., K´ery, M., Zhang, J., & Ma, K. (2009). Factors affecting detection probability in plant distribution studies. Journal of Ecology, 97, 1383–1389. Chen, J.H., Sun, H., Wen, J., & Yang, Y.P. (2010). Molecular phylogeny of Salix L. (Salicaceae) inferred from three chloroplast datasets and its systematic implications. Taxon, 59, 29–37.

111 List of References

Chesson, P. (2000). Mechanisms of maintenance of species diversity. Annual Review of Ecology and Systematics, 31, 343–366. Chiu, C.-H., Jost, L., & Chao, A. (2013). Phylogenetic beta diversity, similarity, and differentiation measures based on Hill numbers. Ecological Monographs, 84, 21–44. Clarke, K.R. & Warwick, R.M. (1998). A taxonomic distinctness index and its statistical properties. Journal of Applied Ecology, 35, 523–531. Clements, F.E. (1916). Plant Succession: An Analysis of the Development of Vegetation. Carnegie Institution of Washington, Washington, DC, United States. Cody, M.L. (1975). Towards a theory of continental species diversities: bird distributions over Mediterranean habitat gradients. In: Ecology and Evolution of Communities. (eds. Cody, M.L. & Diamond, J.M.). The Belknap Press of Harvard University Press, Cambridge, Massachusetts, United States, pp. 214–257. Collins, R.A. & Cruickshank, R.H. (2013). The seven deadly sins of DNA barcoding. Molecular Ecology Resources, 13, 969–975. Colwell, R.K. & Futuyma, D.J. (1971). On the measurement of niche breadth and overlap. Ecology, 52, 567–576. Condit, R., Pitman, N., Leigh, E.G., Chave, J., Terborgh, J., Foster, R.B., N´unez, P., et al. (2002). Beta-diversity in tropical forest trees. Science, 295, 666–669. Connor, E.F. & Simberloff, D. (1979). The assembly of species communities - chance or competition. Ecology, 60, 1132–1140. Cornwell, W.K. & Ackerly, D.D. (2009). Community assembly and shifts in plant trait distributions across an environmental gradient in coastal California. Ecological Monographs, 79, 109–126. Cˆot´e, S.D., Rooney, T.P., Tremblay, J.-P., Dussault, C., & Waller, D.M. (2004). Eco- logical impacts of deer overabundance. Annual Review of Ecology, Evolution, and Systematics, 35, 113–147.

112 List of References

Crist, T.O. & Veech, J.A. (2006). Additive partitioning of rarefaction curves and speciesarea relationships: unifying α–, β–andγ–diversity with sample size and habitat area. Ecology Letters, 9, 923–932. Dabros, A. (2004). Distribution patterns of sedges in subarctic fens: ecological and phylogenetic perspectives. Master’s thesis. McGill University, Montr´eal, Qu´ebec. Dabros, A. & Waterway, M.J. (2008). Segregation of sedge species (Cyperaceae) along environmental gradients in fens of the Schefferville region, northern Quebec. In: Sedges: Uses, Diversity and Systematics of the Cyperaceae. (eds. Naczi, R.F.C. & Ford, B.A.). 108th ed. Monographs in Systematic Botany from the Missouri Botanical Garden, Missouri Botanical Garden Press, pp. 145–161. Darriba, D., Taboada, G.L., Doallo, R., & Posada, D. (2012). jModelTest 2: more models, new heuristics and parallel computing. Nature Methods, 9, 772–772. Darwin, C. (1910). On the Origin of Species. Ward Lock & Co. Limited, London, United Kingdom. Davies, T.J., Kraft, N.J.B., Salamin, N., & Wolkovich, E.M. (2012). Incompletely resolved phylogenetic trees inflate estimates of phylogenetic conservatism. Ecology, 93, 242– 247. Davis, J.I. & Soreng, R.J. (2007). A preliminary phylogenetic analysis of the grass subfamily Pooideae (Poaceae), with attention to structural features of the plastid and nuclear genomes, including an intron loss in GBSSI. Aliso: A Journal of Systematic and Evolutionary Botany, 23, 335–348. De C´aceres, M. & Legendre, P. (2009). Associations between species and groups of sites: indices and statistical inference. Ecology, 90, 3566–3574. de Vere, N., Rich, T.C.G., Ford, C.R., Trinder, S.A., Long, C., Moore, C.W., Satterth- waite, D., et al. (2012). DNA barcoding the native flowering plants and conifers of Wales. PLoS One, 7, e37945.

113 List of References

Degnan, J.H. & Rosenberg, N.A. (2009). Gene tree discordance, phylogenetic inference and the multispecies coalescent. Trends in Ecology & Evolution, 24, 332–340. Dejean, T., Valentini, A., Miquel, C., Taberlet, P., Bellemain, E., & Miaud, C. (2012). Improved detection of an alien invasive species through environmental DNA barcod- ing: the example of the American bullfrog Lithobates catesbeianus. Journal of Applied Ecology, 49, 953–959. Diamond, J.M. (1975). Assembly of species communities. In: Ecology and Evolution of Communities. (eds. Cody, M.L. & Diamond, J.M.). The Belknap Press of Harvard University Press, Cambridge, Massachusetts, United States, pp. 342–444. Diamond, J.M. & Gilpin, M.E. (1982). Examination of the null model of Connor and Simberloff for species co-occurrences on islands. Oecologia, 52, 64–74. Donoghue, M.J. (1994). Progress and prospects in reconstructing plant phylogeny. Annals of the Missouri Botanical Garden, 81, 405–418. Donoghue, M.J. & Edwards, E.J. (2014). Biome shifts and niche evolution in plants. Annual Review of Ecology, Evolution, and Systematics, 45, 547–572. Drummond, A.J., Suchard, M.A., Xie, D., & Rambaut, A. (2012). Bayesian phylogenetics with BEAUti and the BEAST 1.7. Molecular Biology and Evolution, 29, 1969–1973. Dunnett, C.W. (1980). Pairwise multiple comparisons in the homogeneous variance, unequal sample size case. Journal of the American Statistical Association, 75, 789– 795. Dynesius, M. & Jansson, R. (2000). Evolutionary consequences of changes in species’ geographical distributions driven by Milankovitch climate oscillations. Proceedings of the National Academy of Sciences, 97, 9115–9120. Ebach, M.C. & Holdrege, C. (2005). DNA barcoding is no substitute for taxonomy. Nature, 434, 697. Elmendorf, S.C., Henry, G.H.R., Hollister, R.D., Bj¨ork, R.G., Bjorkman, A.D., Callaghan, T.V., Collier, L.S., et al. (2012). Global assessment of experimental climate warming 114 List of References

on tundra vegetation: heterogeneity over space and time. Ecology Letters, 15, 164– 175. Elton, C. (1946). Competition and the structure of ecological communities. The Journal of Animal Ecology, 15, 54–68. Emerson, B.C. & Gillespie, R.G. (2008). Phylogenetic analysis of community assembly and structure over space and time. Trends in Ecology & Evolution, 23, 619–630. Fagan, W.F., Fortin, M.-J., & Soykan, C. (2003). Integrating edge detection and dynamic modeling in quantitative analyses of ecological boundaries. BioScience, 53, 730–738. Faith, D.P. (1992). Conservation evaluation and phylogenetic diversity. Biological Conser- vation, 61, 1–10. Fanelli, D. (2012). Negative results are disappearing from most disciplines and countries. Scientometrics, 90, 891–904. Fazekas, A.J., Burgess, K.S., Kesanakurti, P.R., Graham, S.W., Newmaster, S.G., Hus- band, B.C., Percy, D.M., et al. (2008). Multiple multilocus DNA barcodes from the plastid genome discriminate plant species equally well. PLoS One, 3, e2802. Ferrier, S., Manion, G., Elith, J., & Richardson, K. (2007). Using generalized dissimilarity modelling to analyse and predict patterns of beta diversity in regional biodiversity assessment. Diversity and Distributions, 13, 252–264. Fischer, A.G. (1960). Latitudinal variations in organic diversity. Evolution, 14, 64–81. Fitzpatrick, M.C., Preisser, E.L., Porter, A., Elkinton, J., Waller, L.A., Carlin, B.P., & Ellison, A.M. (2010). Ecological boundary detection using Bayesian areal wombling. Ecology, 91, 3448–3455. Ford, B.A., Iranpour, M., Naczi, R.F.C., Starr, J.R., & Jerome, C.A. (2006). Phylogeny of Carex subg. Vignea (Cyperaceae) based on non-coding nrDNA sequence data. Systematic Botany, 31, 70–82. Fortin, M.-J. & Dale, M.R.T. (2005). Spatial Analysis: A Guide For Ecologists.Cam- bridge University Press, Cambridge, United Kingdom. 115 List of References

Fortin, M.-J. & Drapeau, P. (1995). Delineation of ecological boundaries: comparison of approaches and significance tests. Oikos, 323–332. Fortin, M.-J., Olson, R.J., Ferson, S., Iverson, L., Hunsaker, C., Edwards, G., Levine, D., et al. (2000). Issues related to the detection of boundaries. Landscape Ecology, 15, 453–466. Francis, G. (2004). Biosphere reserves in Canada: ideals and some experience. Environ- ments: A Journal of Interdisciplinary Studies, 32, 1–26. Frezal, L. & Leblois, R. (2008). Four years of DNA barcoding: current advances and prospects. Infection, Genetics and Evolution, 8, 727–736. Futuyma, D.J. & Moreno, G. (1988). The evolution of ecological specialization. Annual Review of Ecology and Systematics, 19, 207–233. Gajewski, K., Payette, S., & Ritchie, J.C. (1993). Holocene vegetation history at the boreal-forest–shrub-tundra transition in north-western Quebec. Journal of Ecology, 433–443. Gaston, K.J. (2000). Global patterns in biodiversity. Nature, 405, 220–227. Gause, G.F. (1934). The Struggle for Existence. Hafner, New York, New York, United States. Gerhold, P., Cahill, J.F., Winter, M., Bartish, I.V., & Prinzing, A. (2015). Phylogenetic patterns are not proxies of community assembly mechanisms (they are far better). Functional Ecology, 29, 600–614. Gernandt, D.S., Magall´on, S., L´opez, G.G., Flores, O.Z., Willyard, A., & Liston, A. (2008). Use of simultaneous analyses to guide fossilbased calibrations of Pinaceae phylogeny. International Journal of Plant Sciences, 169, 1086–1099. Gilbert, B. & Lechowicz, M.J. (2004). Neutrality, niches, and dispersal in a temperate forest understory. Proceedings of the National Academy of Sciences of the United States of America, 101, 7651–7656.

116 List of References

Gilbert, B. & Lechowicz, M.J. (2005). Invasibility and abiotic gradients: the positive correlation between native and exotic plant diversity. Ecology, 86, 1848–1855. Gillespie, E. & Kron, K. (2010). Molecular phylogenetic relationships and a revised classification of the subfamily Ericoideae (Ericaceae). Molecular Phylogenetics and Evolution, 56, 343–354. Gleason, H.A. (1926). The individualistic concept of the plant association. Bulletin of the Torrey Botanical Club, 53, 7–26. Global Carex Group (2015). Making Carex monophyletic (Cyperaceae, tribe Cariceae): a new broader circumscription. Botanical Journal of the Linnean Society, doi:10.1111/boj.12298. Godoy, O., Kraft, N.J.B., & Levine, J.M. (2014). Phylogenetic relatedness and the determinants of competitive outcomes. Ecology Letters, 17, 836–844. Gotelli, N.J. & Colwell, R.K. (2010). Estimating species richness. In: Biological Diversity: Frontiers in Measurement and Assessment. (eds. Magurran, A.E. & McGill, B.J.). Oxford University Press, Oxford, United Kingdom, pp. 39–54. Gotelli, N.J. & Graves, G.R. (1996). Null Models in Ecology. Smithsonian Institution Press, Washington, DC, United States. Govaerts, R., Koopman, J., Simpson, D., Goetghebeur, P., Wilson, K., Egorova, T., & Bruhl, J. (2015). World Checklist of Cyperaceae. http://apps.kew.org/wcsp/. Gower, J.C. (1971). A general coefficient of similarity and some of its properties. Biomet- rics, 857–871. Graham, C.H. & Fine, P.V.A. (2008). Phylogenetic beta diversity: linking ecological and evolutionary processes across space in time. Ecology Letters, 11, 1265–1277. Graham, C.H., Parra, J.L., Rahbek, C., & McGuire, J.A. (2009). Phylogenetic structure in tropical hummingbird communities. Proceedings of the National Academy of Sciences, 106, 19673–19678.

117 List of References

Grant, P.R. (1966). Ecological compatibility of bird species on islands. The American Naturalist, 100, 451–462. Grant, P.R. & Abbott, I. (1980). Interspecific competition, island biogeography and null hypotheses. Evolution, 34, 332–341. Hajibabaei, M., Singer, G.A.C., Hebert, P.D.N., & Hickey, D.A. (2007). DNA barcoding: how it complements taxonomy, molecular phylogenetics and population genetics. Trends in Genetics, 23, 167–172. Hall, T.A. (1999). BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series, 41, 95– 98. Han, J.-P., Shi, L.-C., Chen, X.-C., & Lin, Y.-L. (2012). Comparison of four DNA barcodes in identifying certain medicinal plants of Lamiaceae. Journal of Systematics and Evolution, 50, 227–234. Hardy, O.J. & Senterre, B. (2007). Characterizing the phylogenetic structure of commu- nities by an additive partitioning of phylogenetic diversity. Journal of Ecology, 95, 493–506. Harper, J.L (1977). Population Biology of Plants. Academic Press, Inc., New York, New York, United States. Harrison, S., Ross, S.J., & Lawton, J.H. (1992). Beta diversity on geographic gradients in Britain. Journal of Animal Ecology, 61, 151–158. Hepper, F.N. (1989). Plant Hunting for Kew. HMSO Publications Centre, London, England. Hewitt, G. (2000). The genetic legacy of the Quaternary ice ages. Nature, 405, 907–913. Hill, M.O. (1973). Diversity and evenness: a unifying notation and its consequences. Ecology, 54, 427–432. Hill, S.B. & Kotanen, P.M. (2009). Evidence that phylogenetically novel non-indigenous plants experience less herbivory. Oecologia, 161, 581–590. 118 List of References

HilleRisLambers, J., Adler, P.B., Harpole, W.S., Levine, J.M., & Mayfield, M.M. (2012). Rethinking community assembly through the lens of coexistence theory. Annual Review of Ecology, Evolution, and Systematics, 43, 227–248. Hoagland, K.E. (1996). The taxonomic impediment and the convention on biodiversity. Association of Systematics Collections Newsletter, 24, 61–62. Hollingsworth, P.M., Forrest, L.L., Spouge, J.L., Hajibabaei, M., Ratnasingham, S., van der Bank, M., Chase, M.W., et al. (2009). A DNA barcode for land plants. Proceedings of the National Academy of Sciences, 106, 12794–12797. Hollingsworth, P.M., Graham, S.W., & Little, D.P. (2011). Choosing and using a plant DNA barcode. PLoS One, 6, e19254. Holt, B.G., Lessard, J.P., Borregaard, M.K., Fritz, S.A., Ara´ujo, M.B., Dimitrov, D., Fabre, P.H., et al. (2013). An update of Wallaces zoogeographic regions of the world. Science, 339, 74–78. Horner-Devine, M. C. & Bohannan, B.J.M. (2006). Phylogenetic clustering and overdis- persion in bacterial communities. Ecology, 87, S100–S108. Hubbell, S.P. (2001). The Unified Neutral Theory of Biodiversity and Biogeography. Monographs in Population Biology, 32. Princeton University Press, Princeton. Hubert, L. & Arabie, P. (1985). Comparing partitions. Journal of Classification, 2, 193– 218. Hutchinson, G.E. (1957). Population studies - animal ecology and demography - con- cluding remarks. Cold Spring Harbor Symposia on Quantitative Biology, 22, 415– 427. Ives, J.D. (1960). The deglaciation of Labrador-Ungava, an outline. Cahi´ersdeg´eographie de Qu´ebec, 8, 323–344. Jacquez, G.M., Fortin, M.-J., & Goovaerts, P. (2008). Preface to the special issue on spatial statistics for boundary and patch analysis. Environmental and Ecological Statistics, 15, 365–367. 119 List of References

Janzen, D.H. (2004). Setting up tropical biodiversity for conservation through nondam- aging use: participation by parataxonomists. Journal of Applied Ecology, 41, 181– 187. Janzen, D.H. & Hallwachs, W. (2011). Joining inventory by parataxonomists with DNA barcoding of a large complex tropical conserved wildland in northwestern Costa Rica. PLoS One, 6, e18123. Jinbo, U., Kato, T., & Ito, M. (2011). Current progress in DNA barcoding and future implications for entomology. Entomological Science, 14, 107–124. Johnson, M.T.J. & Stinchcombe, J.R. (2007). An emerging synthesis between community ecology and evolutionary biology. Trends in Ecology & Evolution, 22, 250–257. Jost, L. (2006). Entropy and diversity. Oikos, 113, 363–375. Jost, L. (2007). Partitioning diversity into independent alpha and beta components. Ecology, 88, 2427–2439. Jost, L. (2010). Independence of alpha and beta diversities. Ecology, 91, 1969–1974. Jurasinski, G., Retzer, V., & Beierkuhnlein, C. (2009). Inventory, differentiation, and proportional diversity: a consistent terminology for quantifying species diversity. Oecologia, 159, 15–26. Katoh, K. & Standley, D.M. (2013). MAFFT Multiple Sequence Alignment Software Version 7: Improvements in Performance and Usability. Molecular Biology and Evolution, 30, 772–780. Kaufman, L. & Rousseeuw, P.J. (2009). Finding Groups in Data: An Introduction to Cluster Analysis. Vol. 344. John Wiley & Sons, New Jersey, United States. Keddy, P.A. (1992). Assembly and response rules: two goals for predictive community ecology. Journal of Vegetation Science, 3, 157–164. Kelly, A.E. & Goulden, M.L. (2008). Rapid shifts in plant distribution with recent climate change. Proceedings of the National Academy of Sciences, 105, 11823–11826.

120 List of References

Kembel, S.W. & Hubbell, S.P. (2006). The phylogenetic structure of a neotropical forest tree community. Ecology, 87, S86–S99. Kembel, S.W., Cowan, P.D., Helmus, M.R., Cornwell, W.K., Morlon, H., Ackerly, D.D., Blomberg, S.P., et al. (2010). Picante: R tools for integrating phylogenies and ecology. Bioinformatics, 26, 1463–1464. Kesanakurti, P.R., Fazekas, A.J., Burgess, K.S., Percy, D.M., Newmaster, S.G., Graham, S.W., Barrett, S.C.H., et al. (2011). Spatial patterns of plant diversity below-ground as revealed by DNA barcoding. Molecular Ecology, 20, 1289–1302. Kessler, M., Kluge, J., Hemp, A., & Ohlem¨uller, R. (2011). A global comparative analysis of elevational species richness patterns of ferns. Global Ecology and Biogeography, 20, 868–880. Kikvidze, Z., Suzuki, M., & Brooker, R. (2011). Importance versus intensity of ecological effects: why context matters. Trends in Ecology & Evolution, 26, 383–388. Koleff, P., Gaston, K.J., & Lennon, J.J. (2003). Measuring beta diversity for presence– absence data. Ecology, 72, 367–382. Kool, A., de Boer, H.J., Kr¨uger, A.,˚ Rydberg, A., Abbad, A., Bj¨ork, L., & Martin, G. (2012). Molecular identification of commercialized medicinal plants in southern Morocco. PLoS One, 7, e39459. Kraft, N.J.B., Cornwell, W.K., Webb, C.O., & Ackerly, D.D. (2007). Trait evolution, community assembly, and the phylogenetic structure of ecological communities. American Naturalist, 170, 271–283. Kraft, N.J.B., Adler, P.B., Godoy, O., James, E.C., Fuller, S., & Levine, J.M. (2015a). Community assembly, coexistence and the environmental filtering metaphor. Func- tional Ecology, 29, 592–599. Kraft, N.J.B., Godoy, O., & Levine, J.M. (2015b). Plant functional traits and the mul- tidimensional nature of species coexistence. Proceedings of the National Academy of Sciences, 112, 797–802. 121 List of References

Kress, W.J. & Erickson, D.L. (2007). A two-locus global DNA barcode for land plants: the coding rbcL gene complements the non-coding trnH –psbA spacer region. PLoS One, 2, e508. Kress, W.J., Wurdack, K.J., Zimmer, E.A., Weigt, L.A., & Janzen, D.H. (2005). Use of DNA barcodes to identify flowering plants. Proceedings of the National Academy of Sciences of the United States of America, 102, 8369–8374. Kress, W.J., Erickson, D.L., Jones, F.A., Swenson, N.G., Perez, R., Sanjur, O., & Bermingham, E. (2009). Plant DNA barcodes and a community phylogeny of a tropical forest dynamics plot in Panama. Proceedings of the National Academy of Sciences, 106, 18621–18626. Kress, W.J., Erickson, D.L., Swenson, N.G., Thompson, J., Uriarte, M., & Zimmerman, J.K. (2010). Advances in the use of DNA barcodes to build a community phylogeny for tropical trees in a Puerto Rican forest dynamics plot. Plos One, 5, e15409. Krishnamurthy, P.K. & Francis, R.A. (2012). A critical review on the utility of DNA barcoding in biodiversity conservation. Biodiversity and Conservation, 21, 1901–1919. Kron, K.A. & Luteyn, J.L (2005). Origins and biogeographic patterns in Ericaceae: new insights from recent phylogenetic analyses. In: Plant Diversity and Complexity Patterns: Local, Regional and Global Dimensions. (eds. Friis, I. & Balslev, H.). Proceedings of an International Symposium held at the Royal Danish Academy of Sciences and Letters, Copenhagen, Denmark, pp. 479–500. Kuzmina, M.L., Johnson, K.L., Barron, H.R., & Hebert, P.D.N. (2012). Identification of the vascular plants of Churchill, Manitoba, using a DNA barcode library. BMC Ecology, 12, 25–35. Lahaye, R., van der Bank, M., Bogarin, D., Warner, J., Pupulin, F., Gigot, G., Maurin, O., et al. (2008). DNA barcoding the floras of biodiversity hotspots. Proceedings of the National Academy of Sciences, 105, 2923.

122 List of References

Lalibert´e, E., Legendre, P., & Shipley, B. (2014). FD: measuring functional diversity from multiple traits, and other tools for functional ecology. R package version 1.0-12. Lande, R. (1996). Statistics and partitioning of species diversity, and similarity among multiple communities. Oikos, 5–13. Lawton, J.H. (1999). Are there general laws in ecology? Oikos, 177–192. Lechowicz, M.J. & Adams, M.S. (1978). Diurnal and seasonal structure of the climate at Schefferville, Quebec. Arctic and Alpine Research, 10, 95–104. Legendre, P. & De C´aceres, M. (2013). Beta diversity as the variance of community data: dissimilarity coefficients and partitioning. Ecology Letters, 16, 951–963. Legendre, P. & Legendre, L. (2012a). Chapter 13 - Spatial analysis. In: Numerical Ecology, 3rd Edition. (eds. Legendre, P. & Legendre, L.). Elsevier, Oxford, United Kingdom, pp. 785–858. Legendre, P. & Legendre, L. (2012b). Chapter 9 - Ordination in reduced space. In: Numerical Ecology, 3rd Edition. (eds. Legendre, P. & Legendre, L.). Elsevier, Oxford, United Kingdom, pp. 425–520. Legendre, P., Borcard, D., & Peres-Neto, P.R. (2005). Analyzing beta diversity: partition- ing the spatial variation of community composition data. Ecological Monographs, 75, 435–450. Leponce, M., Meyer, C., Hauser, C., Bouchet, P., Delabie, J.H.C., Weigt, L., & Basset, Y. (2010). Challenges and solutions for planning and implementing large-scale biotic inventories. In: Manual on Field Recording Techniques and Protocols for all Taxa Biodiversity Inventories. ABC Taxa. (eds. Eymann, J., Degreef, J., & Hauser, C.). Belgium Development Cooperation, Brussels, Belgium, pp. 19–49. Lessard, J.-P., Belmaker, J., Myers, J.A., Chase, J.M., & Rahbek, C. (2012). Inferring local ecological processes amid species pool influences. Trends in Ecology & Evolution, 27, 600–607.

123 List of References

Lessard-Therrien, M., Davies, T.J., & Bolmgren, K. (2014). A phylogenetic comparative study of flowering phenology along an elevational gradient in the Canadian subarctic. International Journal of Biometeorology, 58, 455–462. Letten, A.D. & Cornwell, W.K. (2015). Trees, branches and (square) roots: why evolution- ary relatedness is not linearly related to functional distance. Methods in Ecology and Evolution, 6, 439–444. Lewontin, R.C. (1974). The Genetic Basis of Evolutionary Change. Columbia University Press, New York, New York, United States. Li, D.-Z., Gao, L.-M., Li, H.-T., Wang, H., Ge, X.-J., Liu, J.-Q., Chen, Z.-D., et al. (2011a). Comparative analysis of a large dataset indicates that internal transcribed spacer (ITS) should be incorporated into the core barcode for seed plants. Proceedings of the National Academy of Sciences, 108, 19641–19646. Li, M., Cao, H., But, P.P.-H., & Shaw, P.-C. (2011b). Identification of herbal medicinal materials using DNA barcodes. Journal of Systematics and Evolution, 49, 271–283. Li, R., Kraft, N.J.B., Yang, J., & Wang, Y. (2015). A phylogenetically informed delin- eation of floristic regions within a biodiversity hotspot in Yunnan, China. Scientific Reports , 5, 9396. Li, Y., Gao, L.-M., Poudel, R.C., Li, D.-Z., & Forrest, A. (2011c). High universality of matK primers for barcoding gymnosperms. Journal of Systematics and Evolution, 49, 169–175. Loreau, M. (2009). Linking biodiversity and ecosystems: towards a unifying ecological theory. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 365, 49–60. Lortie, C.J., Brooker, R.W., Choler, P., Kikvidze, Z., Michalet, R., Pugnaire, F.I., & Callaway, R.M. (2004). Rethinking plant community theory. Oikos, 107, 433–438. L¨ove, D. (1970). Subarctic and subalpine: where and what? Arctic and Alpine Research, 2, 63–73. 124 List of References

Lozupone, C. & Knight, R. (2005). UniFrac: a new phylogenetic method for comparing microbial communities. Applied and Environmental Microbiology, 71, 8228–8235. Luo, Y., Ogle, K., Tucker, C., Fei, S., Gao, C., LaDeau, S., Clark, J.S., et al. (2011). Eco- logical forecasting and data assimilation in a data-rich era. Ecological Applications, 21, 1429–1442. MacArthur, R. & Levins, R. (1964). Competition, habitat selection, and character displacement in a patchy environment. Proceedings of the National Academy of Sciences of the United States of America, 51, 1207–1210. MacArthur, R. & Levins, R. (1967). The limiting similarity, convergence, and divergence of coexisting species. American Naturalist, 101, 377–385. MacArthur, R.H. (1972). Geographical Ecology: Patterns in the Distribution of Species. Princeton University Press, Princeton, New Jersey. MacArthur, R.H. & Pianka, E.R. (1966). On optimal use of a patchy environment. The American Naturalist, 100, 603–609. Maechler, M., Rousseeuw, P., Struyf, A., Hubert, M., & Hornik, K. (2015). cluster: Cluster Analysis Basics and Extensions. R package version 2.0.1. Maurin, O., Davies, T.J., Burrows, J.E., Daru, B.H., Yessoufou, K., Muasya, A.M., Van der Bank, M., et al. (2014). Savanna fire and the origins of the underground forests of Africa. New Phytologist, 204, 201–214. Maycock, P.F. (1961). Botanical studies on Mont St. Hilaire, Rouville County, Quebec. I - General description of the area and a floristic survey. Canadian Journal of Botany, 39, 1293–1325. Mayfield, M.M. & Levine, J.M. (2010). Opposing effects of competitive exclusion on the phylogenetic structure of communities. Ecology Letters, 13, 1085–1093. McNaughton, S.J. & Wolf, L.L. (1970). Dominance and the niche in ecological systems. Science, 167, 131–139.

125 List of References

Meier, R., Shiyang, K., Vaidya, G., & Ng, P.K.L. (2006). DNA barcoding and taxonomy in Diptera: a tale of high intraspecific variability and low identification success. Systematic Biology, 55, 715–728. Meyer, C.P. & Paulay, G. (2005). DNA barcoding: error rates based on comprehensive sampling. PLoS Biology, 3, e422. Mikl´os, I. & Podani, J. (2004). Randomization of presence-absence matrices: comments and new algorithms. Ecology, 85, 86–92. Miller, A.G. & Nyberg, J.A. (1995). Collecting herbarium vouchers. In: Collecting Plant Genetic Diversity: Technical Guidelines. (eds. Guarino, L., Rao, V.R., & Reid, R.). CAB International, Wallingford, United Kingdom, pp. 561–573. Milligan, G.W. & Cooper, M.C. (1985). An examination of procedures for determining the number of clusters in a data set. Psychometrika, 50, 159–179. Moran, P.A.P. (1950). Notes on continuous stochastic phenomena. Biometrika, 37, 17–23. Morlon, H., Kefi, S., & Martinez, N.D. (2014). Effects of trophic similarity on community composition. Ecology Letters, 17, 1495–1506. Mouillot, D., Simkov´˘ a, A., Morand, S., & Poulin, R. (2005). Parasite species coexistence and limiting similarity: a multiscale look at phylogenetic, functional and reproductive distances. Oecologia, 146, 269–278. Moulton, M.P. & Pimm, S.L. (1987). Morphological assortment in introduced Hawaiian passerines. Evolutionary Ecology, 1, 113–124. Mouquet, N., Devictor, V., Meynard, C.N., Munoz, F., Bersier, L.F., Chave, J., Couteron, P., et al. (2012). Ecophylogenetics: advances and perspectives. Biological Reviews, 87, 769–785. Muasya, A. M., Simpson, D. A., Verboom, G. A., Goetghebeur, P., Naczi, R. F. C., Chase, M. W., & Smets, E. (2009). Phylogeny of Cyperaceae based on DNA sequence data: current progress and future prospects. The Botanical Review, 75, 2–21.

126 List of References

Myers, N. (1988). Threatened biotas: “hot spots” in tropical forests. The Environmental- ist, 8, 187–208. Ndiribe, C., Pellissier, L., Antonelli, S., Dubuis, A., Pottier, J., Vittoz, P., Guisan, A., et al. (2013). Phylogenetic plant community structure along elevation is lineage specific. Ecology and Evolution, 3, 4925–4939. Nipperess, D.A., Faith, D.P., & Barton, K. (2010). Resemblance in phylogenetic diversity among ecological assemblages. Journal of Vegetation Science, 21, 809–820. Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., O’Hara, R.B., Simpson, G.L., et al. (2013). vegan: Community Ecology Package, R package version 2.0-10. Oredsson, A. (2000). Choice of surveyor is vital to the reliability of floristic change studies. Watsonia, 23, 287–292. Palmer, J.D., Jansen, R.K., Michaels, H.J., Chase, M.W., & Manhart, J.R. (1988). Chloroplast DNA variation and plant phylogeny. Annals of the Missouri Botanical Garden, 75, 1180–1206. Palmer, M.W., Earls, P.G., Hoagland, B.W., White, P.S., & Wohlgemuth, T. (2002). Quantitative tools for perfecting species lists. Environmetrics, 13, 121–137. Pamilo, P. & Nei, M. (1988). Relationships between gene trees and species trees. Molecu- lar Biology and Evolution, 5, 568–583. Panero, J.L. & Funk, V.A. (2008). The value of sampling anomalous taxa in phylogenetic studies: major clades of the Asteraceae revealed. Molecular Phylogenetics and Evolution, 47, 757–782. Paradis, E., Claude, J., & Strimmer, K. (2004). APE: analyses of phylogenetics and evolution in R language. Bioinformatics, 20, 289–290. Parmesan, C. (2006). Ecological and evolutionary responses to recent climate change. Annual Review of Ecology, Evolution, and Systematics, 37, 637–669.

127 List of References

Pautasso, M. (2012). Challenges in the conservation and sustainable use of genetic resources. Biology Letters, 8, 321–323. Payette, S., Fortin, M.-J., & Gamache, I. (2001). The subarctic forest-tundra: the struc- ture of a biome in a changing climate. BioScience, 51, 709–718. Pearse, W.D., Purvis, A., Cavender-Bares, J., & Helmus, M.R. (2014). Metrics and models of community phylogenetics. In: Modern Phylogenetic Comparative Methods and Their Application in Evolutionary Biology. (ed. Garamszegi, L.Z.). Springer, pp. 451–464. Pennisi, E. (2007). Wanted: a barcode for plants. Science, 318, 190–191. Peres-Neto, P.R., Leibold, M.A., & Dray, S. (2012). Assessing the effects of spatial contingency and environmental filtering on metacommunity phylogenetics. Ecology, 93, S14–S30. Porsild, A.E. & Cody, W.J. (1980). Vascular Plants of Continental Northwest Territories, Canada. National Museums of Canada, Ottawa, Canada. Post, D.M., Doyle, M.W., Sabo, J.L., & Finlay, J.C. (2007). The problem of boundaries in defining ecosystems: a potential landmine for uniting geomorphology and ecology. Geomorphology, 89, 111–126. Preston, F.W. (1948). The commonness, and rarity, of species. Ecology, 29, 254–283. Pryer, K.M., Schneider, H., Smith, A.R., Cranfill, R., Wolf, P.G., Hunt, J.S., & Sipes, S.D. (2001). Horsetails and ferns are a monophyletic group and the closest living relatives to seed plants. Nature, 409, 618–622. Qian, H. (1999). Floristic analysis of vascular plant genera of North America north of Mexico: characteristics of phytogeography. Journal of Biogeography, 26, 1307–1321. Qian, H. & Ricklefs, R.E. (2007). A latitudinal gradient in large-scale beta diversity for vascular plants in North America. Ecology Letters, 10, 737–744. Qian, H., Swenson, N.G., & Zhang, J. (2013). Phylogenetic beta diversity of angiosperms in North America. Global Ecology and Biogeography, 22, 1152–1161. 128 List of References

R Core Team (2014). R: A Language and Environment for Statistical Computing.R Foundation for Statistical Computing, Vienna, Austria. Rand, W.M. (1971). Objective criteria for the evaluation of clustering methods. Journal of the American Statistical Association, 66, 846–850. Ratnasingham, S. & Hebert, P.D.N. (2007). BOLD: The Barcode of Life Data System (http://www.barcodinglife.org). Molecular Ecology Notes, 7, 355–364. Revell, L.J. (2012). phytools: an R package for phylogenetic comparative biology (and other things). Methods in Ecology and Evolution, 3, 217–223. Ricklefs, R. E. (1987). Community diversity - relative roles of local and regional processes. Science, 235, 167–171. Rieseberg, L.H. & Soltis, D.E. (1991). Phylogenetic consequences of cytoplasmic gene flow in plants. Evolutionary Trends in Plants, 5, 65–84. Rodman, J.E. & Cody, J.H. (2003). The taxonomic impediment overcome: NSF’s Partner- ships for Enhancing Expertise in Taxonomy (PEET) as a model. Systematic Biology, 52, 428–435. Root, H.T. & Nelson, P.R. (2011). Does phylogenetic distance aid in detecting environ- mental gradients related to species composition? Journal of Vegetation Science, 22, 1143–1148. Roughgarden, J. (2009). Is there a general theory of community ecology? Biology and Philosophy, 24, 521–529. Roughgarden, J., & Diamond, J. (1986). Overview: the role of species interactions in community ecology. In: Community Ecology. (eds. Diamond, J. & Case, T.J.). Harper and Row, Publishers, New York, New York, United States, pp. 333–343. Routledge, R.D. (1977). On Whittaker’s components of diversity. Ecology, 58, 1120–1127. Roy, S., Tyagi, A., Shukla, V., Kumar, A., Singh, U.M., Chaudhary, L.B., Datt, B., et al. (2010). Universal plant DNA barcode loci may not work in complex groups: a case study with Indian Berberis species. PLoS One,5. 129 List of References

Rubinoff, D., Cameron, S., & Will, K. (2006). Are plant DNA barcodes a search for the Holy Grail? Trends in Ecology & Evolution, 21, 1–2. Schindel, D. & Miller, S.E. (2005). DNA barcoding a useful tool for taxonomists. Nature, 435, 17. Shipley, B., Vile, D., & Garnier, E. (2006). From plant traits to plant communities: a statistical mechanistic approach to biodiversity. Science, 314, 812–814. Silvertown, J., Dodd, M., Gowing, D., Lawson, C., & McConway, K. (2006). Phylogeny and the hierarchical organization of plant diversity. Ecology, 87, S39–S49. Simberloff, D. (2004). Community ecology: is it time to move on? The American Natural- ist, 163, 787–799. Simberloff, D. & Connor, E.F. (1981). Missing species combinations. The American Naturalist, 118, 215–239. Simberloff, D.S. (1970). Taxonomic diversity of island biotas. Evolution, 24, 23–47. Slingsby, J.A. & Verboom, G.A. (2006). Phylogenetic relatedness limits cooccurrence at fine spatial scales: evidence from the schoenoid sedges (Cyperaceae: Schoeneae) of the Cape Floristic Region, South Africa. The American Naturalist, 168, pp. 14–27. Soltis, D.E., Soltis, P.S., & Doyle, J.J. (1998). Molecular Systematics of Plants II: DNA Sequencing. Klewer Academic Publishers, Norwell, Massachusetts. Starr, J.R., Harris, S.A., & Simpson, D.A. (2008). Phylogeny of the unispicate taxa in Cyperaceae tribe Cariceae II: The limits of Uncinia.In:Sedges: Uses, Diversity, and Systematics of the Cyperaceae. (eds. Naczi, R.F.C. and B.A. Ford). 108th ed. Monographs in Systematic Botany from the Missouri Botanical Garden, Missouri Botanical Garden Press, St. Louis, Missouri, pp. 243–268. Starr, J.R., Naczi, R.F.C., & Chouinard, B.N. (2009). Plant DNA barcodes and species resolution in sedges (Carex, Cyperaceae). Molecular Ecology Resources, 9, 151–163.

130 List of References

Stoeckle, M.Y., Gamble, C.C., Kirpekar, R., Young, G., Ahmed, S., & Little, D.P. (2011). Commercial teas highlight plant DNA barcode identification successes and obstacles. Scientific Reports, 1, 1–7. Straub, S.C.K., Parks, M., Weitemier, K., Fishbein, M., Cronn, R.C., & Liston, A. (2012). Navigating the tip of the genomic iceberg: next-generation sequencing for plant systematics. American Journal of Botany, 99, 349–364. Strayer, D.L., Power, M.E., Fagan, W.F., Pickett, S.T.A., & Belnap, J. (2003). A classifi- cation of ecological boundaries. BioScience, 53, 723–729. Strong, D.R. Jr., Szyska, L.A., & Simberloff, D.S. (1979). Test of community-wide character displacement against null hypotheses. Evolution, 33, 897–913. Sukumaran, J. & Holder, M.T. (2010). DendroPy: a Python library for phylogenetic computing. Bioinformatics, 26, 1569–1571. Sutherland, W.J., Freckleton, R.P., Godfray, H.C.J., Beissinger, S.R., Benton, T., Cameron, D.D., Carmel, Y., et al. (2013). Identification of 100 fundamental eco- logical questions. Journal of Ecology, 101, 58–67. Swenson, N.G. (2009). Phylogenetic resolution and quantifying the phylogenetic diversity and dispersion of communities. PLoS One, 4, e4390. Swenson, N.G., Enquist, B.J., Thompson, J., & Zimmerman, J.K. (2007). The influence of spatial and size scale on phylogenetic relatedness in tropical forest communities. Ecology, 88, 1770–1780. Talavera, G. & Castresana, J. (2007). Improvement of phylogenies after removing diver- gent and ambiguously aligned blocks from protein sequence alignments. Systematic Biology, 56, 564–577. Thompson, K., Gaston, K.J., & Band, S.R. (1999). Range size, dispersal and niche breadth in the herbaceous flora of central England. Journal of Ecology, 87, 150–155.

131 List of References

Timoney, K.P., la Roi, G.H., Zoltai, S.C., & Robinson, A.L. (1992). The high subarctic forest-tundra of northwestern Canada: position, width, and vegetation gradients in relation to climate. Arctic, 45, 1–9. Tisdale, E.W. (1947). The grasslands of the southern interior of British Columbia. Ecology, 28, 346–382. Tripp, E.A. & Hoagland, K.E. (2013). Typifying an era in biology through synthesis of biodiversity information: achievements and impediments. Taxon, 62, 899–911. Tukey, J.W. (1949). Comparing individual means in the analysis of variance. Biometrics, 5, 99–114. Tuomisto, H. (2010). A diversity of beta diversities: straightening up a concept gone awry. Part 1. Defining beta diversity as a function of alpha and gamma diversity. Ecography, 33, 2–22. Udvardy, M.D.F. & Udvardy, M.D.F. (1975). A Classification of the Biogeographical Provinces of the World. Vol. 8. International Union for Conservation of Nature and Natural Resources, Morges, Switzerland. Uriarte, M., Swenson, N.G., Chazdon, R.L., Comita, L.S., Kress, J.W., Erickson, D., Forero-Monta˜na, J., et al. (2010). Trait similarity, shared ancestry and the structure of neighbourhood interactions in a subtropical wet forest: implications for community assembly. Ecology Letters, 13, 1503–1514. Vamosi, S.M., Heard, S.B., Vamosi, J.C., & Webb, C.O. (2009). Emerging patterns in the comparative analysis of phylogenetic community structure. Molecular Ecology, 18, 572–592. van der Valk, A.G. (1981). Succession in wetlands: a Gleasonian approach. Ecology, 62, 688–696. Veech, J.A. & Crist, T.O. (2010). Toward a unified view of diversity partitioning. Ecology, 91, 1988–1992.

132 List of References

Vellend, M. (2001). Do commonly used indices of beta-diversity measure species turnover? Journal of Vegetation Science, 12, 545–552. Vellend, M. (2010). Conceptual synthesis in community ecology. Quarterly Review of Biology, 85, 183–206. Vellend, M., Cornwell, W.K., Magnuson-Ford, K., & Mooers, A.Ø. (2011). Measuring phylogenetic biodiversity. In: Biological Diversity: Frontiers in Measurement and Assessment. (eds. Magurran, A.E. & McGill, B.J.). Oxford University Press, Okford, United Kingdom, pp. 194–207. Vernooy, R., Haribabu, E., Muller, M.R., Vogel, J.H., Hebert, P.D.N., Schindel, D.E., Shimura, J., et al. (2010). Barcoding life to conserve biological diversity: beyond the taxonomic imperative. PLoS Biology, 8, e1000417. Vilela, B. & Villalobos, F. (2015). letsR: tools for data handling and analysis in macroecol- ogy. R package version 2.1. von Cr¨autlein, M., Korpelainen, H., Pietil¨ainen, M., & Rikkinen, J. (2011). DNA bar- coding: a tool for improved taxon identification and detection of species diversity. Biodiversity and Conservation, 20, 373–389. Walesiak, M., Dudek, A., & Dudek, M.A. (2008). clusterSim: Searching for optimal clustering procedure for a data set. R package version 0.44-2. Ward, D.B. (1974). The ”ignorant man” technique of sampling plant populations. Taxon, 23, 325–330. Warwick, R.M. & Clarke, K.R. (1995). New ‘biodiversity’ measures reveal a decrease in taxonomic distinctness with increasing stress. Marine Ecology Progress Series, 129, 301–305. Waterway, M.J. & Starr, J.R. (2007). Phylogenetic relationships in tribe Cariceae (Cyper- aceae) based on nested analyses of four molecular data sets. Aliso: A Journal of Systematic and Evolutionary Botany, 23, 165–192.

133 List of References

Waterway, M.J., Lechowicz, M.J., & Moore, T.R. (1984). Vegetation of the Schefferville region, Nouveau-Quebec. McGill Subarctic Research Paper, 39, 7–20. Waterway, M.J., Hoshino, T., & Masaki, T. (2009). Phylogeny, species richness, and ecological specialization in Cyperaceae tribe Cariceae. The Botanical Review, 75, 138–159. Weaver, J.E. & Fitzpatrick, T.J. (1934). The prairie. Ecological Monographs, 4, 112–295. Webb, C.O. (2000). Exploring the phylogenetic structure of ecological communities: an example for rain forest trees. American Naturalist, 156, 145–155. Webb, C.O. & Donoghue, M.J. (2004). Phylomatic: tree assembly for applied phylogenet- ics. Molecular Ecology Notes, 5, 181–183. Webb, C.O., Ackerly, D.D., McPeek, M.A., & Donoghue, M.J. (2002). Phylogenies and community ecology. Annual Review of Ecology and Systematics, 33, 475–505. Weiher, E. & Keddy, P. (2001). Assembly rules as general constraints on community composition. In: Ecological Assembly Rules: Perspectives, Advances, Retreats. (eds. Weiher, E. & Keddy, P.). Cambridge University Press, Cambridge, United Kingdom, pp. 251–271. Weiher, E., Clarke, G.D.P., & Keddy, P.A. (1998). Community assembly rules, morpholog- ical dispersion, and the coexistence of plant species. Oikos, 81, 309–322. Weitemier, K., Straub, S.C.K., Cronn, R.C., Fishbein, M., Schmickl, R., McDonnell, A., & Liston, A. (2014). Hyb-Seq: combining target enrichment and genome skimming for plant phylogenomics. Applications in Plant Sciences, 2, 1400042. White, P.J.T., McGill, B.J., & Lechowicz, M.J. (2011). Human-disturbance and caterpil- lars in managed forest fragments. Biodiversity and Conservation, 20, 1745–1762. Whittaker, R.H. (1960). Vegetation of the Siskiyou Mountains, Oregon and California. Ecological Monographs, 30, 280–338. Whittaker, R.H. (1972). Evolution and measurement of species diversity. Taxon, 213–251.

134 List of References

Whittaker, R.H. (1975). Communities and Ecosystems. The Macmillan Company, London, England. Wiegert, R.G. (1974). Competition: a theory based on realistic, general equations of population growth. Science, 185, 539–542. Wiens, J.J. (2003). Missing data, incomplete taxa, and phylogenetic accuracy. Systematic Biology, 52, 528–538. Wiens, J.J. & Morrill, M.C. (2011). Missing data in phylogenetic analysis: reconciling results from simulations and empirical data. Systematic Biology, 60, 719–731. Wiens, J.J., Ackerly, D.D., Allen, A.P., Anacker, B.L., Buckley, L.B., Cornell, H.V., Damschen, E.I., et al. (2010). Niche conservatism as an emerging principle in ecology and conservation biology. Ecology Letters, 13, 1310–1324. Willig, M.R., Kaufman, D.M., & Stevens, R.D. (2003). Latitudinal gradients of biodiver- sity: pattern, process, scale, and synthesis. Annual Review of Ecology, Evolution, and Systematics, 34, 273–309. Wilson, D.S. & Yoshimura, J. (1994). On the coexistence of specialists and generalists. The American Naturalist, 144, 692–707. Wilson, M.V. & Shmida, A. (1984). Measuring beta diversity with presence-absence data. Journal of Ecology, 72, 1055–1064. Wisz, M.S., Pottier, J., Kissling, W.D., Pellissier, L., Lenoir, J., Damgaard, C.F., Dor- mann, C.F., et al. (2013). The role of biotic interactions in shaping distributions and realised assemblages of species: implications for species distribution modelling. Biological Reviews, 88, 15–30. Womble, W.H. (1951). Differential systematics. Science, 114, 315–322. Wood, S.N. (2006). Low-rank scale-invariant tensor product smooths for generalized additive mixed models. Biometrics, 62, 1025–1036.

135 List of References

Yarrow, M.M. & Mar´ın, V.H. (2007). Toward conceptual cohesiveness: a historical analysis of the theory and utility of ecological boundaries and transition zones. Ecosystems, 10, 462–476. Yguel, B., Bailey, R., Tosh, N.D., Vialatte, A., Vasseur, C., Vitrac, X., Jean, F., et al. (2011). Phytophagy on phylogenetically isolated trees: why hosts should escape their relatives. Ecology Letters, 14, 1117–1124. Zoltai, S.C. (1988). Ecoclimatic provinces of Canada and man-induced climatic change. In: Newsletter, Canadian Committee on Ecological Land Classification. Vol. 17. Canadian Forestry Service, Ottawa, Canada, pp. 12–15. Zwickl, D.J. (2006). Genetic Algorithm Approaches for the Phylogenetic Analysis of Large Biological Sequence Datasets Under the Maximum Likelihood Criterion. Ph.D dissertation. The University of Texas at Austin, Austin, Texas.

136 Appendices Appendix A

Supporting Information — Chapter 2 Appendix A1: Contributors to the WG1.2 Land Plants iBOL Working Group based on voluntary survey conducted in March and April 2013.

Project Title Taxonomic scope Plant focus Geographic focus

FinBOL Animals, plants, and fungi Seed plants, pteridophtes, Finland and bryophtes (FinBOL - Plants) Dinghushan FDP Angiosperm species Angiosperms Dinghushan FDP NA Angiosperm species Angiosperms Cape Floristic Region VPSBC Vascular Plants Vascular plants Southern British Columbia (below 50 degrees north) — Vascular plants Vascular plants Brazil — Medicinal plants Medicinal plants Unclear (specimens from botanical gardens or seed repositories) CERF Woody plants Trees, shrubs, some lianas NE Queensland, wet tropics, and Cape York Barcoding Australian Tropical Trees Trees NE Queensland and Tropical trees Cape York Peninsula — Endangered medicinal plants Medicinal plants Western and Eastern Ghats of India — Threatened or endangered species Pinaceae, Cupressaceae, Podocarpaceae, Mexico Taxaceae, Agavaceae, Orchidaceae, Cactaceae, and Crassulaceae — Angiosperms Angiosperms Tamil Nadu State — Angiosperms Angiosperm hosts Nilgiris Biosphere Reserve — Biota Vascular plants Chamela Reserve, central Mexico — Vascular Plants Vascular plants Luxembourg REDNO Red-listed plants from Norway Land plants Norway RINGV RINGV Gymnosperms and angiosperms Tronheim, Norway — Medicinal plants Important medicinal plants High altitude Nepal Appendix A2: Updated terrestrial vascular plant species list for the Gault Reserve at Mont St. Hilaire, Qu´ebec as of April 9, 2014. Aquatic vascular plant taxa have been removed from list.

List* Family† Species† Authority† Year Collected Comments Old Pinaceae Abies balsamea (Linnaeus) Miller 2012 Old Euphorbiaceae Acalypha rhomboidea Rafinesque 2012 Old Euphorbiaceae Acalypha virginica 2012 - synonym Synonym for Acalypha rhomboidea Old Euphorbiaceae Acalypha virginica var. rhomboidea (Rafinesque) Cooperrider 2012 - synonym Synonym for Acalypha rhomboidea; variety issue Old Sapindaceae Acer pensylvanicum Linnaeus 2012 Old Sapindaceae Acer rubrum Linnaeus 2012 Old Sapindaceae Acer saccharinum Linnaeus f. Not Collected Reported to not grow in reserve Old Sapindaceae Acer saccharum Marshall 2012 Old Sapindaceae Acer spicatum Lamarck 2012 Old Asteraceae Achillea millefolium Linnaeus 2012 Old Ranunculaceae Actaea pachypoda Elliott 2012 Old Ranunculaceae Actaea rubra (Aiton) Willdenow 2013 Old Ranunculaceae Actaea rubra forma neglecta Not Collected Not in VASCAN database Old Pteridaceae Adiantum pedatum Linnaeus 2012 Old Papaveraceae Adlumia fungosa (Aiton) Green ex Britton, Sterns & Poggenburgh 2013 Old Asteraceae Ageratina altissima var. altissima (Linnaeus) R.M. King & H. Robinson 2012 Old Rosaceae Agrimonia gryposepala Wallroth 2012 Old Poaceae Agropyron repens (Linnaeus) P. Beauvois 2012 - synonym Synonym for Elymus repens Old Poaceae Agropyron trachycaulum (Link) Malte ex H.F. Lewis 2013 - synonym Synonym for Elymus trachycaulus subsp. trachycaulus Old Poaceae Agrostis hyemalis var. scabra (Willdenow) H.L. Blomquist 2012 - synonym Synonym for Agrostis scabra Old Poaceae Agrostis perennans (Walter) Tuckerman 2012 Old Poaceae Agrostis perennaus 2012 - typographical error Spelling - should be Agrostis perennans Old Poaceae Agrostis scabra Willdenow 2012 Old Poaceae Agrostis stolonifera Linnaeus 2012 Old Alismataceae Alisma triviale Pursh 2012 Old Amaryllidaceae Allium schoenoprasum Linnaeus Not Collected Reported to be very rare in reserve Old Amaryllidaceae Allium schoenoprasum var. sibiricum Linnaeus (Hartman) Not Collected Not collected, thus impossible to verify variety/subspecies Old Amaryllidaceae Allium tricoccum var. burdickii Hanes 2013 Old Betulaceae Alnus incana subsp. rugosa (Du Roi) R.T. Clausen 2012 continued (Appendix A2) . . . 140 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Betulaceae Alnus rugosa (DuRoi) Sprengel 2012 - synonym Synonym for Alnus incana subsp. rugosa Old Poaceae Alopecurus aequalis var. aequalis Sobolewski 2012 Listed on original list as Alopecurus aequalis; confirmed from voucher that is Alopecurus aequalis var. aequalis Old Amaranthaceae Amaranthus retroflexus Linnaeus Not Collected Old Asteraceae Ambrosia artemisiifolia Linnaeus 2012 Old Rosaceae Amelanchier arborea (F. Michaux) Fernald 2012 Old Rosaceae Amelanchier bartramiana (Tausch) M. Roemer Not Collected Old Rosaceae Amelanchier laevis Wiegand 2012 Old Rosaceae Amelanchier sanguinea (Pursh) de Candolle 2012 Old Rosaceae Amelanchier stolonifera Wiegand 2012 - synonym Synonym for Amelanchier spicata Old Fabaceae Amphicarpa bracteata (Linnaeus) Fernald nom. Illeg. 2012 - typographical error Synonym for Amphicarpaea bracteata Old Fabaceae Amphicarpaea bracteata (Linnaeus) Fernald 2012 Old Asteraceae Anaphalis margaritacea (Linnaeus) Bentham & Hooker f. 2012 Old Ranunculaceae Anemone acutiloba (de Candolle) G. Lawson 2012 Old Ranunculaceae Anemone americana (de Candolle) H. Hara 2012 Old Ranunculaceae Anemone canadensis Linnaeus 2012 Old Ranunculaceae Anemone hepatica var. acuta Not Collected Not in VASCAN data base; assume synonym for Anemone acutiloba Old Ranunculaceae Anemone virginiana Linnaeus Not Collected Old Asteraceae Antennaria canadensis Greene 2012 - synonym Synonym for Antennaria howellii subsp. canadensis Old Asteraceae Antennaria fallax Greene 2013 - synonym Synonym for Antennaria parlinii subsp. fallax Old Asteraceae Antennaria howellii subsp. canadensis (Greene) R.J. Bayer 2012 Old Asteraceae Antennaria neglecta Greene 2012 Old Asteraceae Antennaria neodioica Greene Not Collected Synonym for Antennaria howelli subsp. neodioica (not on list) Old Asteraceae Antennaria parlinii subsp. fallax (Greene) R.J. Bayer & Stebbins 2013 Old Asteraceae Antennaria plantaginifolia (Linnaeus) Hooker Not Collected Old Apocynaceae Apocynum androsaemifolium Linnaeus 2012 continued (Appendix A2) . . . 141 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Ranunculaceae Aquilegia canadensis Linnaeus 2012 Old Brassicaceae Arabis divaricarpa A. Nelson Not Collected Synonym for Boechera divaricarpa Old Brassicaceae Arabis drummondi A. Gray Not Collected Synonym for Boechera stricata Old Brassicaceae Arabis hirsuta (Linnaeus) Scopoli Not Collected Old Brassicaceae Arabis laevigata (Muhlenberg ex Willdenow) Poiret 2012 - synonym Synonym for Borodinia laevigata Old Araliaceae Aralia hispida Ventenat 2013 Old Araliaceae Aralia nudicaulis Linnaeus 2012 Old Araliaceae Aralia racemosa Linnaeus 2012 Old Asteraceae Arctium minus (Hill) Bernhardi 2012 Old Ericaceae Arctostaphylos uva-ursi (Linnaeus) Sprengel Not Collected Old Caryophyllaceae Arenaria lateriflora Linnaeus Not Collected Synonym for Moehringia lateriflora Old Araceae Arisaema atrorubens (Aiton) Blume 2012 - synonym Synonym for Arisaema triphyllum subsp. triphyllum Old Araceae Arisaema quinata Not Collected Not in VASCAN data base Old Araceae Arisaema triphylla 2012 - typographical error Should be Arisaema triphyllum subsp. triphyllum Old Araceae Arisaema triphylla var. triphylla 2012 - typographical error Spelled incorrectly Old Araceae Arisaema triphyllum subsp. triphyllum (Linnaeus) Schott 2012 Originally on list as Arisaema triphyllum, but subspecies verified with voucher Old Rosaceae Aronia melanocarpa (Michaux) Elliott 2013 Old Aristolochiaceae Asarum canadense Linnaeus 2012 Old Apocynaceae Asclepias syriaca Linnaeus 2012 Old Asparagaceae Asparagus officinalis Linnaeus 2012 Old Asteraceae Aster acuminatus Michaux 2012 - synonym Synonym for Oclemena acuminata Old Asteraceae Aster ciliolatus Lindley 2012 - synonym Synonym for Symphyotrichum ciliolatum Old Asteraceae Aster cordifolius Linnaeus 2012 - synonym Synonym for Symphyotrichum cordifolium Old Asteraceae Aster lateriflorus (Linnaeus) Britton 2012 - synonym Synonym for Symphyotrichum lateriflorum var. lateriflorum Old Asteraceae Aster macrophyllus Linnaeus 2012 - synonym Synonym for Eurybia macrophylla Old Asteraceae Aster novae-angliae Linnaeus 2012 - synonym Synonym for Symphyotrichum novae-angliae Old Asteraceae Aster ontarionis Wiegand Not Collected Synonym for Symphyotrichum ontarionis continued (Appendix A2) . . . 142 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments var. ontarionis Old Asteraceae Aster puniceus Linnaeus 2012 - synonym Synonym for Symphyotrichum puniceum var. puniceum Old Asteraceae Aster simplex Willdenow 2012 - synonym Synonym for Symphyotrichum lanceolatum subsp. lanceolatum var. lanceolatum Old Asteraceae Aster umbellatus Miller 2012 - synonym Synonym for Doellingeria umbellata var. umbellata Old Athyriaceae Athyrium filix-femina (Linnaeus) Roth ex Mertens 2012 Old Diplaziopsidaceae Athyrium pycnocarpon (Sprengel) Tidestrom 2012 - synonym Synonym for Homalosorus pycnocarpos Old Athyriaceae Athyrium thelypterioides 2012 - synonym Spelling error; should be Athyrium thelypteroides; synonym for Deparia acrostichoides Old Poaceae Avenella flexuosa (Linnaeus) Drejer 2012 Old Brassicaceae Barbarea vulgaris W.T. Aiton 2012 Old Berberidaceae Berberis vulgaris Linnaeus Not Collected Old Betulaceae Betula alleghaniensis Britton 2012 Old Betulaceae Betula lutea F. Michaux nom. Illeg. 2012 - synonym Synonym for Betula alleghaniensis Old Betulaceae Betula papyrifera Marshall 2012 Old Betulaceae Betula populifolia Marshall 2012 Old Asteraceae Bidens cernua Linnaeus 2013 Old Asteraceae Bidens frondosa Linnaeus 2013 Old Brassicaceae Boechera grahamii (Lehmann) Windham & Al-Shehbaz 2012 Old Brassicaceae Borodinia laevigata (Muhlenberg ex Willdenow) P.J. Alexander & Windham 2012 Old Ophioglossaceae Botrychium dissectum Sprengel 2012 - synonym Synonym for Sceptridium dissectum Old Ophioglossaceae Botrychium lanceolatum (S.G. Gmelin) Angstrom 2012 Old Ophioglossaceae Botrychium matricariaefolium 2012 - typographical error Name spelled incorrectly Old Ophioglossaceae Botrychium matricariifolium (Doll) A. Braun ex W.D.J. Koch 2013 Old Ophioglossaceae Botrychium multifidum (S.G. Gmelin) Reprecht Not Collected Synonym for Sceptridium multifidum Old Ophioglossaceae Botrychium virginianum (Linnaeus) Swartz 2012 - synonym Synonym for Botrychium virginianum Old Ophioglossaceae Botrypus virginianus (Linnaeus) Michaux 2012 Old Poaceae Brachyelytrum canadensis Not Collected Not in VASCAN database Old Poaceae Brachyelytrum erectum (Schreber) Palisot de Beauvois 2012 Old Poaceae Brachyelytrum erectum var. continued (Appendix A2) . . . 143 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments septentrionale Babel Not Collected Synonym of Brachyelytrum aristosum Old Poaceae Calamagrostis canadensis (Michaux) Palisot de Beauvois Not Collected Old Poaceae Calamagrostis canadensis var. canadensis (Michaux) Palisot de Beauvois 2012 Old Ranunculaceae Caltha palustris Linnaeus 2012 Old Convolvulaceae Calystegia sepium (Linnaeus) R. Brown 2012 Old Campanulaceae Campanula rapunculoides Linnaeus 2012 Old Campanulaceae Campanula rotundifolia Linnaeus Not Collected Old Papaveraceae Capnoides sempervirens (Linnaeus) Borkhausen 2012 Old Brassicaceae Capsella bursa-pastoris (Linnaeus) Medikus 2012 Old Brassicaceae Cardamine diphylla (Michaux) Alph. Wood 2013 Old Brassicaceae Cardamine parviflora Linneaus 2013 Old Brassicaceae Cardamine parviflora var. arenicola (Britton) O.E. Schultz 2013 - synonym Synonym of Cardamine parviflora Old Brassicaceae Cardamine pensylvanica Muhlenberg ex Willdenow 2012 Old Cyperaceae Carex aenea Fernald 2013 - synonym Synonym of Carex foenea Old Cyperaceae Carex albicans Willdenow ex Sprengel Not Collected Old Cyperaceae Carex albursina E. Sheldon 2012 Old Cyperaceae Carex appalachica J.M. Webber & P.W. Ball 2012 Old Cyperaceae Carex arctata Boott 2013 Old Cyperaceae Carex artitecta Mackenzie Not Collected Synonym of Carex albicans var. albicans Old Cyperaceae Carex backii Boott 2013 Old Cyperaceae Carex blanda Dewey Not Collected Old Cyperaceae Carex bromoides subsp. bromoides Schkuhr ex Willdenow 2013 Old Cyperaceae Carex brunnescens (Persoon) Poiret Not Collected Old Cyperaceae Carex canescens Linnaeus 2013 Old Cyperaceae Carex cephaloidea (Dewey) Dewey 2012 Old Cyperaceae Carex cephalophora Muhlenberg ex Willdenow 2012 Old Cyperaceae Carex communis L.H. Bailey 2012 Old Cyperaceae Carex comosa Boott 2013 Old Cyperaceae Carex convoluta Mackenzie 2012 - synonym Synonym of Carex rosea Old Cyperaceae Carex crawfordii Fernald Not Collected Old Cyperaceae Carex crinita var. crinita Lamarck 2013 Old Cyperaceae Carex debilis Michaux 2013 Old Cyperaceae Carex deweyana Schweinitz 2012 Old Cyperaceae Carex digitalis Willdenow 2013 continued (Appendix A2) . . . 144 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Cyperaceae Carex disperma Dewey 2012 Old Cyperaceae Carex foenea Willdenow ex Elliot 2013 Old Cyperaceae Carex gracillima Schweinitz 2013 Old Cyperaceae Carex gynandra Schweinitz 2013 Old Cyperaceae Carex hirta Linnaeus 2012 Old Cyperaceae Carex hirtifolia Mackenzie 2012 Old Cyperaceae Carex hitchcockiana Dewey 2012 Old Cyperaceae Carex houghtoniana Torrey ex Dewey 2013 Old Cyperaceae Carex houghtonii 2013 - synonym Synonym for Carex houghtoniana Old Cyperaceae Carex hystricina Not Collected Spelling error; should be Carex hystericina Old Cyperaceae Carex intumescens Rudge 2012 Old Cyperaceae Carex lacustris Willdenow Not Collected Old Cyperaceae Carex laxiflora Lamarck 2012 Old Cyperaceae Carex leptalea subsp. leptalea Wahlenberg 2012 On original list as Carex leptalea; determined that subsp. leptalea from voucher Old Cyperaceae Carex leptonervia (Fernald) Fernald 2012 Old Cyperaceae Carex lucorum Willdenow ex Link 2013 Old Cyperaceae Carex lupulina Muhlenberg ex Willdenow Not Collected Old Cyperaceae Carex normalis Mackenzie 2012 Old Cyperaceae Carex ormostachys Wiegand Not Collected Spelling error - should be Carex ormostachya Old Cyperaceae Carex ovales Not Collected Old Cyperaceae Carex pallescens Linnaeus 2012 Old Cyperaceae Carex peckii Howe 2013 Old Cyperaceae Carex pedunculata Muhlenberg ex Willdenow 2012 Old Cyperaceae Carex pensylvanica Lamarck 2012 Old Cyperaceae Carex plantaginea Lamarck 2013 Old Cyperaceae Carex platyphylla J. Carey 2012 Old Cyperaceae Carex prasina Wahlenberg 2012 Old Cyperaceae Carex projecta Mackenzie 2012 Old Cyperaceae Carex radiata (Wahlenberg) Small 2012 Old Cyperaceae Carex retrorsa Schweinitz 2012 Old Cyperaceae Carex rosea Schkuhr ex Willdenow 2012 continued (Appendix A2) . . . 145 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Cyperaceae Carex rugosperma Mackenzie 2012 - synonym Synonym of Carex tonsa var. rugosperma Old Cyperaceae Carex scabrata Schweinitz 2012 Old Cyperaceae Carex scoparia Schkuhr ex Willdenow Not Collected Old Cyperaceae Carex sparganioides Muhlenberg ex Willdenow Not Collected Old Cyperaceae Carex sprengelii Dewey ex Sprengel 2012 Old Cyperaceae Carex stipata var. stipata Muhlenberg ex Willdenow 2012 Listed on original list as Carex stipata; confirmed variety with voucher Old Cyperaceae Carex tenera Dewey 2012 Old Cyperaceae Carex tonsa var. rugosperma (Mackenzie) Crins 2012 Old Cyperaceae Carex tribuloides var. tribuloides Wahlenberg 2013 Listed on original list as Carex tribuloides; confirmed variety with voucher Old Cyperaceae Carex trisperma Dewey 2012 Old Cyperaceae Carex vesicaria Linnaeus 2012 Old Cyperaceae Carex vulpinoidea Michaux 2012 Old Betulaceae Carpinus caroliniana Walter 2012 Old Juglandaceae Carya cordiformis (Wangenheim) K. Koch 2012 Spelling error in original list: Carya cordiformus Old Juglandaceae Carya cordiformus 2012 - typographical error Old Juglandaceae Carya ovata (Miller) K. Koch 2012 Old Poaceae Catabrosa sp. Not Collected Perhaps Catabrosa aquatica; no herbarium specimens found from site Old Berberidaceae Caulophyllum thalictroides (Linnaeus) Michaux 2012 Old Celastraceae Celastrus orbiculatus Thunberg Not Collected Old Celastraceae Celastrus scandens Linnaeus 2013 Old Cannabaceae Celtis occidentalis Linnaeus 2012 Old Caryophyllaceae Cerastium fontanum subsp. vulgare (Hartman) Greuter & Burdet 2012 Old Caryophyllaceae Cerastium vulgatum Linnaeus p.p. 2012 - synonym Synonym of Cerastium fontanum subsp. vulgare Old Papaveraceae Chelidonium majus Linnaeus 2012 Old Plantaginaceae Chelone glabra Linnaeus 2012 Old Amaranthaceae Chenopodiastrum simplex (Torrey) S. Fuentes, Uotila & Borsch 2012 Old Amaranthaceae Chenopodium album Linnaeus 2012 continued (Appendix A2) . . . 146 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Amaranthaceae Chenopodium hybridum Torrey 2012 - synonym Synonym of Chenopodiastrum simplex Old Amaranthaceae Chenopodium polyspermum Linnaeus 2012 Synonym of Lipandra polysperma var. polysperma Old Ericaceae Chimaphila umbellata (Linnaeus) W.P.C. Barton Not Collected Old Ericaceae Chimaphila umbellata var. cisatlantica S.F. Blake Not Collected Synonym of Carex umbellata subsp. umbellata Old Asteraceae Chrysanthemum leucanthemum Linnaeus 2012 - synonym Synonym of Leucanthemum vulgare Old Saxifragaceae Chrysosplenium americanum Schweinitz ex Hooker 2012 Old Asteraceae Cichorium intybus Linnaeus 2012 Old Apiaceae Cicuta bulbifera Linnaeus 2012 Old Apiaceae Cicuta maculata Linnaeus Not Collected Old Poaceae Cinna latifolia (Treviranus ex Goppinger) Grisebach 2012 Old Onagraceae Circaea alpina Linnaeus 2012 Old Onagraceae Circaea canadensis subsp. canadensis (Linnaeus) Ascherson & Magnus 2012 Old Onagraceae Circaea quadrisulcata (Linnaeus) A. Love & D. Love 2012 - synonym Synonym of Circaea canadensis subsp. canadensis Old Asteraceae Cirsium arvense (Linnaeus) Scopoli 2012 Old Montiaceae Claytonia caroliniana Michaux 2012 Old Montiaceae Claytonia virginica Linnaeus Not Collected Old Ranunculaceae Clematis verticillaris de Candolle Not Collected Synonym of Clematis occidentalis var. occidentalis Old Ranunculaceae Clematis virginiana Linnaeus Not Collected Old Liliaceae Clintonia borealis (Aiton) Rafinesque 2012 Old Convolvulaceae Convolvulus sepium Sims 2012 - synonym Synonym of Calystegia sepium Old Asteraceae Conyza canadensis (Linnaeus) Cronquist 2012 Synonym of Erigeron canadensis Old Ranunculaceae Coptis groenlandica (Oeder) Fernald 2012 - synonym Synonym of Coptis trifolia Old Ranunculaceae Coptis trifolia (Linnaeus) Salisbury 2012 Old Orchidaceae Corallorhiza maculata (Rafinesque) Rafinesque 2013 Old Orchidaceae Corallorhiza trifida Chatelain Not Collected Old Cornaceae Cornus alternifolia Linnaeus f. 2012 Old Cornaceae Cornus canadensis Linnaeus Not Collected Old Cornaceae Cornus racemosa Lamarck 2012 Old Cornaceae Cornus rugosa Lamarck Not Collected Old Cornaceae Cornus sericea (Michaux) Fosberg 2012 Synonym of Cornus stolonifera Old Cornaceae Cornus stolonifera Michaux 2012 - synonym Old Papaveraceae Corydalis sempervirens (Linnaeus) Persoon 2012 - synonym Synonym of Capnoides sempervirens continued (Appendix A2) . . . 147 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Betulaceae Corylus cornuta Marshall 2012 Old Rosaceae Crataegus brainerdi Sergent Not Collected Old Rosaceae Crataegus chrysocarpa var. phoenicea E.J. Palmer ex J.B. Phipps 2013 Old Rosaceae Crataegus columbiana var. chrysocarpa (Ashe) Dorn 2012 - synonym Synonym of Crataegus chrysocarpa var. chrysocarpa Old Rosaceae Crataegus jackii Sargent Not Collected Synonym of Crataegus lumaria Old Rosaceae Crataegus macrosperma Ashe 2012 Old Rosaceae Crataegus pruinosa (H.L. Wendland) K. Koch Not Collected Old Rosaceae Crataegus punctata Jacquin Not Collected Old Rosaceae Crataegus submollis Sargent Not Collected Old Apiaceae Cryptotaenia canadensis (Linnaeus) de Candolle 2013 Old Cyperaceae Cyperus inflexus Muhlenberg Not Collected Synonym of Cyperus squarrosus Old Cyperaceae Cyperus rivularis Kunth Not Collected Synonym of Cyperus bipartitus Old Orchidaceae Cypripedium acaule Aiton 2013 Old Orchidaceae Cypripedium calceolus Not Collected Two Cypridpedium calceolus varieties in VASCAN; cannot distinguish which one goes with collection record Old Orchidaceae Cypripedium calceolus Not Collected No record in VASCAN var. parviflorum Old Cystopteridaceae Cystopteris bulbifera (Linnaeus) Bernhardi 2012 Old Cystopteridaceae Cystopteris fragilis (Linnaeus) Bernhardi 2012 Old Poaceae Danthonia spicata (Linnaeus) P. Beauvois ex Roemer & Schultes Not Collected Old Apiaceae Daucus carota Linnaeus 2012 Old Dennstaedtiaceae Dennstaedtia punctilobula (Michaux) T. Moore 2012 Old Brassicaceae Dentaria diphylla Michaux 2012 - synonym Synonym of Cardamine diphylla Old Athyriaceae Deparia acrostichoides (Swartz) M. Kato 2012 Old Poaceae Deschampsia flexulosa 2012 - typographical error Name spelled incorrectly Old Poaceae Deschampsia flexuosa (Linnaeus) Trinius 2012 - synonym Old Fabaceae Desmodium canadense (Linnaeus) de Candolle 2012 Old Fabaceae Desmodium glutinosum (Muhlenberg ex Willdenow) Alph. Wood 2012 - synonym Synonym of Hylodesmum glutinosum Old Caryophyllaceae Dianthus armeria subsp. armeria Linnaeus 2012 Record on original list as Dianthus armeria Old Papaveraceae Dicentra canadensis (Goldie) Walpers 2012 Old Papaveraceae Dicentra cucullaria (Linnaeus) Bernhardi 2013 continued (Appendix A2) . . . 148 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Poaceae Dichanthelium latifolium (Linnaeus) Harvill 2012 Old Caprifoliaceae Diervilla lonicera Miller 2012 Old Poaceae Digitalis lanata Ehrhart 2012 Old Poaceae Digitaria ischaemum (Schreber) Muhlenberg Not Collected Old Lycopodiaceae Diphasiastrum complanatum (Linnaeus) Holub 2013 Old Thymelaeaceae Dirca palustris Linnaeus 2012 Old Asteraceae Doellingeria umbellata var. umbellata (Miller) Nees 2012 Old Brassicaceae Draba arabisans Michaux Not Collected Old Dryopteridaceae Dryopteris carthusiana (Villars) H.P. Fuchs 2012 Old Dryopteridaceae Dryopteris clintoniana (D.C. Eaton) Dowell 2013 Old Dryopteridaceae Dryopteris cristata (Linnaeus) A. Gray Not Collected Old Dryopteridaceae Dryopteris cristata var. clintoniana (D.C. Eaton) Underwood 2012 - synonym Synonym of Dryopteris clintoniana Old Dryopteridaceae Dryopteris disjuncta (Ruprecht) C.V. Morton Not Collected Synonym of Gymnocarpium disjunctum Old Dryopteridaceae Dryopteris goldiana (Hooker ex Goldie) A. Gray 2012 Old Dryopteridaceae Dryopteris hexagonoptera (Michaux) C. Christensen 2012 - synonym Synonym of Phegopteris hexagonoptera Old Dryopteridaceae Dryopteris intermedia (Muhlenberg ex Willdenow) A Gray 2012 Old Dryopteridaceae Dryopteris marginalis (Linnaeus) A. Gray 2012 Old Dryopteridaceae Dryopteris noveboracensis (Linnaeus) A. Gray 2012 - synonym Synonym of Thelypteris noveboracensis Old Dryopteridaceae Dryopteris phegopteris (Linnaeus) C. Christensen 2012 - synonym Synonym of Phegopteris connectilis Old Dryopteridaceae Dryopteris spinulosa auct. non (OF. Muller) Watt 2012 - synonym Synonym of Dryopteris carthusiana Old Dryopteridaceae Dryopteris spinulosa var. intermedia (Muhlenberg ex Willdenow) Underwood 2012 - synonym Synonym of Dryopteris intermedia Old Dryopteridaceae Dryopteris thelyptris var. pubescens (Lawson) Weatherby 2012 - synonym Synonym of Thelypteris palustris var. pubescens; spelling wrong in Dryopteris thelypteris in original data sheet Old Dryopteridaceae Dryotperis carthusiana 2012 - typographical error Name spelled incorrectly Old Cyperaceae Dulichium arundinaceum (Linnaeus) Britton 2012 Old Poaceae Echinochloa crusgalli (Linnaeus) Palisot de Beauvois Not Collected Should be Echinochloa crus-galli Old Poaceae Echinochloa muricata (P. Beauvois) Fernald 2012 Old Poaceae Echinochloa pungens (Poiret) Rydberg 2012 - synonym Synonym of Echinochloa muricata var. muricata Old Cyperaceae Eleocharis acicularis (Linnaeus) Roemer & Schultes 2012 Old Cyperaceae Eleocharis obtusa (Willdenow) Schultes 2012 Old Cyperaceae Eleocharis obtusa var. jejuna Fernald 2012 - synonym Variety is synonym continued (Appendix A2) . . . 149 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments of Eleocharis obtusa Old Cyperaceae Eleocharis ovata (Roth) Roemer & Schultes 2012 Old Cyperaceae Eleocharis palustris (Linnaeus) Roemer & Schultes 2013 Old Cyperaceae Eleocharis smallii Britton 2012 - synonym Synonym of Eleocharis palustris Old Poaceae Elymus hystrix Linnaeus 2012 Old Poaceae Elymus repens (Linnaeus) Gould 2013 Old Poaceae Elymus trachycaulus (Link) Gould ex Shinners 2013 Old Orobanchaceae Epifagus virginiana (Linnaeus) W.P.C. Barton 2012 Old Orobanchaceae Epifagus virginianus 2012 - typographical error Should be Epifagus virginiana Old Onagraceae Epilobium angustifolium Linnaeus Not Collected Synonym of Chamerion angustifolium subsp. angustifolium Old Onagraceae Epilobium ciliatum subsp. ciliatum Rafinesque 2012 Originally on list as var. ciliatum Epilobium ciliatum, but voucher verified to be this variety Old Onagraceae Epilobium coloratum Biehler Not Collected Old Onagraceae Epilobium glandulosum Lehmann Not Collected Synonym of Epilobium ciliatum subsp. glandulosum Old Onagraceae Epilobium glandulosum var. (Haussknecht) Fernald 2012 - synonym Synonym of adenocaulon Epilobium ciliatum subsp. ciliatum Old Onagraceae Epilobium hirsutum Linnaeus Not Collected Old Orchidaceae Epipactis helleborine (Linnaeus) Crantz 2012 Old Orchidaceae Epipactus helleborine 2012 - typographical error Name spelled incorrectly Old Equisetaceae Equisetum arvense Linnaeus 2012 Old Equisetaceae Equisetum fluviatile Linnaeus 2012 Old Equisetaceae Equisetum hyemale Linnaeus Not Collected Old Equisetaceae Equisetum hyemale subsp. affine (Engelmann) Calder & Roy L. Taylor 2012 Old Equisetaceae Equisetum pratense Ehrhart 2013 Old Equisetaceae Equisetum scirpoides Michaux Not Collected Old Equisetaceae Equisetum sylvaticum Linnaeus 2012 Old Equisetaceae Equisetum variegatum Schleicher ex F. Weber & D. Mohr 2012 Old Asteraceae Erechtites hieracifolia 2012 - typographical error Name spelled incorrectly Old Asteraceae Erechtites hieraciifolius (Linnaeus) Rafinesque ex de Candolle 2012 Old Asteraceae Erigeron annuus (Linnaeus) Persoon 2012 Old Asteraceae Erigeron canadensis Linnaeus 2012 Old Asteraceae Erigeron philadelphicus Linnaeus Not Collected continued (Appendix A2) . . . 150 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Asteraceae Erigeron philadelphicus Linnaeus 2012 Originally on list as var. philadelphicus Erigeron philadelphicus, but voucher keys to this variety Old Asteraceae Erigeron strigosus Muhlenberg ex Willdenow 2012 Old Liliaceae Erythronium americanum On original list as subsp. americanum Ker Gawler 2012 Erythronium americanum Old Asteraceae Eupatorium maculatum Linnaeus 2012 - synonym Synonym of Eutrochium maculatum var. maculatum Old Asteraceae Eupatorium perfoliatum Linnaeus 2012 Old Asteraceae Eupatorium rugosum Houttuyn 2012 - synonym Synonym of Ageratina altissima var. altissima Old Asteraceae Eurybia macrophylla (Linnaeus) Cassini 2012 Old Asteraceae Euthamia graminifolia (Linnaeus) Nuttall 2012 Old Asteraceae Eutrochium maculatum var. (Linnaeus) E.E. Lamont 2013 maculatum Old Fagaceae Fagus grandifolia Ehrhart 2012 Old Polygonaceae Fallopia cilinodis (Michaux) Holub 2012 Old Polygonaceae Fallopia convolvulus (Linnaeus) A. Love 2012 Old Poaceae Festuca obstusa 2012 - synonym Synonym for Festuca subverticillata; spelling error on list - should be Festuca obtusa Old Poaceae Festuca subverticillata (Persoon) E.B. Alexeev 2012 Old Rosaceae Fragaria vesca subsp.americana (Porter) Staudt 2012 On original list as Fragaria vesca Old Rosaceae Fragaria virginiana Miller 2012 Old Oleaceae Fraxinus americana Linnaeus 2012 Old Oleaceae Fraxinus nigra Marshall 2012 Old Lamiaceae Galeopsis tetrahit Linnaeus 2012 Old Asteraceae Galinsoga ciliata (Rafinesque) S.F. Blake Not Collected Synonym of Galinsoga quadriradiata Old Rubiaceae Galium aparine Linnaeus 2012 Old Rubiaceae Galium asprellum Michaux Not Collected Old Rubiaceae Galium boreale Linnaeus 2013 Old Rubiaceae Galium circaezans Michaux Not Collected Old Rubiaceae Galium circaezans var. hypomalacum Fernald Not Collected Synonym of Galium circaezans continued (Appendix A2) . . . 151 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Rubiaceae Galium labradoricum (Wiegand) Wiegand Not Collected Old Rubiaceae Galium lanceolatum Torrey 2013 Old Rubiaceae Galium obtusum Bigelow Not Collected Old Rubiaceae Galium palustre Linnaeus 2013 Old Rubiaceae Galium tinctorium Linnaeus 2013 Old Rubiaceae Galium triflorum Michaux 2013 Old Rubiaceae Galium trilorum 2012 - typographical error Name spelled incorrectly Old Ericaceae Gaultheria procumbens Linnaeus 2013 Old Ericaceae Gaylussacia baccata (Wangenheim) K. Koch 2013 Old Geraniaceae Geranium robertianum Linnaeus 2012 Old Rosaceae Geum aleppicum Jacquin 2013 Old Rosaceae Geum canadense Jacquin 2013 Old Rosaceae Geum rivale Linnaeus Not Collected Old Lamiaceae Glechoma hederacea Linnaeus 2012 Old Lamiaceae Glecoma hederacea 2012 - typographical error Should be Glechoma hederacea Old Poaceae Glyceria canadensis (Michaux) Trinius 2013 Old Poaceae Glyceria melicaria (Michaux) F.T. Hubbard 2012 Old Poaceae Glyceria septentrionalis Hitchcock 2012 Old Poaceae Glyceria striata (Lamarck) Hitchcock 2012 Old Asteraceae Gnaphalium uliginosum Linnaeus 2012 Old Asteraceae Gnaphalium viscosum auct. non Kunth Not Collected Synonym of Pseudognaphalium macounii Old Plantaginaceae Gratiola virginiana Not Collected not in VASCAN database Old Cystopteridaceae Gymnocarpium dryopteris (Linnaeus) Newman 2012 Old Orchidaceae Habenaria hookeri Torrey ex A. Gray Not Collected Synonym of Platanthera hookeri Old Orchidaceae Habenaria hyperborea (Linnaeus) R. Brown 2012 - synonym Synonym of Platanthera hyperborea Old Orchidaceae Habenaria hyperborea var. huronensis (Nuttall) Farwell Not Collected Synonym of Platanthera huronensis Old Orchidaceae Habenaria lacera (Michaux) R. Brown Not Collected Synonym of Platanthera lacera Old Orchidaceae Habenaria macrophylla Goldie Not Collected Synonym of Platanthera macrophylla Old Orchidaceae Habenaria orbiculata (Pursh) Torrey Not Collected Synonym of Platanthera orbiculata Old Orchidaceae Habenaria psychodes (Linnaeus) Lindley Not Collected Synonym of Platanthera psycodes; original list has spelling mistake in Habenaria psychodes Old Orchidaceae Habenaria viridis var. bracteata (Muhlenberg ex Willdenow) A. Gray Not Collected Synonym of Dactylorhiza viridis Old Boraginaceae Hackelia americana (A. Gray) Fernald Not Collected Synonym of Hackelia deflexa subsp. americana continued (Appendix A2) . . . 152 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Hamamelidaceae Hamamelis virginiana Linnaeus 2013 Old Ranunculaceae Hepatica acutiloba de Candolle 2012 - synonym Synonym of Anemone acutiloba Old Ranunculaceae Hepatica americana (de Candolle) Ker Gawler 2012 - synonym Synonym of Anemone americana Old Ranunculaceae Hepatica nobilis var. obtusa (Pursh) Steyemark 2012 - synonym Synonym of Anemone americana Old Brassicaceae Hesperis matronalis Linnaeus 2012 Old Asteraceae Hieracium aurantiacum Linnaeus 2012 Synonym of Pilosella aurantiaca Old Asteraceae Hieracium caespitosum Dumortier 2012 Synonym of Pilosella caespitosa Old Asteraceae Hieracium canadense Michaux Not Collected Synonym of Hieracium umbellatum Old Asteraceae Hieracium florentinum Allioni Not Collected Synonym of Pilosella piloselloides subsp. piloselloides Old Asteraceae Hieracium paniculatum Linnaeus 2012 Old Asteraceae Hieracium piloselloides Villars Not Collected Synonym of Pilosella piloselloides subsp. piloselloides Old Asteraceae Hieracium pratense Tausch 2012 - synonym Synonym of Pilosella caespitosa Old Asteraceae Hieracium scabrum Michaux 2012 Old Diplaziopsidaceae Homalosorus pycnocarpos (Sprengel) Pichi Sermollii 2012 Old Lycopodiaceae Huperzia lucidula (Michaux) Trevisan 2012 Old Araliaceae Hydrocotyle americana Linnaeus 2012 Old Boraginaceae Hydrophyllum virginianum Linnaeus 2012 Old Fabaceae Hylodesmum glutinosum (Muhlenberg ex Willdenow) H. Ohashi & R.R. Mill 2012 Old Hypericaceae Hypericum ellipticum Hooker Not Collected Old Hypericaceae Hypericum majus (A. Gray) Britton Not Collected Old Hypericaceae Hypericum mutilum Linnaeus 2012 Old Hypericaceae Hypericum perfoliatum Not Collected not in VASCAN database Old Hypericaceae Hypericum perforatum Linnaeus 2012 Old Hypericaceae Hypericum virginicum Linnaeus Not Collected Old Poaceae Hystrix patula Moench 2012 - synonym Synonym of Elymus hystrix Old Poaceae Hystrix patula var. bigeloviana (Fernald) Deam 2012 - synonym Synonym of Elymus hystrix Old Aquifoliaceae Ilex verticillata (Linnaeus) A. Gray 2012 Old Balsaminaceae Impatiens capensis Meerburgh 2012 Old Balsaminaceae Impatiens pallida Nuttall 2012 Old Iridaceae Iris versicolor Linnaeus 2012 Old Isoetaceae Isoetes macrospora Durieu Not Collected Synonym of Isoetes lacustris Old Isoetaceae Isoetes riparia Engelmann ex A. Braun Not Collected Old Juglandaceae Juglans cinerea Linnaeus 2012 continued (Appendix A2) . . . 153 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Juncaceae Juncus effusus Linnaeus 2012 Old Juncaceae Juncus effusus var. solutus Fernald & Weigand 2012 - synonym Synonym of Juncus effusus Old Juncaceae Juncus subtilis E. Meyer Not Collected Old Juncaceae Juncus tenuis Willdenow 2012 Old Cupressaceae Juniperus communis Linnaeus Not Collected Old Cupressaceae Juniperus communis var. depressa Pursh Not Collected Unable to verify if specimens previously observed on mountain is this variety Old Ericaceae Kalmia angustifolia Linnaeus Not Collected Old Asteraceae Lactuca biennis (Moench) Fernald 2013 Old Asteraceae Laportea canadensis (Linnaeus) Weddell 2012 Old Fabaceae Lathyrus palustris Linnaeus Not Collected Old Poaceae Leersia oryzoides (Linnaeus) Swartz 2012 Old Poaceae Leersia virginica Willdenow 2012 Old Asteraceae Leontodon autumnalis Linnaeus 2012 - synonym Synonym of Scorzoneroides autumnalis Old Asteraceae Leucanthemum vulgare Lamarck 2012 Old Liliaceae Lilium lancifolium Thunberg 2013 Old Liliaceae Lilium tigrinum ker Gawler 2013 - synonym Synonym of Lilium lancifolium Old Plantaginaceae Linaria vulgaris Miller 2012 Old Plantaginaceae Linaria vulgaris forma leucantha Not Collected not in VASCAN database Old Caprifoliaceae Linnaea borealis var. americana (J. Forbes) Rehder Not Collected Synonym of Linnaea borealis subsp. longiflora Old Amaranthaceae Lipandra polysperma (Linnaeus) S. Fuentes, Uotila & Borsch var. polysperma Old Campanulaceae Lobelia cardinalis Linnaeus 2012 Old Campanulaceae Lobelia inflata Linnaeus 2012 Old Campanulaceae Lobelia spicata Lamarck Not Collected Old Caprifoliaceae Lonicera canadensis Bartram ex Marshall 2013 Old Caprifoliaceae Lonicera dioica Linnaeus 2013 Old Onagraceae Ludwigia palustris (Linnaeus) Elliott 2012 Old Onagraceae Ludwigia palustris var. americana (de Candolle) Fernald & Griscom Not Collected Old Juncaceae Luzula multiflora (Ehrhart) Lejeune Not Collected Old Caryophyllaceae Lychnis alba Miller 2012 - synonym Synonym of Silene latifolia Old Lycopodiaceae Lycopodium annotinum Linnaeus 2013 Old Lycopodiaceae Lycopodium clavatum Linnaeus 2012 Old Lycopodiaceae Lycopodium complanatum Linnaeus 2012 - synonym Synonym of continued (Appendix A2) . . . 154 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Diphasiastrum complanatum Old Lycopodiaceae Lycopodium dendroideum Michaux 2012 Old Lycopodiaceae Lycopodium lucidulum Michaux 2012 - synonym Synonym of Huperzia lucidula Old Lycopodiaceae Lycopodium obscurum Linnaeus 2012 Old Lycopodiaceae Lycopodium obsurum 2012 - synonym Not in VASCAN database; var. dendroideum assume that is Lycopodium dendroideum Old Lycopodiaceae Lycopodium tristachyum Pursh Not Collected Synonym of Diphasiastrum tristachyum Old Lamiaceae Lycopus americanus Muhlenberg ex W.P.C. Barton 2012 Old Lamiaceae Lycopus uniflorus Michaux 2012 Old Primulaceae Lysimachia borealis (Rafinesque) U. Manns & Anderberg 2012 Old Primulaceae Lysimachia ciliata Linnaeus 2012 Old Primulaceae Lysimachia clethroides Duby Not Collected Old Primulaceae Lysimachia hybrida Michaux Not Collected Old Primulaceae Lysimachia terrestris (Linnaeus) Britton, Sterns 2012 & Poggenberg Old Primulaceae Lysimachia thyrsiflora Linnaeus Not Collected Old Lythraceae Lythrum salicaria Linnaeus 2012 Old Asparagaceae Maianthemum canadense Desfontaines 2012 On original list as subsp. canadense Maiantheum canadense; recent voucher is this subspecies Old Asparagaceae Maianthemum racemosum (Linnaeus) Link 2012 On original list as subsp. racemosum Maiantheum racemosum; recent voucher is this subspecies Old Orchidaceae Malaxis unifolia Michaux 2012 Old Rosaceae Malus pumila Miller 2012 Old Malvaceae Malva moschata Linnaeus 2013 Old Malvaceae Malva neglecta Wallroth Not Collected Old Asteraceae Matricaria discoidea de Candolle 2012 Old Asteraceae Matricaria matricarioides auct. non (Lessing) Porter 2012 - synonym Synonym of Matricaria discoidea Old Onocleaceae Matteuccia struthiopteris (Willdenow) C.V. Morton 2012 On original list as var. pensylvanica Matteuccia struthiopteris; variety determined with recent vouchers Old Onocleaceae Matteucia sthruthiopteris Not Collected Old Liliaceae Medeola virginiana Linnaeus 2013 Old Fabaceae Melilotus alba Not Collected Old Fabaceae Melilotus albus Medikus 2012 continued (Appendix A2) . . . 155 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Fabaceae Melilotus officinalis (Linnaeus) Pallas 2012 Old Fabaceae Melilotus officinalis subsp. alba Not Collected Not in VASCAN database Old Lamiaceae Mentha arvensis Linnaeus 2012 Old Menyanthaceae Menyanthes trifoliata Linnaeus 2012 Old Menyanthaceae Menyanthes trifoliata var. minor Fernald Not Collected Old Saxifragaceae Micranthes virginiensis (Michaux) Small 2012 Old Poaceae Milium effusum var. cisatlanticum Fernald 2012 On original list as Milium effusum; voucher verified to be this variety Old Phrymaceae Mimulus ringens Linnaeus 2012 Old Rubiaceae Mitchella repens Linnaeus 2012 Old Rubiaceae Mitchella ripens 2012 - typographical error Spelled incorrectly Old Saxifragaceae Mitella diphylla Linnaeus 2012 Old Saxifragaceae Mitella nuda Linnaeus 2012 Old Molluginaceae Mollugo verticillata Linnaeus Not Collected Old Ericaceae Monotropa hypopithys Not Collected Not in VASCAN database Old Ericaceae Monotropa uniflora Linnaeus 2013 Old Myricaceae Myrica gale Linnaeus 2012 Old Asteraceae Nabalus altissimus (Linnaeus) Hooker 2012 Old Nymphaeaceae Nuphar lutea subsp. variegata (Durand) E.O. Beal 2012 - synonym Synonym for Nuphar variegata Old Nymphaeaceae Nuphar variegata Durand 2012 Old Nymphaeaceae Nuphar variegatum 2012 - typographical error Should be Nuphar variegata Old Asteraceae Oclemena acuminata (Michaux) Greene 2012 Old Onagraceae Oenothera biennis Linnaeus 2012 Old Onagraceae Oenothera perennis Linnaeus 2012 Old Onocleaceae Onoclea sensibilis Linnaeus 2012 Old Orchidaceae Orchis spectabilis (Linnaeus) Rafinesque Not Collected Synonym of Galearis spectabilis Old Poaceae Oryzopsis asperifolia Michaux 2012 Old Poaceae Oryzopsis racemosa (Smith) Ricker ex Hitchcock 2012 - synonym Synonym of Patis racemosa Old Apiaceae Osmorhiza claytoni (Michaux) C.B. Clarke 2012 Old Osmundaceae Osmunda cinnamomea Linnaeus Not Collected Old Osmundaceae Osmunda claytoniana Linnaeus 2012 Old Osmundaceae Osmunda regalis Linnaeus 2012 Old Osmundaceae Osmundastrum cinnamomeum (Linnaeus) C. Presl 2012 Old Betulaceae Ostrya virginiana (Miller) K. Koch 2012 Old Oxalidaceae Oxalis acetosella (Rafinesque) Hulten 2012 - synonym Synonym of Oxalis montana continued (Appendix A2) . . . 156 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Oxalidaceae Oxalis europaea Jordan 2012 - synonym Synonym of Oxalis stricta Old Oxalidaceae Oxalis montana Rafinesque 2012 Old Oxalidaceae Oxalis stricta Linnaeus 2012 Old Araliaceae Panax quinquefolius Linnaeus Not Collected Old Araliaceae Panax trifolius Linnaeus 2012 Old Poaceae Panicum capillare subsp. capillare Linnaeus 2013 Old Poaceae Panicum latifolium Linnaeus 2012 - synonym Synonym of Dichanthelium latifolium Old Poaceae Panicum linearifolium Scribner 2012 - synonym Synonym of Dichanthelium linearifolium Old Poaceae Panicum philadelphicum subsp. philadelphicum Bernhardi ex Trinius 2012 Old Poaceae Panicum tuckermani Fernald 2012 - synonym Synonym of Panicum philadelphicum subsp. philadelphicum Old Thelypteridaceae Parathelypteris noveboracensis (Linnaeus) Ching 2012 - synonym Synonym of Thelypteris noveboracensis Old Vitaceae Parthenocissus quinquefolia (Linnaeus) Planchon ex de Candolle 2013 Old Poaceae Patis racemosa (Smith) Romaschenko, 2012 P.M. Peterson & Soreng Old Orobanchaceae Pedicularis canadensis Linnaeus 2012 Old Penthoraceae Penthorum sedoides Linnaeus 2012 Old Polygonaceae Persicaria hydropiper (Linnaeus) Delarbre 2012 Old Polygonaceae Persicaria lapathifolia (Linnaeus) Delarbre 2013 Old Polygonaceae Persicaria pensylvanica (Linnaeus) M. Gomez de la Maza 2012 Old Polygonaceae Persicaria sagittata (Linnaeus) H. Gross 2012 Old Poaceae Phalaris arundinacea Linnaeus 2012 Old Thelypteridaceae Phegopteris connectilis (Michaux) Watt 2012 Old Thelypteridaceae Phegopteris hexagonoptera (Michaux) Fee 2013 Old Poaceae Phleum pratense Linnaeus 2012 Old Polemoniaceae Phlox subulata Linnaeus Not Collected Old Phrymaceae Phryma leptostachya Linnaeus 2012 Old Pinaceae Picea glauca (Moench) Voss 2012 Old Urticaceae Pilea pumila (Linnaeus) A. Gray 2012 Old Pinaceae Pinus resinosa Aiton 2012 Old Pinaceae Pinus strobus Linnaeus 2012 Old Plantaginaceae Plantago eriopoda Torrey Not Collected Old Plantaginaceae Plantago major Linnaeus 2013 Old Plantaginaceae Plantago rugelii Decaisne 2012 continued (Appendix A2) . . . 157 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Orchidaceae Platanthera hyperborea (Linnaeus) Lindley 2012 Old Poaceae Poa annua Linnaeus 2012 Old Poaceae Poa compressa Linnaeus 2012 Old Poaceae Poa glauca Vahl Not Collected Old Poaceae Poa languida A. Hitchcock 2012 - synonym Synonym of Poa saltuensis subsp. languida Old Poaceae Poa palustris Linnaeus 2012 Old Poaceae Poa pratensis Linnaeus 2012 Old Poaceae Poa saltuensis Fernald & Weigand 2012 Old Asparagaceae Polygonatum pubescens (Willdenow) Pursh 2012 Old Polygonaceae Polygonum arifolium Linnaeus Not Collected Synonym of Persicaria arifolia Old Polygonaceae Polygonum aviculare subsp. aviculare Linnaeus 2012 On original list as Polygonum aviculare; other subspecies on mountain Old Polygonaceae Polygonum cilinode Michaux 2012 - synonym Synonym of Fallopia cilinodis Old Polygonaceae Polygonum convolvulus Linnaeus 2012 - synonym Synonym of Fallopia convolvulus Old Polygonaceae Polygonum hydropiper Linnaeus 2012 - synonym Synonym of Persicaria hydropiper Old Polygonaceae Polygonum lapathifolium Linnaeus 2012 - synonym Synonym of Persicaria lapathifolia Old Polygonaceae Polygonum pensylvanicum Linnaeus 2012 - synonym Synonym of Persicaria pensylvanica Old Polypodiaceae Polypodium appalachianum Haufler & Windham Not Collected Old Polypodiaceae Polypodium virginianum Linnaeus 2012 Old Dryopteridaceae Polystichum acrostichoides (Michaux) Schott 2012 Old Dryopteridaceae Polystichum braunii (Spenner) Fee 2013 Old Pontederiaceae Pontederia cordata Linnaeus Not Collected Old Salicaceae Populus alba Linnaeus Not Collected Old Salicaceae Populus balsamifera Linnaeus 2012 Old Salicaceae Populus deltoides W. Bartram ex Marshall 2012 Old Salicaceae Populus grandidentata Michaux 2012 Old Salicaceae Populus tremula subsp. grandidentata (Michaux) A. Love & D. Love 2012 - synonym Synonym of Populus grandidentata Old Salicaceae Populus tremuloides Michaux 2012 Old Portulacaceae Portulaca oleracea Linnaeus 2012 Old Rosaceae Potentilla anserina Linnaeus 2012 Old Rosaceae Potentilla argentea Linnaeus 2013 Old Rosaceae Potentilla arguta Pursh Not Collected Synonym of Drymocallis arguta Old Rosaceae Potentilla norvegica Linnaeus 2012 Old Rosaceae Potentilla palustris (Linnaeus) Scopoli Not Collected Synonym of Comarum palustre continued (Appendix A2) . . . 158 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Rosaceae Potentilla recta Linnaeus 2012 Old Rosaceae Potentilla tridentata Aiton 2012 - synonym Synonym of Sibbaldia tridentata Old Asteraceae Prenanthes alba Linnaeus Not Collected Synonym of Nabalus albus Old Asteraceae Prenanthes altissima Linnaeus 2012 - synonym Synonym of Nabalus altissimus Old Lamiaceae Prunella vulgaris Linnaeus 2012 Old Lamiaceae Prunella vulgaris forma albiflora Not Collected Not in VASCAN database Old Rosaceae Prunus pensylvanica Linnaeus f. 2012 Old Rosaceae Prunus serotina Ehrhart 2012 Old Rosaceae Prunus virginiana var. virginiana Linnaeus 2012 Listed under Prunus virginiana for original species list; variety determined from recent voucher Old Onocleaceae Pteretis pensylvanica (Willdenow) Fernald 2012 - synonym Synonym of Matteuccia struthiopteris var. pensylvanica Old Dennstaedtiaceae Pteridium aquilinum (Linnaeus) Kuhn Not Collected Old Dennstaedtiaceae Pteridium aquilinum var. latiusculum (Desvaux) Underwood ex A. Heller 2012 Both listed in original species list, so kept both entries in this list Old Ericaceae Pyrola asarifolia Michaux Not Collected Old Ericaceae Pyrola elliptica Nuttall 2013 Old Ericaceae Pyrola secunda Linnaeus Not Collected Synonym of Orthilia secunda Old Ericaceae Pyrola virens Schweigger & Korte Not Collected Synonym of Pyrola chlorantha Old Rosaceae Pyrus melanocarpa (Michaux) Willdenow 2013 - synonym Synonym of Aronia melanocarpa Old Fagaceae Quercus macrocarpa Michaux 2012 Old Fagaceae Quercus rubra Linnaeus 2012 Old Fagaceae Quercus rubra var. borealis (F. Michaux) Farwell 2012 - synonym Synonym of Quercus rubra Old Ranunculaceae Ranunculus abortivus Linnaeus 2012 Old Ranunculaceae Ranunculus acris Linnaeus 2012 Old Ranunculaceae Ranunculus flabellaris Rafinesque Not Collected Old Ranunculaceae Ranunculus flammula var. ovalis (J.M. Bigelow) L.D. Benson 2012 Listed under Ranunculus flammula in original list for site; variety determined with recent voucher Old Ranunculaceae Ranunculus pensylvanicus Linnaeus f. 2013 Old Ranunculaceae Ranunculus recurvatus Poiret 2012 continued (Appendix A2) . . . 159 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Ranunculaceae Ranunculus reptans Linnaeus 2012 - synonym Synonym of Ranunculus flammula var. reptans; is a variety of Ranunculus flammula, which is already listed Old Rhamnaceae Rhamnus alnifolia L’Heritier Not Collected Old Anacardiaceae Rhus radicans Linnaeus 2012 - synonym Synonym of Toxicodendron radicans var. radicans Old Anacardiaceae Rhus typhina Linnaeus 2012 Old Grossulariaceae Ribes americanum Miller Not Collected Old Grossulariaceae Ribes cynosbati Linnaeus 2012 Old Grossulariaceae Ribes glandulosum Grauer 2012 Old Grossulariaceae Ribes lacustre (Persoon) Poiret 2013 Old Grossulariaceae Ribes triste Pallas Not Collected Old Brassicaceae Rorippa islandica (Oeder) Borbas Not Collected Old Brassicaceae Rorippa palustris subsp. hispida (Desvaux) Jonsell 2012 Old Rosaceae Rosa carolina Linnaeus Not Collected Old Rosaceae Rosa nitida Willdenow Not Collected Old Rosaceae Rubus allegheniensis Porter 2012 Old Rosaceae Rubus idaeus Linnaeus 2012 Old Rosaceae Rubus occidentalis Linnaeus 2012 Old Rosaceae Rubus odoratus Linnaeus 2012 Old Rosaceae Rubus pubescens Rafinesque 2012 Old Asteraceae Rudbeckia hirta var. pulcherrima Farwell Not Collected Old Asteraceae Rudbeckia serotina Nuttall 2012 Synonym of Rudbeckia hirta var. pulcherrima Old Polygonaceae Rumex acetosella Linnaeus 2012 Old Polygonaceae Rumex obtusifolius Linnaeus 2012 Old Alismataceae Sagittaria latifolia Willdenow 2012 Old Salicaceae Salix alba Linnaeus Not Collected Old Salicaceae Salix bebbiana Sargent 2012 Old Salicaceae Salix discolor Muhlenberg 2012 Old Salicaceae Salix eriocephala Michaux 2012 Old Salicaceae Salix gracilis Andersson 2012 - synonym Synonym of Salix petiolaris Old Salicaceae Salix humilis var. humilis Marshall 2013 On original site list as Salix humilis; verified variety with recent voucher continued (Appendix A2) . . . 160 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Salicaceae Salix myricoides Muhlenberg Not Collected Old Salicaceae Salix nigra Marshall Not Collected Old Salicaceae Salix petiolaris J.E. Smith 2012 Old Salicaceae Salix rigida Muhlenberg 2012 - synonym Synonym of Salix eriocephala Old Adoxaceae Sambucus canadensis Linnaeus 2012 Old Adoxaceae Sambucus pubens Michaux 2012 - synonym Synonym of Sambucus racemosa subsp. pubens var. pubens Old Adoxaceae Sambucus racemosa Linnaeus 2012 Old Papaveraceae Sanguinaria canadensis Linnaeus 2012 Old Apiaceae Sanicula gregaria E.P. Bicknell 2012 - synonym Synonym of Sanicula odorata Old Apiaceae Sanicula marilandica Linnaeus Not Collected Old Apiaceae Sanicula odorata (Rafinesque) Pryer & Phillippe 2013 Old Apiaceae Sanicula trifoliata E.P. Bicknell 2012 Old Lamiaceae Satureja vulgaris var. neogaea Fernald Not Collected Synonym of Clinopodium vulgare Old Saxifragaceae Saxifraga virginiensis Michaux 2012 - synonym Synonym of Micranthes virginiensis Old Ophioglossaceae Sceptridium dissectum (Sprengel) Lyon Not Collected Old Poaceae Schizachne purpurascens (Torrey) Swallen 2012 Old Cyperaceae Schoenoplectus tabernaemontani (C.C. Gmelin) Palla 2012 Old Cyperaceae Scirpus atrovirens Willdenow 2012 Old Cyperaceae Scirpus cyperinus (Linnaeus) Kunth 2012 Old Cyperaceae Scirpus microcarpus J. Presl & C. Presl 2013 Old Cyperaceae Scirpus validus Vahl 2012 - synonym Synonym of Schoenoplectus tabernaemontani Old Asteraceae Scorzoneroides autumnalis (Linnaeus) Moench 2012 Old Scrophulariaceae Scrophularia lanceolata Pursh 2012 Old Lamiaceae Scutellaria epilobiifolia A. Hamilton Not Collected Synonym of Scutellaria galericulata var. pubescens Old Lamiaceae Scutellaria galericulata Linnaeus 2012 Old Lamiaceae Scutellaria lateriflora Linnaeus 2012 Old Lamiaceae Scutellaria parvula Michaux Not Collected Old Crassulaceae Sedum purpureum (Linnaeus) Schultes Not Collected Synonym of Hylotelephium telephium Old Selaginellaceae Selaginella rupestris (Linnaeus) Spring Not Collected Old Selaginellaceae Selaginella sellowii 2012 Not in VASCAN database Old Asteraceae Senecio viscosus Linnaeus Not Collected Old Poaceae Setaria glauca auct. non (Linnaeus) P. Beauvois 2012 - synonym Synonym of continued (Appendix A2) . . . 161 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Setaria pumila subsp. pumila Old Poaceae Setaria pumila subsp. pumila (Poiret) Roemer & Schultes 2012 On original list as Setaria pumila; subspecies verified with recent voucher Old Poaceae Setaria viridis var. viridis (Linnaeus) Palisot de Beauvois 2012 On original list as Seteria viridis; verified variety with recent voucher Old Rosaceae Sibbaldia tridentata (Aiton) Paule & Sojak 2012 Old Caryophyllaceae Silene antirrhina Linnaeus Not Collected Old Caryophyllaceae Silene cucubalus Wibel nom. Illeg. 2012 - synonym Synonym of Silene vulgaris Old Caryophyllaceae Silene flos-cuculi (Linnaeus) Clairville 2012 Old Caryophyllaceae Silene latifolia Poiret 2012 Old Caryophyllaceae Silene noctiflora Linnaeus Not Collected Old Caryophyllaceae Silene vulgaris (Moench) Garcke 2012 Old Iridaceae Sisyrinchium montanum Greene 2012 Old Apiaceae Sium suave Walter 2012 Old Asparagaceae Smilacina racemosa (Linnaeus) Desfontaines 2012 - synonym Synonym of Maianthemum racemosum subsp. racemosum Old Solanaceae Solanum dulcamara Linnaeus 2012 Old Asteraceae Solidago altissima subsp. altissima Linnaeus 2012 Old Asteraceae Solidago altissima var. altissima Linnaeus 2012 - synonym Synonym of Solidago altissima subsp. altissima Old Asteraceae Solidago caesia Linnaeus 2012 Old Asteraceae Solidago canadensis var. canadensis Linnaeus 2012 On original list as Solidago canadensis; confirmed variety with recent voucher Old Asteraceae Solidago canadensis var. scabra (Muhlenberg ex Willdenow) 2012 - synonym Synonym of Torrey & A. Gray Solidago altissima subsp. altissima Old Asteraceae Solidago flexicaulis Linnaeus 2012 Old Asteraceae Solidago graminifolia (Linnaeus) Salisbury 2012 - synonym Synonym of Euthamia graminifolia Old Asteraceae Solidago hispida Muhlenberg ex Willdenow 2012 Old Asteraceae Solidago juncea Aiton 2012 Old Asteraceae Solidago rugosa Miller 2012 Old Asteraceae Solidago squarrosa Muhlenberg ex Nuttall 2012 Old Asteraceae Solidao flexicaulis 2012 - typographical error continued (Appendix A2) . . . 162 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Asteraceae Sonchus arvensis Linnaeus 2012 Old Asteraceae Sonchus asper (Linnaeus) Hill 2013 Old Asteraceae Sonchus oleraceus Linnaeus Not Collected Old Rosaceae Sorbus decora (Sargent) C.K. Schneider 2012 Old Typhaceae Sparganium androcladum (Engelmann) Morong Not Collected Old Typhaceae Sparganium chlorocarpum Rydberg 2012 - synonym Synonym of Sparganium emersum Old Typhaceae Sparganium emersum Rehmann 2012 Old Typhaceae Sparganium eurycarpum Engelmann 2012 Old Poaceae Sphenopholis intermedia (Rydberg) Rydberg 2012 Old Rosaceae Spiraea alba Du Roi Not Collected Old Rosaceae Spiraea japonica Linnaeus f. 2012 Old Rosaceae Spiraea latifolia (Aiton) Borkhausen 2013 Old Rosaceae Spiraea tomentosa Linnaeus 2012 Old Rosaceae Spiraea tomentosa var. rosea (Rafinesque) Fernald 2012 - synonym Old Orchidaceae Spiranthes cernua (Linnaeus) Richard Not Collected Old Orchidaceae Spiranthes gracilis (Bigelow) L.C. Beck Not Collected Synonym of Spiranthes lacera var. gracilis Old Staphyleaceae Staphylea trifolia Linnaeus Not Collected Old Caryophyllaceae Stellaria graminea Linnaeus 2012 Old Colchicaceae Streptopus amplexifolius (Linnaeus) de Candolle Not Collected Old Colchicaceae Streptopus lanceolatus (Aiton) Reveal 2012 Old Colchicaceae Streptopus roseus Michaux 2012 - synonym Synonym of Streptopus lanceolatus var. lanceolatus Old Caprifoliaceae Symphoricarpos albus (Linnaeus) S.F. Blake Not Collected Old Asteraceae Symphyotrichum ciliolatum (Lindley) A. Love & D. Love 2012 Old Asteraceae Symphyotrichum cordifolium (Linnaeus) G.L. Nesom 2012 Old Asteraceae Symphyotrichum lanceolatum (Willdenow) G.L. Nesom 2012 Old Asteraceae Symphyotrichum lateriflorum (Linnaeus) A. Love & D. Love 2012 Old Asteraceae Symphyotrichum novae-angliae (Linnaeus) G.L. Nesom 2012 Old Asteraceae Symphyotrichum puniceum (Linnaeus) A. Love & D. Love 2012 var. puniceum Old Boraginaceae Symphytum officinale Linnaeus Not Collected Old Asteraceae Tanacetum officinale Not Collected Not in VASCAN database Old Asteraceae Tanacetum vulgare Linnaeus 2012 Old Asteraceae Taraxacum officinale F.H. Wiggers 2012 Old Asteraceae Taraxacum offinale 2012 - typographical error continued (Appendix A2) . . . 163 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Taxaceae Taxus canadensis Marshall Not Collected Old Ranunculaceae Thalictrum dioicum Linnaeus 2012 Old Ranunculaceae Thalictrum polygamum Muhlenberg ex Sprengel 2012 Synonym of Thalictrum pubescens Old Ranunculaceae Thalictrum pubescens Pursh 2012 - synonym Old Thelypteridaceae Thelypteris hexagonoptera (Michaux) Weatherby 2013 - synonym Synonym of Phegopteris hexagonoptera Old Thelypteridaceae Thelypteris noveboracencis 2012 Old Thelypteridaceae Thelypteris palustris Schott 2012 Old Thelypteridaceae Thelypteris phegopteris (Linnaeus) Slosson 2012 - synonym Synonym of Phegopteris connectilis Old Brassicaceae Thlaspi arvense Linnaeus 2012 Old Cupressaceae Thuja occidentalis Linnaeus 2012 Old Saxifragaceae Tiarella cordifolia Linnaeus 2012 Old Malvaceae Tilia americana Linnaeus 2012 Old Anacardiaceae Toxicodendron radicans (Linnaeus) Kuntze 2012 Old Asteraceae Tragopogon pratensis Linnaeus 2012 Old Primulaceae Trientalis borealis Rafinesque 2012 - synonym Synonym of Lysimachia borealis Old Fabaceae Trifolium agrarium Linnaeus nom. rej. 2012 - synonym Synonym of Trifolium aureum Old Fabaceae Trifolium arvense Linnaeus 2012 Old Fabaceae Trifolium aureum Pollich 2012 Old Fabaceae Trifolium campestre Schreber 2012 Old Fabaceae Trifolium hybridum Linnaeus 2012 Old Fabaceae Trifolium pratense Linnaeus 2012 Old Fabaceae Trifolium procumbens Linnaeus nom. rej. 2012 - synonym Synonym of Trifolum campestre Old Fabaceae Trifolium repens Linnaeus 2012 Old Melanthiaceae Trillium erectum Linnaeus 2012 Old Melanthiaceae Trillium erectum forma viridiflorum Not Collected Not in VASCAN database Old Melanthiaceae Trillium grandiflorum (Michaux) Salisbury 2012 Old Melanthiaceae Trillium undulatum Willdenow 2013 Old Caprifoliaceae Triosteum aurantiacum E.P. Bicknell Not Collected Old Pinaceae Tsuga canadensis (Linnaeus) Carriere 2012 Old Asteraceae Tussilago farfara Linnaeus 2012 Old Typhaceae Typha latifolia Linnaeus 2012 Old Ulmaceae Ulmus americana Linnaeus 2012 Old Ulmaceae Ulmus rubra Muhlenberg 2012 Old Colchicaceae Uvularia grandiflora J.E. Smith 2012 Old Colchicaceae Uvularia sessilifolia Linnaeus 2012 Old Ericaceae Vaccinium angustifolium Aiton 2012 continued (Appendix A2) . . . 164 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Ericaceae Vaccinium angustifolium (Michaux) House 2012 - synonym Synonym of var. myrtilloides Vaccinium myrilloides Old Ericaceae Vaccinium corymbosum Linnaeus Not Collected Old Ericaceae Vaccinium myrtilloides Michaux 2012 Old Caprifoliaceae Valeriana officinalis Linnaeus 2012 Old Scrophulariaceae Verbascum thapsus Linnaeus 2012 Old Verbenaceae Verbena hastata Linnaeus 2012 Old Plantaginaceae Veronica americana (Rafinesque) Schweinitz ex Bentham 2012 Old Plantaginaceae Veronica officinalis Linnaeus 2012 Old Plantaginaceae Veronica scutellata Linnaeus 2012 Old Plantaginaceae Veronica serpyllifolia Linnaeus 2012 Old Adoxaceae Viburnum acerifolium Linnaeus 2012 Old Adoxaceae Viburnum alnifolium auct. non Marshall 2012 - synonym Synonym of Viburnum lantanoides Old Adoxaceae Viburnum cassinoides Linnaeus Not Collected Synonym of Viburnum nudum var. cassinoides Old Adoxaceae Viburnum edule (Michaux) Rafinesque Not Collected Old Adoxaceae Viburnum lantanoides Michaux 2012 Old Adoxaceae Viburnum opulus Linnaeus 2012 Old Adoxaceae Viburnum trilobum Marshall 2013 Synonym of Viburnum opulus subsp. trilobum var. americanum Old Fabaceae Vicia cracca Linnaeus 2012 Old Violaceae Viola canadensis Linnaeus 2012 Old Violaceae Viola conspersa Reichenback Not Collected Synonym of Viola labradorica Old Violaceae Viola cucullata Aiton 2012 Old Violaceae Viola incognita Brainard Not Collected Synonym of Viola blanda Old Violaceae Viola macloskeyi F.E. Lloyd 2012 Old Violaceae Viola pallens ex De Candolle 2012 - synonym Synonym of Viola macloskeyi Old Violaceae Viola papilionacea Pursh 2012 - synonym Synonym of Viola sororia var. sororia Old Violaceae Viola pensylvanica Michaux 2012 - synonym Synonym of Viola pubescens var. pubescens Old Violaceae Viola pubescens Aiton 2012 Old Violaceae Viola renifolia A. Gray Not Collected Old Violaceae Viola rostrata Pursh 2013 Old Violaceae Viola selkirkii Pursh ex Goldie 2013 Old Violaceae Viola septentrionalis Greene Not Collected continued (Appendix A2) . . . 165 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Old Violaceae Viola sororia Willdenow 2012 Old Violaceae Viola sororia forma papilionaceae 2012 - synonym No record in VASCAN Old Violaceae Viola sororia forma septentrionalis 2012 - synonym No record in VASCAN Old Vitaceae Vitis riparia Michaux 2012 Old Vitaceae Vitus riparia Not Collected Spelled incorrectly Old Rosaceae Waldsteinia fragarioides (Michaux) Trattinnick Not Collected Synonym of Geum fragarioides Old Woodsiaceae Woodsia ilvensis (Linnaeus) R. Brown 2012 Addition Sapindaceae Acer tataricum subsp. ginnala (Maximowicz) Wesmael 2013 Addition Orobanchaceae Agalinis purpurea var. parviflora (Bentham) B. Boivin 2013 Addition Rosaceae Agrimonia striata Michaux 2013 Addition Rosaceae Amelanchier spicata (Lamarck) K. Koch 2012 Addition Poaceae Anthoxanthum odoratum Linnaeus 2013 Addition Apiaceae Anthriscus sylvestris (Linnaeus) Hoffmann 2012 Addition Rosaceae Aruncus dioicus var. vulgaris (Maximowicz) H. Hara 2012 Addition Aspleniaceae Asplenium trichomanes Linnaeus 2013 Addition Asteraceae Bidens connata Muhlenberg ex Willdenow 2013 Addition Poaceae Bromus inermis Leysser 2013 Addition Cyperaceae Carex lurida Wahlenberg 2012 Addition Cyperaceae Carex pseudocyperus Linnaeus 2013 Addition Plantaginaceae Chaenorhinum minus (Linnaeus) Lange 2012 Addition Asteraceae Cirsium vulgare (Savi) Tenore 2012 Addition Cyperaceae Cyperus diandrus Torrey 2012 Addition Poaceae Dactylis glomerata Linnaeus 2012 Addition Poaceae Danthonia compressa Austin 2012 Addition Caryophyllaceae Dianthus barbatus subsp. barbatus Linnaeus 2013 Addition Poaceae Dichanthelium acuminatum (Scribner) Freckmann & Lelong 2012 subsp. implicatum Addition Poaceae Dichanthelium linearifolium (Scribner) Gould 2012 Addition Poaceae Dichanthelium xanthophysum (A. Gray) Freckmann 2013 Addition Poaceae Digitaria sanguinalis (Linnaeus) Scopoli 2012 Addition Dryopteridaceae Dryopteris campyloptera (Kunze) Clarkson 2013 Addition Dryopteridaceae Dryopteris filix-mas (Linnaeus) Schott 2013 Addition Cyperaceae Eleocharis erythropoda Steudel 2013 Addition Onagraceae Epilobium leptophyllum Rafinesque 2013 Addition Poaceae Eragrostis minor Host 2012 Addition Poaceae Eragrostis pectinacea var. pectinacea (Michaux) Nees 2012 continued (Appendix A2) . . . 166 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Addition Brassicaceae Erysimum cheiranthoides Linnaeus 2012 Addition Celastraceae Euonymus alatus (Thunberg) Siebold 2012 Addition Euphorbiaceae Euphorbia helioscopia Linnaeus 2012 Addition Euphorbiaceae Euphorbia maculata Linnaeus 2012 Addition Euphorbiaceae Euphorbia vermiculata Rafinesque 2012 Addition Polygonaceae Fallopia japonica (Houttuyn) Ronse-Decraene 2012 Addition Poaceae Festuca rubra subsp. rubra Linnaeus 2012 Addition Poaceae Festuca trachyphylla (Hackel) Kajina 2012 Addition Rubiaceae Galium mollugo Linnaeus 2013 Addition Poaceae Glyceria grandis S. Watson 2012 Addition Plantaginaceae Gratiola neglecta Torrey 2012 Addition Lamiaceae Hedeoma pulegioides (Linnaeus) Persoon 2012 Addition Xanthorrhoeaceae Hemerocallis fulva (Linnaeus) Linnaeus 2012 Addition Poaceae Hordeum jubatum Linnaeus 2012 Addition Poaceae Hordeum vulgare Linnaeus 2012 Addition Hydrangeaceae Hydrangea arborescens Linnaeus 2012 Addition Fabaceae Hylodesmum nudiflorum (Linnaeus) H. Ohashi & R.R. Mill 2012 Addition Hypericaceae Hypericum punctatum Lamarck 2012 Addition Hypericaceae Hypericum virginicum Linnaeus 2012 Addition Juncaceae Juncus brevicaudatus (Engelmann) Fernald 2012 Addition Juncaceae Juncus bufonius Linnaeus 2012 Addition Juncaceae Juncus filiformis Linnaeus 2012 Addition Asteraceae Lactuca canadensis Linnaeus 2012 Addition Asteraceae Lapsana communis Linnaeus 2013 Addition Fabaceae Lathyrus latifolius Linnaeus 2012 Addition Lamiaceae Leonurus cardiaca Linnaeus 2013 Addition Brassicaceae Lepidium densiflorum Schrader 2012 Addition Boraginaceae Lithospermum officinale Linnaeus 2013 Addition Caprifoliaceae Lonicera morrowii A. Gray 2012 Addition Caprifoliaceae Lonicera x bella Zabel 2012 Addition Fabaceae Lotus corniculatus Linnaeus 2012 Addition Juncaceae Luzula pallescens Swartz 2012 Addition Primulaceae Lysimachia nummularia Linnaeus 2012 Addition Primulaceae Lysimachia vulgaris Linnaeus 2012 Addition Fabaceae Medicago lupulina Linnaeus 2013 Addition Poaceae Muhlenbergia mexicana (Linnaeus) Trinius 2012 continued (Appendix A2) . . . 167 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Addition Boraginaceae Myosotis laxa Lehmann 2012 Addition Poaceae Panicum dichotomiflorum subsp. Michaux 2012 dichotomiflorum Addition Vitaceae Parthenocissus inserta (A. Kerner) Fritsch 2013 Addition Poaceae Phragmites australis (Cavanilles) Trinius ex Steudel 2012 Addition Solanaceae Physalis heterophylla Nees 2012 Addition Rosaceae Physocarpus opulifolius (Linnaeus) Maximowicz 2012 Addition Phytolaccaceae Phytolacca americana Linnaeus 2012 Addition Pinaceae Picea mariana (Miller) Britton, Sterns 2012 & Poggenburgh Addition Pinaceae Picea rubens Sargent 2012 Addition Plantaginaceae Plantago lanceolata Linnaeus 2012 Addition Orchidaceae Platanthera grandiflora (Bigelow) Lindley 2013 Addition Poaceae Poa nemoralis Linnaeus 2013 Addition Salicaceae Populus nigra Linnaeus 2013 Addition Primulaceae Primula beesiana 2013 Garden escape; not in VASCAN database Addition Fagaceae Quercus alba Linnaeus 2012 Addition Grossulariaceae Ribes alpinum Linnaeus 2013 Addition Grossulariaceae Ribes rubrum Linnaeus 2013 Addition Fabaceae Robinia pseudoacacia Linnaeus 2012 Addition Rosaceae Rosa blanda Aiton 2012 Addition Asteraceae Rudbeckia laciniata Linnaeus 2013 Addition Polygonaceae Rumex crispus Linnaeus 2012 Addition Salicaceae Salix daphnoides Villars 2013 Addition Salicaceae Salix interior Rowlee 2012 Addition Salicaceae Salix serissima (L.H. Bailey) Fernald 2013 Addition Salicaceae Salix x fragilis Linnaeus 2013 Addition Caryophyllaceae Saponaria officinalis Linnaeus 2012 Addition Asteraceae Senecio vulgaris Linnaeus 2012 Addition Caryophyllaceae Silene coeli-rosa 2013 Not in VASCAN database; garden escape Addition Brassicaceae Sinapis arvensis Linnaeus 2012 Addition Solanaceae Solanum americanum P. Miller 2012 Addition Asteraceae Solidago nemoralis subsp. nemoralis Aiton 2012 Addition Poaceae Spartina pectinata Link 2012 continued (Appendix A2) . . . 168 ...continued (Appendix A2) List* Family† Species† Authority† Year Collected Comments Addition Caryophyllaceae Spergularia salina J. Presl & C. Presl 2012 Addition Poaceae Sphenopholis obtusata (Michaux) Scribner 2013 Addition Poaceae Sporobolus vaginiflorus var. vaginiflorus (Torrey ex A.Gray) Alph. Wood 2012 Addition Caryophyllaceae Stellaria pallida (Dumortier) Crepin 2013 Addition Oleaceae Syringa vulgaris Linnaeus 2012 Addition Poaceae Torreyochloa pallida (Torrey) G.L. Church 2012 Addition Asteraceae Tragopogon dubius Scopoli 2012 Addition Asteraceae Tripleurospermum inodorum (Linnaeus) Schultz-Bipontinus 2012 Addition Brassicaceae Turritis glabra Linnaeus 2013 Addition Typhaceae Typha angustifolia Linnaeus 2012 Addition Urticaceae Urtica dioica subsp. gracilis (Aiton) Selander 2013 Addition Plantaginaceae Veronica arvensis Linnaeus 2012 Addition Caprifoliaceae Weigela floribunda C.A. Mey 2013 No record in VASCAN; garden escape Addition Asteraceae Xanthium strumarium Linnaeus 2012

* Old — refers to species on previous list of vascular plant species for the Gault Nature Reserve made prior to 2012; Addition — refers to new species added to the list of vascular plant species for the Gault Nature Reserve during the 2012 and 2013 field seasons † Family, species, and authority names are in accordance with VASCAN (Brouillet et al. 2010) as of April 9, 2014.

References

Bouillet L., Coursol F., Meades S.J., Favreau M., Anions M., Belilse P. & Desmet P. (2010+). VASCAN, the Database of Vascular Plants of Canada. http://data.canadensys.net/vascan/ (consulted on 2014-04-09)

169 Appendix A3: Time and cost estimates for two regional barcoding efforts in Qu´ebec, Canada. The Mont St. Hilaire estimates are based on experiences barcoding that flora during 2012 and 2013, estimates for Schefferville are based on lead author’s research experience with other botanical projects in that region. All price estimates are in Canadian dollars, and all time estimates (hours) assume a team of two workers.

Location Easily accessible region - Mont St. Hilaire Remote region - Schefferville Approximate size of flora† 600 species (unpublished data, Elliott) 381 species (Makinen and Kallio*) Approximate size of study area (ha) 1000 245000

Time estimates Field preparation (preparing lists, 160.00 120.00 reviewing herbarium specimens, buying supplies)‡ Approximate sampling-hours 560.00 650.00 Species verification (0.5 hour per sample) 300.00 190.50 Data organization and submission 300.00 190.50 (0.5 hour per sample) Pressing plants and getting tissue samples 150.00 95.25 (0.25 hour per sample) Processing of plant vouchers 420.00 285.75 (mounting, accessioning, labelling, photos) (0.75 hour per sample) Subtotal: 1890.00 1532.00 Cost estimates Lodging and food (for two people)§ $455.000 $9000.00 Transportation¶ $780.00 $14850.00 Field Equipment** $865.00 $725.00 DNA barcoding†† $30000.00 $21000.00 Shipment of specimens to taxonomic experts§§ $250.00 $200.00 Wages for assistants/professionals ($20 CN/hour)¶¶ $37800.00 $30640.00 Total (Canadian Dollars) $74245.00 $76415.00

*Makinen Y. & Kallio P. (1980). A preliminary checklist of the vascular plants of Schefferville area, Qu´ebec-Labrador. (Predvaritel’nyi, spisok sosudistykh rastenii raiona Sheffervill, poluostrova Kvebek Labrador.) McGill Sub Arctic Research Report, 30, 17-36. † Assumes collection, processing of vouchers and submission of samples for one individual specimen per species ‡ Estimates are approximately one month for flora of 600 species and 3 weeks for smaller flora of 381 species; based on lead author’s experiences § Mont St. Hilaire cost based on $35/day for 65 days per worker for food and lodging at Mont St. Hilaire research chalets; Schefferville cost is $60/day for 75 days for each worker to stay at McGill Subarctic Research Station ¶ Mont St. Hilaire transportation cost based on $60/work week for team of two workers; Schefferville costs includes return plane transportation to site ($3400, including excess baggage per worker) as well as $115 truck rental and gas per day while at the site ** Estimate includes equipment for two plant presses, one field press, one bottle silica gel per 50 samples and $150 incidentals per site †† Based on price of $3000 per batch of 95 samples (price calculated without iBOLD discount); assumes resubmission of three boxes for Mont St. Hilaire samples and two boxes for Schefferville samples to cover first attempt failures §§ Shipment fees vary depending on weight of specimens and destination ¶¶ Estimation does not include compensation for time required by taxonomic experts; we recommend that future studies consider incorporating this fee into their budgets 170 Appendix B

Supporting Information — Chapter 3 Appendix B1: Distribution of 700 focal sedge plots across nine different habitat types near Schefferville, Qu´ebec in the Canadian subarctic. Twenty individuals of each focal species were located across the Schefferville landscape. Species codes are described in Appendix B4, whereas the classification of habitat types follows the descriptions in Waterway et al. (1984).

Species Disturbed Fen Heath Lichen Rocky Shore– Spruce– Tundra Tundra code forest lines moss pond forest

AQU 0 18 1 0 0 1 0 0 0 BIG 0 0 1 0 0 1 0 18 0 LEN 0 3 0 0 1 11 2 1 2 OLI 0 19 0 0 0 1 0 0 0 ROS 0 20 0 0 0 0 0 0 0 SAX 0 12 1 0 0 4 0 1 2 UTR 0 16 0 0 0 4 0 0 0 VES 0 2 0 0 0 18 0 0 0 BRU 2 4 2 1 2 2 4 3 0 CAN 0 15 0 0 1 3 0 1 0 HEL 0 19 0 0 0 0 1 0 0 TEN 0 20 0 0 0 0 0 0 0 TRI 0 15 0 0 0 0 5 0 0 ANG 0 18 1 0 0 0 0 1 1 EVA 0 7 0 2 1 2 5 2 1 VIR 0 20 0 0 0 0 0 0 0 ALP 0 19 0 0 1 0 0 0 0 CAP 0 6 0 0 0 0 0 14 0

continued (Appendix B1) ...... continued (Appendix B1)

Species Disturbed Fen Heath Lichen Rocky Shore– Spruce– Tundra Tundra code forest lines moss pond forest

CES 0 15 0 0 0 1 0 4 0 CHO 0 20 0 0 0 0 0 0 0 DEF 2 0 2 3 3 5 2 3 0 DIS 0 15 0 0 0 1 4 0 0 ECH 0 16 0 1 0 3 0 0 0 EXI 0 20 0 0 0 0 0 0 0 GYN 0 16 0 0 0 1 3 0 0 LEP 0 15 0 0 0 2 3 0 0 LIM 0 19 0 0 0 1 0 0 0 LIV 0 20 0 0 0 0 0 0 0 MAG 0 10 0 1 0 3 4 0 1 PAU 0 19 0 0 0 0 1 0 0 RAR 0 17 0 0 0 0 1 1 1 RUS 0 18 0 0 0 1 0 0 1 SCI 0 2 1 0 0 1 4 12 0 STY 0 9 0 1 0 7 2 0 1 VAG 0 4 2 2 0 1 8 2 1 Total 4 468 11 11 9 74 49 63 11

References

Waterway, M.J., Lechowicz, M.J., & Moore, T.R. (1984). Vegetation of the Schefferville region, Nouveau-Qu´ebec. McGill Subarctic Research Paper, 39, 7–20.

173 Appendix B2: Locating and marking focal sedge plants in the Schefferville area.

Sampling protocol

Focal plant sampling was based on previous knowledge of the habitats around Schefferville and the presence/absence of different sedge species in these different habitats. Focal species were selected using the ‘Ignorant Man’ method, with a maximum of ninety-nine paces between sampling points. ‘Random walks’ were used when the ‘Ignorant Man’ did not yield additional focal species. Sampling priority of focal sedge species at a site was determined by local abundance patterns. Rare species in the Schefferville area were sampled along the ‘Ignorant Man’ path as they were encountered (e.g. Carex heleonastes, Eriophorum vaginatum). Locally common species such as C. magellanica and C. pauciflora were sampled only at the stopping points of ‘Ignorant Man’ walks, whereas a species with low abundance at a site (e.g. C. disperma) was sampled at any point along the ‘Ignorant Man’ walk. Common species such as C. aquatilis, C. limosa, C. trisperma, Trichophorum cespitosum, C. vaginata and C. utriculata were omitted from the selection process during the first round of sampling. For these common focal species, a maximum of one GPS point per site was recorded for each so that twenty focal plants of these species could be randomly selected for marking in the second round of data collection. The sampling area started 0.5 metres in front of where the sampler stopped and extended in a 1.5 metre diameter semi-circle, without changing directions from the ‘Ignorant Man’ walk. The nearest non-trampled area was selected for sampling in cases where potential sampling areas had been previously stepped on. If several different sedge species were found within the two metre diameter semi- circle, the following order of sampling preferences was followed based on local abundance patterns: 1) Rare sedge species were given sampling priority (e.g. C. deflexa or Eriophorum vaginatum) and were sampled at any point along an ‘Ignorant Man’ walk. 2) Medium priority was given to species with more than a few observations at a site. 3) Common species such as C. aquatilis, C. limosa, C. trisperma, Trichophorum cespitosum, C. vaginata and C. utriculata were given low priority. Medium priority species were sampled before low priority species. If more than one species in a priority class was found at the end of an ‘Ignorant Man’ walk, the focal plant species was selected by first assigning numbers to them and then selecting the number at random. To be eligible for sampling, potential plants had to meet the following criteria: – mature plants of the 35 focal species, – were not hybrids, – had reproductive culm(s) on the current year’s growth, – had no evidence of disease or herbivory, – were not trampled by myself during my walk to the sampling spot, and – were more than five metres from other focal plants to avoid trampling. In the case of clonal species (e.g. Carex aquatilis or C. limosa), the focal plant corresponded to an individual reproductive ramet. Focal plants were the entire plant tussock in the case of caespitose species. Upon reaching the end of each ‘Ignorant Man’ walk, we numbered the eligible plants of the focal species within the sampler’s physical reach. A number was drawn from the random number table and called out to select the focal plant. In cases where there were greater than 15 reproductive culms of the focal species in the sampling area, the sampling area was divided into four mini-quadrats and assigned random numbers between one and four to select one of the mini-quadrats. The focal plant was then determined by the sampling protocol outlined earlier in this paragraph.

175 Sampling populations of locally rare species

Many species were represented by very few localized populations at a site (e.g. Eriopho- rum vaginatum, Carex utriculata). However, these localized populations often had a high number of individuals of the species to be sampled. Populations of these species were first identified during either the ‘Ignorant Man’ or random walk. Once a suitable population was identified and roughly delineated, we estimated the approximate size in metres of the diameter of the population. The size of the population then determined the maximum number of random steps the sampler could take during the truncated ‘Ignorant Man’ walk.

Random walks

We would have missed several species at a site by solely following the sampling rules of the ‘Ignorant Man’ walk. For that reason, we followed ‘Ignorant Man’ walks with ‘random walks’ if the stopping point yielded no additional species. Once the sampler was directed to stop at a point that had no new species, we would spend up to five minutes conducting a random walk to look for species that had either not been sampled at that site or had been sampled at that site, but previous data from other researchers indicated that these species were present at a low number of sites in the Schefferville area. We would then either begin a new ‘Ignorant Man’ walk or conduct the sampling protocol for a “small population of locally rare species” at a site if we found a new or relatively rare species during the random walk.

176 Appendix B3: Detailed description of methods for phylogenetic reconstruction.

The phylogeny for the group was reconstructed from gene sequences of four plastid regions (rbcL protein coding region, matK protein coding region, trnL intron, and the trnL-F spacer) and three nuclear ribosomal spacers (ITS1, ITS2, and ETS-1f) commonly used in phylogenetic reconstructions within the Cyperaceae family (Muasya et al. 2009; Starr et al. 2009; Waterway et al. 2009). The phylogenetic analysis included 51 taxa from four genera of Cyperaceae and two species of Juncaceae as an outgroup. Sequences were retrieved from GenBank (Benson et al. 2010) and BOLD (Ratnasingham & Hebert 2007) giving a matrix of 4737 base pairs with 31 missing sequences from 24 different species (see Appendix B5). The sequences were aligned using MAFFT version 7 (Katoh & Standley 2013) and edited manually in BioEdit version 7.0.5.3 (Hall 1999). The optimal model of evolution for each gene region was determined by assessing 88 different models in jMod- elTest 2.1.1 using a fixed BIONJ-JC starting tree and the Akaike information criterion (Darriba et al. 2012). A general time-reversible model estimating rate variation across sites using a gamma distribution (GTR + G) was chosen for ETS-1f, ITS, trnL and matK, whereas the HKY + G model was selected as optimal for rbcL. Two independent calibra- tion points were set in BEAUti v1.8.0 (Drummond et al. 2012), 86 million years for the stem age of the Cyperaceae (Anderson & Janssen 2009) and 61.1 million years for the stem age of Carex (Escudero et al. 2012), assuming a normal distribution for the Cyper- aceae calibration and lognormal distribution for the Carex calibration. Bayesian inference was conducted on a Yule process starting tree with an uncorrelated lognormal relaxed clock model in Beast v1.8.0 (Drummond et al. 2012) using partitioned data for each gene region. Six independent analyses were run for 150,000,000 generations each and conver- gence was visually assessed by examining estimated sample sizes, marginal probability distributions and traces in Tracer (version 1.6, http://tree.bio.ed.ac.uk/software/tracer/). Node posterior probabilities were calculated using the SumTrees package in the DendroPy 3.12.0 phylogenetic computing library (Sukumaran & Holder 2010, see Appendix B6). Appendix B7 shows the age estimations for each node in the phylogeny.

References

Anderson, C.L. & Janssen, T. (2009). Monocots. In: The Timetree of Life. (eds. Hedges, S.B & Kumar, S.). Oxford University Press, Oxford, UK, pp. 203–212. Benson, D.A., Karsch-Mizrachi, I., Lipman, D.J., Ostell, J., & Sayers, E.W. (2010). GenBank. Nucleic Acids Research, 38, D46–D51. Darriba, D., Taboada, G.L., Doallo, R., & Posada, D. (2012). jModelTest 2: more models, new heuristics and parallel computing. Nature Methods, 9, 772–772. Drummond, A.J., Suchard, M.A., Xie, D., & Rambaut, A. (2012). Bayesian phylogenetics with BEAUti and the BEAST 1.7. Molecular Biology and Evolution, 29, 1969–1973. Escudero, M., Hipp, A.L., Waterway, M.J., & Valente, L.M. (2012). Diversification rates and chromosome evolution in the most diverse angiosperm genus of the temperate zone (Carex, Cyperaceae). Molecular Phylogenetics and Evolution, 63, 650–655. Hall, T.A. (1999). BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series, 41, 95– 98. Katoh, K. & Standley, D.M. (2013). MAFFT Multiple Sequence Alignment Software Version 7: Improvements in Performance and Usability. Molecular Biology and Evolution, 30, 772–780. Muasya, A. M., Simpson, D. A., Verboom, G. A., Goetghebeur, P., Naczi, R. F. C., Chase, M. W., & Smets, E. (2009). Phylogeny of Cyperaceae based on DNA sequence data: current progress and future prospects. The Botanical Review, 75, 2–21. Ratnasingham, S. & Hebert, P.D.N. (2007). BOLD: The Barcode of Life Data System

178 (http://www.barcodinglife.org). Molecular Ecology Notes, 7, 355–364. Starr, J.R., Naczi, R.F.C., & Chouinard, B.N. (2009). Plant DNA barcodes and species resolution in sedges (Carex, Cyperaceae). Molecular Ecology Resources, 9, 151–163. Sukumaran, J. & Holder, M.T. (2010). DendroPy: a Python library for phylogenetic computing. Bioinformatics, 26, 1569–1571. Waterway, M.J., Hoshino, T., & Masaki, T. (2009). Phylogeny, species richness, and ecological specialization in Cyperaceae tribe Cariceae. The Botanical Review, 75, 138–159.

179 Appendix B4: Regional species pool of Cyperaceae from the subarctic Qu´ebec and adjacent Labrador region of Canada. Names and authorities are in accordance with VASCAN. Species, Genus, and Clade Codes refer to the clade to which each species was assigned for different analyses. The number of plots for each focal species included in Cyperaceae analyses are listed under # plots Cyp, whereas # plots Carex gives the number of plots for each focal species included in the Carex-only analyses.

Species Authority Species Genus # plots Clade # plots Code Code Cyp Code Carex Carex aquatilis Wahlenb. AQU CAR 17 CC 17 Carex arcta Boott ARC— ——— Carex atratiformis Britton ATR — — — — Carex bigelowii Torr. ex Schwein BIG CAR 12 CC 9 Carex brunnescens (Pers.) Poir. BRU CAR 12 VIG 12 Carex buxbaumii Wahlenb. BUX — — — — Carex canescens L. CAN CAR 16 VIG 16 Carex capillaris L. CAP CAR 17 CC 17 Carex capitata Sol. CAT — — — — Carex chordorrhiza L.f. CHO CAR 20 VIG 20 Carex deflexa Hornem. DEF CAR 4 CC 4 Carex diandra Schrank DIA — — — — Carex disperma Dewey DIS CAR 20 VIG 20 Carex echinata subsp. echinata Murray ECH CAR 20 VIG 20 Carex exilis Dewey EXI CAR 20 VIG 19 Carex foenea Willd. FOE — — — — Carex garberi Fernald GAR — — — — Carex glacialis Mack. GLA— ——— continued (Appendix B4) . . . 180 . . . continued (Appendix B4) Species Authority Species Genus # plots Clade # plots Code Code Cyp Code Carex Carex gynocrates Wormsk. GYN CAR 20 VIG 20 Carex heleonastes Ehrh. ex L.f. HEL CAR 20 VIG 20 Carex lachenalii Schkuhr LAC — — — — Carex lenticularis Michx. LEN CAR 15 CC 15 Carex leptalea Wahlenb. LEP CAR 20 CRC 20 Carex limosa L. LIM CAR 20 CC 17 Carex livida (Wahlenb.) Willd. LIV CAR 20 CC 18 Carex magellanica subsp. irrigua (Wahlenb.) Hiitonen MAG CAR 16 CC 16 Carex media R.Br. MED — — — — Carex nardina (Hornem.) Fr. NAR — — — — Carex oligosperma Michx. OLI CAR 16 CC 13 Carex pauciflora Lightf. PAU CAR 18 CRC 18 Carex rariflora (Wahlenb.) Sm. RAR CAR 19 CC 17 Carex rostrata Stokes ROS CAR 19 CC 18 Carex saxatilis L. SAX CAR 18 CC 17 Carex scirpoidea Michx. SCI CAR 16 CC 16 Carex stylosa C.A. Mey STY CAR 19 CC 18 Carex tenuiflora Wahlenb. TEN CAR 20 VIG 20 Carex trisperma Dewey TRI CAR 17 VIG 17 Carex utriculata Boott UTR CAR 17 CC 17 Carex vaginata Tausch VAG CAR 11 CC 11 Carex vesicaria L. VES CAR 17 CC 17 continued (Appendix B4) . . . 181 . . . continued (Appendix B4) Species Authority Species Genus # plots Clade # plots Code Code Cyp Code Carex Carex viridula Michx. CVI — — — — Carex williamsii Britton WIL — — — — Eleocharis acicularis (L.) Roem. & Schult. ACI — — — — Eleocharis nitida Fernald NIT— ——— Eriophorum angustifolium Honck. ANG ERI 21 — — Eriophorum brachyantherum Trautv. & C.A. Mey BRA — — — — Eriophorum chamissonis C.A. Meyer CHA ERI 20 — — Eriophorum vaginatum L. EVA ERI 16 — — Eriophorum viridicarinatum (Engelm.) Fernald VIR ERI 20 — — Trichophorum alpinum (L.) Pers. ALP TRI 20 — — Trichophorum cespitosum (L.) Hartm. CES TRI 20 — — Juncus arcticus Willd. JARC Outgroup — — — Juncus trifidus L. JTRI Outgroup — — —

182 Appendix B5: Sequence data used to create a regional phylogeny of sedges found near Schefferville, in subarctic Qu´ebec. Missing data from the sequences matrix are represented by ‘NA’ values.

Sequence Species ETS-1f ITS matK rbcL trnL intron; trnL–trnF intergenic spacer; trnF gene

Carex aquatilis Accession number AY757651.1 JN999014 CAREX087-08 JN965331.1 AY757517.1 Source GenBank GenBank BOLD GenBank GenBank Carex arcta Accession number JN903107.1 JN903131.1 KP980054 NA NA Source GenBank GenBank GenBank NA NA Carex atratiformis Accession number KR902926 KR902921 KR902908 NA KR902939 Source GenBank GenBank GenBank NA GenBank Carex bigelowii Accession number AY770438.1 AY278303 CAREX048-08 MLTVP024-11 GQ244699.1 Source GenBank GenBank BOLD BOLD GenBank Carex brunnescens Accession number DQ115115.1 EU541872 KSR318-07 NA AY757481.1 Source GenBank GenBank BOLD NA GenBank Carex buxbaumii Accession number KR902927 EU288545 KR902907 BBYUK2109-12 EU288428 Source GenBank GenBank GenBank BOLD GenBank Carex canescens Accession number DQ115117.1 JX644817 JN895021.1 GBVD236-11 AY757480.1 Source GenBank GenBank GenBank BOLD GenBank Carex capillaris Accession number NA JN999026.1 CAREX052-08.matK MKPCH414-09 DQ998958.1 Source NA GenBank BOLD BOLD GenBank Carex capitata Accession number DQ115119.1 DQ115118.1 PLCHU246-09 PLCHU245-09 GU176109.1 Source GenBank GenBank BOLD BOLD GenBank Carex chordorrhiza Accession number DQ115125.1 DQ115126 MKPCH525-09 JN965354 AY757485.1 Source GenBank GenBank BOLD GenBank GenBank continued (Appendix B5) . . . 183 . . . continued (Appendix B5) Sequence Species ETS-1f ITS matK rbcL trnL intron; trnL–trnF intergenic spacer; trnF gene Carex deflexa Accession number KR902928 KR902922 KR902909 NA KR902940 Source GenBank GenBank GenBank NA GenBank Carex diandra Accession number DQ115145.1 EU000972 POWMA116-10 JN893397.1 AY757473.1 Source GenBank GenBank BOLD GenBank GenBank Carex disperma Accession number DQ115149.1 EU000976 FCA2003-11 KR902904 AY757489.1 Source GenBank GenBank BOLD GenBank GenBank Carex echinata Accession number DQ115161.1 DQ461104 POWNA2439-12 JN892572.1 AY757477.1 subsp. echinata Source GenBank GenBank BOLD GenBank GenBank Carex exilis Accession number DQ115169 DQ115168 KR902912 NA AY757478.1 Source GenBank GenBank GenBank NA GenBank Carex foenea Accession number DQ460975.1 AY779100.1 GBVD536-11 NA NA Source GenBank GenBank BOLD NA NA Carex garberi Accession number KR902929 NA KC474320.1 JN965360.1 KR902941 Source GenBank NA GenBank GenBank GenBank Carex glacialis Accession number AY757685 AY757625 FJ548096.1 CAN589963 AY757553.1 Source GenBank GenBank GenBank BOLD GenBank Carex gynocrates Accession number DQ115177.1 DQ115176.1 KC474326.1 JN965375.1 AY757479.1 Source GenBank GenBank GenBank GenBank GenBank Carex heleonastes Accession number AY757388.1 AY757418.1 KP980092 NA AY757484.1 Source GenBank GenBank GenBank NA GenBank Carex lachenalii Accession number NA EU541869.1 CAREX025-08 CAN579671 EU288440.1 Source NA GenBank BOLD BOLD GenBank Carex lenticularis Accession number AY770447 AY770475 KR902911 NA KR902942 Source GenBank GenBank GenBank NA GenBank continued (Appendix B5) . . . 184 . . . continued (Appendix B5) Sequence Species ETS-1f ITS matK rbcL trnL intron; trnL–trnF intergenic spacer; trnF gene Carex leptalea Accession number AY757690.1 AY757630 KR902913 NA AY757559.1 Source GenBank GenBank GenBank NA GenBank Carex limosa Accession number AY757656 JX644831.1 POWNA1375-12 JN891210.1 AY757522.1 Source GenBank GenBank BOLD GenBank GenBank Carex livida Accession number AY757688 AY757628 KC474343.1 JN965380.1 AY757556.1 Source GenBank GenBank GenBank GenBank GenBank Carex magellanica Accession number AY757655.1 FJ694731 POWNA131-10 JN965395 AF284917.1 subsp. irrigua Source GenBank GenBank BOLD GenBank GenBank Carex media Accession number NA DQ998932 MKPCH5415-09 JN965388.1 DQ998985.1 Source NA GenBank BOLD GenBank GenBank Carex nardina Accession number DQ115221.1 AY241973 FJ548124.1 CAN589952 GU176114.1 Source GenBank GenBank GenBank BOLD GenBank Carex oligosperma Accession number AY757640 AY757578.1 KR902910 NA AY757505.1 Source GenBank GenBank GenBank NA GenBank Carex pauciflora Accession number JN893827.1 AY757631 POWNA139-10 JN893827.1 AY757569.1 Source GenBank GenBank BOLD GenBank GenBank Carex rariflora Accession number KR902930 AY278305 FJ548129.1 CAN586948 KR902943 Source GenBank GenBank GenBank BOLD GenBank Carex rostrata Accession number KR902931 EU288564 POWNA3130-12 JN893765.1 EU288448.1 Source GenBank GenBank BOLD GenBank GenBank Carex saxatilis Accession number KR902932 KR902919 FJ548134.1 CAN583275 DQ860539.1 Source GenBank GenBank GenBank BOLD GenBank Carex scirpoidea Accession number AY241991 AY757582 FJ548137.1 CAN589957 AY757509.1 Source GenBank GenBank GenBank BOLD GenBank continued (Appendix B5) . . . 185 . . . continued (Appendix B5) Sequence Species ETS-1f ITS matK rbcL trnL intron; trnL–trnF intergenic spacer; trnF gene Carex stylosa Accession number AY757652 AY757591.1 NA KR902906 AY757518.1 Source GenBank GenBank NA GenBank GenBank Carex tenuiflora Accession number JN903126 JX644848.1 KP980009 JX644649 AY757482.1 Source GenBank GenBank GenBank GenBank GenBank Carex trisperma Accession number DQ115299 AY757429 KP980079 NA AY757483.1 Source GenBank GenBank GenBank NA GenBank Carex utriculata Accession number KR902933 JN999091 JN966210.1 NA KR902944 Source GenBank GenBank GenBank NA GenBank Carex vaginata Accession number AY757689 AY757629 KC575510.1 JN965414 AY757557.1 Source GenBank GenBank GenBank GenBank GenBank Carex vesicaria Accession number KR902934 AY278289 JN314505 JN893673.1 KR902945 Source GenBank GenBank GenBank GenBank GenBank Carex viridula Accession number AY757658 JN634666 POWNA2309-12 JN893603.1 AY757524.1 Source GenBank GenBank BOLD GenBank GenBank Carex williamsii Accession number KR902935 KR902920 FJ548144.1 JN965422.1 GQ244759.1 Source GenBank GenBank GenBank GenBank GenBank Eleocharis Accession number NA GU110751 POWNA231-10 JN892607 JX644718.1 acicularis Source NA GenBank BOLD GenBank GenBank Eleocharis nitida Accession number NA GU110848.1 NA NA NA Source NA GenBank NA NA NA Eriophorum Accession number NA DQ998950 POWNA1957-12 JN892176 DQ999003.1 angustifolium Source NA GenBank BOLD GenBank GenBank Eriophorum Accession number KR902936 KR902924 KR902916 KC482800.1 GQ244945.1 brachyantherum Source GenBank GenBank GenBank GenBank GenBank continued (Appendix B5) . . . 186 . . . continued (Appendix B5) Sequence Species ETS-1f ITS matK rbcL trnL intron; trnL–trnF intergenic spacer; trnF gene Eriophorum Accession number KR902937 KR902923 KR902914 KR902905 KR902946 chamissonis Source GenBank GenBank GenBank GenBank GenBank Eriophorum Accession number AY242009 JX566737.1 POWNA1758-12 JN892385.1 AY757692.1 vaginatum Source GenBank GenBank BOLD GenBank GenBank Eriophorum Accession number KR902938 KR902925 KR902915 U49230 KR902947 viridicarinatum Source GenBank GenBank GenBank GenBank GenBank Trichophorum Accession number AY757400 JF313184 KR902917 JF313189 AY757496.1 alpinum Source GenBank GenBank GenBank GenBank GenBank Trichophorum Accession number NA DQ998951 POWNA3148-12 MLTVP003-11 DQ999004.1 cespitosum Source NA GenBank BOLD BOLD GenBank Juncus arcticus Accession number NA JN999264 JN966338.1 FCA083-09 AY437972.1 Source NA GenBank GenBank GenBank GenBank Juncus trifidus Accession number NA AY973508.1 GBVD2391-11 FCA1489-11 AY437971.1 Source NA GenBank BOLD GenBank GenBank

187 Appendix B6: Phylogeny of the Cyperaceae species of the subarctic Qu´ebec and adjacent Labrador region of Canada based on bayesian analysis showing bayesian node posterior probabilities. Species names are in accordance with VASCAN. Appendix B7: Phylogeny of the Cyperaceae species of the subarctic Qu´ebec and adjacent Labrador region of Canada based on bayesian analysis with blue node bars indicating estimated divergence times in millions of years. Species names are in accordance with VASCAN. Appendix B8: Mean net relatedness index to focal plant (FNRI) for the 35 Cyperaceae and 29 Carex focal species. The species included in each analysis are listed in Appendix B4.

Metric FNRI FNRI Cyperaceae† Carex Cyperaceae† Carex Species Code

AQU 0.073 0.371 0.267 0.428 BIG -0.079 0.421 -0.086 0.438 BRU 0.496 0.421 0.487 0.427 CAN 0.092 -0.179 0.184 -0.446 CAP 0.661 0.771 0.612 0.844 CHO -0.645 -0.721 -0.218 -0.72) DEF 0.612 0.833 0.276 0.832 DIS 0.128 -0.242 0.288 -0.329 ECH -0.141 -0.052 -0.394 -0.29) EXI -1.145 -0.720 -1.331 -0.636 GYN -0.460 -0.654 0.124 -0.507 HEL -0.787 -0.590 -0.334 -0.671 LEN 0.427 0.400 0.439 0.317 LEP 0.540 -0.043 0.522 0.419 LIM -0.341 0.837 -0.365 0.919 LIV -0.385 0.678 -0.407 0.877 MAG 0.002 -0.055 0.249 0.298 OLI -0.210 0.422 -0.530 0.494 PAU 0.094 -0.076 0.079 0.321 RAR 0.127 0.826 0.320 1.063

continued (Appendix B8) ...... continued (Appendix B8)

Metric FNRI FNRI Cyperaceae† Carex Cyperaceae† Carex Species Code

ROS -0.139 0.596 0.255 0.783 SAX -0.206 0.448 -0.157 0.594 SCI 0.654 0.637 0.424 0.863 STY 0.032 0.025 0.016 0.100 TEN -0.287 -0.575 0.069 -0.695 TRI 0.001 -0.375 0.173 -0.579 UTR 0.259 -0.091 0.259 -0.011 VAG 0.478 -0.087 0.560 0.178 VES -0.001 -0.248 0.501 -0.033 ALP 1.511 2.369 CES -0.114 -0.124 ANG -0.137 -0.324 RUS 0.106 -0.104 EVA -0.354 -0.351 VIR 0.197 0.423

weighted by percent cover estimates † this analysis includes all co-occurring Cyperaceae species

191 Appendix B9: Mean nearest taxon index to focal plant (FNTI) for the 35 Cyperaceae and 29 Carex focal species. The species included in each analysis are listed in Appendix B4.

Metric FNTI FNTI Cyperaceae† Carex Cyperaceae† Carex Species Code

AQU 0.315 0.345 0.198 0.261 BIG -0.015 0.617 0.009 0.450 BRU 0.374 0.233 0.363 0.124 CAN 0.229 0.003 0.203 0.077 CAP 0.533 0.523 0.524 0.528 CHO -0.646 -0.579 -0.463 -0.304 DEF 0.425 0.270 0.451 0.405 DIS 0.119 -0.066 0.366 0.210 ECH 0.091 0.114 0.158 0.107 EXI -0.929 -0.723 0.169 -0.543 GYN -0.574 -0.667 -0.724 -0.650 HEL -0.601 -0.479 -0.466 -0.284 LEN 0.770 0.692 0.706 0.652 LEP 0.143 0.006 -0.131 -0.128 LIM 0.290 0.946 0.207 0.657 LIV -0.036 0.391 -0.375 -0.073 MAG 0.085 0.012 -0.172 -0.252 OLI 0.136 0.624 0.114 0.480 PAU 0.039 -0.061 -0.048 -0.008 RAR 0.316 0.831 0.379 0.635

continued (Appendix B9) ...... continued (Appendix B9)

Metric FNTI FNTI Cyperaceae† Carex Cyperaceae† Carex Species Code

ROS 0.230 0.523 -0.009 0.273 SAX 0.162 0.414 0.210 0.439 SCI 0.418 0.362 0.421 0.439 STY 0.262 0.330 0.318 0.179 TEN -0.375 -0.429 -0.198 -0.206 TRI -0.144 -0.320 -0.071 -0.259 UTR 0.134 -0.036 0.046 -0.331 VAG 0.368 0.239 0.111 -0.089 VES 0.245 0.086 0.015 -0.113 ALP 0.883 0.207 CES -0.196 0.044 ANG 0.133 -0.022 RUS 0.055 0.062 EVA -0.320 -0.365 VIR 0.188 -0.011

weighted by percent cover estimates † this analysis includes all co-occurring Cyperaceae species

193 Appendix B10: Average species richness per focal species and sampling frequency of the 35 focal sedge species included in the analyses. Average species richness was calculated based on the 20 focal plots per species. Species, Genus, and Carex clade codes are the same as those described in Appendix B4.

Species Genus Carex Average Average Sampling frequency code code clade species richness species richness across all plots code (all Cyperaceae/plot) (all Carex/plot)† AQU CAR CC 3.1 2.65 218 BIG CAR CC 1.7 1.45 51 BRU CAR VIG 1.8 1.75 28 CAN CAR VIG 3.1 2.95 74 CAP CAR CC 2.9 2.60 27 CHO CAR VIG 5.4 4.50 46 DEF CAR CC 1.4 1.40 23 DIS CAR VIG 3.6 3.55 31 ECH CAR VIG 4.4 3.75 56 EXI CAR VIG 3.4 3.15 29 GYN CAR VIG 3.6 3.10 77 HEL CAR VIG 4.8 3.75 26 LEN CAR CC 2.6 2.45 30 LEP CAR CRC 4.0 3.80 79 LIM CAR CC 3.3 2.35 167 LIV CAR CC 4.0 2.95 77 MAG CAR CC 2.8 2.50 202 OLI CAR CC 3.1 2.55 40 PAU CAR CRC 3.2 2.75 75 continued (Appendix B10) . . . 194 . . . continued (Appendix B10) Species Genus Carex Average* Average Sampling frequency code code clade species richness species richness across all plots code (all Cyperaceae/plot) (all Carex/plot) RAR CAR CC 3.7 3.00 60 ROS CAR CC 3.4 2.65 66 SAX CAR CC 3.6 2.75 33 SCI CAR CC 2.8 2.60 41 STY CAR CC 2.7 2.40 30 TEN CAR VIG 4.8 4.35 62 TRI CAR VIG 2.8 2.75 60 UTR CAR CC 3.6 3.40 66 VAG CAR CC 2 1.95 145 VES CAR CC 2.8 2.60 40 ALP TRI — 5.0 76 CES TRI — 3.9 190 ANG ERI — 4.0 57 RUS ERI — 3.1 62 EVA ERI — 3.8 38 VIR ERI — 5.7 38

Calculation based on the number of Cyperaceae species present within the plots of the 35 different Cyperaceae focal species †Calculation based on the number of Carex species present within the plots of the 29 different Carex focal species

195 Appendix B11: Mean net related index to focal (FNRI) and nearest taxon index to focal (FNTI) compared to average species richness.

Table B11–1 Mean net related index to focal (FNRI) and nearest taxon index to fo- cal (FNTI) compared to average species richness in the plots of each focal species. The Cyperaceae analyses include the 35 Cyperaceae focal species and co-occurring Cyperaceae plants, whereas the Carex set of analysis were restricted to the 29 Carex focal species and only co-occurring Carex. Species richness values per each focal species are listed in Appendix B10.

Taxa included Parameter FRNI FNRI FNTI FNTI

Cyperaceae df 33 33 33 33 Slope -0.60 -0.67 -0.55 -0.68 2 R A 0.05 0.06 0.06 0.01 P value 0.11 0.09 0.09 0.26

Carex df 27 27 27 27 Slope -0.41 -0.44 -0.34 -0.23 2 R A 0.35 0.32 0.32 0.20 P value < 0.01 < 0.01 < 0.01 < 0.01

Weighted by percent cover Table B11–2 Mean net related index to focal (FNRI) and nearest taxon index to focal (FNTI) compared to sampling frequency of each focal species across all 700 plots. The Cyperaceae analyses include the 35 Cyperaceae focal species and co-occurring Cyperaceae plants, whereas the Carex set of analysis were restricted to the 29 Carex focal species and co-occurring Carex. The sampling frequency of each focal species is listed in Appendix B10.

Taxa included Parameter FRNI FNRI FNTI FNTI

Cyperaceae df 33 33 33 33 Slope 0.00 0.00 0.01 0.00 2 R A 0.02 0.00 0.25 0.19 P value 0.19 0.33 < 0.01 < 0.01

Carex df 27 27 27 27 Slope 0.00 0.00 0.00 0.00 2 R A -0.01 0.03 0.00 -0.03 P value 0.42 0.19 0.31 0.68

Weighted by percent cover

197 Figure B11–1 Relationship between species richness and two community phylogenetics metrics to focal for sedges found near Schefferville, Qu´ebec. The results for 35 different focal Cyperaceae species are shown in (a) and (b), whereas the results for 29 focal Carex species are indicated in (c) and (d). Species richness was calculated as the average number of co-occurring Cyperaceae and Carex species found with each of the 35 and 29 different focal species, respectively. Significance levels < 0.05 are indicated with a regression line.

198 Figure B11–2 Relationship between sampling frequency across 700 plots and two com- munity phylogenetics metrics to focal for sedges found near Schefferville, Qu´ebec. The results for 35 different focal Cyperaceae species are shown in (a) and (b), whereas the results for 29 focal Carex species are indicated in (c) and (d). Sampling frequency was cal- culated as the total number of Cyperaceae and Carex plots that each focal sedge species was sampled, respectively. Significance levels < 0.05 are indicated with a regression line.

199 Figure B11–3 Net relatedness index to focal for three different Carex focal clades, cal- culated based on the ‘trialswap’ algorithm that randomizes the Carex community data matrix, while maintaining species occurrence frequency and richness at each site. Signif- icant differences among clades (Tukey-Kramer multiple comparison of means, P < 0.05) are indicated by different letters. Medians for each plot are represented by thick lines, the boundaries of each box show the 25th and 75th percentiles and whiskers above and below each plot represent the 10th and 90th percentiles. Outlying data is indicated by hollow circles and abbreviations for each clade are the same as those in Appendix B4. The species included in each analysis are listed in Appendix B4.

200 Appendix B12: Net relatedness to focal (FNRI) comparisons among different habitat types.

Net relatedness index to focal (FNRI) values for Cyperaceae focal plants sampled in the tundra were significantly greater than zero, indicating a trend towards phylogenetic clustering in this habitat (Table B12–2; Fig. B12–1); however, FNRI values did not differ when compared between habitats (Table B12–1). On the other hand, FNRI values for the other three habitats did not differ significantly from zero (Table B12–2; Fig. B12–1). Further dividing focal plants into major clades revealed more nuanced patterns, with significant clade-level patterns in both the Cyperaceae and Carex within fens (Table B12–3). Net relatedness to focal values were greater than zero within the Trichophorum in fens, suggesting a trend towards phylogenetic clustering in that genus (Table B12–4; Fig. B12–2). In contrast, FNRI values were significantly less than zero in Carex and in particular, the Vignea clade, within fens (Table B12–4; Fig. B12–2). Within shoreline and spruce moss forest habitats, Eriophorum had FNRI values that were significantly less than zero, suggesting a trend towards phylogenetic overdispersion (Table B12–4; Fig. B12–2). In contrast, Carex in the Caricoid clade had FNRI that were significantly greater than zero within spruce-moss forests (Table B12–4; Fig B12–2). Within the tundra, FNRI values in both the Carex and Core Carex clade were significantly greater than zero, showing a trend towards phylogenetic clustering, whereas FNRI values in the Trichophorum were significantly less than zero (Table B12–4; Fig. B12–2). Contrasting trends towards phylogenetic clustering and overdispersion were evident in fens, with FNRI values significantly higher in the Trichophorum compared to the Eriophorum and Carex (Table B12–5; Fig. B12–2), whereas Carex in the Vignea clade had lower FNRI values than observed in both the Core Carex and Caricoid clades (Table B12–5; Fig B12–2). Table B12–1 Differences in net relatedness index to focal index (FNRI) for Cyperaceae focal plants compared between four different habitats. The difference in the mean values between clades is represented by ‘Difference’ with significance levels from the Tukey- Kramer multiple comparisons of means test. All of the pairwise habitat comparisons were non-significant. Habitats with less than 10 focal plants were omitted from the analysis.

Comparison Difference t value P value Fen — Shoreline 0.209 1.589 0.370 Fen — Spruce-moss forest 0.247 1.590 0.365 Fen — Tundra 0.326 2.161 0.128 Shoreline — Spruce-moss forest 0.038 0.195 0.997 Shoreline — Tundra 0.117 0.616 0.922 Spruce-moss forest — Tundra 0.079 0.385 0.979

202 Table B12–2 Net relatedness index to focal (FNRI) for Cyperaceae focal plants divided into four habitats. The difference between the means and zero is represented with P val- ues from one-sample t tests. Habitats with less than 10 focal plants were omitted from the analysis. Values for Cyperaceae focal plants sampled in the tundra were significantly greater than zero.

Habitat df Mean t value P value Fen 445 -0.075 -1.649 0.100 Shoreline 56 0.134 1.182 0.242 Spruce-moss forest 39 0.172 1.174 0.247 Tundra 41 0.251 2.047 0.047

203 Table B12–3 Among-clade ANOVA comparisons of mean net relatedness index to focal (FNRI) across four different habitat types in the Schefferville region of northern Canada. The Cyperaceae analyses include the 35 Cyperaceae focal species and co-occurring Cyper- aceae plants, whereas the Carex set of analysis were restricted to the 29 Carex focal species and only co-occurring Carex. Significant results (i.e. P value < 0.05) suggest clade-level patterns in FNRI for a particular habitat. Differences in FNRI were evident among both the Cyperaceae and Carex within fens.

Habitat Taxa included df F value P value Fen Cyperaceae 443 10.71 < 0.001 Fen Carex 347 7.349 < 0.001 Shoreline Cyperaceae 54 0.796 0.456 Shoreline Carex 44 0.380 0.686 Spruce-moss Cyperaceae 38 3.094 0.087 Spruce-moss Carex 33 0.550 0.582 Tundra Cyperaceae 39 1.964 0.154 Tundra Carex 34 0.766 0.388

204 Table B12–4 Net relatedness index to focal (FNRI) for Cyperaceae focal plants across four different habitats divided into major clades. The difference between the means and zero is represented with P values from one-sample t tests. Habitats with less than 10 focal plants were omitted from the analysis. Values for Cyperaceae focal plants sampled in the tundra were significantly greater than zero. The species included in each clade are listed in Appendix B4. In fens, Trichophorum showed a trend towards phylogenetic clustering (FNRI values significantly greater than zero) and Carex showed a trend towards phyloge- netic overdispersion (FNRI values significantly less than zero). Carex in the Vignea clade had FNRI values significantly less than zero, indicating a trend towards overdispersion. Within tundra habitats, Trichophorum, Carex and Core Carex sedges all had FNRI values significantly greater than zero, showing a trend towards phylogenetic clustering.

Habitat Clade df Mean t value P value Fen TRI 33 0.583 3.104 0.004 ERI 61 0.068 0.679 0.500 CAR 349 -0.164 -3.247 0.001 CC 162 -0.080 -1.016 0.311 VIG 153 -0.347 -5.019 < 0.001 CRC 1 -0.516 -3.443 0.180 Shoreline ERI 3 -0.581 -7.043 0.006 CAR 35 0.255 1.634 0.111 CC 17 0.284 1.354 0.196 VIG 14 0.124 0.444 0.664 CRC 2 0.745 5.305 0.119 Spruce-moss forest ERI 3 -0.581 -7.043 0.009 CAR 35 0.255 1.634 0.111 CC 17 0.284 1.354 0.194 VIG 14 0.134 0.444 0.664 CRC 2 0.7445 8.690 0.013 Tundra TRI 3 -0.469 -7.191 0.006 ERI 1 0.132 0.594 0.658 CAR 35 0.338 2.478 0.018 CC 33 0.367 2.568 0.015 VIG 1 -0.156 -2.550 0.238

205 Table B12–5 Among-clade ANOVA comparisons of mean net relatedness index to focal (FNRI) across four different habitat types in the Schefferville region of northern Canada. The Cyperaceae analyses include the 35 Cyperaceae focal species and co-occurring Cyper- aceae plants, whereas the Carex set of analysis were restricted to the 29 Carex focal species and only co-occurring Carex. Significant results (i.e. P value < 0.05) indicate clade-level patterns in FNRI for a particular habitat. Within fens, values of FNRI were significantly higher in the Trichophorum than in the Carex and Eriophorum,whereas Carex in the Vignea clade had significantly lower FNRI values than the Core Carex and Caricoid species.

Habitat Taxa included Comparison t value P value Fen Cyperaceae TRI—ERI 2.582 0.026 ERI—CAR -1.795 0.166 TRI—CAR -4.440 < 0.001 Carex CC—VIG -2.556 0.028 CC—CRC 2.007 0.107 VIG—CRC 3.495 0.002 Shoreline Cyperaceae TRI—ERI -0.121 0.991 ERI—CAR 1.103 0.501 TRI—CAR 0.640 0.790 Carex CC—VIG -0.057 0.998 CC—CRC 0.862 0.655 VIG—CRC 0.809 0.688 Spruce-moss forest Cyperaceae ERI—CAR 1.759 0.087 Carex CC—VIG -0.481 0.877 CC—CRC 0.778 0.711 VIG—CRC 1.033 0.552 Tundra Cyperaceae TRI—ERI 0.892 0.638 ERI—CAR 0.365 0.926 TRI—CAR 1.969 0.127 Carex CC—VIG -0.875 0.388

206 Figure B12–1 Net relatedness index to focal (FNRI) for 35 focal Cyperaceae species across four habitats found near Schefferville, Qu´ebec. Only habitats with sample sizes greater than 10 focal species were included in the analysis. Medians for each plot are rep- resented by thick lines, the boundaries of each box show the 25th and 75th percentiles and whiskers above and below each plot represent the 10th and 90th percentiles. Outlying data is indicated by hollow circles, and the following abbreviations are used—Shore (shoreline) and SM (spruce-moss forest). Sample sizes for the number of focal species found in each habitat are given above each plot.

207 Figure B12–2 Clade-level comparisons of net relatedness index to focal (FNRI) for 35 focal Cyperaceae species across four habitats found near Schefferville, Qu´ebec. Habitats with less than 10 observations were omitted because of low power to detect differences. Medians for each plot are represented by thick lines, the boundaries of each box show the 25th and 75th percentiles and whiskers above and below each plot represent the 10th and 90th percentiles. Outlying data is indicated by hollow circles and abbreviations for each clade are the same as those in Appendix B4. The species included in each analysis are listed in Appendix B4. In fens (a), significant clade-level differences (P < 0.05) were found between Trichophorum and the other two genera in fens, whereas Core Carex and Vignea clades differed significantly within Carex.

208 Appendix C

Supporting Information — Chapter 4 Appendix C1: Descriptions of four different specialization-generalization indices considered in the study.

Three different niche width indices were calculated and compared before choosing one to use in the study. First, we calculated the Extent metric by generating a convex hull using QGIS 2.2 (Quantum GIS Development Team 2014) from global occurrence data for each focal species downloaded from the Global Biodiversity Information Facility website (GBIF 2014). While downloading data, we ensured that the species authority matched those given in Chapter 3. We then square root transformed the convex hull for each species. Specialist species were considered those focal species with the smallest convex hull. Second, we generated an index of habitat diversity for each species as described in the Methods. Third, the among habitat Co-occurrence metric was calculated by finding the mean Jaccard similarity of all plots containing a particular focal species, as suggested in Fridley et al. (2007) and Manthey & Fridley (2008), using the vegdist function in the vegan package of R. We consider specialists as species with the lowest similarity values. Figure C1–1 The maximum clade credibility tree created from the posterior distribu- tion of 36,006 phylogenies for sedges in subarctic Qu´ebec and Labrador, Canada showing relative differences in the specialization-generalization indices for each focal species. The length of the bars besides each focal species corresponds to the relative level of special- ization for each metric; with shorter bars representing more specialized species. Pairwise correlations show that the three among habitat specialization-generalization metrics are not correlated. The degree of specialization for each focal species does not have a phyloge- netic signal.

211 References (Appendix C1)

Copenhagen: Global Biodiversity Information Facility. 2014. GBIF Data Agreement. http://www.gbif.org/disclaimer/datause. Fridley, J.D., Vandermast, D.B., Kuppinger, D.M., Manthey, M. & and Peet, R.K. (2007). Co-occurrence based assessment of habitat generalists and specialists: a new approach for the measurement of niche width. Journal of Ecology, 95, 707–722. Manthey, M. & Fridley, J.D. (2008). Beta diversity metrics and the estimation of niche width via species co-occurrence data: reply to Zeleny. Journal of Ecology, 97, 18–22. Quantum GIS Development Team. 2014. Open Source Geospatial Foundation Project. http://www.qgis.org.

212 Appendix C2: Justification for exclusion of one focal species from the final analyses.

Carex deflexa Hornem

The primary reason for excluding C. deflexa from the final was that it had few neigh- boring species. Only 23 plots out of the 700 plots sampled in the study had C. deflexa, skewing the results of the ‘Co-occurrence’ index of specialization-generalization so that it appeared that all other focal species were generalists in comparison to C. deflexa (Fig. C2-1). The low co-occurrence of C. deflexa with other species could be because it occurs in a couple of habitats (i.e. coniferous woodlands and rock outcrops) that only a few other focal species included in the study occasionally occur (Table C2-1). However, C. deflexa does not occur in wet meadows, swamps, or peatlands, where many of the other focal sedge species included in the study occur (Table C2-1). Table C2–1 Habitat descriptions for 35 focal sedge species in subarctic Qu´ebec and Labrador.

Focal species Subspecies Variety Habitat descriptions Carex aquatilis aquatilis Marshes; bogs wet meadows; shallow water along shores; usually in acidic substrates

Carex bigelowii bigelowii Dry to moist alpine or arctic tundra

Carex brunnescens Heaths; rocky slopes; temporarily dry areas; thin-peated mires; thickets; woodlands

Carex canescens Base-poor habitats: sphagnum bogs; moist coniferous forests and meadows; from lowlands to near the timberline in mountains

Carex capillaris Mesic to moist tundra; seeps on cliffs, rocks, and slopes; fens; meadows; shores; prairie sloughs; edges of Sphagnum mats; moist woods

Carex chorrdorrhiza Floating mats on lakeshores; emergent sedge marshes; usually in very wet sites; often in shallow water

Carex deflexa deflexa Moist to dry, open or shaded, mixed and coniferous woodlands; talus slopes; ridges; rock outcrops; burns; clearings; fields; banks

Carex disperma Swamps; bogs; wet meadows; mossy and shady coniferous woods

Carex echinata echinata Bogs; swamps; peaty or sandy shores of streams or lakes; wet meadows; usually in acidic soils

Carex exilis Fens; bogs

Carex gynocrates Chiefly boreal; wet peaty ground; usually in opening in coniferous swamps and conifer-hardwood stands; less often in poor fens; boggy swales (flarks) and alder thickets; also subalpine meadows; tundra; outwash gravel continued (Appendix C2: Table C2–1)...

214 ...continued (Appendix C2: Table C2–1) Focal species Subspecies Variety Habitat descriptions

and seepage areas; usually on calcareous substrates

Carex heleonastes Mires; damp meadows; lowlands

Carex lenticularis lenticularis Seasonally flooded river and lakeshores

Carex leptalea Mossy or wet woods; conifer swamps and bogs; wet, often calcareous (including subalpine) meadows; fens; swales; lakeshores; stream banks; also damp; shaded rock ledges; marsh fields; and swampy ditches

Carex limosa Sphagnum bogs; wet meadows; shores

Carex livida Boreal fens; calcareous floating mats

Carex magellanica irrigua Bogs; fens; marshes; usually associated with Sphagnum

Carex oligosperma Bogs, often forming extensive stands in Sphagnum-dominated areas; poor fens; sometimes in acidic, sandy, or peaty soils in open swamps, marshes, lakeshores, and riverbanks

Carex pauciflora Sphagnum bogs and acidic peat (damp mossy tundra, dryish heaths, alpine quagmires, moist forests); usually on open mats; also in partial shade of conifers

Carex rariflora Open bogs; meadows; seepage slopes; heath

Carex rostrata Fens, especially in flarks in patterned fens; bogs and bog pools; lake and stream shores; often in shallow water or on floating mats

Carex saxatilis Fens; bogs; wet tundra; roadside ditches; shores of lakes; ponds; and slow moving streams often in shallow water Carex scirpoidea scirpoidea Calcareous soils continued (Appendix C2: Table C2–1)... 215 ...continued (Appendix C2: Table C2–1) Focal species Subspecies Variety Habitat descriptions

Carex stylosa Meadows; heaths; bogs; lakeshores

Carex tenuiflora Mires; especially Sphagnum bogs and woodlands; lowlands

Carex trisperma billingsii Mires; especially Sphagnum bogs and woodlands; lowlands

Carex utriculata Open swamps; wet thicket; marshes; sedge meadows; bogs; fens; streams; ponds; and lakeshores

Carex vaginata Calcareous swamps; boggy thickets; and woods

Carex vesicaria Swamps; wet thickets; wet depressions in forests; marshes; sedge meadows; bogs; streams; ponds; lakeshores; often in sites inundated in spring and dry during summer

Eriophorum angustifolium angustifolium Marshes; bogs; fens; meadows; shores

Eriophorum chamissonis Peat bogs; marshes; muskegs

Eriophorum vaginatum Bogs; meadows; swales; tundra; wet places; peaty soils

Eriophorum viridicarinatum Marshes; meadows; bogs; fens; wet woods

Trichophorum alpinum Open or shaded, wet, peaty or gravelly fens; bogs; sheltered banks of lakes; ponds and streams; tending to occur on lime-rich substrates

Trichophorum cespitosum Open, wet, rocky or peaty meadows; fens; bogs; shores

216 Figure C2–1 One of 36,000 phylogenies randomly drawn from the Bayesian posterior distribution of phylogenies for 35 focal sedge species in subarctic Qu´ebec and Labrador, Canada showing the ‘Co-occurrence’ index of specialization. After scaling all results be- tween 0 and 1, the ‘Co-occurrence’ index for C. deflexa (represented by DEF) was much lower than the 34 other focal species; therefore, it appears as though all other species are generalists in comparison.

217 References (Appendix C2)

Ball, P., Reznicek, A. & Murray, D. (2002). Cyperaceae Jussieu. In: Flora of North America; North of Mexico. (eds. Ball, P., Gandhi, K., Kiger, R., Murray, D., Zarucchi, J., Reznicek, A., & Strother, J.). Oxford University Press, Oxford, United Kingdom, pp. 3–8.

218 Appendix C3: Results from ordinary least squares regressions among pairwise phylogenetic distances and the co-occurrence of generalist and specialist species.

Figure C3–1 Relationship among phylogenetic distance, niche breadth and species co-occurrence for the ‘Jaccard’ index of niche width. Contour lines are on the scale of standardized co-occurrences fitted with a generalized additive model (see main text for details). Predicted co-occurrence varies from high (dark) to low (light). (a) shows the comparisons among all species pairs included, whereas (b) includes only those species pairs separated by less than 30 million years. Figure C3–2 Relationship among phylogenetic distance, niche breadth and species co-occurrence for the ‘Extent’ index of niche width. Contour lines are on the scale of standardized co-occurrences fitted with a generalized additive model (see main text for details). Predicted co-occurrence varies from high (dark) to low (light). (a) shows the comparisons among all species pairs included, whereas (b) includes only those species pairs separated by less than 30 million years.

220 Table C3–1 Results of ordinary least squares regression models comparing pairwise species co-occurrences, absolute differences in the ‘Habitat’ index of niche width and phy- logenetic distances for the 34 focal species used in the analyses. The first four models in- clude all species pairs, whereas only those species pairs with a phylogenetic distance of less than 30 million years are included in the final four models. ‘Co-occur’ represents standard- ized co-occurrence values for the 34 focal species used in the analyses. ‘Gen/spec’ are the absolute differences in niche width values for the 560 species pairs in the analysis, whereas ‘phy.dist’ is the square root transformed pairwise phylogenetic difference between each of the focal species included in the analyses. Significant P values are shown in boldface type, 2 2 and adjusted R values are reported (RA).

2 Maximum Model df Slope RA P value phylogenetic distance (my) 123.2 Co-occur ∼ gen/spec 558 -1.6 0.02 < 0.01 123.2 Co-occur ∼ phy.dist 528 0.11 0.01 0.01 123.2 Gen/spec ∼ phy.dist 558 0 0.01 0.02 123.2 Co-occur ∼ 557 0.02 < 0.01 phy.dist + gen/spec -1.50 (Gen/spec) < 0.01 (Gen/spec) 0.10 (phy.dist) 0.03 (phy.dist) 123.2 Co-occur ∼ 526 phy.dist × gen/spec 0.05 < 0.01 123.2 phy.dist × gen/spec 0.76 < 0.01

30 Co-occur ∼ gen/spec 120 -2.62 0.06 < 0.01 30 Co-occur ∼ phy.dist 120 0.37 0.01 0.2 30 Gen/spec ∼ phy.dist 120 0.01 0 0.7 30 Co-occur ∼ 119 0.07 < 0.01 phy.dist + gen/spec -2.67 (Gen/spec) < 0.01 (Gen/spec) 0.40 (phy.dist) 0.15 (phy.dist) 30 Co-occur ∼ 118 phy.dist × gen/spec 0.07 0.01 30 phy.dist × gen/spec -0.95 0.35

221 Appendix D

Supporting Information — Chapter 5 Appendix D1: Taxonomic sampling and background information used in phylogenetic reconstruction.

Table D1–1 List of vascular plant species found in 176 plot on Mnt Irony, Labrador during 2013 field sampling. Orders, families, species names and authorities are in accordance with Vascan (Brouillet et al. 2013) as of April 2015.

Order Family Species name Authority MTMG MT No. Sequences accession accession of plots added BOLD/Genbank Asparagales Asparagaceae Maianthemum trifolium (Linnaeus) Sloboda 138332 187660 7 N Asparagales Orchidaceae Neottia cordata (Linnaeus) Richard 138356 187687 7 N Asterales Asteraceae Arnica angustifolia Vahl 138335 187663 1 N subsp. angustifolia Asterales Asteraceae Achillea millefolium Linnaeus 138365 187701 7 N Asterales Asteraceae Eurybia radula (Aiton) G.L. Nesom 137547 187752 12 Y Asterales Asteraceae Solidago macrophylla Banks ex Pursh 137550 187761 39 Y Asterales Asteraceae Solidago multiradiata Aiton 138366 187704 18 N Asterales Asteraceae Packera aurea (Linnaeus) A.´ L¨ove & D. L¨ove 137543 187737 1 Y Asterales Asteraceae Petasites frigidus (Aiton) Cronquist 187755 17 N var. palmatus Asterales Asteraceae Antennaria alpina (Linnaeus) Gaernter 137524 187667 1 Y Carylophyllales Caryophyllaceae Cerastium arcticum Lange 138352 187682 1 N Carylophyllales Caryophyllaceae Minuartia biflora (Linnaeus) Schinz 137523 187665 1 Y & Thellung Carylophyllales Caryophyllaceae Stellaria borealis Bigelow 137540 187732 2 Y subsp. borealis continued (Appendix D1: Table D1–1)... 223 ...continued (Appendix D1: Table D1–1) Order Family Species name Authority MTMG MT No. Sequences accession accession of plots added BOLD/Genbank

Carylophyllales Polygonaceae Bistorta vivipara (Linnaeus) Delarbre 138398 187758 17 N Cornales Cornaceae Cornus canadensis Linnaeus 138326 187652 83 N Cupressales Cupressaceae Juniperus communis Linnaeus 138286 187610 8 N Dipsacales Adoxaceae Viburnum edule (Michaux) Rafinesque 138308 187633 4 N Dipsacales Caprifoliaceae Linnaea borealis Linnaeus 187696 76 N Dipsacales Caprifoliaceae Lonicera villosa (Michaux) Roemer 138301 187626 5 N &Schultes Equisetales Equisetaceae Equisetum arvense Linnaeus 138328 187655 19 N Equisetales Equisetaceae Equisetum scirpoides Michaux 138291 187615 4 N Equisetales Equisetaceae Equisetum sylvaticum Linnaeus 138310 187635 36 N Equisetales Equisetaceae Equisetum variegatum Schleicher ex 138357 187688 5 N subsp. variegatum F. Weber & D. Mohr Ericales Diapensiaceae Diapensia lapponica Linnaeus 138282 187603 2 N Ericales Ericaceae Andromeda polifolia Aiton 138302 187626 6 N var. latifolia Ericales Ericaceae Arctous alpina (Linnaeus) Niedenzu 138295 187620 12 N Ericales Ericaceae Empetrum nigrum Linnaeus 138343 187673 64 N Ericales Ericaceae Gaultheria hispidula (Linnaeus) Muhlenberg 138312 187637 23 N ex Bigelow Ericales Ericaceae Kalmia polifolia Wangenheim 138283 187622 13 N Ericales Ericaceae Moneses uniflora (Linnaeus) A. Gray 138363 187698 1 N Ericales Ericaceae Orthilia secunda (Linnaeus) House 138401 187763 23 N Ericales Ericaceae Pyrola asarifolia Michaux 137535 187719 3 Y continued (Appendix D1: Table D1–1)... 224 ...continued (Appendix D1: Table D1–1) Order Family Species name Authority MTMG MT No. Sequences accession accession of plots added BOLD/Genbank

subsp. asarifolia Ericales Ericaceae Pyrola grandiflora Radius 138345 187675 3 N Ericales Ericaceae Rhododendron groenlandicum (Oeder) Kron 138330 187658 91 N & Judd Ericales Ericaceae Rhododendron lapponicum (Linnaeus) Wahlenberg 138298 187623 7 N Ericales Ericaceae Vaccinium caespitosum Michaux 137521 187653 34 Y Ericales Ericaceae Vaccinium myrtilloides Michaux 138305 187630 31 N Ericales Ericaceae Vaccinium oxycoccos Linnaeus 138378 187721 18 N Ericales Ericaceae Vaccinium uliginosum Linnaeus 138270 187589 79 N Ericales Ericaceae Vaccinium vitis-idaea Linnaeus 138340 187670 107 N Ericales Primulaceae Lysimachia borealis (Rafinesque) 138325 187651 36 N U. Manns & Anderberg Fagales Betulaceae Alnus viridis (Aiton) Turrill 137518 187608 10 Y subsp. crispa Fagales Betulaceae Betula glandulosa Michaux 138285 187609 70 N Fagales Myricaceae Myrica gale Linnaeus 138317 187642 3 N Lamiales Orobanchaceae Bartsia alpina Linnaeus 138348 187678 7 N Lamiales Orobanchaceae Castilleja septentrionalis Lindley 137532 187709 2 Y Lamiales Orobanchaceae Rhinanthus minor Linnaeus 138395 187751 2 N Liliopsida Tofieldiaceae Tofieldia pusilla (Michaux) Persoon 138367 187705 14 N Lycopodiales Lycopodiaceae Diphasiastrum complanatum (Linnaeus) Holub 138278 187599 3 N Lycopodiales Lycopodiaceae Huperzia appressa (Desvaux) 138300 187625 3 N A.´ L¨ove & D. L¨ove continued (Appendix D1: Table D1–1)... 225 ...continued (Appendix D1: Table D1–1) Order Family Species name Authority MTMG MT No. Sequences accession accession of plots added BOLD/Genbank

Lycopodiales Lycopodiaceae Lycopodium annotinum Linnaeus 138290 187614 42 N Malpighales Salicaceae Salix arctophila Cockerell ex A. Heller 138333 187661 13 N Malpighales Salicaceae Salix argyrocarpa Andersson 138324 187649 1 N Malpighales Salicaceae Salix glauca (Pursh) Dorn 138313 187639 8 N var. cordifolia Malpighales Salicaceae Salix humilis Marshall 138304 187629 1 N var. humilis Malpighales Salicaceae Salix pedicellaris Pursh 138384 187736 1 N Malpighales Salicaceae Salix planifolia Pursh 138292 187617 8 N Malpighales Salicaceae Salix uva-ursi Pursh 138284 187607 5 Y Malpighales Salicaceae Salix vestita Pursh 138268 187588 55 N Malpighales Violaceae Viola adunca Smith 138296 187621 21 N Malpighales Violaceae Viola renifolia A. Gray 138279 187600 17 N Myrtales Onagraceae Chamerion angustifolium (Linnaeus) Holub 138388 187760 2 N subsp. angustifolium Myrtales Onagraceae Epilobium hornemannii Reichenbach 137541 187733 1 Y Pinales Pinaceae Abies balsamea (Linnaeus) Miller 137548 187754 5 Y Pinales Pinaceae Larix laricina (Du Roi) K. Koch 187720 14 N Pinales Pinaceae Picea glauca (Moench) Voss 187697 40 N Pinales Pinaceae Picea mariana (Miller) Britton, 138306 187631 62 N Sterns & Poggenburgh Poales Cyperaceae Carex aquatilis Wahlenberg 138320 187645 7 N var. aquatilis continued (Appendix D1: Table D1–1)... 226 ...continued (Appendix D1: Table D1–1) Order Family Species name Authority MTMG MT No. Sequences accession accession of plots added BOLD/Genbank

Poales Cyperaceae Carex bigelowii TorreyexSchweinitz 137337 187666 16 N Poales Cyperaceae Carex brunnescens (Persoon) Poiret 137527 187693 2 Y Poales Cyperaceae Carex capillaris Linnaeus 138334 187662 7 N Poales Cyperaceae Carex capitata Linnaeus 138372 187715 1 N Poales Cyperaceae Carex deflexa Hornemann 137519 187618 2 Y var. deflexa Poales Cyperaceae Carex disperma Dewey 137538 187730 3 Y Poales Cyperaceae Carex gynocrates Wormskjold ex Drejer 138303 187628 31 N Poales Cyperaceae Carex leptalea Wahlenberg 137536 187723 1 Y Poales Cyperaceae Carex limosa Linnaeus 6 N Poales Cyperaceae Carex magellanica Lamarck 138390 187745 2 N Poales Cyperaceae Carex scirpoidea Michaux 138271 187590 32 N subsp. scirpoidea Poales Cyperaceae Carex trisperma Dewey 137546 187747 2 Y Poales Cyperaceae Carex utriculata Boott 137545 187744 1 Y Poales Cyperaceae Carex vaginata Tausch 138289 187613 68 N Poales Cyperaceae Eriophorum viridicarinatum (Engelmann) Fernald 138319 187644 2 N Poales Cyperaceae Trichophorum alpinum (Linnaeus) Persoon 138323 187648 1 N Poales Cyperaceae Trichophorum cespitosum (Linnaeus) Hartman 138322 187647 16 N Poales Juncaceae Oreojuncus trifidus (Linnaeus) Z´avesk´a 138399 187759 3 N Dr´abkov´a & Kirschner Poales Juncaceae Luzula parviflora (Ehrhart) Desvaux 137539 187731 2 Y Poales Poaceae Agrostis mertensii Trinius 137530 187702 2 Y continued (Appendix D1: Table D1–1)... 227 ...continued (Appendix D1: Table D1–1) Order Family Species name Authority MTMG MT No. Sequences accession accession of plots added BOLD/Genbank

Poales Poaceae Anthoxanthum monticola (Bigelow) Veldkamp 138336 187664 1 N Poales Poaceae Calamagrostis canadensis (Michaux) P. Beauvois 138400 187762 10 N Poales Poaceae Avenella flexuosa (Linnaeus) Drejer 138389 187742 25 N Poales Poaceae Elymus trachycaulus (Link) Gould 138393 187749 5 N subsp. trachycaulus ex Shinners Poales Poaceae Poa arctica R. Brown 138364 187700 1 N subsp. arctica Poales Poaceae Schizachne purpurascens (Torrey) Swallen 138307 187632 2 N Poales Poaceae Trisetum spicatum (Linnaeus) K.Richter 138368 187706 1 N Polypodiales Dryopteridaceae Dryopteris expansa (C. Presl) 138375 187618 2 N Fraser-Jenkins & Jermy Polypodiales Woodsiaceae Cystopteris montana (Lamarck) Bernhardi 138359 187690 4 N ex Desvaux Ranunculales Ranunculaceae Anemone parviflora Michaux 138294 187619 12 N Ranunculales Ranunculaceae Coptis trifolia (Linnaeus) Salisbury 138309 187634 40 N Rosales Rosaceae Amelanchier bartramiana (Tausch) M. Roemer 138311 187636 1 N Rosales Rosaceae Dasiphora fruticosa (Linnaeus) Rydberg 138381 187727 1 N Rosales Rosaceae Dryas integrifolia Vahl 138299 187624 12 N Rosales Rosaceae Fragaria virginiana Miller 138329 187656 2 N Rosales Rosaceae Rubus arcticus (Michaux) Focke 138318 187643 27 N subsp. acaulis Rosales Rosaceae Rubus chamaemorus Linnaeus 138316 187641 19 N Rosales Rosaceae Rubus idaeus Linnaeus 138361 187694 6 N continued (Appendix D1: Table D1–1)... 228 ...continued (Appendix D1: Table D1–1) Order Family Species name Authority MTMG MT No. Sequences accession accession of plots added BOLD/Genbank

Santalales Santalaceae Geocaulon lividum (Richardson) Fernald 138288 187637 49 N Saxifragales Grossulariaceae Ribes glandulosum Grauer 138287 187611 8 N

References

Brouillet, L., Coursol, F., Meades, S.F, Favreau, M., Anions, M., Belisle, P., & Desmet, P. (2013). VASCAN, the Database of Vascular Plants of Canada. http://data.canadensys.net/vascan/.

229 Table D1–2 Sequence data used to create a regional phylogeny of vascular plants sampled on Mnt Irony, in Labrador. Missing data from the sequences matrix are represented by ‘NA’ values. Contributions to GenBank by the authors are indicated with an asterisk (*).

Species Sequence ITS matK rbcL Maianthemum trifolium Accession number NA KC251185 KC251460 Source GenBank GenBank GenBank Corallorhiza trifida Accession number EU391324 KC474485 KC482466 Source GenBank GenBank GenBank Neottia cordata var. cordata Accession number JN999304 JN966362 JN965648 Source GenBank GenBank GenBank Arnica angustifolia subsp. angustifolia Accession number JN998952 KC474106 KC482034 Source GenBank GenBank GenBank Achillea millefolium Accession number AF046939 KC473958 KC481883 Source GenBank GenBank GenBank Eurybia radula Accession number EU200207 NA KJ841319 Source GenBank NA GenBank* Solidago macrophylla Accession number NA KJ841015 KJ841584 Source NA GenBank* GenBank* Solidago multiradiata Accession number HQ142575 KC475916 KC484137 Source GenBank GenBank GenBank Packera aurea Accession number KJ418347 KJ840955 KJ841440 Source GenBank GenBank* GenBank* Petasites frigidus var. palmatus Accession number JN999371 KC475256 KC483416 Source GenBank GenBank GenBank continued (Appendix D1: Table D1–2)...

230 ...continued (Appendix D1: Table D1–2) Species Sequence ITS matK rbcL Antennaria alpina Accession number NA KC474014 KC481941 Source NA GenBank GenBank Cardamine bellidifolia Accession number EU819310 KC474274 KC482205 Source GenBank GenBank GenBank Arenaria humifusa Accession number JN998949 KC474089 KC482021 Source GenBank GenBank GenBank Cerastium arcticum Accession number KC691714 KC474439 KC482413 Source GenBank GenBank GenBank Minuartia biflora Accession number KC958800 KC475004 KC483175 Source GenBank GenBank GenBank Moehringia macrophylla Accession number NA MLTVP039-11 MLTVP039-11 Source NA BOLD BOLD Stellaria borealis subsp. borealis Accession number JN589064 NA VEMSH689-13 Source GenBank NA BOLD Stellaria longipes subsp. longipes Accession number JN589086 KC475951 KC484175 Source GenBank GenBank GenBank Bistorta vivipara Accession number GQ339919 KC474192 KC482118 Source GenBank GenBank GenBank Parnassia kotzebuei Accession number JF811074 KC475149 KC483315 Source GenBank GenBank GenBank Cornus canadensis Accession number AY530913 MLTVP038-11 MLTVP38-11 Source GenBank BOLD BOLD Cornus alternifolia Accession number DQ340526 KJ840899 KJ841246 Source GenBank GenBank* GenBank* continued (Appendix D1: Table D1–2)... 231 ...continued (Appendix D1: Table D1–2) Species Sequence ITS matK rbcL Cornus stolonifera Accession number DQ343139 EU749307 KJ841248 Source GenBank GenBank GenBank* Hydrangea arborescens Accession number DQ006012 VEMSH499-13.matK DQ006098 Source GenBank BOLD GenBank Juniperus communis Accession number EU277677 EU749466 AY664859 Source GenBank GenBank GenBank Dillenia indica Accession number AY09030 AB24752 AB25515 Source GenBank GenBank GenBank Dillenia ovata Accession number JX852686 AB92954 AB925579 Source GenBank GenBank GenBank Davilla nitida Accession number NA JQ587393 JQ591320 Source NA GenBank GenBank Viburnum edule Accession number AY265123 HQ591577 MLTVP033-11 Source GenBank GenBank BOLD Linnaea borealis Accession number AY236181 KC474956 KC483091 Source GenBank GenBank GenBank Lonicera villosa Accession number EU240677 NA NA Source GenBank BOLD BOLD Equisetum arvense Accession number Y11471 AM883546 HQ590081 Source GenBank GenBank GenBank Equisetum scirpoides Accession number EU372663 AM883534 KC482737 Source GenBank GenBank GenBank Equisetum sylvaticum Accession number EU328342 AM883553 KC482739 Source GenBank GenBank GenBank continued (Appendix D1: Table D1–2)... 232 ...continued (Appendix D1: Table D1–2) Species Sequence ITS matK rbcL Equisetum variegatum subsp. variegatum Accession number DQ377155 AM883537 KJ841306 Source GenBank GenBank GenBank* Diapensia lapponica Accession number AF396235 KC474547 KC482536 Source GenBank GenBank GenBank Andromeda polifolia var. latifolia Accession number AF358872 KC473989 KC481912 Source GenBank GenBank GenBank Arctous alpina Accession number JN998944 JN966112 JN965267 Source GenBank GenBank GenBank Empetrum nigrum Accession number GU176626 KC474704 KC482698 Source GenBank GenBank GenBank Gaultheria hispidula Accession number JF801562 MLTVP069-11 MLTVP069-11 Source GenBank BOLD BOLD Kalmia polifolia Accession number JN999277 JN966342 JN965614 Source GenBank GenBank GenBank Moneses uniflora Accession number FJ378568 JN966382 JN965683 Source GenBank GenBank GenBank Orthilia secunda Accession number FJ378569 KC475046 KC483216 Source GenBank GenBank GenBank Phyllodoce caerulea Accession number GU176630 KC475275 KC483434 Source GenBank GenBank GenBank Pyrola asarifolia subsp. asarifolia Accession number AF133736 KJ840976 KJ841503 Source GenBank GenBank* GenBank* Pyrola grandiflora Accession number HM021772 KC475648 KC483818 Source GenBank GenBank GenBank continued (Appendix D1: Table D1–2)... 233 ...continued (Appendix D1: Table D1–2) Species Sequence ITS matK rbcL Rhododendron groenlandicum Accession number JN999452 JN966493 JN965806 Source GenBank GenBank GenBank Rhododendron lapponicum Accession number JN999454 KC475682 KC483875 Source GenBank GenBank GenBank Vaccinium caespitosum Accession number AF419775 AF419703 KJ841646 Source GenBank GenBank GenBank* Vaccinium myrtilloides Accession number NA MLTVP045-11 MLTVP045-11 Source NA BOLD BOLD Vaccinium oxycoccos Accession number NA JN895357 JN893514 Source NA GenBank GenBank Vaccinium uliginosum Accession number AY274575 KC476089 KC484313 Source GenBank GenBank GenBank Vaccinium vitis-idaea Accession number AF382743 KC476096 KC484320 Source GenBank GenBank GenBank Lysimachia borealis Accession number AY855156 HQ593472 HQ590306 Source GenBank GenBank GenBank Alnus viridis subsp. crispa Accession number AJ251681 KC473970 KC481895 Source GenBank GenBank GenBank Betula glandulosa Accession number AY761110 KC474175 KC482101 Source GenBank GenBank GenBank Betula minor Accession number NA KJ840872 KJ841128 Source NA GenBank* GenBank* Betula pumila Accession number AY761131 GU373395 GU373377 Source GenBank GenBank GenBank continued (Appendix D1: Table D1–2)... 234 ...continued (Appendix D1: Table D1–2) Species Sequence ITS matK rbcL Myrica gale Accession number DQ501423 JN966385 JN965686 Source GenBank GenBank GenBank Bartsia alpina Accession number JF900505 KC474166 KC482094 Source GenBank GenBank GenBank Castilleja septentrionalis Accession number NA NA KJ841223 Source NA NA GenBank* Rhinanthus minor Accession number KC480393 JN966492 JN965805 Source GenBank GenBank GenBank Tofieldia pusilla Accession number JN999675 KC476051 KC484269 Source GenBank GenBank GenBank Diphasiastrum complanatum Accession number NA NA AB574627 Source NA NA GenBank Huperzia appressa Accession number NA DQ465956 DQ464220 Source NA GenBank GenBank Lycopodium annotinum Accession number KF977441 NA KC483144 Source GenBank NA GenBank Salix arctophila Accession number JN999496 KC475742 KC483935 Source GenBank GenBank GenBank Salix argyrocarpa Accession number NA VEMSH669-13 VEMSH669-13 Source NA BOLD BOLD Salix glauca var. cordifolia Accession number JN999543 KC475754 KC483946 Source GenBank GenBank GenBank Salix herbacea Accession number EF060384 KC475764 KC483957 Source GenBank GenBank GenBank continued (Appendix D1: Table D1–2)... 235 ...continued (Appendix D1: Table D1–2) Species Sequence ITS matK rbcL Salix humilis var. humilis Accession number NA SALIX330-08 SALIX330-08 Source NA BOLD BOLD Salix myricoides Accession number NA SALIX401-08 SALIX401-08 Source NA BOLD BOLD Salix pedicellaris Accession number JN999564 JN966602 JN965922 Source GenBank GenBank GenBank Salix planifolia Accession number JN999592 KC475774 KC483969 Source GenBank GenBank GenBank Salix uva-ursi Accession number NA KC475815 KC484004 Source NA GenBank GenBank Salix vestita Accession number JN999612 JN966659 JN965981 Source GenBank GenBank GenBank Viola adunca Accession number JN999694 JN966732 JN966059 Source GenBank GenBank GenBank Viola renifolia Accession number JN999696 JN966733 JN966060 Source GenBank GenBank GenBank Chamerion angustifolium subsp. angustifolium Accession number JF976296 KC474459 KC482434 Source GenBank GenBank GenBank Epilobium hornemannii Accession number NA NA KJ841298 Source NA NA GenBank* Abies balsamea Accession number EF057709 NA JN935605 Source GenBank NA GenBank Larix laricina Accession number AF041348 NA KC483075 Source GenBank NA GenBank continued (Appendix D1: Table D1–2)... 236 ...continued (Appendix D1: Table D1–2) Species Sequence ITS matK rbcL Picea glauca Accession number AF136621 EU749473 JN965724 Source GenBank GenBank GenBank Picea mariana Accession number AF136617 EU749475 KC483441 Source GenBank GenBank GenBank Carex aquatilis var. aquatilis Accession number FJ904610 KC474284 KC482222 Source GenBank GenBank GenBank Carex bigelowii Accession number AY278303 JN895168 JN892344 Source GenBank GenBank GenBank Carex brunnescens Accession number EU541872 HQ593204 KJ841158 Source GenBank GenBank GenBank Carex capillaris Accession number DQ998905 FJ548087 FJ548255 Source GenBank GenBank GenBank Carex capitata Accession number DQ115118 JX065080 JN965353 Source GenBank GenBank GenBank Carex deflexa var. deflexa Accession number AY686720 KJ840880 KJ841168 Source GenBank GenBank* GenBank* Carex disperma Accession number DQ115150 KC474314 KC482265 Source GenBank GenBank GenBank Carex glacialis Accession number AY757625 FJ548096 FJ548259 Source GenBank GenBank GenBank Carex gynocrates Accession number AY757417 KC474326 KC482281 Source GenBank GenBank GenBank Carex leptalea Accession number AY241979 NA VEMSH689-13 Source GenBank GenBank BOLD continued (Appendix D1: Table D1–2)... 237 ...continued (Appendix D1: Table D1–2) Species Sequence ITS matK rbcL Carex limosa Accession number AY757595 JN896254 JN893833 Source GenBank GenBank GenBank Carex magellanica Accession number AY757594 JN896255&AY757594 JN893834 Source GenBank GenBank GenBank Carex scirpoidea subsp. scirpoidea Accession number AY757582 KC474394 KC482363 Source GenBank GenBank GenBank Carex trisperma Accession number AY757429 KJ840888 KJ841216 Source GenBank GenBank* GenBank* Carex utriculata Accession number JN999091 JN966210 KJ841217 Source GenBank GenBank GenBank* Carex vaginata Accession number AY278285 KC474414 KC482387 Source GenBank GenBank GenBank Eriophorum viridicarinatum Accession number NA JX074652 U49230 Source NA GenBank GenBank Trichophorum alpinum Accession number JF313184 JX065093 AJ810999 Source GenBank GenBank GenBank Trichophorum cespitosum Accession number DQ998951 KC476058 KC484276 Source GenBank GenBank GenBank Oreojuncus trifidus Accession number AY727770 AY973526 KC483037 Source GenBank GenBank GenBank Luzula parviflora Accession number FJ873788 KJ840935 KJ841390 Source GenBank GenBank* GenBank* Luzula spicata Accession number FJ213865 KJ840936 KJ841391 Source GenBank GenBank* GenBank* continued (Appendix D1: Table D1–2)... 238 ...continued (Appendix D1: Table D1–2) Species Sequence ITS matK rbcL Agrostis mertensii Accession number GQ324467 KC473966 KC481891 Source GenBank GenBank GenBank Anthoxanthum monticola Accession number EF577511 JN966327 JN965577 Source GenBank GenBank GenBank Calamagrostis canadensis Accession number FJ377632 KC474230 KC482155 Source GenBank GenBank GenBank Avenella flexuosa Accession number JQ972936 JN894954 JN892037 Source GenBank GenBank GenBank Elymus trachycaulus subsp. trachycaulus Accession number FJ040168 KC474693 KC482688 Source GenBank GenBank GenBank Poa alpina subsp. alpina Accession number GQ324483 KC475318 KC483480 Source GenBank GenBank GenBank Poa arctica subsp. arctica Accession number GQ324487 KC475341 KC483504 Source GenBank GenBank GenBank Schizachne purpurascens Accession number FM179432 KJ841007 KJ841556 Source GenBank GenBank* GenBank* Trisetum spicatum Accession number FJ377674 KC476086 KC484304 Source GenBank GenBank GenBank Dryopteris expansa Accession number NA NA KF539809 Source NA NA GenBank Cystopteris montana Accession number NA NA MLTVP070-11 Source NA NA BOLD Anemone parviflora Accession number FJ639887 KC474010 KC481934 Source GenBank GenBank GenBank continued (Appendix D1: Table D1–2)... 239 ...continued (Appendix D1: Table D1–2) Species Sequence ITS matK rbcL Coptidium lapponicum Accession number AY680194 KC474484 KC482462 Source GenBank GenBank GenBank Coptis trifolia Accession number AB695608 AB695565 AF093730 Source GenBank GenBank GenBank Amelanchier bartramiana Accession number JQ392368 MLTVP016-11 MLTVP016-11 Source GenBank BOLD BOLD Dasiphora fruticosa Accession number GU444027 KC474486 KC482473 Source GenBank GenBank GenBank Dryas integrifolia Accession number NA KC474668 KC482658 Source NA GenBank GenBank Fragaria virginiana Accession number AF163479 HQ593298 JN965555 Source GenBank GenBank GenBank Rubus arcticus subsp. acaulis Accession number AF055741 KC475689 KC483881 Source GenBank GenBank GenBank Rubus chamaemorus Accession number AF055740 KC475694 KC483885 Source GenBank GenBank GenBank Rubus idaeus Accession number AF055757 JN966515 JN892122 Source GenBank GenBank GenBank Ximenia americana Accession number NA JF270999 KF496540 Source NA GenBank GenBank Geocaulon lividum Accession number JN999241 KC474864 KC482932 Source GenBank GenBank GenBank Comandra umbellata Accession number NA DQ329192 JX848536 Source GenBank GenBank GenBank continued (Appendix D1: Table D1–2)... 240 ...continued (Appendix D1: Table D1–2) Species Sequence ITS matK rbcL Thesium humifusum Accession number GU256780 JN895002 JM892108 Source GenBank GenBank GenBank Ribes glandulosum Accession number AF426345 KJ840984 KJ841518 Source GenBank GenBank* GenBank* Mitella nuda Accession number JN999326 JN966378 JN965679 Source GenBank GenBank GenBank Selaginella selaginoides Accession number AF419000 NA AB574651 Source GenBank NA GenBank

241 Table D1–3 Fossil priors used in Beast to create the phylogenetic reconstruction of vascular plants sampled on Mnt Irony, Labrador.

Taxa Reference Stem or Fossil estimation Beast calibration crown calibration Vascular plants Magallon & Sanderson (2001) Crown 408 MY Mean = 6.012, Sigma = 0.03; log normal

Eudicots Magallon & Sanderson (2001) Crown 121.0 MY Mean = 4.792, Sigma = 0.125; log normal

Coniferae Miller (1999) Crown 310 MY Mean = 5.738, Sigma = 0.04; log normal Lycopsida Wikstr¨om & Kenrick (2001) Crown 390 MY Mean = 5.965, Sigma = 0.035; log normal

Ferns — Clarke et al. (2011) Stem 454 MY (soft) - 388.2 MY Mean = 6.043, Monolophyta & Spermatopyta Sigma= 0.05; log normal

Euasterids Martinez-Millan (2010) Stem 83.5 MY Mean = 4.425, Sigma = 0.2; log normal

Asterids Mart´ınez-Mill´an (2010) Stem 89.5 MY Mean = 4.4945; Sigma = 0.2; log normal

Monocots from Eudicots Clarke et al. (2011) Stem 248.4 MY (soft) - 124 MY Mean = 5.225; + Ceratophyllum Sigma = 0.22; log normal

Cariceae Escudero et al. (2012) Stem 61.1 MY Mean = 4.1125; Sigma = 0.01; log normal

242 References (Table D1–3)

Clarke, J. T., Warnock, R., & Donoghue, P.C.J. (2011). Establishing a timescale for plant evolution. New Phytologist, 192, 266-301.

Escudero, M., Hipp, A.L., Waterway, M.J., & Valente, L.M. (2012). Diversification rates and chromosome evolution the most diverse angiosperm genus of the temperate zone (Carex, Cyperaceae). Molecular Phylogenetics and Evolution, 63, 650–655.

Magallon, S. & Sanderson, M. J. (2001). Absolute diversification rates in angiosperm clades. Evolution, 55, 1762-1780.

Mart´ınez-Mill´an, M. (2010). Fossil record and age of the Asteridae. The Botanical Review, 76, 83-135.

Miller, C.N. Jr. 1999. Implications of fossil conifers for the phylogenetic relationships of living families. The Botanical Review, 65, 239-277.

Wikstr¨om, N. & Kenrick, P. 2001. Evolution of Lycopodiaceae (Lycopsida): estimating divergence times from rbcL gene sequences by use of nonparametric rate smoothing. Molecular Phylogenetics and Evolution, 19, 177-186.

243 Figure D1–1 Phylogeny of vascular plant species sampled on Mnt Irony, Labrador based on maximum likelihood phylogenetic reconstruction. Species names are in accordance with Vascan, where red shading represents low bootstrap support for species placement, and black shading represents high support. 244 Figure D1–2 Phylogeny of vascular plant species sampled on Mnt Irony, Labrador based on bayesian phylogenetic reconstruction. Clade names are in accordance with Vascan, and blue node bards indicate estimated divergence times in millions of years. 245 Appendix D2: Supplementary five cluster analyses.

Table D2–1 The highest cluster membership support coefficients compared between different clustering methods with the Tukey’s range test, where P values less than 0.05 are considered significant. The following abbreviations are used to represent the different clustering methods: Gower dissimilarity (SiteEnv), vascular beta diversity (VascBD), vas- cular phylogenetic beta diversity (VascPhBD), angiosperm beta diversity (AngioBD) and angiosperm phylogenetic beta diversity (AngioPhBD).

Comparison Difference P value SiteEnv — VascBD 0.425 < 0.001 SiteEnv — VascPhBD 0.020 < 0.001 SiteEnv — AngioBD 0.322 < 0.001 SiteEnv — AngioPhBD 0.005 0.999 VascBD — VascPhBD -0.305 < 0.001 VascBD — AngioBD -0.102 < 0.001 VascBD — AngioPhBD -0.420 < 0.001 VascPhBD — AngioBD 0.202 < 0.001 VascPhBD — AngioPhBD -0.115 < 0.001 AngioBD — AngioPhBD -0.318 < 0.001 Table D2–2 Spatial associations in plot assignments to clusters for vascular plants and angiosperms, as estimated by Moran’s I. The following abbreviations are used: Gower dis- similarity (SiteEnv), vascular beta diversity (VascBD), vascular phylogenetic beta diversity (VascPhBD), angiosperm beta diversity (AngioBD) and angiosperm phylogenetic beta di- versity (AngioPhBD). Pairwise distances (metres) between plots were divided into 16 bins, with between 1924 and 1926 observations per bin. The P values for SiteEnv, VascBD, VascPhBD, AngioBD and AngioPhBD were calculated from randomization tests.

Bin Mean Count SiteEnv VascBD VascPhBD AngioBD AngioPhBD distance (metres) 1 96.74 1926 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 2 198.62 1924 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 3 272.69 1926 < 0.001 0.210 0.028 < 0.001 < 0.001 4 336.51 1924 < 0.001 0.055 0.982 < 0.001 < 0.001 5 395.4 1926 < 0.001 0.080 0.516 0.458 0.611 6 456.6 1924 0.002 0.096 0.309 0.620 0.088 7 517.77 1926 0.767 0.395 0.366 0.849 0.197 8 583.27 1924 0.752 0.207 0.863 0.018 0.176 9 655.12 1926 0.743 0.900 0.593 0.981 0.733 10 733.57 1924 0.002 0.826 0.560 0.860 0.348 11 821.21 1926 < 0.001 0.078 0.729 0.573 0.421 12 921.79 1924 < 0.001 0.904 0.575 0.624 0.293 13 1040.01 1926 < 0.001 0.411 0.063 0.060 0.237 14 1185.67 1924 < 0.001 0.003 < 0.001 0.061 0.038 15 1388.76 1926 < 0.001 < 0.001 < 0.001 0.001 < 0.001 16 2080 1924 < 0.001 < 0.001 < 0.001 < 0.001 0.016

247 Table D2–3 Phylogenetic signal, as assessed by Blomberg’s K statistic, and associated significance values for five clusters based on Gower’s dissimilarities (SiteEnv). Signifi- cance was assessed as the variance of phylogenetically independent contrasts relative to a random tip shuffling algorithm, where P values < 0.05 are considered significant.

SiteEnv K SiteEnv P Cluster 1 0.019 0.347 Cluster 2 0.027 0.238 Cluster 3 0.016 0.663 Cluster 4 0.019 0.532 Cluster 5 0.019 0.502

248 Figure D2–1 Species richness patterns across 176 plots on Mnt Irony, Labrador. Species richness per plot is indicated in panel (a), with the highest richness values given in black and the lowest values in light gray. Panel (b) indicates the average species richness per sampling bands of similar elevation, where the highest elevation sampling bands are rep- resented on the left-hand side of the plot and the lowest elevation plots on the right-hand side.

249 Figure D2–2 Clusters of the 176 plots on Mont Irony, Labrador, shaded by cluster mem- bership. The top three panels show the results for the Gower dissimilarity (SiteEnv) dis- tances. The middle three panels show beta diversity distances. The bottom three panels show results for phylogenetic beta diversity distances. Panels (a), (d) and (g) show prin- cipal coordinate ordinations (PCoA) of five K-means clusters, gray lines indicate the loca- tion of plots in ordination space. The optimal number of K-means clusters estimated by the Calinski and Harabasz index are shown in panels (b), (e) and (h), with black circles in- dicating the optimal and gray circles showing the second highest recommended number of clusters for each analysis. Panels (c), (f) and (i) show the location of the plots correspond- ing to each of the five clusters on the sampling grid, with grayscale shading indicating cluster membership. All results are based on vascular plant data.

250 Figure D2–3 Relative abundance patterns across 176 plots on Mnt Irony, Labrador for ferns, gymnosperms and lycophytes. Average relative frequencies per plot for ferns (a), gymnosperms (b) and lycophytes (c), where the highest abundance values are given in black and the lowest values are shown in light gray. Panels (d), (e) and (f) indicate the average relative abundance per sampling bands of similar elevation, where the highest elevation sampling bands are represented on the left-hand side of the plot and the lowest elevation plots on the right-hand side. Panel (d) shows the average relative abundance per sampling band for ferns, whereas Panels (e) and (f) indicate the corresponding values for gymnosperms and lycophytes, respectively.

251 Figure D2–4 The highest cluster membership support coefficients across all clusters com- pared among three cluster types. Medians for each plot are represented by thick lines, the boundaries of each box show the 25th and 75th percentiles, and whiskers above and below each plot represent the 10th and 90th percentiles. Outlying data is indicated by hollow cir- cles and abbreviations, and the following abbreviations are used to represent the different clustering methods: Gower dissimilarity (SiteEnv), angiosperm beta diversity (AngioBD) and angiosperm phylogenetic beta diversity (AngioPhBD).

252 Figure D2–5 Correlation indices of the vascular plant species to the five different SiteEnv clusters, where each column of bars represents a different cluster. Bars extending to the right of each column centre represent positive correlations with clusters, whereas leftward extending bars indicate negative correlations.

253 Appendix D3: Supplementary three and seven cluster analyses.

Table D3–1 Comparison among three clusters using Gower dissimilarity (SiteEnv), beta diversity (BD) and phylogenetic beta diversity distances (PhBD). Values range between 0 and 1 and are based on the adjusted Rand index, where 1.00 indicates that the cluster- ing of plots was the same. Two sets of beta and phylogenetic beta diversity analyses are included, one with all vasculars (Vasc) and a second to angiosperms only (Angio).

SiteEnv VascBD VascPhBD AngioBD AngioPhBD SiteEnv 1.00 0.23 0.11 0.12 0.17 VascBD 0.23 1.00 0.34 0.40 0.26 VascPhBD 0.11 0.34 1.00 0.14 0.29 AngioBD 0.12 0.40 0.14 1.00 0.34 AngioPhBD 0.17 0.26 0.29 0.34 1.00 Table D3–2 Comparison among seven clusters using Gower dissimilarity (SiteEnv), beta diversity (BD) and phylogenetic beta diversity distances (PhBD). Values range between 0 and 1 and are based on the adjusted Rand index, where 1.00 indicates that the cluster- ing of plots was the same. Two sets of beta and phylogenetic beta diversity analyses are included, one with all vasculars (Vasc) and a second to angiosperms only (Angio).

SiteEnv VascBD VascPhBD AngioBD AngioPhBD SiteEnv 1.00 0.15 0.13 0.14 0.11 VascBD 0.15 1.00 0.27 0.61 0.23 VascPhBD 0.13 0.27 1.00 0.21 0.25 AngioBD 0.14 0.61 0.21 1.00 0.24 AngioPhBD 0.11 0.23 0.25 0.24 1.00

255 Table D3–3 The highest cluster membership support coefficients compared between dif- ferent cluster types with the Tukey’s range test, where P values less than 0.05 are consid- ered significant. The following abbreviations are used to represent the different clustering methods: Gower dissimilarity (SiteEnv), vascular beta diversity (VascBD), vascular phylo- genetic beta diversity (VascPhBD), angiosperm beta diversity (AngioBD) and angiosperm phylogenetic beta diversity (AngioPhBD). All values are for the three cluster analysis, and distance types varied in their support coefficients for cluster membership showing which

metric produced the most well-defined clusters (ANOVA, F 4,875 = 127.7, P < 0.001).

Comparison Difference P value SiteEnv — VascBD 0.282 < 0.001 SiteEnv — VascPhBD 0.158 < 0.001 SiteEnv — AngioBD 0.263 < 0.001 SiteEnv — AngioPhBD 0.047 0.025 VascBD — VascPhBD -0.123 < 0.001 VascBD — AngioBD -0.019 0.749 VascBD — AngioPhBD -0.235 < 0.001 VascPhBD — AngioBD 0.104 < 0.001 VascPhBD — AngioPhBD -0.111 < 0.001 AngioBD — AngioPhBD -0.216 < 0.001

256 Table D3–4 The highest cluster membership support coefficients compared between dif- ferent cluster types with the Tukey’s range test, where P values less than 0.05 are consid- ered significant. The following abbreviations are used to represent the different clustering methods: Gower dissimilarity (SiteEnv), vascular beta diversity (VascBD), vascular phylo- genetic beta diversity (VascPhBD), angiosperm beta diversity (AngioBD) and angiosperm phylogenetic beta diversity (AngioPhBD). All values are for the seven cluster analysis, and distance types varied in their support coefficients for cluster membership showing which

metric produced the most well-defined clusters (ANOVA, F 4,875 = 103.4, P < 0.001).

Comparison Difference P value SiteEnv — VascBD 0.312 < 0.001 SiteEnv — VascPhBD 0.086 0.001 SiteEnv — AngioBD 0.336 < 0.001 SiteEnv — AngioPhBD 0.034 0.814 VascBD — VascPhBD -0.227 < 0.001 VascBD — AngioBD 0.024 0.814 VascBD — AngioPhBD -0.279 < 0.001 VascPhBD — AngioBD 0.251 < 0.001 VascPhBD — AngioPhBD -0.052 0.131 AngioBD — AngioPhBD -0.303 < 0.001

257 Table D3–5 Spatial associations in plot assignments to three clusters for vascular plants and angiosperms compared to Gower dissimilarity (SiteEnv) clusters, as estimated by Moran’s I. The following abbreviations are used: vascular beta diversity (VascBD), vas- cular phylogenetic beta diversity (VascPhBD), angiosperm beta diversity (AngioBD) and angiosperm phylogenetic beta diversity (AngioPhBD). Pairwise distances (metres) between plots were divided into 16 bins, with between 1924 and 1926 observations per bin. The P values for SiteEnv, VascBD, VascPhBD, AngioBD and AngioPhBD were calculated from randomization tests, where values less than 0.05 are considered significant.

Bin Mean Count SiteEnv VascBD VascPhBD AngioBD AngioPhBD distance (metres) 1 96.74 1926 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 2 198.62 1924 < 0.001 < 0.001 < 0.001 0.017 < 0.001 3 272.69 1926 < 0.001 < 0.001 < 0.001 0.081 0.025 4 336.51 1924 < 0.001 < 0.001 0.001 0.176 0.283 5 395.4 1926 < 0.001 0.523 < 0.001 0.423 0.608 6 456.6 1924 < 0.001 0.206 0.174 0.518 0.197 7 517.77 1926 0.084 0.114 0.216 0.797 0.422 8 583.27 1924 0.325 0.043 0.090 0.125 0.498 9 655.12 1926 0.560 0.497 0.237 0.688 0.804 10 733.57 1924 0.619 0.541 0.950 0.79 0.524 11 821.21 1926 0.024 0.162 0.360 0.286 0.411 12 921.79 1924 < 0.001 0.644 0.205 0.861 0.878 13 1040.01 1926 < 0.001 0.477 < 0.001 0.483 0.180 14 1185.67 1924 < 0.001 < 0.001 < 0.001 0.138 0.143 15 1388.76 1926 < 0.001 < 0.001 < 0.001 0.003 < 0.001 16 2080 1924 < 0.001 < 0.001 < 0.001 0.002 0.002

258 Table D3–6 Spatial associations in plot assignments to seven clusters for vascular plants and angiosperms compared to Gower dissimilarity (SiteEnv) clusters, as estimated by Moran’s I. The following abbreviations are used: vascular beta diversity (VascBD), vas- cular phylogenetic beta diversity (VascPhBD), angiosperm beta diversity (AngioBD) and angiosperm phylogenetic beta diversity (AngioPhBD). Pairwise distances (metres) between plots were divided into 16 bins, with between 1924 and 1926 observations per bin. The P values for SiteEnv, VascBD, VascPhBD, AngioBD and AngioPhBD were calculated from randomization tests, where values less than 0.05 are considered significant.

Bin Mean Count SiteEnv VascBD VascPhBD AngioBD AngioPhBD distance (metres) 1 96.74 1926 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 2 198.62 1924 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 3 272.69 1926 < 0.001 < 0.001 0.012 < 0.001 0.002 4 336.51 1924 < 0.001 < 0.001 0.595 < 0.001 0.028 5 395.4 1926 < 0.001 < 0.001 0.246 < 0.001 0.498 6 456.6 1924 < 0.001 0.156 0.827 0.005 0.032 7 517.77 1926 0.006 0.415 0.874 0.076 0.143 8 583.27 1924 0.813 0.480 0.309 0.442 < 0.001 9 655.12 1926 0.825 0.385 0.722 0.524 0.506 10 733.57 1924 0.886 0.660 0.953 0.890 0.561 11 821.21 1926 0.013 0.351 0.420 0.582 0.165 12 921.79 1924 < 0.001 0.038 0.418 0.009 0.836 13 1040.01 1926 < 0.001 < 0.001 0.004 < 0.001 0.142 14 1185.67 1924 < 0.001 < 0.001 < 0.001 < 0.001 0.008 15 1388.76 1926 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 16 2080 1924 < 0.001 < 0.001 0.010 < 0.001 0.051

259 Table D3–7 Phylogenetic signal, as assessed by Blomberg’s K statistic, and associated significance values for three clusters based on Gower’s dissimilarities (SiteEnv). Signifi- cance was assessed as the variance of phylogenetically independent contrasts relative to a random tip shuffling algorithm, where P values < 0.05 are considered significant.

SiteEnv K SiteEnv P Cluster 1 0.019 0.407 Cluster 2 0.013 0.735 Cluster 3 0.020 0.400

260 Table D3–8 Phylogenetic signal, as assessed by Blomberg’s K statistic, and associated significance values for seven clusters based on Gower’s dissimilarities (SiteEnv). Signifi- cance was assessed as the variance of phylogenetically independent contrasts relative to a random tip shuffling algorithm, where P values < 0.05 are considered significant.

SiteEnv K SiteEnv P Cluster 1 0.015 0.590 Cluster 2 0.017 0.483 Cluster 3 0.026 0.339 Cluster 4 0.015 0.694 Cluster 5 0.018 0.551 Cluster 6 0.030 0.213 Cluster 7 0.015 0.652

261 Figure D3–1 Clusters of the 176 plots on Mont Irony, Labrador, shaded by cluster mem- bership for three clusters. The top three panels show the results for the Gower dissimi- larity (SiteEnv) distances. The middle three panels show beta diversity distances. The bottom three panels show results for phylogenetic beta diversity distances. Panels (a), (d) and (g) show principal coordinate ordinations (PCoA) of five K-means clusters, grey lines indicate the location of plots in ordination space. The optimal number of K-means clusters estimated by the Calinski and Harabasz index are shown in panels (b), (e) and (h), with black circles indicating the optimal and grey circles showing the second highest recom- mended number of clusters for each analysis. Panels (c), (f) and (i) show the location of the plots corresponding to each of the five clusters on the sampling grid, with greyscale shading indicating cluster membership. All results are based on angiosperm data.

262 Figure D3–2 Clusters of the 176 plots on Mont Irony, Labrador, shaded by cluster mem- bership for seven clusters. The top three panels show the results for the Gower dissim- ilarity (SiteEnv) distances. The middle three panels show beta diversity distances. The bottom three panels show results for phylogenetic beta diversity distances. Panels (a), (d) and (g) show principal coordinate ordinations (PCoA) of five K-means clusters, grey lines indicate the location of plots in ordination space. The optimal number of K-means clusters estimated by the Calinski and Harabasz index are shown in panels (b), (e) and (h), with black circles indicating the optimal and grey circles showing the second highest recom- mended number of clusters for each analysis. Panels (c), (f) and (i) show the location of the plots corresponding to each of the five clusters on the sampling grid, with greyscale shading indicating cluster membership. All results are based on angiosperm data.

263 Figure D3–3 Spatial associations in plot assignments to three clusters (panels (a) and (b)) and seven clusters (panels (c) and (d)) for vascular plants and angiosperms compared to Gower dissimilarity (SiteEnv) clusters, as estimated by Moran’s I. Panels (a) and (c) show values for vascular plants, whereas angiosperm values are indicated in panels (b) and (d). Values above zero indicate positive spatial associations in cluster membership across plots, whereas negative values show negative spatial associations. Pairwise distances (metres) were divided into 16 bins with between 1924 and 1926 observations per bin, and the following abbreviations are used: vascular beta diversity (VascBD), vascular phyloge- netic beta diversity (VascPhBD), angiosperm beta diversity (AngioBD) and angiosperm phylogenetic beta diversity (AngioPhBD).

264 Figure D3–4 The proportion of shared species and branch lengths within three (pan- els (a) and (b)) and seven (panels (c) and (d)) Gower dissimilarity (SiteEnv) clus- ters (BDwithin and PhBDwithin, respectively) compared to among cluster beta diversity

(BDamong) and phylogenetic beta diversity(PhBDamong) averages. Panels (a) and (c) show values for vascular plants, whereas angiosperm values are indicated in panels (b) and (d). The black dashed line indicates among-cluster beta diversity, whereas among-cluster phy- logenetic beta diversity is shown in light grey. Black circles indicate within-cluster beta diversity, and light grey circles show within cluster phylogenetic beta diversity.

265 Figure D3–5 Species contributions to beta diversity values (SCBD). The length of in- dividual bar corresponds to each SCBD, with longer bars corresponding to higher SCBD values and short bars representing low SCBD. Bars (a) indicate SCBD among the three Gower dissimilarity (SiteEnv) clusters, whereas bars (b) and (c) show SCBD values among the three beta and phylogenetic beta diversity clusters for vascular plants, respectively. Considering only angiosperms, bar (d) indicate SCBD among the three SiteEnv clusters, while bars (e) and (f) indicate SCBD values among the angiosperm beta and phylogenetic beta diversity clusters.

266 Figure D3–6 Species contributions to beta diversity values (SCBD). The length of in- dividual bar corresponds to each SCBD, with longer bars corresponding to higher SCBD values and short bars representing low SCBD. Bars (a) indicate SCBD among the seven Gower dissimilarity (SiteEnv) clusters, whereas bars (b) and (c) show SCBD values among the seven beta and phylogenetic beta diversity clusters for vascular plants, respectively. Considering only angiosperms, bar (d) indicate SCBD among the seven SiteEnv clusters, while bars (e) and (f) indicate SCBD values among the angiosperm beta and phylogenetic beta diversity clusters.

267