Citation
Phylogenetic Signal to Invasion and Rarity in Florida's Imperiled Pine Rockland Flora

Material Information

Title:
Phylogenetic Signal to Invasion and Rarity in Florida's Imperiled Pine Rockland Flora
Creator:
Trotta, Lauren B
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (68 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Wildlife Ecology and Conservation
Committee Chair:
BAISER,BENJAMIN H
Committee Co-Chair:
SESSA,EMILY
Committee Members:
BRUNA,EMILIO M,III
Graduation Date:
12/17/2016

Subjects

Subjects / Keywords:
Distance functions ( jstor )
Ecosystems ( jstor )
Endangered species ( jstor )
Endemic species ( jstor )
Invasive species ( jstor )
Phylogenetics ( jstor )
Phylogeny ( jstor )
Species ( jstor )
Taxa ( jstor )
Threatened species ( jstor )
Wildlife Ecology and Conservation -- Dissertations, Academic -- UF
community -- phylogeny
Miami metropolitan area ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Wildlife Ecology and Conservation thesis, M.S.

Notes

Abstract:
Community phylogenetic approaches are used to examine the relatedness of groups of species with in a larger regional pool. I have used a community phylogenetic approach to understand the phylogenetic relatedness of invasive, threatened and endangered, and endemic species found in the globally imperiled Pine Rockland ecosystem. I found that invasive species are randomly related, while threatened and endangered, and endemic species are under-dispersed. Finally, by comparing nearest neighbor metrics I found that invasive species are more closely related to native species than threatened and endangered species. This indicates that invasive species share similar life history traits to native species that have been able to adapt to changing environmental conditions in the Pine Rockland habitat rather than threatened and endangered species that have detrimental life history traits under the new Pine Rockland environmental regime. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (M.S.)--University of Florida, 2016.
Local:
Adviser: BAISER,BENJAMIN H.
Local:
Co-adviser: SESSA,EMILY.
Statement of Responsibility:
by Lauren B Trotta.

Record Information

Source Institution:
UFRGP
Rights Management:
Copyright Trotta, Lauren B. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Classification:
LD1780 2016 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

PHYLOGENETIC SIGNAL TO INVASION AND RARITY IN PINE ROCKLAND FLORA By LAUREN B. TROTTA A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR TH E DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2016

PAGE 2

2016 Lauren B. Trotta

PAGE 3

To Cee, Fred, Cynthia, and Marilyn

PAGE 4

4 ACKNOWLEDGMENTS I would like to express my deepest gratitude to my advisors Ben jamin Baiser and Emily Sessa for their guidance and unending enthusiasm for rocking the pine rocklands. I would further like to acknowledge my Mom, Dad, Nana and Gr ammy for their lifelong support that has facilitated my passion to pursue my studies Zachary Siders, thank you, not only for your expert technical support, but also your kind and steady encourag ement. I would like to express my and sorrows throughout my tenure at UF. Lastly, thank you to Patrick Ewanchuk for fostering my inte rest in research, and being i nstrumental in shaping the path that I have taken and the person that I am today. I would like to thank the Eppley Foundation for Research who graciously funded this work. Thank you also to, the Miami Dade Endangered Environmental Lands Program for assist ance with permitting and access to many crucial field sites. I would like to express my great appreciation to Paul and Judy Seigel for graciously and repeatedly hosting the pine rockland research crew to in the most amazing field station one could hope for Finally, none of this work would be possible without the tremendous help from our fabulous botanists: Jenn ifer Possley, James Lange and Sarah Martin. Their skills and knowledge as well as those contributed by the Institute for Regional Conservation and Fair child Tropical Botanic Garden, are the basis upon which this work is built.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIGURE S ................................ ................................ ................................ ......................... 8 LIST OF OBJECTS ................................ ................................ ................................ ......................... 9 LIST OF ABBREVIATIONS ................................ ................................ ................................ ........ 10 ABSTRACT ................................ ................................ ................................ ................................ ... 11 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 13 Community Phylogenetic Approach ................................ ................................ ....................... 13 Phylogenetic Signal ................................ ................................ ................................ ................ 13 Phylogenetic Signal of Invasion ................................ ................................ ...................... 15 Phylogenetic Signal of Extinction Threat ................................ ................................ ........ 16 Pine Rockland Ecosystem ................................ ................................ ................................ ....... 17 Research Goals and Questions ................................ ................................ ................................ 19 2 METHODS ................................ ................................ ................................ ............................. 23 Phylogenetic Reconstruction ................................ ................................ ................................ .. 23 Taxon Sampling and Field Collection ................................ ................................ ............. 23 DNA Extraction, Amplification and Sequencing ................................ ............................ 24 Sequence Alignment and Phylogenetic Analyses ................................ ........................... 24 Community Phylogenetic Analyses ................................ ................................ ........................ 26 Species Designations ................................ ................................ ................................ ....... 26 Metrics of Community Relatedness ................................ ................................ ................ 27 Null Model for Metrics of Community Relatedness ................................ ....................... 28 3 RESULTS ................................ ................................ ................................ ............................... 37 Phylogenetic Recon struction ................................ ................................ ................................ .. 37 Taxon Sampling, DNA Extraction, Amplification and Sequencing ............................... 37 Sequence Alignment and Phylogenetic Analyses ................................ ........................... 37 Community Phylogenetic Analyses ................................ ................................ ........................ 38 Species Designations ................................ ................................ ................................ ....... 38 Phylogenetic Relatedness of Invasive Species ................................ ................................ 38 Phylogenetic Relatedness of Threatened and Endangered Species ................................ 39 Nearest Neighbor Analyses for Invasive Species ................................ ............................ 39

PAGE 6

6 4 DISCUSSION ................................ ................................ ................................ ......................... 51 Community Phylogenetic Analyses ................................ ................................ ........................ 51 Invasive Species ................................ ................................ ................................ ...................... 51 Threatened, Endangered, and Endemic Species ................................ ................................ ..... 52 ................................ ................................ ................. 54 Relatedness of Invasive to Threatened and Endangered, and Native Species ........................ 56 Issues with Phylogenet ic Analyses ................................ ................................ ......................... 58 Future Directions ................................ ................................ ................................ .................... 59 LIST OF REFERENCES ................................ ................................ ................................ ............... 63 BIOGRAPHIC AL SKETCH ................................ ................................ ................................ ......... 68

PAGE 7

7 LIST OF TABLES Table page 2 1 Species of interest within the 527 species pine rockland regional pool. ........................... 31 3 1 Z score and p values from SES community relatedness metrics. ................................ ...... 41

PAGE 8

8 LIST OF FIGURES FIGURE page 1 1 Map of pine rockland his toric and current distributions ................................ ................... 22 2 1 Graphical representation of three community relatedn ess metrics ................................ .... 35 2 2 Schematic for calculating n earest neighbor observed values ................................ ........... 36 3 1 Best ML phylogeny of 527 pine r ockland species colored by rank ................................ .. 42 3 2 Best ML phylogen y of 527 pine rockland species ................................ ............................ 43 3 3 B I consensus phylogeny of 527 pine rockland species ................................ ..................... 44 3 4 Diversit y metrics (PD, MPD and MNTD) ................................ ................................ ....... 45 3 5 Best ML phylogeny with spec ies of special interest denoted ................................ ........... 46 3 6 Best ML phylogeny with moderate and severe invasive speci es denoted ........................ 47 3 7 Best ML phylogeny with US threatened and endangered spec ies denoted ...................... 48 3 8 Best ML phylogeny with FL threatened, endnagered, and endemic species denoted ..... 49 3 9 Boxplot observed nearest neighbor distances. ................................ ................................ ... 50 4 1 Representatives of threatened and endangered species found in pine rockland s .............. 62

PAGE 9

9 LIST OF OBJECTS Object page 2 1 Table of collection and accession numbe rs for 527 species ................................ .............. 30

PAGE 10

10 LIST OF ABBREVIATIONS AIC Akaike Information Criterion APG Angiosperm Phylogeny Group BI Bayesian Inference BOL Barcode of Life BS Bootstrap FISF Floristic Inventory of South Florida FLAS University of Florida Herbarium FLDACS Florida Department of Agriculture and Consumer Services FLEPPC Florida Exotic Pest Plant Council FLMNH Florida Museum of Natural History FTBG Fairchild Tropical Botanic Garden IRC Institute for Regional Conservation ML Maximum Likelihood MNTD Mean Nearest Taxon Distance MPD Mean Pairwise Distance NN Nearest Neighbor PCR Polymerase Chain Reaction PD Phylogenetic Diversity PP Posterior Probability SES Standardized Effect Size USFWS U.S. Fish and Wildlife Service

PAGE 11

11 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science PINE ROCKLAND FLORA By Lauren B. Trotta December 2016 Chair: Benjamin Baiser Cochair: Emily B. Sessa Major: Wildlife Ecology and Conservation Commun ity phylogenetic ap proaches incorporate the evolution ary history between co occurring species in order to understand the drivers of community assembly. These approaches are commonly used to examine the relatedness between species that are either becoming invasive, or are at risk of extinction. In this study, I used a community phylogenetic approach to understand phylogenetic relatedness within invasive, threatened e ndangered, and endemic species group f ound in the globally imperiled pine rockland ecosystem. flora is an ideal system to use for broaching these questions because a multitude sp ecies belonging to each group are know n to directly interact in the habitat. I found that in vasive species are randomly related, while threatened and endangered, and end emic species are more closely related than we expected indicating that different factors are driving the establishment or persistence of these specie s in pine rockland ecosystems I also sought to understand the relatedness between invasive species and their threatened and endangered, and native counterparts. Using a nearest neighbor approach I found that invasive species are more closely relate d to native species than threatened and endangered

PAGE 12

12 species. This indicates that invasive species may share similar life history traits to native species that have been able to adapt to changing environmental conditions in the pine rockland habitat.

PAGE 13

13 CHAPTER 1 INTROD U C TION Community Phylogenetic Approach Species richness, historically quantified in terms of taxonomic diversity, provides an incomplete picture of biodiversity in natural systems because it fails to take into account the evolutionary relationship s between species (Olden 2006; Cavender Bares et al. 2009) As a result, community phylogenetic approaches have become a powerful way to explore evolutionary relationships betwee n species interacting on an ecological timescale (Olden 2006; Cavender Bares et al. 2009; Vamosi et al. 2009) Phylogenetic relatedness between species can also pose as a proxy for functional similarity adding to the depth of information we can glean from community phylogenetic techniques (Cavender Bares et al. 2004; Cadotte et al. 2009a; Swenson 2013; Marx et al. 2015) This multifaceted approach to biodiversity anal ysis is essential to elucidate drivers of community assembly, persistence, and disassembly because both environmental influence and competitive interactions leave behind phylogenetic signals in the observed community (Cavender Bares et al. 2009) Furt hermore, community phylogenetic approaches have been used to detect evolutionary signal underlying species extirpations and invas ions in the face of global climate change (MacArthur & Wilson 1967; Olden 2006; Morlon et al. 2011) Here, we investigate relatedness within invasive, as well as t hreatened and endangered species groups found in a single globally imperiled ecosystem: the pine rocklands. Phylogenetic Signal From a phylogenetic standpoint, a community phylogeny is constructed for one or more assemblage of species (Webb et al. 2002) Generally communities are considered groups of spe cies that co occur within a local area. These species are a subset of those found in a larger regional pool and are treated as their own unit when assessing rela tedness (Purvis 2008; Vamosi

PAGE 14

14 et al. 2009) Groups of species that share certain traits or designations can also be used to subset species from within a regional pool (Tucker et al. 2016) In this study I am interested in the groups of species within the pine rockland community phylogeny that have been designated as invasive, threatened, endangered or endemic at e ither a state for federal level. C ommunity phylogenetic approaches are a powerful way to query relatedness between species in a local community or designated group (e.g., invasive species) in comparison to the regional pool of species There are a myriad of metrics available to calculate relatedness between species that share a spatially defined community or species designation (Vamosi et al. 2009; Tucker et al. 2016) C om monly these metrics are used to infer processes driving community assembly by comparing patterns of phylogenetic dispersion compared to a null model (Webb et al. 2002) Species that are more distantly related to each other than we would expect by chance are considered to be over dispersed (Webb et al. 2002) Over dispersion can be a signal of competitive interactions in a habitat because closely related species are thought to compete more strongly and cannot co exist (Gause 1932; Hutchinson 1959; MacArthur & Levins 1967; Lack 1973) Cape Floristic region the regional pool of species is dominated by relatively few clades, however these clad es are incredibly speciose as a result of large species radiations (Slingsby & Verboom 2007; Davies et al. 2011) At a local scale overdispersion is seen in these communities because of high interspecific competition between phylogeneti cally closely related species (Slingsby & Verboom 2007) Species that are more closely related than we would expect at random are dee med under disper sed within the community phylogeny. Underdispersion can result from environmental filtering where only closely related species with certain adaptations can survive the

PAGE 15

15 environmental conditions (Webb et al. 2002) Verd et al ., ( 2007 ) found that plant species in a fire dominated Mediterranean habitat exhibit phylogenetic underdispersion potentially because species must share traits (i.e., resinous cones, thick bark) that al low them to cope with the harsh environmental conditions (Verd & Pausas 2007) Recent work suggests that overdispersion and underdispersion do not necessarily indicate competition and environmental filtering explicitly, but rather interactions between both processes (Cadotte et al. 2009b; Mayfield & Levine 2010; Marx et al. 2015) Furthermore, functional trait dispersion does not always mirror phylogenetic dispersion. Differences in functional and phylogenetic relatedness can alter our interpretation of over and underdispersion of species (Cavender Bares et al. 2004) Phylogenetic Signal of Inv asion Community phylogenetic approaches have been widely used when considering relationships between invasive species and a native community (Fridley & Sax 2014) These studies focu s on determining whether invasive species are under dispersed, over dis persed or randomly related to natives in a community. This ma y indicate the level of relatedness to the native pool required for a species to become invasi ve. Underdispersion of invader s to the native pool may be required to allow invasive species to pass through the environmental filter. However, closely related species may competitively exclude phylogenetically similar invaders leading to patterns of overdispersion. This juxtaposed exp ectation of relatedness between invasive and native species was first posed by Darwin and has been a major center of recent research Experiments with plant communities have shown that a single potential invasive species is unlikely to invade if it is dist antly related to the native community (Jiang et al. 2010) However, in microbe communities studies have sho wn that successful invaders are distantly related to native species pools and can take advantage of different niches and available resources

PAGE 16

16 (Peay et al. 2012) Natural systems have also been explored and scale seems to play an important role in the resulting patterns I n Austrialia, invasive speices show a pattern o f random dispersion at local scale s and a pattern of underdispersion at contient wide scales potentially because they share certain traits (Cadotte et al. 2009b) Experimental and field approaches have been applied to tackle this debate and the results have been highly dependent on scale and taxa (Cavender Bares et al. 2006; Cadotte et al. 2009b; Marx et al. 2015) Phylogenetic Signal of Extinction Threat Inve stigations of phylogenetic signal of extinction or species' tendency toward risk of extinction are also based on community phylogenetic methods (Purvis et al. 2000a, 2000b; Bielby et al. 2008; Cardillo et al. 2008; Purvis 2008; Davies et al. 2011) These studies generally focus on patterns of extinction across large scales and within single taxonomic groups. They seek to understand whether extinct species or species endangered by extinction are closely or distantly related and how this information can be used to determine species at risk of extinction. Ge nerally, extinction risk is greater in closely related species that share certain life history traits (Purvis et al. 2000a) For example, large bodi ed, slow growing and slow to reproduce birds, mammals and amphibians have all become increasing rare and in danger of going extinct (McKinney 1997; Bielby et al. 2008; Cardillo et al. 2008) Further, amphibians as well as shallow water reef forming corals have shown that changes in environ ment increase the risk of threat and extinction (Bielby et al. 2008; Carpenter et al. 2008; Huang & Roy 2015) While the focus of studies of phylogenetic signal of invasion and extinction stem from similar questions about relatedness between different groups of species, rarely are both invasive and endangered species considered concu rrently or in the same system

PAGE 17

17 Pine Rockland Ecosystem An ideal system to combine both approaches phylogenetic relatedness to invasion and extinction risk in the globally imperiled pine rockland ecosystem. The pine rockland ecosystem is a savannah like fo rest found along the Miami Rock Ridge in Miami Dade County It occurs from North Miami Beach in the north to Long Pine Key and Everglades National Park in the south and west, as well as in the Florida Key s, the Caribbean, and Cuba ( Figure 1 1 ) (MSRP 1999) In Florida, t h is hyperdiverse ecosystem is only found on the limestone outcroppings of the Miami Rock Ridge and provides habitat for a host of endemic, threatened, and endangered bat, insect, and plant species. The pine rockland plant community represents a confluence of tropical Caribbean and temperate North American plant taxa, resulting in a flora not found anywh ere else in the United States (MSRP 1999) This habitat has a characteristic upper canopy composed solely of slash pine ( Pinus elliottii var. d ensa ), a midstory of palms ( e.g., Sabal palmetto, Sabal etonia, and Serenoa repens ) and woody shrubs (e.g., Lantana involucrat a Tetrazygia bicolor ) and an understory of unique herbaceous ground cover that includes many end emic species. Among the endemic species eight are f ederally endangered, one is federally threatened and another eight are candidates for federa l listing (Possley et al. 2008; Powell & Maschi nski 2012) P ine rockland habitat also hosts nearly 40 plant sp ecies considered to be threatened or endangered in the state of Florida (FLDACS 2015). The rapid growth of infrastructure and development across the Miami Rock Ridge, which at 2.33 m above sea level constitutes the highest elevation land in south Florida (MSRP 1999) and the explosive expansion of the greater Miami metropolitan area, has resulted in the reduction of the pine rockland ecosystem to an occupancy today of less than of 2% of its historical range outside Everglades National Park (MRSP 1999 ) The remnants of this endangered habitat are

PAGE 18

18 largest remaining fragment, found in Everglades Nat ional Park, is 8,029 ha and the combined total area of habitat outside of par k amounts to less than 920 ha (Possley et al. 2008) H abi tat loss and fragmentation, which define the current state of the pine rockland ecosystem have been identified as the greatest threat to biodiversity for ecosystems worldwide (Fahrig 2003) Furthermore, in the pine rockland ecosystem, habi tat loss due to fragmentation is compounded by the introduction of invasive plant species, fire suppression, and changes in hydrology (MRSP 1999 ). Along with the multitude of threatened and endangered species found in this ecosystem there are nearly 50 spe cies ranked as moderately or severely invasive in Florida according to the Florida Exotic Pest Plant Council (FLEPPC 2015). The preponderance of invasive, threatened, endangered and endemic species that co occur in the same habitat, make the pine rockland ecosystem ideal to study the phylogenetic relatedness of these species groups. The Floristic Inventory of South Florida (FISF) was performed by the Institute for Regional Conser vation (IRC) for all habitats in south Florida including pine rockland (Ga nn e t al. 2001). The FISF database identifies 532 species a s being found in pine rockland habitat This list includes nat ive, threatened, endangered, and invasive species as defined at both FL state and US federal levels. Fine scale local data such as those fo und in the FISF database provide the starting point for building community phylogenies for the habitat as a whole (referred to as the regional pool) or for subsets of species found within the community. Community phylogenetic approaches have become a popul ar technique for exploring evolutionary relationships and evaluating relatedness between species in a community (Webb et al. 2002; Caven der Bares et al. 2009)

PAGE 19

19 Research Goals and Questions In this study, I incorporated 527 taxa from the pine rockland species list into a community phylogeny. I then used this phylogeny to examine patterns of relatedness within threatened and endangered, e ndemic, and invasive species groups in order to gain insight into pine rockland community assembly. By investigating relationships within individual groups we can determine whether environmental filtering or competitive interactions are mainly driving spec ies persistence in the community. For example, if invasive species are under dispersed this indicates that they share evolutionarily derived traits that allow them to pass through the (Faith 1992) However, overdispersion a mong invasive species may indicate that in order to establish and persist in pine rockland communities invasive species needed to possess novel traits that allowed them to outcompete for avoid competition with already established species (Faith 1992) Sim ilarly, patterns of relatedness with threatened and endangered species may indicate why these species are becoming increasingly in danger of local or global extinction. If threatened and endangered species found in pine rocklands are under dispersed their risk of extinction may be due to sharing highly similar, specialized life history traits that are no longer supported in the habitat following anthropogenic impacts. Alternatively, if threatened and endangered species are overdispersed this may indicate th at they fill dissimilar niches in the habitat that once allowed them to escape competition, but now put them at a disadvantage following regime shifts in the ecosystem. Further, endemic species are a particularly interesting group in the pine rocklands be cause they are known to have evolved in this harsh fire driven habitat. Many of these endemic species are also listed as threatened or endangered, but the threatened and endangered lists also include numerous species that are not endemic to the pine rockland ecosystem. Comparing the

PAGE 20

20 phylogenetic relatedness of these similar, but not mutually exclusive groups, may illuminate subtly differ ent stories about how these communities establish and persist We expect to find that endemic pine rockland species are under dispersed because they evolved in a frequently burned ecosystem. This historical environmental regime in combi nation with the phys ical attributes of the ecosystem (e.g., bare, limestone substrate, high light open canopy) might have acted as an environmental filter, which selected for traits that performed well in these conditions. However, phylogenetic relatedness analyses may indica te that endemic species are overdispersed because community assembly was driven by competition. In this scenario species evolving in pine rockland habitat evolved to fill disparate niches that allowed them to avoid strong competitive interactions common am ong closely related congeners. While examining the phylogenetic relatedness of species within their own groups of interest may inform our understanding of the community assembly it fails to provide a picture of how these species groups interact. To expl ore these interactions I queried the relative phylogenetic relatedness between invasive species with the threatened and endangered species group and the native species group. This approach used a nearest neighbor metric to determine whether invasive specie s were more closely related to the threatened and endangered group or the native group. If the invasive species are more closely related to the threatened and endangered group we would conclude that the invasive species share similar life history traits wi th threatened and endangered species and thus can replace them within a niche. An alternate mechanism for this same result could be that threatened and endangered species are disappearing regardless of invasive species, but closely related invasive species can advantageously occupy the open niches. However, if we find that invasive species are more closely related to native species we would conclude that invasive species have similar life history traits to those species

PAGE 21

21 that are not becoming increasingly at risk of extinction. Native species have been able to adapt to a changing environmental regime, unlike the threatened and endangered species, and the invasive species that establish and persist may closely related and share similar traits with the native s pecies. I sought to answer four questions regarding the phylogenetic relatedness of threatened and endangered species and invasive species the globally imperiled pine rockland ecosystem: 1) Are invasive species more or less closely related than would expec t at random? 2) Are threatened and endangered species more or less closely related than we expect at random? 3) Are endemic species more or less closely related that we expect at random? 4) Are invasive species more closely related to threatened and endang ered species or native species in pine rockland ecosystems ?

PAGE 22

22 Figure 1 1. Map of pine rockland historic and current distributions Dark and light green represent the historic distribution of pine rockland ecosystems. Light green areas represent the modern day extent of pine rocklands within Everglades National Park. Red areas represent remaining pine rockland fragments outside of Everglades National Park, which occ upy less than 2% of the historical distribution. Map adapted from IRC. Historic pine rockland Everglades National Park Current pine rockland

PAGE 23

23 CHAPTER 2 METHODS Phylogenetic Reconstruction Taxon Sampling and Field Collection The FISF pine rockland list served as a guide for taxa collection (Gann et al. 2001) Additional speci es were collected if encountered in a pine rockland habitat, regardless of their inclusion in FISF. This community phylogeny represents 527 pine rockland taxa; I collected 350 taxa in the field, 17 taxa from pre existing herbarium vouchers located at the F airchild Tropical Botanic Garden (F TBG) h erbarium (Miami, FL, USA) or the Universi ty of Florida h erbarium (FLAS), and 58 taxa from field collected plants in cultivation at the F T BG greenhouses (Object 2 1 ). I sampled all field collected taxa over the cours e of five sampling trips to public and privately owned pine rockland fragments in Miami Dade County. These sampling trips also spanned a range of seasons and successional habitat stages in order to incorporate the greatest number of taxa possible. I suppl emented new collections with additional sequence data for 143 taxa included in the Flora of Florida project performed at Florida Museum of Natural History (FLMNH) (Neubig et al, personal communication) (Object 2 1 ). In total, this study includes data from 381 taxa sequenced de novo and 143 taxa with pre existing sequences (Object 2 1 ). I collected all field and cultivated specimens in fruit or flower, and a botanical expert identified taxa to species level or below. I preserved up to 100 g of fresh leaf ma terial in silica (Absorbent Industries, LLC, Harrisburg, NC, USA) for later DNA extraction, and I collected a representative portion of the plant (entire above ground structure if small or a small branch with fruit or flower if large) as voucher material w hich I submitted to FLAS.

PAGE 24

24 DNA Extraction, Amplification and Sequencing I extracted total genomic DNA from the silica dried leaf material collected from field, cultivated, and herbarium collections using the Qiagen DNeasy Mini Plant Kit (Qiagen, Valencia, California, USA). I modified the standard protocol to use 3.5uL of RNAase, extend the incubation step to 25 minutes and reduce the final elution volume to 100mL in order to maximize final DNA yield. Final DNA concentrations were verified with a Qubit spect rometer. I amplified three chloroplast gene regions two coding regions, rbcL and matK and one intergenic spacer, psbA trnH using polymerase chain reaction (PCR). These primers are recommended by the Smithsonian Barcode of Life (BOL) to resolve evolutionary relationships across the spectrum of vascular plants. In combination these loci capture genetic differences in a highly conserved gene ( rbcL ), a moderately conserved gene ( matK ), and a highly variable non coding region ( psbA trnH) (Kress et al. 2009, 2010) I prepared 20uL PCR reactions using Phusion High Fidelity DNA Polymerase (New Engl and BioLabs, Inc., Ipswich, Massachusetts, USA), following the manufacturer's protocol. I also followed Phusion recommendations for Thermocycler programming with the appropriate annealing temperature for each marker. I confirmed successful amplification wi th gel electrophoresis. I sequenced only amplifications that yielded a solitary band during gel electrophoresis I used Sanger sequencing due to the necessity of obtaining long, high quality reads. PCR products were cleaned and sequenced in both forward a nd reverse directions by Beckman Coulter Genomics (Cambridge, Massachusetts, USA) and Genewiz (South Plainfield, New Jersey, USA). Sequence Alignment and Phylogenetic Analyses I used Geneious R9 (Biomatters, Auckland, New Zealand) to assemble contigs for individual gene regions from newly generated sequences. I then aligned newly assembled and

PAGE 25

25 pre existing sequence using the MAFFT 1.3.5 plugin in Geneious. I trimmed pre existing sequences to match the length of the de novo sequences if they contained gene regions more extensive gene regions than new sequences. I checked individual alignments by eye and then concatenated them into a final alignment. The chloroplast is a single, non recombining molecule and thus behaves as a single locus, which is why concat enation of loci is appropriate. In total, I included sequences from 527 taxa in the phylogenetic analyses. I performed maximum likelihood (ML) and Bayesian inference (BI) analyses to determine phylogenetic relationships of the included species. I determi ned the optimal partitioning scheme and model of evolution for each of the three markers using PartitionFinder and the Akaike information criterion (AIC) (Lanfear et al. 2012) The optimal scheme and set of models were used in both the ML and BI analyses. Additionally, I constrained both analyses at the family level in accordance with the classifications in the Angios perm Phylogeny Group IV (APG) (The Angiosperm Phylogeny Group 2016) for angiosperms, and Pteridophyte Phylogeny Group (PPG) for monilophytes (Pterid ophyte Phylogeny Group in press). Based on well established phylogenetic relationships of vascular plants, my phylogenetic analyses are rooted with two lycophyte species ( Selaginella arenicola and S. eatonii ) (Wickett et al. 2014) All analyses were performed on the HiPerGator 2.0 supercomputing cluster at the University of Florida. I used RAxML to perform ML analysis (Stamatakis 2006) with 1000 bootstraps to determine clade support. I selected the tree with the highest likelihood score for use in downstream analyses. I used MrBayes to execute BI analyses (Huelsenbeck et al. 2001) This analysis used two runs with four c hains and 50,000,000 generations each, and default settings. Trees were sampled every 1000 generations and the initial 25% of trees were discarded as burn

PAGE 26

26 in. I combined trees from each of the two runs using MrBayes and summaries the PP results on the BI consensus tree. Community Phylogenetic Analyses I used a three step approach to evaluate relatedness of threatened and endangered, and invasive species in the imperiled pine rockland ecosystem. First, I desi gnated species as threatened, endangered, endemic or invasive Next, I calculated three metrics of phylogenetic relatedness within each species designation group as well as nearest neighbor distances from invasive to threatened and endangered and invasive to native species Finally, I d etermined the sig nificance of these metrics using null models or comparative tests. Species Designations I determined species designations for a) invasive b) threatened or endangered, and c) endemic species using state and federal level rank ings I considered invasive spe cies highly or moderately invasive based on the Florida Exotic Plant Pest Council (FLEPPC) rankings (2015). Highly invasive species (FLEPPC level I) have already altered the structure of the communities in which they are found, while moderately invasive sp ecies (FLEPPC level II) have not yet altered the structure of the communities in which they are found (FLEPPC 2015) I designated threatened and endangered species at the Florida state level following the Florida Department of Agriculture and Consumer Serv ices (FLDACS, referred to as FL threatened and endangered ), and at the federal level following U.S. Fish and Wildlife Servic e (USFWS, referred to as US threatened and endangered ). Finally, South Florida endemic species designations are based on a Florida N atural Area Inventory Multi Species Recovery P lan for south Florida (MRSP 1999 ). I also used broader, more i nclusive rankings of invasive, threatened and endangered, and native for nearest n eighbor (NN) analyses. All species that were not a) invasive, b) t hreatened or endangered, or c) endemic were considered n ative. I considered each of these species designation groups as a

PAGE 27

27 absence data matrix of species by community (x = species, y = community). I used thi s matrix in combination with the best ML tree in downstream analyses ( Table 2 1 ) Metrics of Community Relatedness There are a myriad of metrics available to calculate relatedness between species that share a spatially defined community or species designa tion ( Tucker et al. 2016) The most common metric of relatedness within a community or group is Phylogenetic Di versity (PD) or (Faith 1992) PD metric sums the branch lengths shared by sp ecies within a group. Shared branch lengths indicate relatedness among species in a group; species that share short branch lengths are more closely related than species that share long branch lengths ( Figure 2 1 ) This metric provides a measure of how clos ely related all species are within a group If the species are closely related they will share short branch lengths between then, while if they are distantly related the branch lengths co nnecting to species will be long Community phylogenetic methods als o commonly employ pairwise distance between species to determine relatedness. Mean pairwise distance (MPD) is used to calculate diversity at the base of the tree, while mean nearest taxon distance (MNTD) is used to calculate diversity at the tips of the tr ee ( Figure 2 2 ) ( Swenson 2014; Tucker et al. 2016) MPD is considered a measure of basal diversity for a group of species because it averages the distances between all pairs of species in the tree, capturing the branch lengths that connect even the most distantly related species in the group. MNTD on the other hand averages the distance between each species and its nearest neighbor in the group, thus only the shortest branch lengths between species in a group are being averaged. I used the picante package in R to calculate PD, MPD and MNTD metrics for invasive, threatened and endangered and endemic species groups to answer question s 1, 2 and 3 (Kembel et al 2010)

PAGE 28

28 To answer question 4 I used three more broadly inclusive categories to designated species as threatened and endangered, invasive, or native. I used the nearest neighbor (NN) distance, a nother commonly used metric to comp are phylogenetic relatedness of species NN metrics determine the most closely relating species to a species of interest ( Figure 2 2 ) In order to determine whether invasive speci es are more closely related to native or threatened and e ndangered species I d etermined the NN from both the native and threatened and e n dangered species pool for each i nvasive species. I then compared the distance to NN between the both groups and tested the significance of the dif ference with a paired Wilcoxon rank sum t est ( Figu re 2 2 ) This non parametric test allows us to control for the correlation of each invasive species yielding a distance for the threatened and endangered, and native groups (Lam & Longnecker 1983) Null Model for Metrics of Community Relatedness I used a null model to determine if species within our invasive, threatened and endangered and endemic groups were significantly over dispersed, u nder dispersed, or randomly dispersed within the community phylogeny. I selected a tip swapping null model, which shuffles the names at the tips of the phylogen y while maintaining the evolutionary re lationships. (Swenson 2014) I then randomly dre w the equ ivalent number of species as that found in each group and recalculated the metrics. I repeated these randomizations 1000 times in order to build a distribution of random metric scores. Standardized effect size (SES): was calculated for each species and then a veraged within a group. SES calculations translate the observed values into z scores. Assuming a n alpha of 0.05 under a two tailed test observed SES values 2 indicate significant overdispersion (i.e., species within a group are more distantly

PAGE 29

29 related th an we would expect at random). SES 2.0 indicate significant underdispersion (i.e., species within a group are more closely related than we would expect at random), and scores falling between 2.0 and 2.0 indicate that species are no more closely of dist antly related than we would randomly expect.

PAGE 30

30 Object 2 1 Table of collection and accession numbers for 527 pine rockland species (.pdf file 149 KB)

PAGE 31

31 Tabl e 2 1. Species of interest within the 527 species pine rockland regional pool. Species US Endang US Threat FL Endang FL Threat Endemic Severe Invasive Moderate Invasive Brickellia mosieri x x x Galactia smallii x x x Linum carteri x x x Amorpha herbacea v. crenulata x Warea carteri x Chamaesyce garberi x x Argythamnia blodgetti x x Cham aesyce deltoidea ssp. pinetorum x x Chamaesyce porteriana x x Dalea carthagenesis v. floridana x x Lantana depressa x x Linum arenicola x x Poinsettia pinetorum x x Tephrosia angustissima v. corallicola x x Alvaradoa amorphoides x Chamaesyce deltoidea x Colubrina cubensis v. floridana x Ernodea cokeri x Glandularia maritima x Ipomoea microdactyla x Ipomoea tenuissima x Jacquemontia pentathos x Koanophyllon villosum x Lantana canescens x Phyla stoechadifolia x Scutellaria havanensis x Selaginella eatonii x Spiranthes torta x Thrinax radiata x Tillandsia faciculata v denispicata x Tillandsia utriculata x Trema lamarckianum x Crossopetalum ilicifolium x x Jacquemontia curtisii x x Melanthera parvifolia x x

PAGE 32

32 Table 2 1. Continued Species US Endang US Threat FL Endang FL Threat Endemic Severe Invasive Moderate Invasive Tragia saxifolia x x Angadenia berteroi x Bletia purpurea x Byrsonima lucida x Chaptalia albicans x Chrysophyllum oliviforme x Coccothrinax argentata x Crossopetalum rhacoma x Cynanchum blodgetti x Digitaria filiformis v. dolchophylla x Ilex krugiana x Pithecellobium keyense x Psidium longipes x Pteris bahamensis x Rhynchosia parvifolia x Sachsia polycephala x Senna mexicana v. chapmanii x Smilax havanensis x Solanum donianum x Swietenia mahagoni x Tetrazygia bicolor x Tillandsia balbisiana x Tillandsia flexuosa x Tillandsia variabilis x Tripsacum floridanum x Chamaesyce conferta x Elytraria caroliniensis v. angustifolia x Hedyotis nigricans x Phyllanthus pentaphyllus v. floridana x Ruellia succulenta x Sideroxylon reclinatum x Abrus precatorius x Acacia auriculformis x Albizia lebbeck x Ardisia elliptica x Asparagus aethipicus x

PAGE 33

33 Table 2 1. Continued Species US Endang US Threat FL Endang FL Threat Endemic Severe Invasive Moderate Invasive Bauhinia variegata x Bischofia javanica x Dioscorea alata x Dioscorea bulbifera x Eugenia uniflora x Ficus microcarpa x Imperata brasilensis x Jasminum dichotomum x Jasminum fluminense x Lantana camara x Lygodium japonicum x Melaleuca quinquenervia x Melinis repens x Neyraudia reynaudiana x Paederia cruddasiana x Psidium guajava x Schefflera actinophylla x Schinus terebinthifolius x Syngonium podophyllum x Urena lobata x Agave sisalana x Antigonon leptopus x Callisia fragrans x Cecropia palmata x Crassocephalum crepidioides x Dalbergia sissoo x Eulophia graminea x Ficus altissima x Flacourtia indica x Kalanchoe pinnata x Leucaena leucocephala x Macroptillum lathyroides x Melaleuca viminalis x Melia azedarach x Melinis minutiflora x Momordica charantia x Passiflora biflora x Pteris vittata x

PAGE 34

34 Table 2 1. Continued Species US Endang US Threat FL Endang FL Threat Endemic Severe Invasive Moderate Invasive Richardia grandiflora x Ricinus communis x Spermacoce verticillata x Sphagneticola trilobata x Terminalia catappa x Tradescantia spathacea x

PAGE 35

35 Figure 2 1 Graphical representation of three community relatedness metrics. Top: Phylogenetic Diversity (PD) is the summation of all branch lengths in a community. Bottom left: Mean Pairwise Distance (MPD) is the average of all pairwise distances between species in a community. Bottom right: Mean Nearest Taxon Distance (M NTD) is the average of distances from each species to its nearest neighbor for all species in a community.

PAGE 36

36 Figure 2 2 Schematic for calculating nearest neighbor observed values. Top: Complete pa irwise distance matrix for all invasive (orange), threatened and endangered (magenta), and n ative (teal) species based on phylogenetic relationships. Bottom: Nearest neighbor determined for a single species (large orange triangle) in phylogeny (right) and based on pairwise distances (left).

PAGE 37

37 CHAPTER 3 R ESULTS Phylogenetic Reconstruction Taxon Sampling DNA Extraction, Amplification and Sequencing The aligned matrix for rbcL included 466 taxa (328 sequenced de novo 138 pre existing sequences). The matK matrix included 384 taxa (256 sequenced de novo 128 pre existing sequences), and the psbA trnH matrix included 214 taxa (214 sequenced de novo 0 pre existing sequences). 166 taxa had all three markers, 284 taxa had two regions ( rbcL + matK : 157, rbcL + psbA trnH : 82, matK + psbA trnH : 45) and 77 taxa had only one region ( rbcL : 38, matK : 12, psbA trnH : 19). For taxa below the species level (e.g., with variety or subspecies rankings) that I was unable to sequence de novo or obtain pre existing sequences for, I used sequence data from the FLAS dataset at the species level for that taxon when available. In total, the 527 included taxa represent 101 families, from 40 orders. Sequence Alignment and Phylogenetic Analyses The final concatenated matrix of a ll three loci was 4750 bp long. P artitionFinder analysis det ermined that the GTR +G+ I model best explained the evolution of all three loci with each locus in its own partition In the ML and BI analyses this scheme was used, with GTR+G+I assigned to each partition and model parameters allowed to vary independently for each of them. ML analyses with a family level co nstraint resulted in a best ML tree with a likelihood s core of ln 151287.11 ( Figure 3 1 ). Bootstrap values for the resulting ML phylogeny show strong support throughout the backbone of the tree because most inner branches were constrained by family ( Figure 3 2 ). Overall, the ML analysis found that most con generic species were monophyletic based on sequence data (see D iscussion). PP values were also very high throughout the phylogeny ( Figure 3 3 ).

PAGE 38

38 Comm unity Phylogenetic Analyses Species Designations Among the 527 species included in the phylogeny there were a total of 49 invasive species (Table 3 1 ) ; 25 species were designated as severe invasive species and 24 species were de signated as moderate invasi ve species. There were also 28 FL threatened species, 29 FL endangered s pecies, 1 US threatened species, 5 US endangered species and 22 south FL endemic species ( Figure 3 5, 3 6, 3 7, 3 8) (Table 2 1 ) For the purposes of these analyses I have treated eac h group as its own entity as well as combined the threatened and endangered species at the FL leve l into one group and at the US level into another group. For the PD, MPD and MNTD diversity metrics the FL threatened group does not have any results because there is only one species in the group the community relatedness metrics need multiple species in order to calculate distances Phylogenetic Relatedness of Invasive Species Both moderately and severely invasive species showed random dispersion across t he community phylogeny. Invasive species are no more or less closely related to each other than we would expect at random for PD, MPD and MNTD metrics ( Figure 3 4 ). PD, the sum of branch length within each community, indicates random to non significant und erdispersion for all communities. MPD calculates the basal diversity of a group of species in a community phylogeny by including all branch lengths connecting the group of species. This metric averages all pairwise distances in the group. MNTD metrics are the group are averaged. All three metrics resulted in similar SES values for severely invasive species (range: 0.5 t o 1.1). Moderately invasive species displayed similar PD and MNTD values (PD: 0.8, MNTD: 0.69) while SES MPD values were lower (MPD: 0.06) (Table 3 1 ).

PAGE 39

39 Phylogenetic Relatedness of Threatened and Endangered Species Overall, threatened and endangered spe cies show random dispersion or underdispersion for all three community metrics of relatedness: PD, MPD and MNTD ( Figure 3 4 ). US endangered species SES PD metrics show that FL endangered species are significantly under dispersed (z: 1.99, p: 0.016), and c ombined FL threatened and endangered species and endemic species are also under dispersed, but not significantly (z: 1.90, p: 0. 024; z: 1.67, p:0.038) (Table 3 1). US endangered and combined US threatened and endangered species show patterns of random di spersion within the community phylogeny. All metrics for US threatened species could not be performed because there is only one species in this group. Standardized metrics for MPD indicate combined FL threatened and endangered species as well as endemic sp ecies were significantly under dispersed (z: 2.07, p: 0.019, z: 2.18, p: 0.007, respectively). Individually, FL threatened and FL endangered species showed a trend towards underdispersion, but not significantly (z: 1.48, p: 0.0 60; z: 1.61, p: 0.046) (T able 3 1). US endangered and combined US threatened and endangered species show patterns of random dispersion. Finally, standardized MNTD metrics show that FL endangered species are significantly under dispersed (z: 2.13, p: 0.07) while endemic species a nd combined FL threatened and endangered species and show trends of underdispersion, but not significantly (z: 1.65, p: 0.035; z: 1.95, p: 0.021, respectively) (Table 3 1). US endangered, combined US threatened and endangered, and FL threatened species al l show patterns of random dispersion within the community phylogeny. Nearest Neighbor Analyses for Invasive Species In order to determine whether invasive species are more closely related to native species or threatened and endangered species I determine d distance to nearest neighbor for each invasive

PAGE 40

40 species. 77.6% of invasive species were more closely related to native species, while 22.4% of invasive species were more closely related to threatened and endangered species based on observed NN values ( Fig ure 3 9). Using a paired Wicoxon rank sum test I determined the NN distances to threatened and endangered, and native groups were significantly different (p value <0.0001) (Figure 3 9).

PAGE 41

41 Table 3 1. Z score and p values from SES community relatedness metrics for each group of interest (PD, MPD, MNTD) denotes significant SES values. PD MPD MNTD z score p value z score p value z score p value US Endangered 0.34 0.681 0.66 0.255 0.18 0.437 US Threatened US Threatened and Endangered 0.28 0.644 0.75 0.22 0.01 0.51 FL Endangered 1.99 0.016 1.61 0.046 2.13 0.007 FL Threatened 0.92 0.177 1.48 0.06 1.09 0.125 FL Threatened and Endangered 1.90 0.024 2.07 0.019 1.95 0.021 Endemic 1.67 0.028 2.18 0.007 1.65 0.035 Severe Invasive 0.68 0.274 1.01 0.15 0.50 0.321 Moderate Invasive 0.80 0.785 0.06 0.505 .69 0.769

PAGE 42

42 Figure 3 1 Best ML phylog eny of 527 pine rockland species colored by rank. Inner ring includes: Lycophytes (blue) Monocots (light green), Monilophytes (dark green), Gymnosperms (maroon), Asterids (orange), Sister to Eudicots (teal), Rosids (light blue), Magnolids (purple) Outer ring includes breakdown of Asterids (orange) in to Lamiids (pink) and Campanulids (red), and breakdown of Rosids (aqua) into Fabids (violet) and Malvids (dark purple).

PAGE 43

43 Figure 3 2 Best ML phylogeny containing 527 pine rockland s species with bootstrap support values (red: high support, blue: low support).

PAGE 44

44 Figure 3 3 BI consensus phylogeny containing 527 pine rockland s species with posterior probability support values (red: high support, blue: low support).

PAGE 45

45 Figure 3 4 Diversity metrics (PD, MPD and MNTD) for US threatened and endangered species (green), FL threatened and endan gered species (blue), endemic species (orange) and moderate and severe invasive species (purple). Observed z score values below 2 indicate significant underdisperson, values between 2 and 2 in dicate random dispersion and values above 2 indicate overdispe rsion.

PAGE 46

46 Figure 3 5 Best ML phylogeny with species of special interest denoted with dots at the tips. Inner ring: US threatened and endangered species, green. Second ring: FL threatene d and endangered species, blue Third ring: south Florida endemic species, orange. Outer ring: moderate a nd severe invasive species, purple

PAGE 47

47 Figure 3 6 Best ML phylogeny with moderate and severe invasiv e species tips denoted with purple dots.

PAGE 48

48 Figure 3 7 Best ML phylogeny with US threatened and endangered spec ies tips denoted with green dots.

PAGE 49

49 Figure 3 8 Best ML phylogeny with FL threatened and endang ered species tips denoted with blue dots (inner circle), and south Florida endemic species denoted with orange dots (outer circle).

PAGE 50

50 Figure 3 9 Boxplot observed n earest neighbor distances from i nvasive to n ative species (teal, left), and invasive to threatened and e ndangered species (magenta, right)

PAGE 51

51 CHAPTER 4 DISCUSSION Community Phylogenetic Analyses Community phylogenetic analyses of Pine Rocklands c an be used to answer questions of community assembly as well as dispersion of traits and clades within the regional phylogeny. Pine Rocklands are an exceptionally unique and globally endangered system with invasive, threatened, endangered, and endemic spec ies persisting in the same local area. This system allows us to ask questions about the phylogenetic dispersion of species in a regional pool where invasive species are interacting the federally threatened and endangered species. Further, Pine Rockland eco systems were historically more continuous and fire driven, but are now highly fragmented and fire suppressed. We can explore the community assembly of this habitat in historical evolutionary context while also trying to understand how modern day communitie s are being maintained or disassembling with anthropogenic change and invasive species. Using a newly constructed community phylogeny of 527 species from the globally imperil ed pine rockland ecosystem (Figure 3 1 ) I took a four step approach to evaluate r elatedness between invasive, and threatened and endangered species. I found that invasive species are randomly dispersed throughout the community phylogeny while threatened, endangered, and endemic species are under dispersed. Finally, I determined that in vasive species are more closely related to their nearest native species than they are to their nearest threatened and endangered species. Invasive Species Exploring the relatedness of moderate and severe invasive species within the pine rockland regional p ool allows us to understand if certain groups of species are primarily responsible for invasion across the ecosystem. Using three metrics of community relatedness

PAGE 52

52 PD, MPD and MNTD I found that invasive species found in pine rockland ecosystems are no m ore or less closely related to each other than we would expect (Figure 3 4, 3 6 ). It has been shown in Australian ecosystems that at local scales invasive plant species are also randomly dispersed, while at the continental scale these same invasive species are under dispersed (Cadotte et al. 2009b) This difference in pattern across scales may be the result of competition for resources between species at a local scale while at larger scales the signature of human sel ection for certain traits (e.g, large blooms) drive introductions and persistence of invasive species (Cadotte et al. 2009b) A similar trend may be occurring with in the pine rockland ecosystem. Many of the invas ive species are introduced horticultural species from the surrounding Miami urban areas. To better explore the drivers of invasion we need to incorporate functional trait and local scale information because while species show random phylogenetic dispersion trait dispersion may be non random (Cavender Bares et al. 2004, 2009; Swenson 2013) Non ran dom trait dispersion could indicate that certain traits, such as ability to recover quickly from fire, or high dispersal ability determine whether or not a species is a good invader in the pine rockland ecosystem. Further, we need to correlate trait inform ation with communities under different fire regimes or across fragments of different sizes. It may be that some the species with certain traits (e.g., rapid recovery from fire) dominate in well burned fragments while other species with different traits (e. g., fast growth and woodiness) dominate in fragments with low fire frequency (Cavender Bares et al. 2004; Cadotte et al. 2009a) Threatened, Endangered, and Endemic Species The globally imperiled pine rockland ecosystem to home to a multitude of threatened and endangered species I was interested in determining if there was a phylogenetic signal to their threat ened or endangered status (MSRP 1999). I found that three community relatedness metrics

PAGE 53

53 PD, MPD, and MNTD reveal similar values for all rankings of the threatened and endang ered species groups (Figure 3 4, 3 7 ). US threatened and endangered species show trends toward underdispersion, but not significantly, while FL threatened and endangered species as a whole are significan tly under dispersed (Figure 3 4, 3 8 ). The US threatened and endangered groups consist of 6 spe cies while there are 60 FL threatened and endangered species with four taxonomic families contributing heavily to the list: Euphorbiaceae, Fabaceae, Orchidacea e, and Bromeliaceae (Figure 4 1 ). Previous studies examining phylogenetic signal of extinction an d risk of extinction have shown that species in danger of extinction are very closely related to each other and share life history traits that endanger them (Purvis et al. 2000a, 2000b; Purvis 2008) Birds and mammals that have gone extinct or are in danger of becoming extinct are generally large bodied, s low to grow, and slow to reproduce (McKinney 1997; Cardillo et al. 2008) Frogs and corals with high extinction risks also exhibit traits of slow growth, but their risk of extinction is compounded by requirements o f narrow, optimal environmental conditions and low plasticity in ability to adapt to changing environmental conditions (Bielby et al. 2008; Carpenter et al. 2008; Huang & Roy 2015) In another globally imperiled region, the Cape Floristic Region in South Africa, Davies et al., 2011 showed plant taxa within rapidly diversifying clades were more likely to go extinct. As a result the Cape Floristic Region is referred to as both a nursery and a graveyard of plant diversity (Davies et al. 2011) Many of these same traits may be driving pine rockland plant species to become increasingly rare. Species representing the Orchidaceae and Bromeliacae families are on the threatened and endangered lists be cause they are highly sought out for personal collections throughout the state (Figure 4 1). These clades are both grow slowly, bloom infrequently, and require very precise e for orchid

PAGE 54

54 and bromeliad specimens warrants placing them on the threatened and endangered list. Pine rockland ecosystem also hosts a great diversity of endemic and native Fabaceae species (Figure 4 1). Many of these threatened and endangered Fabaceae spe cies are very closely related and have limited ranges. As pine rockland habitat is lost to fragmentation, urban encroachment, and fire suppression so too is habitat available to these species (Figure 4 1). Finally, species in the Euphorbiaceae family are arguably the most iconic pine rockland esque group of species in the FL threatened and endangered list (Figure 4 1). Many of these of the high light environment atop bare limestone that is available after frequent fires (Figure 4 1). Pine rockland Euphorbiaceae species are very small in size, and require precise environmental conditions to flourish (Herndon 1993) Increased fragmentation, changes in fire regime, and encroachment by fast growing large invasive species will bring about an increasing risk of extinction for all species that exhi bit these specialized traits. Overall, threatened and endangered species may share a suite of evolutionarily derived traits that allows them to persist in the difficult pine rockland habitat, a relic of environmental filtering (Cavendar Bares et a l. 2004; Verd & Pausas 2007) Such traits include small size, preference for open canopy high light conditions, and an ability to thrive in xeric limestone based habitats. These species may also share traits such a slow growth, fire dependence, or poor dispersal that endanger them in this increasingly fragmented and fire suppressed habitat (Cardillo et al. 2008; Carpenter et al. 2008) South Florida endemic species also showed patterns of underdispersion (significant for MPD, but not quite for PD and MNTD) (Table 3 1; Figure 3 4 ). These species, in most cases, are a subset of those species found in the FL threatened and endangered list (Figure 3 8 ). They are

PAGE 55

55 Endemic species have the specialized traits required for persistence in pine rockland habitat, and therefor e set the standard for species of special concern within this habitat. Similarly, species designations as threatened and endangered, and invasive are b ased on the subjective human decisions FL threatened and endangered species can originate in any habita t in Florida and are considered equally threatened or endangered across the state of Florida by FLDACS. Many of the species included in the pine rockland list and FL state threatened an d endangered list are actually rockland h ammock specializing species th at are only found in highly fire suppressed pine rockland fragments. We have used FL state threatened and endangered lists here as a way to denote species of importance, but if we are interested in maintaining the biodiversity of a specific habitat we need to inform the analyses with more fine grain information. South Florida endemic species may be better indicators of species susceptible to risk of extinction in pine rockland ecosystems than species that have been approved by a committee to be listed. Many species in the current analysis listed as threatened and endangered or invasive are truly rockland hammock species, such as the orchids and bromeliads (Figure 4 1). When pine rockland fails to burn for 20 or more years the habitat succeeds to rockland ham mock. Unlike pine rocklands which have an open canopy, extremely high light conditions, exposed limestone, and a rich understory of herbaceous species, rockland hammock have a closed canopy, low light conditions, a relatively depauperate understory and ric h diversity of woody species. Orchids and bromeliads thrive in the cooler, more humid conditions of rockland hammocks, and only appear in pine rocklands that are highly disturbed or not recently burned.

PAGE 56

56 Inclusion of rockland hammock specialist species in the threatened and endangered group may skew this analysis. The repercussions of this sort of inclusion range from obscuring important patterns to altering our ability to predict which species face future risk of extinction. Multiple species of closely rel ated orchids and bromeliads drive some of the underdispersion pattern that we see with the FL threatened and endangered group. Importantly, these species are not good indicators of pine rockland threatened and endangered species; endemic species paint a mo re accurate picture, even though they do not yield significant result s under all metrics (Table 3 1; Figure 3 4 ) (Diez et al. 2009; Schaefer et al. 20 11) Invasive species also need to be more tightly informed by the habitat and botanical experts of that habitat. Quite a few species included in this phylogeny were categorized as native for analyses, but were designated as invasive by local experts. Previous studies have considered all non native species to be invasive (Schaefer et al. 2011; Marx et al. 2015) while others have advocated for considering abundances and disturbance to better select invasive species (Cavender Bares et al. 2009; Diez et al. 2009; Schaefer et al. 2011; Swenson 2014) It is essential to inform phylogenetic analyses with high level, habi tat specific natural history information to make analyses biologically relevant and robust. Relatedness of Invasive to Threatened and Endangered, and Native Species Ecologists have long been interested in the relationships of invasive species to a group of native species. Darwin proposed that we might expect invasive species to be closely related to the native pool in order to overcome environmental filters put forth by the habitat and allow persistence of the invasive species (Darwin 1859) However, if an invasive species is too closely related to the native pool, competition for limited resources may exclude the invasive species (Darwin 1859) From this conundrum we can imagine two possible routes for successful invasion: 1) Invasive species will be very closely related to the native species pool if

PAGE 57

57 environmental filtering drives community assembly, or 2) invasive species will be distantly r elated to the native pool in order to either a) take advantage of available, unexploited resources or b) outcompete species in the pre existing community (Darwin 1859; Fridley & Sax 2014) Here, I consider the relative relatedness of invasive species to not only the native species pool, but also the threatened and endangered species. Globally imperiled pine rockland habitat is a n ideal system for exploring this interaction because as I have shown threatened, endangered, and endemic species in this ecosystem are closely related and may share traits that allowed them to persist under very specific environm ental conditions. Howeve r, with anthropogenic change conditions, invasive species may not experience the historical environmental filters once imposed on the native, and threatened and endangered species pool. Instead, invasive species may have novel transitional habitat types av ailable to them in increasingly fragmented and fire suppressed pine rockland ecosystem. I determined nearest neighbor distances from invasive species to threatened and endangered species and invasive species to native species for each of the 49 designated invasive species in the regional pool. I found that 77.6% of invasive species were more closely related to their nearest native neighbor than their nearest threatened and endangered neighbor (Figure 3 9 ). I confirmed that differences in NN distances were s ignificant (p value < 0.0001) using a pair Wilcoxon rank sum test (Figure 3 9). My results indicate that invasive species exhibit life history traits that are more similar to native species than those of the threatened and endangered species. Non rare nati ve species have been able to adapt to changing environmental conditions in pine rockland habitats. Successful invasive species may share these adaptive strategies life history traits) rather than the strategies of threatened and endangered species that are

PAGE 58

58 species were closely related to threatened and endangered species it might follow that invasive species share similar traits and are cau sing threatened and endangered species to become increasingly rare, however this does not seem to be the case. Rather, invasive species are taking advantage of pine rockland habitat that is less frequently burned, more over grown and hammock like the exa threatened, endangered and e n demic species. Issues with Phylogenetic Analyses The pine rockland community phylogeny presented here is the key to many possible community phylogenetic analyses. Thus, it is essential that relationships within the phylogeny are as accurate as possible. To assure proper placement of taxa with the phylogeny I imposed a family level constraint to the phylogenetic ML and BI analyses (Muscarella et al. 2014) Within a family, species topology and branch lengt hs were allowed to vary and in the resulting phylogenies many of the sub family relationships have strong boots trap and PP support (Figure 3 2, 3 3 ). Community phylogenetic analyses evaluate branch lengths or determine nearest neighbor distances to species of interest. In this study I focused on evaluating relatedness within threatened, endangered, endemic and invasive species within theses species are inappropriately placed in the phylogeny it will skew the analyses. Topological issues involving threatened, endangered or invasive species include: Ipomoea microdactyla (threatened or endangered) was not monophyletic with the remainder of the Ipomoea spp. (BS: 100). Two Phyla spp. (threatened or endangered) were found to be successive sisters to the Lantana spp. (invasive, and threatened or endangered) clade (BS: 42). The clade including Spermacoce spp. (invasive), Ernodea spp. (threatened or endangered), Richadia spp. (threatened or endangered) and Diodia teris has stro ng bootstrap support (BS: 100), but within the clade the Ernodea spp. and the Spermacoce spp. are both polyphyletic with

PAGE 59

59 much lower internal bootstrap values (BS: 62, 14). Sachia polycephala (invasive) was determined to fall within the Liatris spp. clade i n this phylogenetic analysis (BS: 81). Finally, within the clade containing Chamaesyce spp. (threatened or endangered), Euphorbia pinetorum and Poinsettia spp. (threatened or endangered) congeneric species are not determined to be sister. Support at the ba se of this clade is strong (BS: 100), but internal clade support is very low (BS : 28 41). These species have a very nebulous taxonomy thus low support and congeneric mixing is not wholly unexpected (Herndon 1993; Bradley & Gann 1999) In the BI phylogeny one clade of species ( Carica sp., Polanisia sp,, Warea sp., and sister clade) appears to be sister to monilophytes rather than within the angiosperm clade where we would expect them to be (Wickett et al. 2014) This topology is probably a result a mistake in constructing the family level constraint. Inconsistencies in the BI analysis will o nly affect the PP support values shown for this clade, not downstream analyses because the best ML tree was used to calculated community relatedness metrics. I hope to improve taxon and sequence sampling in future iterations of this study. Kress et al., ( 2009) showed that it was possible to resolve a similar phylogenetic analyses with 69% sequence sampling using only rbcL matK and psbA trnH (Kress et al. 2009) Sequence sampling in this study is closer to 41%, and bolstering the number of sequences included per taxon will cla rify some of these problematic relationships. Limited sequence sampling for a few taxon might cause issues because matK and psbA trnH sequences are so variable that for species where that is the only sequence included sequence data may place them in the tr ee improperly (Kress et al. 2010) Future Directions Pine rockland ecosystem is an ideal locale to study interactions of threatened and endangered species with invasive species. This globally imperiled ecosystem has, not only a

PAGE 60

60 plethora of endemic, and threatened and endangered species, but also history of fire suppression and fra gmentation that has lead to the persistence of a multitude of invasive species. Here, I examined relatedness of species groups based on different designations, but in future studies I aim to assess these patterns at a local level. Fragments of this habitat vary greatly in time since most recent fire, fire frequency, fragment size and fragment location. By evaluating patterns of relatedness among all species with a certain designations (i.e., all invasive species) I have essentially conglomerated all species interactions within fragments to one global habitat. Thus, the patterns of relatedness we see at a global level may not match the patterns we observed for species intera cting one on one in the field. Fragmented pine rockland habitats also present a uniqu e opportunity to explore interactions between species groups of high interest invasive species and threatened and endangered species within a single isolated habitat type. We can take advantage of this unique system by incorporating abundances data int o my analyses (Richardson Cavendar Bares et al. 2009; Swenson 2014) When we consider only global groups of species or presence based species data we assume that all species are equally abundant, and equally interacting, though we know this to be untrue. If invasi ve species are much more abundant than threatened and endangered species we could better weight community phylogenetic analyses of relatedness such as this one. We aim to predict which species could present a threat of invasion or which species are at risk of local extinction based on the relatedness of already known invasive or threatened and endangered species. We may be able to improve our ability to determine invasion threat based on the most abundant existing invasive species or species to watch for co nservation based on relatedness and abundance of known threatened and endangered species. We may also be able to use the traits of abundant native species to predict future

PAGE 61

61 invaders because I have shown in this study that they are closely related. If succe ssful pine rockland invaders share life history traits of successful native species, we may be able to predict future successful invaders based on phylogenetic relatedness as well as trait similarities. With more fine scale data on both the invasive and th reatened and endangered species that play a major role in pine rockland habitat we can improve community phylogenetic analyses in the hope of being able to make predictions about future invasive and threatened species. By exploring relationships between in vasive, and threatened and endangered species in a community phylogenetic context in pine rockland we are able to understand broadly how communities are maintained, change over time and may appear in the future. Phylogenetic signal to invasions and extinct ion have generally been handled separately across a variety of taxa and a range of scales. Assessing phylogenetic signal of invasion and risk of extinction in one, globally imperiled, highly disturbed, spatially limited ecosystem we can incorporate both of these fields to understand the interactions between multiple species groups of interest. This approach would be applicable and useful in many other ecosystems where invasive and threatened and endangered species interact. Here, I present the first ever co mmunity phylogeny for globally imperiled pine rockland taxa. This phylo geny includes sequences from 381 newly collected field samples supplemented wit h pre existing sequences for 143 plant taxa. The pine rockland community phylogeny I have developed will i nform a plethora of analysis including those that examine patterns of community assembly, maintenance and disassembly across groups of taxa and ac ross local and regional scales.

PAGE 62

62 Figure 4 1 Representatives of threatened and endangered species found in pine rockland habitat. Clockwise from top right: Galactia smallii (Fabaceae), Ency cli a tampensis (Orchidaceae), Tillandsia fasiculata var. densipicata (Bromeliaceae), Chamaesyce deltoidea (Euphorbiaceae) Photo credit: Emily B. Sessa, and Zachary Siders.

PAGE 63

63 LIST OF REFERENCE S Bielby J, Cooper N, Cunningham AA, Garner TWJ, Purvis A. 2008. Predicting susceptibility to 1 :82 90. Bradley KA, Gann GD 1999. Status summaries of 12 rockland plant taxa in southern Florida. Citeseer. Available from http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.669.496&rep=rep1&type=pdf (accessed October 14, 2016). Cadotte MW, Cavender Bares J, Tilman D, Oakley TH. 2009a. Using phylogenetic, functional and trait diversity to understand patterns of plant community productivity. PloS one 4 :e5695. Cadotte MW, Hamilton MA, Murray BR. 2009b. Phylogenetic relatedness and plant invader success across two spatial scales. Diversity and Distributions 15 :481 488. Cardillo M, Mace GM, Gittleman JL, Jones KE, Bielby J, Purvis A. 2008. The predictability of extinction: biological and external correlates of decline in mammals. Proceedings of the Royal Society of London B: Biolog ical Sciences 275 :1441 1448. Candidate Species Report. (n.d.). Retrieved March 5, 2015, from U.S. Fish and Wildlife Service, Environmental Conservation Online System, http://ecos.fws.gov/tess_public/reports/candidate species report. Carpenter KE et al. 20 08. One third of reef building corals face elevated extinction risk from climate change and local impacts. Science 321 :560 563. Cavender Bares J, Ackerly DD, Baum DA, Bazzaz FA. 2004. Phylogenetic overdispersion in Floridian oak communities. The American N aturalist 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 KH, Fine PV, Kembel SW. 2009. The merging of community ecology and phylogenetic biology. Ecology letters 12 :693 715. Darwin C. 1859. On the origin ofspecies by means of natural selection, or the preservation of favoured races in the struggle for life. Davies TJ et al. 2011. Extinction risk and diversification are linked in a plant biodiversity hotspot. PLoS Biol 9 :e1000620. Diez JM, Williams PA, Randall RP, Sullivan JJ, Hulme PE, Duncan RP. 2009. Learning from failures: testing broad taxonomic hypotheses about plant naturalization. Ecology Letters 12 :1174 1183.

PAGE 64

64 Endangered, Threatened and Commercially Exploited Plants of Florida. (October 2, 2015). Retrieved July 15, 2016 from Florida Department of Agriculture and Consumer Services, ht tp://www.freshfromflorida.com/Divisions Offices/Plant Industry/Bureaus and Services/Bureau of Entomology Nematology Plant Pathology/Botany/Florida s Endangered Plants/Endangered Threatened and Commercially Exploited Plants of Florida Fahrig L. 2003. Effe cts of habitat fragmentation on biodiversity. Annual review of ecology, evolution, and systematics:487 515. Faith DP. 1992. Conservation evaluation and phylogenetic diversity. Biological conservation 61 :1 10. f Invasive Plant Species. (2015). Retrieved April 13, 2015, from Florida Exotic Pest Plant Council, http:// www. fleppc.org Fridley JD, Sax DF. 2014. The imbalance of nature: revisiting a Darwinian framework for invasion biology. Global Ecology and Biogeography 23 :1157 1166. Gann, G.D. et al. (2001 2015). The Floristic Inventory of South Florida Database Online, from the Instit ute of Regional Conservation, http://regionalconservation.org. Gann, G.D. 2015. Vascular Plant Species of Management Concern in Everglades National Park: Final Report. The Institute of Regional Conservation. pp. i 216. Gause GF. 1932. Experimental studie s on the struggle for existence. Journal of experimental biology 9 :389 402. Herndon A. 1993. Notes on Chamaesyce (Euphorbiaceae) in Florida. Rhodora:352 368. Huang D, Roy K. 2015. The future of evolutionary diversity in reef corals. Philosophical Transacti ons of the Royal Society of London B: Biological Sciences 370 :20140010. Huelsenbeck JP, Ronquist F, others. 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17 :754 755. Hutchinson GE. 1959. Homage to Santa Rosalia or why are there so many kinds of animals? The American Naturalist 93 :145 159. Naturalization Hypothesis. The American Naturalist 175 :415 423. Kembel SW, Cowan PD, Helmus MR, Cornwell WK, Mo rlon H, Ackerly DD, Blomberg SP, Webb CO. 2010. Picante: R tools for integrating phylogenies and ecology. Bioinformatics 26 :1463 1464.

PAGE 65

65 Kress WJ, Erickson DL, Jones FA, Swenson NG, Perez R, Sanjur O, Bermingham E. 2009. Plant DNA barcodes and a community ph ylogeny of a tropical forest dynamics plot in Panama. Proceedings of the National Academy of Sciences 106 :18621 18626. Kress WJ, Erickson DL, Swenson NG, Thompson J, Uriarte M, Zimmerman JK. 2010. Advances in the use of DNA barcodes to build a community ph ylogeny for tropical trees in a Puerto Rican forest dynamics plot. PLoS one 5 :e15409. Lack D. 1973. My life as an amateur ornithologist. Ibis 115 :421 431. Lam FC, Longnecker MT. 1983. A modified Wilcoxon rank sum test for paired data. Biometrika 70 :510 513 Lanfear R, Calcott B, Ho SY, Guindon S. 2012. PartitionFinder: combined selection of partitioning schemes and substitution models for phylogenetic analyses. Molecular biology and evolution 29 :1695 1701. MacArthur RH, Wilson EO. 1967. The theory of island biography. Princeton, New jersey. MacArthur R, Levins R. 1967. The limiting similarity, convergence, and divergence of coexisting species. American naturalist:377 385. Conundrum in the San Juan Islands using community phylogenetics and functional traits. Diversity and Distributions. Available from http://onlinelibrary.wiley.com/doi/10.1111/ddi.12401/pdf (accessed September 29, 2016). Mayfield MM, Levine JM. 2010. Opposi ng effects of competitive exclusion on the phylogenetic structure of communities. Ecology Letters 13 :1085 1093. McKinney ML. 1997. Extinction vulnerability and selectivity: combining ecological and paleontological views. Annual Review of Ecology and System atics:495 516. Morlon H, Schwilk DW, Bryant JA, Marquet PA, Rebelo AG, Tauss C, Bohannan BJ, Green JL. 2011. Spatial patterns of phylogenetic diversity. Ecology letters 14 :141 149. Multi Species Recovery Plant for South Florida: Pine Rocklands: U.S. Fish and Wildlife Service. pp. 3 161 3 194. Muscarella R, Uriarte M, Erickson DL, Swenson NG, Zimmerman JK, Kress WJ. 2014. A Well Resolved Phylogeny of the Trees of Puerto Rico Based on DNA Barcode Sequence Data: e112843. PLoS One 9 :e112843. Olden JD. 2006. Bi otic homogenization: a new research agenda for conservation biogeography. Journal of Biogeography 33 :2027 2039. Peay KG, Belisle M, Fukami T. 2012. Phylogenetic relatedness predicts priority effects in nectar yeast communities. Proceedings: Biological Scie nces 279 :749 758.

PAGE 66

66 Possley J, Woodmansee SW, Maschinski J. 2008. Patterns of plant composition in fragments of globally imperiled pine rockland forest: effects of soil type, recent fire frequency, and fragment size. Natural Areas Journal 28 :379 394. Powell D, Maschinski J. 2012. Connecting fragments of the pine rockland ecosystem of South Florida: the connect to protect network. Ecological Restoration 30 :285 288. Pteridophyte Phylogeny Group. A community derived classification of extant lycopods and ferns. J ournal of Systematics and Evolution, in review. Purvis A. 2008. Phylogenetic approaches to the study of extinction. Annual Review of Ecology, Evolution, and Systematics:301 319. Purvis A, Agapow P M, Gittleman JL, Mace GM. 2000a. Nonrandom extinction and t he loss of evolutionary history. Science 288 :328 330. Purvis A, Jones KE, Mace GM. 2000b. Extinction. BioEssays 22 :1123 1133. and community invasibility. Progress i n Physical Geography 30 :409 431. naturalization hypothesis in the Azores. Ecology Letters 14 :389 396. Slingsby JA, Verboom GA. 2007. Phylogenetic relatedness limits co occu rrence at fine spatial scales: Evidence from the Schoenoid Sedges (Cyperaceae: Schoeneae). South African Journal of Botany 73 :313. Stamatakis A. 2006. RAxML VI HPC: maximum likelihood based phylogenetic analyses with thousands of taxa and mixed models. Bio informatics 22 :2688 2690. Swenson NG. 2013. The assembly of tropical tree communities the advances and shortcomings of phylogenetic and functional trait analyses. Ecography 36 :264 276. Swenson NG. 2014. Functional and phylogenetic ecology in R. Springer. Available from http://link.springer.com/content/pdf/10.1007/978 1 4614 9542 0.pdf (accessed April 3, 2016). The Angiosperm Phylogeny Group. 2016. An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: A PG IV. Botanical Journal of the Linnean Society 181 :1 20. Threatened and Endangered. (January 2014). Retrieved March 5, 2015, from United States Department of Agriculture Natural Resources Conservation Service, http://plants.usda.gov.

PAGE 67

67 Tucker CM et al. 2016. A guide to phylogenetic metrics for conservation, community ecology and macroecology. Biological Reviews. Available from http://onlinelibrary.wiley.com/doi/10.1111/brv.12252/pdf (accessed October 14, 2016). Vamosi SM, Heard SB, Vamosi JC, Webb CO. 20 09. Emerging patterns in the comparative analysis of phylogenetic community structure. Molecular ecology 18 :572 592. Verd M, Pausas JG. 2007. Fire drives phylogenetic clustering in Mediterranean Basin woody plant communities. Journal of Ecology 95 :1316 13 23. Webb CO, Ackerly DD, McPeek MA, Donoghue MJ. 2002. Phylogenies and community ecology. Annual review of ecology and systematics:475 505. Wickett NJ et al. 2014. Phylotranscriptomic analysis of the origin and early diversification of land plants. Proceed ings of the National Academy of Sciences 111 :E4859 E4868.

PAGE 68

68 BIOGRAPHICAL SKETCH Lauren B. Trotta, originally from Gui lford, CT, received her B.S in b iology from Providence College in 2013. As an undergraduate she fostered her lifelong fascination i n ecology by taking advantage of opportunities to travel, explore and learn about marine eco systems ranging from Puget Sound in Washington to Narragansett Bay in Rhode Island to the barrier reef in Belize. Lauren began her MS in Wildlife Ecology and Conse rvation at the University of Florida under the co advisement of Dr. Benjamin Baiser and Dr. Emily B. Sessa in 2014 and has continued to expand her exploration of ecosystems. Her study system, sou rocklands, has proven to be an exc eptionall y inspiring landscape and she looks forward continuing her investigations of this globally imperiled habitat in the future.