University of Florida Thesis Or Dissertation Formatting

GENETIC AND ENVIRONMENTAL CONSIDERATIONS FOR ACROPORID CORAL RESTORATION

KATHRYN ELAINE LOHR

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2018

To my family

ACKNOWLEDGMENTS

I would first like to thank my advisor, Dr. Joshua Patterson for giving me the opportunity to complete my studies at UF. The guidance and mentorship he has provided over the past four years have been invaluable, and his advice and support have been deeply rewarding for my studies and my career. Sincere thanks also go to my supervisory committee members, Drs. Matt DiMaggio, Tom Frazer, Lyn Gettys, and

Margaret Miller, for providing constructive advice and direction that have greatly improved the quality of my work. I would also like to thank Dr. Mark Flint for offering valuable early feedback on my work.

My graduate research would not have been possible without the assistance of a number of individuals and institutions. First, I would like to thank the Tampa Electric

Company for providing me with office space at the Florida Conservation and

Technology Center during my time as a student. Chapters 2-4 of this dissertation would not have been possible without the Coral Restoration Foundation (CRF), who allowed me to conduct research involving their nursery and corals. Nursery-based studies were conducted under permit FKNMS-2011-159-A3 issued to CRF (Chapter 2), and permit

FKNMS-2016-129-A1 issued to Dr. Joshua Patterson (Chapter 3). I would particularly like to thank Kayla Ripple for organizing field logistics throughout and for collaborating on Chapter 4. I’d also like to thank CRF for providing supplementary outplant data for

Chapter 4, and Amelia Moura for coordinating data sharing. The Florida Aquarium provided field and logistical support for the initial outplanting work in Chapter 4, and specific thanks go to Keri O’Neil, Tim Stripling, Chris O’Neil, and Scott Graves for their assistance. Outplanting for this study was conducted under permit FKNMS-2011-159-

A4 issued to CRF. Sincere thanks also go to the University of Florida’s Southeast

Center for Integrated Metabolomics for funding the metabolomics work in Chapter 3, as well as Drs. Tim Garrett, Matt Merritt, Ram Khattri, and Joy Guingab-Cagmat for their collaboration and feedback on this work. A number of University of Florida, CRF, and

Florida Aquarium volunteers also provided indispensable field support for Chapters 2-4 of this dissertation.

Field work for Chapter 5 was conducted in collaboration with the Central

Caribbean Marine Institute (CCMI) in Little Cayman. Thanks to Dr. Carrie Manfrino for assistance in coordinating this collaboration, and to a number of CCMI staff members and students for assistance with field work. Particular thanks go to Aimee Cook McNab and Lowell Forbes.This work was supported by the National Science Foundation

Research Experiences for Undergraduates program (OCE-1358600), CCMI, and the

University of Florida. I also thank Krystan Wilkinson of the Chicago Zoological Society c/o Mote Marine Laboratory and the University of Florida for creating Figure 5-2.

I would like to sincerely thank Drs. Emma Camp, David Suggett, and

Unnikrishnan Kuzhiumparambil from the University of Technology, Sydney and Dr. Bill

Leggat from the University of Newcastle for providing intellectual guidance, logistical support, and the research experience of a lifetime on the Great Barrier Reef (the results of which are detailed in Chapter 6). I’d also like to thank the staff of the University of

Queensland’s Heron Island Research Station and Gus Fordyce for field and lab assistance on this study. Finally, I’d like to thank Dr. Adrian Lutz of the University of

Melbourne for providing critical input on the metabolomic data presented in Chapter 6.

Funding for the work in Chapter 6 was provided by the Australian Research Council

(ARC; DP160100271 issued to Drs. David Suggett and Bill Leggat), with additional

support from an ARC Future Fellowship to Dr. David Suggett (FT130100202) and the

University of Florida Foundation through the Evelyn Timm Koenig Memorial Endowment

Fund. Coral collections in Chapter 6 were performed under permits G15/37488.1 and

G16/38534.1 issued to Dr. Bill Leggat.

A vast number of University of Florida staff, interns, and volunteers have been key to the completion of my studies. I’d especially like to thank the SFRC team for administrative assistance throughout my time as a student, Cheryl Thacker for assistance with dive logistics, and all my JPL labmates – Kelli O’Donnell, Rebecca

Lucas, Kailee Schulz, Joe Henry, and Aaron Pilnick – for helping out with everything from field work to figure preparation. My unique office location provided me with a number of other coworkers who have provided support, guidance, and friendship over the years, so thanks also go to the entire FCTC crew. I’d also like to thank all the incredible friends who provided advice and encouragement throughout my academic experience.

In closing, I’d like to thank my parents, Richard and Jill, who have always supported me throughout my academic journey, my brother, Kevin, who has one last chance to show up to one of my graduation ceremonies, and of course, Stanley and

Copernicus. Finally, I’d like to say a huge thank you to Luke Miller. He has gone above and beyond for me these past years, and I appreciate his love and support more than words can express.

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... 4

LIST OF TABLES ...... 9

LIST OF FIGURES ...... 10

LIST OF ABBREVIATIONS ...... 12

ABSTRACT ...... 13

CHAPTER

1 GENERAL INTRODUCTION ...... 15

2 INTRASPECIFIC VARIATION IN PHENOTYPE AMONG NURSERY-REARED ACROPORA CERVICORNIS ...... 26

Introduction ...... 26 Methods ...... 28 Results ...... 31 Discussion ...... 37

3 METABOLOMIC PROFILES DIFFER AMONG UNIQUE ACROPORA CERVICORNIS GENOTYPES ...... 43

Introduction ...... 43 Methods ...... 46 Sample Collection and Extraction ...... 46 1H-NMR Profiling ...... 48 LC-MS Global Metabolomics ...... 49 Statistical Analysis ...... 50 Results ...... 51 1H-NMR Profiling ...... 52 LC-MS Global Metabolomics ...... 52 Discussion ...... 63

4 DIFFERENTIAL DISTURBANCE EFFECTS AND PHENOTYPIC PLASTICITY AMONG RESTORED ACROPORA CERVICORNIS AT PATCH AND FORE REEF SITES ...... 72

Introduction ...... 72 Methods ...... 76 Outplanting Experiment ...... 76 Expanded Dataset ...... 78

Data Analysis ...... 78 Results ...... 81 Discussion ...... 87

5 ASSESSMENT OF WILD AND RESTORED ACROPORA CERVICORNIS ACROSS THREE REEF ZONES IN THE CAYMAN ISLANDS ...... 95

Introduction ...... 95 Methods ...... 99 Wild Population Surveys...... 99 Outplanting Experiment ...... 100 Statistical Analysis ...... 102 Results ...... 103 Wild Population Surveys...... 103 Outplanting Experiment ...... 103 Discussion ...... 110

6 RESOLVING THE DYNAMICS OF CORAL PHOTOACCLIMATION THROUGH COUPLED PHOTOPHYSIOLOGICAL AND METABOLOMIC PROFILING ...... 115

Introduction ...... 115 Methods ...... 118 Experimental Design ...... 118 Photophysiology ...... 120 Metabolome Profiling ...... 121 GC-MS ...... 122 LC-MS ...... 123 Data Analysis ...... 124 Results ...... 127 Photophysiology ...... 127 Metabolomic Responses to Light Shifts...... 128 Discussion ...... 146

7 CONCLUSIONS ...... 156

APPENDIX: REPRINT PERMISSIONS ...... 162

LIST OF REFERENCES ...... 163

BIOGRAPHICAL SKETCH ...... 192

LIST OF TABLES

Table page

2-1 Statistical results for analyses of total linear extension (TLE) and bleaching prevalence among replicate colonies sourced from varying donor colonies (n = 3) for each genotype...... 33

5-1 Characteristics of extant wild A. cervicornis in Little Cayman ...... 105

5-2 Results for outplanted A. cervicornis survival, growth, and condition across three reef zones in Little Cayman...... 106

LIST OF FIGURES

Figure page

2-1 Net total linear extension (TLE), daily mean temperature, and net buoyant weight of ten experimental genotypes over time...... 34

2-2 Total linear extension (TLE) and buoyant weight for each of the ten experimental genotypes evaluated...... 35

2-3 Visual differences in bleaching response among replicate Acropora cervicornis colonies of varying genotype...... 36

3-1 Principal component analysis (PCA) model comparing 1H-NMR metabolomic profiles among three unique genotypes of A. cervicornis...... 54

3-2 Partial least squares discriminant analysis (PLS-DA) model comparing 1H- NMR metabolomic profiles among three unique genotypes of A. cervicornis. .... 55

3-3 Compounds driving separation of 1H-NMR metabolomic profiles among genotypes determined by partial least square discriminant analysis...... 56

3-4 Heat map showing differences in concentrations among genotypes for the top 50 most significant features identified via analysis of variance from LC- MS ...... 57

3-5 Principal component analysis (PCA) model comparing LC-MS metabolomic profiles among three unique genotypes of A. cervicornis...... 59

3-6 Partial least squares discriminant analysis (PLS-DA) model comparing LC- MS metabolomic profiles among three unique genotypes of A. cervicornis ...... 61

4-1 Map displaying the location of experimental outplant sites and the nursery of origin...... 83

4-2 Standardized change in ellipsoid volume (EV) for each genotype among sites at t = 51...... 84

4-3 Visual differences post-hurricane between sites...... 85

4-4 Kaplan-Meier survival plot illustrating differences in percent survival of outplanted colonies among sites over 365 days...... 86

5-1 Zonation by depths in Little Cayman, modified from Logan (1994)...... 107

5-2 Outplant experiment location in context of the Caribbean region, with points indiciating the location of the nursery and each study site...... 108

5-3 Total linear extension (TLE) of outplanted Acropora cervicornis in each reef zone over time...... 109

6-1 A schematic illustrating experimental design for this study...... 132

6-2 Maximum photochemical efficiency of PSII (Fq’/Fm’(max)) and sub-saturation irradiance (Ek) over time...... 133

6-3 Maximum relative electron transport rate (rETRMAX) over time...... 134

6-4 Principal component analysis (PCA) results illustrating shifts in the metabolome over time based on GC-MS data...... 135

6-5 Partial least squares discriminant analysis (PLS-DA) models based on GC- MS data illustrating shifts in the metabolome over time ...... 137

6-6 Heat map comparing significant entities identified by GC-MS and ANOVA ...... 139

6-7 Principal component analysis (PCA) results illustrating shifts in the metabolome over time based on LC-MS data ...... 141

6-8 Partial least squares discriminant analysis (PLS-DA) models based on LC- MS data illustrating shifts in the metabolome over time...... 143

6-9 Heat map comparing significant entities identified by LC-MS and ANOVA...... 145

LIST OF ABBREVIATIONS

1H-NMR Proton-nuclear magnetic resonance

Ek Sub-saturation irradiance

EV Ellipsoid volume

Fq’/Fm’(max) Maximum photochemical efficiency of photosystem II

GC-MS Gas chromatography-mass spectrometry

HL High light

LC-MS Liquid chromatography-mass spectrometry

LL Low light

PAM Pulse amplitude modulation

PSII Photosystem II rETRmax Maximum relative electron transport rate

RLC Rapid light curve

TLE Total linear extension

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

GENETIC AND ENVIRONMENTAL CONSIDERATIONS FOR ACROPORID CORAL RESTORATION

Kathryn Elaine Lohr

December 2018

Chair: Joshua Patterson Major: Fisheries and Aquatic Sciences

Interest in and capacity for coral restoration are increasing globally. Many

Caribbean practitioners are focused on restoring acroporid species, which are threatened throughout their range and well-suited to nursery culture. Better understanding the roles of genotype and site selection could improve overall success of restoration activities. This dissertation explored the role of genetic and environmental factors on performance of acroporid corals. Intraspecific variation in growth and thermotolerance phenotypes was tested for 10 Acropora cervicornis genotypes in a common garden. Differences among genotypes were found for each phenotype monitored, including linear extension, calcification, branching, and bleaching prevalence. Additionally, thermal stress significantly reduced growth rate, and a potential tradeoff between skeletal density and linear extension was observed. 1H-NMR and LC-MS metabolomic profiling on a subset of three phenotypically-different genotypes then revealed unique metabolite “fingerprints” for each genotype.

Additionally, a subset of six previously-studied genotypes were outplanted to Florida

Keys fore and patch reef sites to determine the effect of outplanting on phenotype.

While phenotypic plasticity in linear extension was found for one genotype early in the experiment, long-term growth results were confounded by the passage of Hurricane

Irma. Hurricane disturbance differentially affected fore and patch reef sites; while ~50% of outplants survived at patch reef sites, no colonies were located at fore reef sites.

Performance of restored A. cervicornis among reef zones was also explored in a

Cayman Islands system. Growth and survival of outplanted A. cervicornis on intermediate spur-and-groove reefs closest to the nursery of origin were higher than those on shallow back reefs and deep reef terraces. Breakage and thermal variability likely affected performance at shallow sites, however factors driving low survival at deep sites were unclear. To determine whether reduced light may have negatively affected deep outplants, capacity for photoacclimation was investigated in a model acroporid.

Photophysiological parameters indicated acclimation to both increased and decreased light over a 21 day period, however LC-MS and GC-MS metabolomic shifts were less apparent for low light-exposed corals. Together, these results provide new insight into the effects of genotype and site on acroporid coral performance with many applications for ongoing coral restoration efforts.

CHAPTER 1 GENERAL INTRODUCTION

Coral reefs are highly diverse ecosystems (Connell, 1978) that contain an estimated 35% of the world’s marine species despite occupying only 0.2% of Earth’s oceans (Reaka-Kudla, 2005). The architects of coral reefs are hermatypic corals, which build calcium carbonate skeletons that create structure and habitat for other species

(Yonge, 1973). Reef-building is facilitated by in hospite dinoflagellate endosymbionts of the family Symbiodiniaceae (LaJeunesse et al. 2018), which provide hermatypic corals with critical nutrition through the organic carbon products of photosynthesis in otherwise nutrient-poor tropical regions (Muscatine and Porter, 1977; Birkeland, 2015). Corals also associate with a diverse community of microorganisms, which are believed to play a role in physiology and immune function (Rosenberg et al., 2007). Together, the coral host, endosymbionts, and associated microbial community comprise the coral holobiont

(Rohwer et al., 2002; Blackall et al., 2015). In addition to habitat provision, these holobionts and the reefs they form provide critical services to human populations, including employment in fishing and tourism (Cinner, 2014), shoreline protection

(Ferrario et al., 2014), and pharmaceutical compounds (Adey, 2000). Coral reefs also have recreational and cultural value to many human communities (Hicks et al., 2013).

Acropora is the most speciose genus of extant hermatypic corals, with an estimated 180 species recognized globally (Veron, 2000). The vast majority of these species are found in the Indo-Pacific (Veron, 2000), with only two true species,

Acropora cervicornis and A. palmata, present in the Caribbean bioregion (Vollmer &

Palumbi, 2002). Despite the lack of diversity within this genus in the Caribbean, these species have functional importance on many Caribbean reefs. Caribbean acroporids

were spatially dominant for hundreds of thousands of years (Pandolfi and Jackson,

2006), and formed vast, often monospecific thickets across large areas of reef through much of the 20th century (Goreau, 1959; Bellwood et al., 2004). Acropora is considered to be the fastest-growing coral taxon in the western Atlantic (Gladfelter et al., 1978;

Lirman et al., 2014) as well as the Indo-Pacific (Anderson et al., 2012). For example, A. cervicornis has been found to exceed 5 cm of annual linear growth per cm of existing coral (Lirman et al., 2014). The fast growth rate and branching morphology of acroporids make them critical to building functional three-dimensional structure on reefs

(Pandolfi and Jackson, 2006). This complex structure creates habitat for ecologically and economically important reef fish and invertebrates (Lirman, 1999; Bellwood et al.,

2004) and contributes to wave attenuation, especially on shallow reefs (Ferrario et al.,

2014). Acroporids are also believed to play a role in stabilizing reefs by binding rubble and sediment (Gilmore and Hall, 1976; Mercado-Molina et al., 2015). Functional redundancy is limited on Caribbean reefs as A. palmata and A. cervicornis are the only two large, open-branching corals present in the region; it is therefore unlikely that another Caribbean species could adequately fill their unique ecological role (Bellwood et al., 2004; Pandolfi and Jackson, 2006).

Unfortunately, coral reefs have declined considerably over recent decades due to a combination of global and local stressors (Knowlton and Jackson, 2008). Reef decline has been particularly severe in the Caribbean (Roff and Mumby, 2012). In the Florida

Keys, 93% and 98% declines in A. palmata and A. cervicornis cover, respectively, were reported on a reef between 1983 and 2000 (Miller et al., 2002). This is consistent with broader trends for the Caribbean, where both acroporid species have declined by an

estimated 95% or more since the late 1970s (Precht et al., 2002). These once-dominant species became the first corals listed under the U.S. Endangered Species Act in 2006

(National Marine Fisheries Service, 2006), and are also considered critically endangered by the International Union for Conservation of Nature (Aronson et al.,

2008a, 2008b). Disease is an important driver of the decline in Caribbean acroporids

(Gladfelter, 1982a; Aronson and Precht, 2001; Vollmer and Kline, 2008; Sutherland et al., 2011; Miller et al., 2014a). Outbreaks of progressive tissue loss have been reported throughout the Caribbean in both acroporid species (Aronson and Precht, 2001), with some locations reporting disease prevalence as high as ~70% (Williams and Miller,

2005; Miller et al., 2014a) and tissue loss rates of ~4 cm d-1 during outbreaks (Williams and Miller, 2005). Although increases in coral disease prevalence do not always correspond directly to increased sea temperature (e.g. Miller et al., 2014a), evidence suggests that long-term increases in temperature due to climate change (Randall and van Woesik, 2015) and El Niño-driven temperature anomalies (Harvell et al., 2002) have likely driven many disease outbreaks.

Climate change has also significantly contributed to increases in the frequency and severity of bleaching events among corals globally (Eakin et al., 2009; Hughes et al., 2017a, 2017b, 2018). Bleaching refers to the expulsion of endosymbiotic dinoflagellates from coral host tissues, resulting in a loss of pigmentation and reduced nutrition to the host (Brown, 1997; Bosch and Miller, 2016). Mass coral bleaching events are almost always linked to elevated sea temperature (Hoegh-Guldberg, 1999; Eakin et al., 2009, Hughes et al., 2017b), although bleaching can also occur due to stressors such as excessive solar radiation, reduced water temperature or salinity, and disease

(Brown, 1997). Branching corals are particularly susceptible to thermal stress (Brown and Suharsono, 1990; Hughes et al., 2018), potentially due to thin tissue and slower mass transfer compared to massive species (Loya et al., 2001), and bleaching has resulted in direct mortality among acroporids (Quinn and Kojis, 2008; Williams et al.,

2017; Hughes et al., 2018). Coral bleaching has also been shown to trigger disease outbreaks in acroporids (Muller et al., 2008). However, corals are capable of surviving bleaching events and recovering zooxanthellae if conditions improve relatively quickly

(Bosch and Miller, 2016). In addition to causing bleaching events and related consequences, climate change is also expected to increase the frequency of severe tropical cyclones (Emanuel, 2005; Bender et al., 2010), which can result in significant structural damage to acroporids (Woodley et al., 1981; Hughes and Connell, 1997).

Tropical cyclones can also cause secondary problems for Acropora, including increased predation prevalence (Knowlton et al., 1981, 1990) and disease outbreaks (Miller et al.,

2014a).

In addition to global stressors, coral reefs have also been negatively impacted by more localized stressors. In the Caribbean, reef decline has been partially linked to the region-wide die-off of the keystone herbivore Diadema antillarum (Lessios et al., 1984;

Lessios, 1988). The decline of D. antillarum is a driver of increased cover of macroalgae, which can outcompete corals for space (Hughes, 1994). Such effects are particularly severe on overfished reefs where herbivorous fish species are also depleted

(Hughes, 1994). Overfishing has long affected coral reefs through removal of fish that control algae and corallivores (Jackson et al., 2001). However, fishing is also known to directly reduce live coral cover via destructive fishing practices, particularly in Southeast

Asia (McManus 1997; White et al., 2000; Bailey and Sumaila, 2015). In addition to the effects of fishing, corals reefs are affected on local and regional scales by poor water quality (Fabricius, 2005; De’ath and Fabricius, 2010). Water quality is often negatively impacted by terrestrial nutrient runoff, particularly as a result of agricultural activities and coastal population growth (Fabricius, 2005; De’ath and Fabricius, 2010; Kroon et al.,

2012). Increased nutrient inputs lead to increased growth of competitive algae on coral reefs (Fabricius, 2005), and have also been linked to increased susceptibility of corals to bleaching and disease (Wiedenmann et al., 2013; Vega Thurber et al., 2014).

Furthermore, sedimentation resulting from coastal development can reduce coral recruitment, cover, growth, and species richness (Rogers et al., 1990; Fabricius, 2005).

Both Caribbean acroporid species are severely depleted following major population declines due to a confluence of these natural and anthropogenic stressors

(Aronson et al., 2008a, 2008b). In contrast, many Indo-Pacific acroporids are still considered relatively abundant (Richards et al., 2014), although populations of some species are decreasing (e.g. Richards et al., 2008, 2014; Hughes et al., 2018).

Localized natural recovery (Lidz and Zawada, 2013; Lucas and Weil, 2015) and persistence of large thickets (Lirman et al., 2010) have been reported from multiple

Caribbean locations, but large-scale recovery of Acropora is limited by low rates of sexual recruitment and potentially also by Allee effects following widespread decline of reproductive colonies (Aronson and Precht, 2001; Lirman and Schopmeyer, 2016).

Despite the fact that remnant populations are generally sparse, high genetic diversity has been found among extant A. cervicornis (Drury et al., 2016). Genetic diversity is critically important for persistence of a species, as it facilitates adaptation to a changing

environment (Hoffmann and Sgrò, 2011) and resistance to disturbance (Hughes and

Stachowicz, 2004). Genetic diversity also ensures avoidance of problems such as genetic bottlenecks and inbreeding depression (Baums, 2008). High genetic diversity in

Caribbean acroporids is therefore a sign that recovery is possible due to the retention of adaptive alleles in remnant populations (Drury et al., 2016).

Despite this potential for recovery among Caribbean acroporids, substantial limitations to natural reproduction of these species also exist. Research has therefore increasingly focused on management strategies that enhance live coral cover through direct human intervention (Rinkevich, 2015; van Oppen et al., 2015). One such strategy is coral restoration, a practice that involves seeding degraded natural reefs with cultured coral colonies, fragments, or sexual recruits (Rinkevich, 1995). Coral restoration has become increasingly popular since the mid-1990s, with dozens of practitioners operating at reef sites globally (Young et al., 2012; Rinkevich, 2014). One common method of coral restoration, known as coral gardening, consists of asexually propagating donor fragments from healthy, wild colonies (Rinkevich, 1995; Bowden-

Kerby, 2001). The primary goal of coral gardening is to create a self-sustaining stock of maricultured coral within a nursery, which can later be transplanted to restore degraded reefs (Rinkevich, 1995; Epstein et al., 2001, 2003; Young et al., 2012). Hundreds of thousands of colonies have been cultured worldwide using this method (Rinkevich,

2014). Transplantation from nurseries to reefs, known as outplanting, immediately increases reef structural complexity and local abundance of target coral species at the outplanted site. The addition of heathy, genotypically diverse populations to depleted

reefs also creates opportunities for future sexual reproduction, which can increase genetic diversity and enhance natural recovery (Johnson et al., 2011).

Restoration efforts in the Caribbean and western Atlantic have focused largely on acroporids, and particularly A. cervicornis, due to its ecological importance,profound range-wide decline, and amenable life history characteristics (Young et al., 2012;

Lirman and Schopmeyer, 2016). Its fast growth rate, branching morphology, and ability to reproduce readily bye asexual fragmentation make A. cervicornis an ideal candidate for nursery propagation (Herlan and Lirman, 2008; Johnson et al., 2011). Although a wide variety of strategies have been successfully developed for rearing A. cervicornis in nurseries (Johnson et al., 2011; Young et al., 2012; Lirman and Schopmeyer, 2016), the

PVC tree method (Nedimyer et al., 2011), in which cultured colonies are freely suspended in the water column, has been widely adopted. PVC trees have been shown to enhance linear extension compared to culture methods in which colonies are fixed to a static surface, such as a cinder block; this method can generate large amounts of biomass for restoration in a short period of time (Johnson et al., 2011; Nedimyer et al.,

2011; O’Donnell et al., 2017; Kuffner et al., 2017).

Although a variety of methods have been developed to successfully grow A. cervicornis colonies in ocean-based nurseries, questions remain regarding the performance of these colonies following transplantation to natural reefs. Common metrics used to assess post-outplant performance of A. cervicornis include survivorship and growth (Bruckner and Bruckner, 2001; Bowden-Kerby and Carne, 2012; Ross,

2014; Mercado-Molina et al., 2015). Colonies are typically individually monitored for a fixed period (generally 1 year) following outplanting, and in this time mortality and

growth can vary widely among restored populations, even those at neighboring sites

(Bowden-Kerby and Carne, 2012; Ross, 2014; Mercado-Molina et al., 2015).

Practitioners are also concerned with negative impacts to outplanted colonies, particularly disease (Miller et al., 2014a), bleaching (Bowden-Kerby and Carne, 2012), breakage (Bowden-Kerby, 2008), and predation (Miller et al., 2014b), each of which can be detrimental to restored populations, but highly site-specific and difficult to predict.

Improvements in environmental conditions are clearly needed to mitigate the stressors affecting restored populations, such as climate change and disease (National Marine

Fisheries Service, 2015). Despite these ongoing issues, restoration efforts have resulted in significant positive effects on local abundance of A. cervicornis (Miller et al.,

2016), and the scale of restoration efforts has increased to the point that ecologically significant goals can be attained (Lirman and Schopmeyer, 2016). However, additional exploration of genetic and environmental effects on corals in both nursery and outplant settings could improve coral restoration outcomes and increase the utility of restoration as a reef management tool (Rogers et al., 2015).

Given that genetic diversity is critically important for species recovery and conservation, it is also a key consideration in coral restoration efforts (Baums, 2008;

Shearer et al., 2009; Baums et al., 2010; Drury et al., 2016) and most restoration practitioners strive to maximize diversity (Johnson et al., 2011; Hunt and Sharp, 2014).

Although practitioners generally incorporate a high number of genotypes in restoration programs to increase diversity and provide opportunities for sexual reproduction

(Johnson et al., 2011), genotypes are rarely systematically evaluated to determine whether they possess desirable traits. However, selective use of such genotypes in

breeding programs could increase the prevalence of desirable phenotypes that could enhance survival in restored populations (van Oppen et al., 2015, 2017). This strategy could also extend to selective outplanting, in which genotypes with key traits such as thermotolerance, high skeletal density, or fast growth rate are outplanted more frequently to enhance overall survival of restored coral populations. To implement such a strategy, it is first necessary to determine the extent to which phenotypes vary in a common garden for species of interest, such as A. cervicornis. It is equally important to determine whether phenotypes measured in a common garden setting are consistent once a given genotype is outplanted across multiple sites. Although there is evidence to suggest certain traits, like calcification, may be conserved among Caribbean acroporid genotypes (e.g. Kuffner et al., 2017), high phenotypic plasticity has been documented for other traits, such as linear extension, in these species (Drury et al., 2017). A lack of consistency in phenotype across space and/or time could complicate trait-based restoration efforts.

Another limitation of this strategy is that identifying robust genotypes typically involves conducting labor- and cost-intensive in situ measurements, such as total linear extension (Lirman et al., 2014), potentially making this strategy impractical for restoration practitioners. However, molecular techniques could aid in streamlining genotype selection for coral restoration. Metabolomic profiling is a method for identifying the suite of small molecules involved in cell metabolism and signaling. Metabolomic profiles have been closely linked to phenotype in a number of organisms (Steinfath et al., 2010; Matsuda et al., 2012; Riedelsheimer et al., 2012), suggesting these may also differ predictably among coral genotypes with unique phenotypes. However, no study

has yet determined whether metabolomic profiles vary among corals of the same species with unique genotypes and phenotypes.

In addition to genetic considerations, site selection is key in determining restoration outcomes. Best practice recommendations with regard to site selection for coral restoration include outplanting to areas that once harbored high densities of target species and selecting sites at depths similar to site of origin (Edwards and Gomez,

2007; Johnson et al., 2011). These best practice recommendations may inadvertently limit the scope and effectiveness of coral restoration. For example, many sites that once harbored high densities of target species have been structurally altered over time

(Alvarez-Filip et al., 2009), which could affect the ability of target species to survive at such sites. In contrast, reefs not typically targeted for outplanting could have value as restoration sites. Following the widespread decline of A. cervicornis, relatively high abundances of wild colonies have been observed at patch reef sites in Florida, which were not historically dominated by this species (Miller et al., 2008). This suggests that outplants may also succeed on patch reefs. Additionally, outplanting to a single depth could limit the breadth of restoration efforts. A. cervicornis typically occurs at depths ranging from 1 – 25 m (Aronson et al., 2008a), and can have differing ecological functions based on the depth and reef zone it inhabits (e.g. Ferrario et al., 2014).

Although avoiding large shifts in depth/light during outplanting is recommended

(Edwards and Gomez, 2007; Johnson et al., 2011), evidence suggests that many coral species are capable of acclimating to relatively large shifts in light availability (e.g.

Langlois and Hoogenboom, 2014; Cohen and Dubinsky, 2015). New information on the

capacity for photoacclimation in acroporid corals could determine whether a wider range of depths should be considered for restoration of these species.

The overarching objectives of this research were to (1) assess the utility of a trait- based restoration strategy for A. cervicornis and (2) provide information that could improve site selection for restoration of this species. Chapter 2 investigates intraspecific variation in phenotype among A. cervicornis cultured within a common garden setting.

This experiment provides practical information on differences in desirable phenotypes that could be used in selection of coral genotypes for restoration. Chapter 3 builds on these results by exploring differences in metabolomic profiles among A. cervicornis genotypes known to differ phenotypically. Chapter 4 investigates the short-term effects of outplanting on phenotype in nursery-reared A. cervicornis, and explores the longer- term effect of site type on restoration outcomes. Chapter 5 also provides information on site selection for restoration by investigating A. cervicornis outplant performance among diverse reef zones. This experiment identifies potential limitations in current restoration practices that could be addressed to better restore each type of habitat where A. cervicornis occurs. Chapter 6 employs photophysiological and metabolomic techniques to identify the capacity for rapid light acclimation in the model acroporid A. muricata.

These data can assist site selection and aid in interpreting previously observed differences in restored coral performance among reef zones. Chapter 7 summarizes the major findings of this dissertation and their implications for ongoing coral reef restoration efforts. This dissertation adds to the existing body of research on coral reef restoration by exploring strategies for genotype and site selection that could improve restoration outcomes for acroporid corals.

CHAPTER 2 INTRASPECIFIC VARIATION IN PHENOTYPE AMONG NURSERY-REARED ACROPORA CERVICORNIS*

Introduction

The unprecedented decline in abundance of staghorn coral Acropora cervicornis due to a confluence of natural and anthropogenic factors (Greenstein et al., 1998;

Bruckner et al., 2002; Miller et al., 2002) has resulted in a critical loss of ecosystem services for Caribbean coral reefs (Bellwood et al., 2004; Alvarez-Filip et al., 2009). In recent decades, the coral gardening method was developed and adopted throughout the wider Caribbean to culture and reestablish lost populations of A. cervicornis and restore their associated services (Nedimyer et al., 2011; Young et al., 2012). This method is characterized by the asexual propagation of A. cervicornis within in situ nurseries with the ultimate goal of conducting population enhancement on local reefs

(Rinkevich, 1995; Young et al., 2012). Fragmenting portions of A. cervicornis donor colonies in a nursery to create new, smaller colonies can rapidly increase the amount of tissue and skeleton available for restoration purposes (Johnson et al., 2011; Lohr et al.,

2015). Worldwide, hundreds of thousands of coral colonies have been farmed and gradually transplanted to natural reefs (Rinkevich, 2014). Nursery practitioners are now poised to scale up transplantation of nursery-reared colonies to match historical abundances of farmed coral species on target reefs (Rinkevich, 2014).

* This chapter is reprinted from Lohr, K.E., Patterson, J.T., 2017. Intraspecific variation in phenotype among nursery-reared staghorn coral Acropora cervicornis (Lamarck, 1816). J. Exp. Mar. Biol. Ecol. 486, 87–92.

Although the coral gardening method has been largely successful, individual coral restoration efforts are currently small in scale (Edwards and Gomez, 2007) and can have variable outcomes (e.g. Bowden-Kerby and Carne, 2012; Ross, 2014).

Sources of variability in A. cervicornis restoration outcomes include disease (Miller et al., 2014a), bleaching (Bowden-Kerby and Carne, 2012), breakage (Bowden-Kerby,

2008), and predation (Miller et al., 2014b). In addition, nursery-reared, outplanted colonies can vary in performance between sites with differing environmental parameters

(e.g. depth, reef zone; Ross, 2014). There is also evidence to suggest that survivorship of restored Acropora may decline long-term (i.e. after 2 years; Bruckner and Bruckner,

2001; Ware, 2015). Given these ongoing challenges, it is apparent that, in conjunction with improvements in environmental conditions (National Marine Fisheries Service,

2015), new information is required to advance strategies for coral restoration (Rogers et al., 2015).

The role of genetics in coral restoration efforts has generated particular interest among researchers, managers, and practitioners (Baums, 2008; Shearer et al., 2009).

Although published guides for coral restoration recommend outplanting genetically diverse populations (Shearer et al., 2009; Johnson et al., 2011), selection of genotypes for outplanting is typically not based on a systematic evaluation of phenotype. Previous studies have documented differences in phenotype, particularly growth rate, among known genotypes of nursery-cultured A. cervicornis (Bowden-Kerby, 2008; Lirman et al.,

2014). Some A. cervicornis genotypes have also been shown to resist infection by disease (Vollmer and Kline, 2008; Libro and Vollmer, 2016). In addition, differences in

bleaching resistance among host genotypes have been documented in other scleractinian coral species (Edmunds, 1994; Baird et al., 2009; Kenkel et al., 2013).

Understanding how phenotype can vary among nursery stocks of A. cervicornis is critical for improving restoration outcomes given that differences in phenotype could influence the ability of outplants to cope with stressors and grow to maturity. Such information represents a necessary first step in the development of a trait-based system to select genotypes for use in population enhancement activities, which has been identified as an important strategy for improving coral restoration outcomes (Hunt and

Sharp, 2014; National Marine Fisheries Service, 2015). A 13-month experiment was therefore conducted to quantify and compare linear growth, branching, calcification, and bleaching prevalence among ten known genotypes of A. cervicornis in an established coral nursery in the Florida Keys.

Methods

This study was conducted within an established coral nursery operated by the

Coral Restoration Foundation (CRF) located four miles offshore from Tavernier, FL.

CRF maintains 150 known genotypes of Acropora cervicornis (CRF, unpublished data), which were identified using microsatellites developed by Baums et al. (2009). Ten genotypes that had been previously outplanted according to CRF records and were sufficiently stocked in the nursery were selected for use. Genotypes used in this study were sourced from sites within 30 km of the nursery between 1996 and 2011. These genotypes had acclimated to the nursery location for a minimum of 3.5 years prior to the initiation of this study. For each genotype, four 5-cm non-branching apical tips were clipped from each of three existing nursery donor colonies (n = 12 tips per genotype) to control for any intracolonial variation in genotype (Schweinsberg et al., 2015). The tree

nursery method (Nedimyer et al., 2011) was used in this study due to its popularity for

A. cervicornis grow-out at many Caribbean nurseries (e.g. Lohr et al., 2015; Meesters et al., 2015; Johnson et al., 2011). All replicate colonies (n = 120) were tagged for identification (genotype, donor colony, replicate) and haphazardly distributed across four PVC tree nurseries (n = 30 fragments per tree) installed at the same location within the source nursery. Fragments were affixed to trees in rows of 10 at ~10 cm intervals using monofilament and aluminum crimps. After 162 days, spacing was increased to

~25 cm (4 fragments per row) to ensure growing fragments did not touch. A HOBO

Pendant® data logger (Onset Computer Corporation, USA) was installed on each of two tree nurseries to continuously record temperature in situ.

Replicate coral colonies were transported from the nursery to shore and buoyantly weighed (Jokiel et al., 1978) on day 0, day 122, and day 390 at the conclusion of the experiment. Total linear extension (TLE) and number of branches ≥1 cm TLE for each replicate colony were also recorded in situ at the start of the experiment and then at approximately 45 day intervals throughout the duration of the study. At each interval, replicate colonies were examined for signs of disease following

Miller et al. (2014a) and presence or absence of bleaching was also documented for each replicate colony. Bleaching was defined as any visible loss of pigmentation from a coral, following the Atlantic and Gulf Rapid Reef Assessment protocol version 5.4 (Lang et al., 2010). Observed loss of pigmentation was recorded in the field and verified by comparing photographs of apparently bleached replicate colonies to photographs of the same colony from the previous sampling interval. Specific growth rate was calculated for each replicate colony twice: before bleaching (the time period prior to and including

the last interval at which colonies were not affected by bleaching, day 0 – 162) and after bleaching (the time period beginning at the first observation of bleaching up to and including the first interval at which replicate colonies were fully recovered, day 207 –

291). Specific growth rate was calculated as

ln(푓) − ln⁡(푖) SGR = ⁡푥⁡100 푡 (2-1) where f is the final TLE and i is the initial TLE for a given colony in a given time period and t is the duration of the time period in days.

Statistical analysis was performed using R statistical software (v. 3.1.2, R Core

Team, 2014). All statistical tests were conducted at a significance level of α = 0.05. Data were assessed for normality and homogeneity of variance using the Shapiro-Wilk test and Levene’s test, respectively. Differences in TLE among replicate colonies sourced from varying donor colonies were tested for each genotype after 291 days of growth using analysis of variance (ANOVA) with Tukey HSD for pairwise comparisons. Values for net buoyant weight and net TLE were calculated by subtracting initial values (t = 0 days) from final values (t = 390 days) and compared among genotypes using ANOVA with Tukey HSD. The relationship between net buoyant weight and net TLE was analyzed using simple linear regression. A ratio of buoyant weight to TLE was calculated for each replicate colony and compared among genotypes using a Kruskal-

Wallis test with a Dunn post hoc test. The relationship between this ratio and TLE was assessed using a Pearson correlation. Final data on branch number per replicate colony were collected on day 291 and were compared among genotypes using a Kruskal-

Wallis test with a Dunn post hoc test. Broad sense heritability (H2) was calculated for

TLE, buoyant weight, and branching following Császár et al. (2010) and Toker (2004).

Genotypes found to differ in growth by donor colony were excluded from this analysis.

Prevalence of bleaching was analyzed among genotypes using chi-squared tests, and among donor colonies within each genotype using Fisher exact tests. Specific growth rate values before and after bleaching were compared across all genotypes using a paired t-test.

Results

During the first 291 days of the experiment, one replicate colony was lost, and four others experienced complete (n = 1) or partial (n = 3) mortality due to unknown causes. Growth data for these colonies were excluded from analysis. No signs of active disease were observed during the study period. Between day 291 and day 390, two tree nurseries were lost. Therefore, data from the remaining 58 colonies were used for analysis of net buoyant weight and net total linear extension (TLE). Installed HOBO

Pendant® data loggers were lost along with the two tree nurseries, therefore data from an adjacent National Oceanic and Atmospheric Administration monitoring station

(National Data Buoy Center station MLRF1, ~6.8 km from nursery) were used to supplement available HOBO data.

Net buoyant weight (F9,48 = 2.54, p = 0.02) and net TLE (F9,48 = 4.02, p < 0.001) varied among genotypes after 390 days of growth (Fig. 2-1). Differences in TLE among replicate colonies originating from different donor colonies were detected for two genotypes (K2 and U41; Table 2-1). Buoyant weight was significantly predicted by TLE

(β = 0.46, t56 = 14.45, p < 0.001) and TLE explained a significant proportion of variance

2 in buoyant weight (F1,56 = 208.9, R = 0.79, p < 0.001). The ratio of buoyant weight to

TLE varied among genotypes (H9 = 17.23, p < 0.05; Fig. 2-2) and decreased with increasing TLE (r56 = -0.67, p < 0.001). The number of branches per replicate colony on

day 291 ranged from 8 to 30 and mean branch number varied among genotypes (H9 =

35.29, p < 0.001). For TLE, buoyant weight, and branching, H2 was 0.28, 0.27, and

0.25, respectively.

Sub-lethal bleaching was observed among replicate colonies on day 207 following two 10-day periods in which daily mean temperature exceeded 31°C (Fig. 2-

1A). Specific growth rate for all genotypes during the period preceding the bleaching event was significantly higher compared to specific growth rate during the period immediately following bleaching (t112 = 30.15, p < 0.001; Fig. 2-1). All affected replicate colonies had visually recovered by day 291. Genotype had a significant effect on bleaching frequency (X2 = 42.81, p < 0.001), and frequency of bleaching was lowest among replicate colonies of genotype U44. All replicate colonies of genotype U25 were fully bleached. Figure 2-3 illustrates visual differences in bleaching response among a subset of replicates from three experimental genotypes. Differences in bleaching prevalence among replicate colonies from varying donor colonies were detected for two genotypes (K2 and K3; Table 2-1).

Table 2-1. Statistical results for analyses of total linear extension (TLE) and bleaching prevalence among replicate colonies sourced from varying donor colonies (n = 3) for each genotype. TLE data for U41 was log transformed to meet the assumption of normality. ANOVA was performed on TLE data collected on day 291 and Fisher exact tests were performed on bleaching prevalence data collected on day 207 (first observation of bleaching). Statistically significant results are highlighted in bold.

Data Test Value Genotype

K1 K2 K3 U25 U41 U44 U47 U73 U77 U78

TLE ANOVA F 0.62 5.10 0.48 0.21 5.98 1.36 1.51 1.43 0.80 0.71

df 2, 9 2, 9 2, 9 2, 8 2, 8 2, 8 2, 9 2, 9 2, 7 2, 9 p 0.53 0.03 0.63 0.82 0.03 0.31 0.27 0.29 0.49 0.52

Bleaching Fisher’s p 1.00 0.01 0.01 1.00 1.00 0.07 0.77 1.00 1.00 1.00 Prevalence Exact Test

Figure 2-1. Net total linear extension (TLE) and daily mean temperature (dashed line) (A) and net buoyant weight (B) of ten experimental genotypes over time. Error bars represent SE. Both net TLE (F9,48 = 4.02, p < 0.001) and net buoyant weight (F9,48=2.54, p = 0.02) varied among genotypes after 390 days of growth. Letters denote differences among genotypes as identified by Tukey HSD. Temperature data from day 0 through day 290 was recorded on HOBO data loggers attached to nursery tree structures, and temperature data from day 291 through day 390 was sourced from the National Oceanographic and Atmospheric Administration National Data Buoy Center monitoring station MLRF1.

Figure 2-2. Total linear extension (TLE) and buoyant weight for each of the ten experimental genotypes evaluated, illustrating differences in the ratio of buoyant weight to TLE among genotypes. Error bars represent SE. The ratio of TLE to buoyant weight differed significantly among genotypes (H9 = 17.23, p < 0.05). Letters denote differences in the ratio of buoyant weight to TLE among genotypes as identified by the Dunn post hoc test. The ratio of TLE to buoyant weight decreased with increasing TLE (r56 = -0.67, p < 0.001).

Figure 2-3. Visual differences in bleaching response among replicate Acropora cervicornis colonies of varying genotype. Photos are labeled as genotype- donor colony-replicate number. Bleaching prevalence differed significantly among genotypes (X2 = 42.81, p < 0.001). Photos courtesy of author.

Discussion

Results of this study suggest that differences in multiple traits may be discerned among known genotypes of Acropora cervicornis in a nursery setting. Furthermore, these differences can be identified using simple measurements and monitoring techniques that could be adopted by coral nursery managers. These findings support the potential development of a trait-based system for use in A. cervicornis restoration activities. Nursery stocks, and subsequently outplanted colonies, typically consist of a subset of locally available genotypes. It is unlikely, however, that all available local genotypes are well-suited for restoring reefs facing ongoing changes due to stressors

(van Oppen et al., 2015). Therefore, developing a trait-based system could help guide restoration decisionmaking and aid in ideniftying genotypes that are better able to tolerate current stressors (Hunt and Sharp, 2014). However, efforts to incorporate trait information into restoration strategy must simultaneously strive to maintain genetic diversity in nursery and restored populations in order to facilitate adaptation to future conditions (Webster et al., 2017).

The genotypes systematically monitored in this study were found to differ in both net TLE and net buoyant weight. For the population investigated in this study, genetic variation was found to account for approximately one quarter of observed variability in growth phenotypes (TLE, buoyant weight, branching). Although this contributes to an understanding of genetic factors for this species, broad sense heritability (H2) measured in this study is most useful in clonally propagated organisms and should not be considered a measure of the likelihood that a trait will be passed on to sexually- produced offspring (Holland et al., 2003). For clonally propagated A. cervicornis, faster growth rates have been proposed as an indicator of overall coral fitness (Lirman et al.,

2014), and therefore faster growing genotypes could potentially outperform their slower growing counterparts in restoration scenarios. Tradeoffs have, however, been identified between coral growth and thermal tolerance (Jones and Berkelmans, 2010), and have also been found for growth and disease resistance in other organisms (e.g. Huot et al.,

2014). Data collected in the present study suggest a tradeoff between faster growth and the ratio of buoyant weight to TLE, whereby genotypes that grew faster had less buoyant weight per centimeter TLE. These results could be explained by growth rate- based differences in colony morphology and/or skeletal density. Interestingly, a tradeoff between linear extension and skeletal density was reported for A. cervicornis reared grown under two nursery conditions (Kuffner et al., 2017). If slower-growing genotypes in the present study are indeed characterized by higher skeletal density, they could potentially outperform faster-growing genotypes in shallow, high-energy environments characterized by high rates of breakage in restored populations (Hunt and Sharp, 2014;

Kuffner et al., 2017). Fine-scale analysis of colony surface area and volume would help to clarify how morphology and skeletal density vary among genotypes and with changes in growth and buoyant weight.

Further investigation is needed to determine how phenotype in nursery-reared A. cervicornis may change following outplanting. If phenotype is consistent following outplanting, observed differences in both growth and branching among known A. cervicornis genotypes could have implications for habitat provision following restoration.

Branching corals such as A. cervicornis create important habitat for other reef species

(Huntington et al., 2017). Fish abundance has been shown to increase with branching coral size and complexity (Holbrook et al., 2002) and colony density (Huntington et al.,

2017). In particular, the probability of damselfish occupation has been shown to increase with A. cervicornis colony size and number of branches; although damselfish can cause coral mortality through the formation of algal lawns, they can also deter corallivores such as Hermodice carunculata and Coralliophila abbreviata (Schopmeyer and Lirman, 2015). Outplanting faster-growing, rapidly-branching genotypes could therefore result in more rapid habitat provision for reef fishes and potentially reduce the impacts of corallivores by attracting damselfishes; however, the potential benefits of outplanting rapid-growing genotypes should be weighed against the risk of damselfish- induced mortality at sites where damselfish are prevalent (Schopmeyer and Lirman,

2015). Furthermore, traits such as fast growth should be considered in the context of larger restoration goals; for example, a higher proportion of fast-growing genotypes could be used when restoration of fish habitat is a priority, while more equal representation of growth traits could be best for general A. cervicornis restoration.

In addition to genotype-based differences in growth, specific growth rate was also found to decrease across all genotypes following bleaching. This decrease is consistent with the findings of Jones and Berkelmans (2010), who reported diminished growth for up to 18 months following bleaching in A. millepora. In addition, bleaching prevalence varied among genotypes of nursery-reared A. cervicornis, which is consistent with published evidence suggesting that coral host genotype plays a considerable role in determining thermotolerance (Edmunds, 1994; Baird et al., 2009;

Kenkel et al., 2013). Differences in bleaching prevalence among known genotypes of nursery-reared A. cervicornis suggest the presence of possible bleaching-resistant genotypes, the identification of which could have implications for outplant selection. For

example, given that coral bleaching can cause extensive mortality (Baird and Marshall,

2002), increasing the proportion of bleaching-resistant genotypes during outplanting could improve overall survivorship of outplanted A. cervicornis cohorts and therefore increase the likelihood that outplants will grow to maturity. In addition, Meyer et al.

(2009) and Dixon et al. (2015) found evidence for heritability of thermal tolerance among A. millepora, suggesting that selective breeding of bleaching-resistant genotypes could potentially confer advantageous traits governing thermotolerance to offspring (see van Oppen et al., 2015). In contrast, Császár et al. (2010) found that heritable traits related to thermal tolerance were present primarily in Symbiodinium rather than host A. millepora. Heritability of thermotolerance in A. cervicornis also warrants additional study.

Although symbiont genotype was not analyzed in this study, analysis of symbiont identity could provide valuable complementary information on bleaching dynamics among known A. cervicornis genotypes. Finer-scale assessments of bleaching such as those using PAM fluorometry (Roth and Deheyn, 2013) or digital photographic analysis

(Chow et al., 2016) could also provide more detailed information on bleaching dynamics among genotypes. Better understanding of how thermotolerance varies among both host and symbiont genotypes in A. cervicornis is critical for incorporating measures of bleaching tolerance into existing restoration methods and potentially for future selective breeding of this species.

Although growth and bleaching phenotypes were consistent regardless of donor colony for most genotypes, some measurements differed among replicates from varying donor colonies of the same genotype (Table 2-1). Several potential factors could cause this variation. First, it is possible that one or more donor colonies selected for use in this

study were mislabeled with regard to genotype. Alternatively, donor colonies used in this study may be genetically heterogeneous as observed among multiple Indo-Pacific coral species by Schweinsberg et al. (2015), likely resulting in intracolonial variation in phenotype. Intensive intracolonial genetic sampling (e.g. Schweinsberg et al., 2015) could reveal genetic heterogeneity within A. cervicornis donor colonies. With respect to bleaching, it is well-documented that some colonies that bleach and recover can resist bleaching during subsequent warming events, potentially by taking up more thermotolerant symbionts upon recovery (i.e. adaptive bleaching; Maynard et al., 2008;

Silverstein et al., 2014). Thus, variation in bleaching history among donor colonies could have resulted in the observed differences in bleaching prevalence among replicate colonies from varying donor colonies of a single genotype. Identification of symbionts and multi-year monitoring of nursery bleaching dynamics could identify any adaptive response among nursery colonies. Alternatively, observed differences in phenotype among donor colonies of the same genotype could be the result of epigenetic changes resulting from prior exposure to differing environmental conditions (see Jaenisch and

Bird, 2003). Additional evidence is required to confirm the presence of epigenetic effects in corals, however these processes have been tentatively suggested to play a role in observed coral phenotype (van Oppen et al., 2015; Marsh et al., 2016). Ultimately, differences in phenotype among colonies of the same presumed genotype warrant further investigation.

These results confirm that differences in phenotype can be discerned among varying genotypes of nursery-reared A. cervicornis. Additional information is required to fully assess the utility of a trait-based system for A. cervicornis restoration. In particular,

it is critical to determine whether observed phenotype within a nursery is consistent across multiple years of study, including additional fragmentation events. It is also necessary to ascertain whether observed phenotype in a nursery is predictive of colony phenotype following outplanting. Finally, it is important to determine to what degree phenotype observed in a nursery reflects acclimation to local conditions versus an adaptation for similar performance across a range of conditions (Palumbi et al., 2014).

Nursery managers should also consider assessing phenotypes among local A. cervicornis genotypes within their nurseries to understand how performance may vary under local conditions.

CHAPTER 3 METABOLOMIC PROFILES DIFFER AMONG UNIQUE ACROPORA CERVICORNIS GENOTYPES

Introduction

The global decline of coral reefs demands novel tools to understand these ecosystems and improve their ability to persist under present and future environmental conditions (Hughes et al., 2017a; van Oppen et al., 2017, 2015). Mitigation of local stressors and global reductions in CO2 emissions are critical to ensuring broad-scale preservation of coral reefs (Hughes et al., 2017a, 2017b); however, complementary, progressive strategies, including selective breeding, transplantation, and coral restoration, have received increasing attention for their potential to contribute to coral reef conservation (van Oppen et al., 2017, 2015). Many strategies that involve human intervention may benefit from targeted use of robust genotypes that possess desirable traits. Major threats to the survival of coral reefs include increasing global sea temperature (Hughes et al., 2018, 2017b), prevalence of diseases (Maynard et al.,

2015), and frequency and intensity of tropical storms (Emanuel, 2005, 2013). Thus, desirable traits for use in restoration or other interventional strategies include thermotolerance, disease resistance, and high skeletal density (van Oppen et al., 2017).

Identifying genotypes that possess such traits can involve painstaking repeated measurements in situ (Lohr and Patterson, 2017; Mercado-Molina et al., 2016) or long- term manipulative experiments (Kenkel et al., 2013; Vollmer and Kline, 2008), which are time-consuming and costly. Identifying broad metabolomic profiles or specific metabolites associated with desirable genotypes could assist in streamlining this process.

The metabolome is the set of small chemical compounds (metabolites) within an organism that are involved in metabolic processes. These metabolites are the end product of gene transcription and translation as well as post-translational physiological processes. The metabolome is shaped by both genetic and environmental effects, and is therefore linked to phenotype (Fiehn 2002; Goulitquer et al., 2012). Metabolomic profiling, the identification and quantification of metabolites within an organism (Fiehn

2002), has only recently been applied to the reef-building coral holobiont (Gordon et al.,

2013; Gordon and Leggat, 2010; Quinn et al., 2016; Sogin et al., 2014). To date, metabolomic profiling has reliably shown differences in the suite of metabolites among coral species (Sogin et al., 2014). Metabolomic profiling has also been used to explore the role of Symbiodinium and the coral-associated microbial community in coral biology and physiology (Gordon and Leggat, 2010; Hillyer et al., 2017; Sogin et al., 2017).

However, the coral metabolome is also known to change in response to external stimuli, such as exposure to abiotic conditions consistent with predicted climate change (Sogin et al., 2016) and contact with competitive algae (Greff et al., 2017; Quinn et al., 2016).

Changes in metabolomic profiles can be a result of altered gene expression (Hollywood et al., 2006; Sogin et al., 2014), but may also be due to post-transcriptional processes

(Patti et al., 2012). Given that genetically distinct individuals can respond differently to the same environment (Lohr and Patterson, 2017) and the link between genome and metabolome, it is reasonable to anticipate that unique genotypes in a common garden will exhibit differing suites of metabolites.

The link between genotype, phenotype, and metabolome has been explored in plant systems (Carreno-Quintero et al., 2013; Keurentjes, 2009). For example,

metabolomic profiling has demonstrated a relationship between cultivar and metabolites linked to desirable traits in a number of agriculturally-important species (Stewart et al.,

2007; Steinfath et al., 2010; Matsuda et al., 2012; Riedelsheimer et al., 2012).

Metabolomics can also be a useful tool for predicting pathological phenotypes in humans, and therefore has an important function in medicine (Sha et al., 2010). Despite potential important conservation applications, exploration of the relationship between genotype, phenotype, and metabolome in corals has been limited to date. A previous study found unique sesquiterpene signatures among species of the soft coral Sinularia with different morphological and anatomical traits, suggesting phenotype is also linked to metabolomic profile in corals (Kashman et al., 1982). However, no study has yet demonstrated unique metabolomic profiles among coral genotypes of the same species in a common garden.

To determine whether metabolomic profiles differ among reef-building coral genotypes, we employed proton-nuclear magnetic resonance spectroscopy (1H-NMR) and liquid chromatography-mass spectrometry (LC-MS) to identify and compare metabolomic profiles for three unique genotypes of the threatened staghorn coral

Acropora cervicornis in a common garden coral nursery. While 1H-NMR is less sensitive than LC-MS, it is ideal for quantifying, and identifying the structure of, unknown compounds (Markley et al., 2017), and is therefore useful for untargeted metabolic studies (Emwas, 2015), particularly of organisms such as corals that lack extensive metabolite databases for compound identification. However, LC-MS is far more sensitive, and can therefore resolve a higher number of metabolites, particularly less abundant compounds like secondarly metabolites (Emwas, 2015). Application of both

1H-NMR and LC-MS in tandem therefore improves coverage of the metabolome and enhances both metabolite resolution and identification. This exploratory work builds upon a previous study in which multiple phenotypes were characterized among these A. cervicornis genotypes at the same nursery (Lohr and Patterson, 2017). During this previous study, four traits were assessed and compared among genotypes, three related to growth (total linear extension, branch formation, calcification), and one related to thermotolerance (bleaching prevalence following a natural period of elevated sea temperature) (Lohr and Patterson, 2017). Three of these ten genotypes were selected for use in the present study: U41 (rapid growth, moderate bleaching prevalence), U44

(moderate growth, low bleaching prevalence), and U25 (slow growth, high bleaching prevalence) (Lohr and Patterson, 2017). We hypothesized that each of the three genotypes tested would have unique metabolomic profiles. These data could inform coral reef conservation and management and also provide useful information for future studies of metabolites in corals.

Methods

Sample Collection and Extraction

Corals used in this study were collected from an established coral nursery operated by the Coral Restoration Foundation (CRF) and located four miles offshore of

Tavernier, FL. All A. cervicornis genotypes in this nursery were previously determined to be unique via microsatellite genotyping performed by the Baums lab at Penn State

University (unpublished data).

In December 2016, two colonies from each of three genotypes (U25, U41, and

U44) were removed from grow-out structures at a depth of approximately 8 m in the nursery and brought to the surface intact. Corals remained submerged during sample

collection to minimize stress. Diagonal pliers were used to clip ~3 cm actively growing branch tips on each colony. A total of five replicate tips were collected from genotypes

U25 and U44, with three tips collected from the first replicate colony (A) and two collected from the second replicate colony (B). Six replicate tips were collected from genotype U41, with four tips collected from the first replicate colony (A) and two collected from the second replicate colony (B). Tips were placed in 20 mL scintillation vials containing 10 mL of 100% methanol, which were then immediately placed on ice in a cooler. Immersion in methanol has been shown to be an effective method for quenching metabolic activity (Canelas et al., 2008; Teng et al., 2009), and was more conducive to offshore field collections compared to snap freezing in liquid nitrogen.

Following sample collection, colonies were returned to the nursery and an existing logger (HOBO Pendant® UA-002-64, Onset Corporation) was downloadeded to determine temeprature at the time of and preceding sample collection. Samples were transported to shore and stored at -20°C overnight. Samples were transported back to the laboratory on ice and were again stored at -20°C overnight.

Methods for sample processing were modified from published methods (Gordon et al., 2013). The next day, intact fragments in methanol were agitated for 5 minutes, then allowed to settle for one hour in the -20°C freezer. One mL of extract from each sample was transferred to clean 1.5 mL microcentrifuge tubes and centrifuged at 20,000 g for 5 minutes. The supernatant was then transferred to a new 1.5 mL microcentrifuge tube and stored at -80°C until processing.

1H-NMR Profiling

All metabolomic analyses were performed at the Southeast Center for Integrated

Metabolomics (SECIM) at the University of Florida. Coral extract (in methanol) was added to double distilled water (1:2 v/v of sample to water), then flash freeze lyophilized

(Labconco) until dry. Lyophilized dry powder was re-suspended in phosphate buffer in

1 deuterium oxide (D2O) at pH 7. The final volume for the H-NMR samples was 60 μL (in a 1.5 mL tube) with 90 % (v/v) of deuterated 50 mM sodium phosphate buffer (pH 7) with 2 mM of ethylene diamine tetra-acetic acid (EDTA). The remaining 10 % (v/v) was occupied by an internal standard [5 mM D6-4,4-dimethyl-4-silapentane-1-sulfonic acid

(DSS-D6) and 0.2% sodium azide (NaN3) in D2O; Chenomx, Inc.].

All 1H-NMR spectra were collected with a 14.1 T NMR system, equipped with a

CP TXI CryoProbe and Avance II Console (Bruker Biospin). The first slice of a NOESY pulse sequence (tnnoesy) (Ravanbakhsh et al., 2015) was used to acquire proton spectra consisting of 1s relaxation delay (d1), 64 scans (nt), 100 ms mixing time, with 4s acquisition time over a spectral window (sw) of 7211.54 Hz. Samples were acquired at room temperature (25 oC). Before Fourier transformation, acquired spectra were further processed with a line-broadening factor of 0.5 Hz and zero filling to 65,536 points.

MestReNova 11.0.0-17609 (Mestrelab Research S.L.) was used to process the spectra.

Calibration on proton spectra was performed with respect to a DSS-D6 resonance peak at 0 ppm chemical shift. These spectra were phase corrected, followed by spline baseline correction to adjust spectral baselines.

Identification and quantification of the metabolites from the 1H-NMR spectra was done using Chenomx NMR Suite 8.2 (Chenomx, Inc.).

LC-MS Global Metabolomics

LC-MS global metabolomics samples were prepared by protein precipitation.

Briefly, 5 µL of internal standard mixture prepared in-house consisting of labeled amino acids was spiked into each 25 µL sample. Extraction was done by adding 200 µL of

8:1:1 Acetonitrile:Methanol:Acetone to the sample. Samples were held at 2-8 °C for 30 min to allow protein precipitation. Samples were centrifuged at 20,000 xg for 10 minutes at 4°C. From each sample, 190 µL supernatant was collected and dried completely under nitrogen at 30°C. Samples were reconstituted with 25 µL of reconstitution solution containing injection standards. Samples were mixed thoroughly, held at 2-8 °C for 10 min, and centrifuged at 20,000 xg for 10 min at 4 °C. Supernatants were transferred to vials for LC-MS analysis.

Global metabolomics profiling was performed on a Q Exactive Orbitrap mass spectrometer with UltiMate 3000 UHPLC (Thermo Fisher Scientific). All samples were analyzed in positive and negative heated electrospray ionization with a mass resolution of 35,000 at m/z 200 as separate injections. Separation was achieved on an ACE 18-

PFP 100 x 2.1 mm, 2 µm column (MAC-MOD Analytical) with mobile phase A as 0.1% formic acid in water and mobile phase B as acetonitrile. This is a polar embedded stationary phase that provides comprehensive coverage, but does have some limitation in the coverage of very polar species. The flow rate was 350 µL/min with a column temperature of 25 °C. 4 µL was injected for negative ions and 2 µL for positive ions.

MSConvert (ProteoWizard 3.0) was used to convert raw files to open format.

MZmine 2 was used to identify features, deisotope, align features, and perform gap filling to fill in any features that may have been missed in the first alignment algorithm.

All adducts and complexes were identified and removed from the data set. The data

were searched against SECIM’s internal retention time metabolite library of 1100 compounds and subsequently searched against KEGG for putative identification.

Mummichog, a program available on metaboanalyst.ca, was used to identify metabolic pathways driving separation among genotypes (see Li et al., 2013). The p- value and t.score along with the m/z value were used for pathway searching with a p- value cutoff of 0.01 and mass accuracy of 5 ppm in positive mode and 10 ppm in negative mode. A metabolic pathway reference library for the model species

Arabidopsis thaliana was used in this analysis, as no coral libraries were available.

Statistical Analysis

All statistical tests were conducted at a significance level of α = 0.05. For 1H-

NMR data, abundance of metabolites among three different genotypes of corals was analyzed via multivariate statistical analysis. The analysis used 15 spectra for a set (five samples per genotype). The spectra were binned (0.04 ppm) and locally aligned at several regions using MestReNova 11.0.0-17609. To reduce any bias that might have arisen because of sample handling and potential variability in the total amount of tissue per sample, normalization and scaling were performed on the data integrals prior to multivariate statistical analysis. Probability quotient normalization was used to reduce any possible variation in total signal intensity between the groups (see Dieterle et al.,

2006). Pareto scaling (mean centered and divided by the square root of standard deviation of each variable) was used to provide equivalent weight among the variables.

The web-based metabolomics data processing tool MetaboAnalyst 3.0 (Xia et al.,

2015) was used to perform one-way analysis of variance (ANOVA), principal component analysis (PCA), and partial least square discriminant analysis (PLS-DA) for both LC-MS

and 1H-NMR data. PCA is an unsupervised dimension reduction method that seeks to explain the maximum variation in a multivariate dataset without a priori information on sample groups (Want and Masson, 2010). PCA therefore provides an overview of variation in datasets, but principal components may not identify variables driving maximum separation among groups, as treatments/classes are not accounted for in the

PCA algorithm (Want and Masson, 2010). In contrast, PLS-DA is a supervised method that seeks to maximize separation among known groups (i.e. genotypes in the present study; Want and Masson, 2010). PLS-DA models maximum covariance between variables (i.e. metabolites) and treatmeant groups (i.e. genotypes) to best understand factors driving separation (Want and Masson, 2010). Thus, use of both PCA and PLS-

DA provides a more comprehensive analysis of patterns in metabolomic profiles among genotypes. For LC-MS, data from positive and negative ion modes were analyzed separately. Robustness of both PLS-DA models was validated by calculating Q2.

Variable Importance in Projection (VIP) was used to summarize the importance of each variable (i.e. metabolite) in driving separation among treatments (i.e. genotypes) in the

PLS-DA models (Davis et al., 2013; Eriksson et al., 2006). Compounds with a VIP values of >1 are generally considered to influential in PLS-DA models (e.g. Davis et al.,

2013; He et al., 2014). The present study used the conservative cutoff value of >2 to identify highly important compounds driving separation.

Results

During sample collection (~2 hours), temperature at the depth of the nursery trees was 26.5 – 26.6 °C. Logged data showed that mean daily temperature was < 31°C during the months preceding sample collection (i.e. Jul – Dec).

1H-NMR Profiling

ANOVA identified 59 compounds that differed significantly among genotypes (p <

0.05). Combined, PCA components 1 and 2 explained 71.6% of the total variance among genotypes (Fig. 3-1). PCA indicated relatively separate clustering for genotypes

U25 and U44, however U41 had greater within-group variability compared to the other two genotypes, and overlapped with both (Fig. 3-1). The PLS-DA model was well- validated (Q2 > 0.6). PLS-DA revealed separate clustering of all three genotypes, and components 1 and 2 explained 70.8% of the total variance among genotypes (Fig. 3-2).

Genotype U41 had a greater spread across components 1 and 2 compared to the other two genotypes, again suggesting relatively greater within-group variability. Compounds driving separation in the PLS-DA model (with a VIP score >2) are presented in Figure 3-

3. A full list of compounds identified by 1H-NMR, including putative metabolite identifications, is provided as supplementary data.

LC-MS Global Metabolomics

LC-MS detected a total of 1763 mass features in the positive mode and 718 mass features in the negative mode. ANOVA identified metabolites that differed significantly among genotypes (p < 0.05) in the positive ion mode (n = 508) and in the negative ion mode (n = 222). Figure 3-4 illustrates variation in the top 50 features identified by ANOVA among genotypes, including those that could be putatively attributed to known metabolites. U41 displayed greater variability among replicates in metabolite concentration compared to genotypes U25 and U44, but this variability did not appear to relate to donor colony (A versus B; Fig. 3-4). PCA components 1 and 2 described 43.6% (positive ion mode) and 44.6% (negative ion mode) of the total

variance among genotypes (Fig. 3-5). In PCA models derived from both the positive and negative ion mode, U44 clustered separately from both U25 and U41, however U41 had greater within-group variability compared to the other two genotypes (Fig. 3-5). PLS-DA results are presented in Figure 3-6. The PLS-DA model was valid (Q2 > 0.6), and revealed distinct clustering of metabolomic profiles among genotypes (Fig. 3-6).

Components 1 and 2 explained 37.3% (positive ion mode) and 40.3% (negative ion mode) of the total variance among genotypes. Greater spread of genotype U41 across components 1 and 2 compared to the other two genotypes also suggests higher within- group variability (Fig. 3-6). No specific features identified by PLS-DA were considered highly significant (with a VIP score >2). A full list of compounds resolved by LC-MS, primarily consisting of level 1 (confirmed structure), level 3 (tentative candidates), and level 5 identifications (exact mass, m/z) (Schymanski et al., 2014), is provided as supplementary data.

Mummichog pathway enrichment analysis for both positive and negative mode

LC-MS datasets identified aminoacyl-tRNA biosynthesis as a pathway that varied substantially among genotypes. Lysine biosynthesis (positive mode), as well as phenylalanine, tyrosine, and tryptophan biosynthesis (negative mode) differed based on genotype. In addition, metabolism of the amino acids arginine, proline, cysteine and methionine (positive mode) as well as glycine, serine, and threonine (negative mode) varied among genotypes. Purine (positive mode) and pyrimidine (negative mode) metabolism were also identified as important metabolic pathways driving separation among genotypes. A full list of metabolic pathways identified by Mummichog is provided as supplementary data.

Figure 3-1. Principal component analysis (PCA) model comparing 1H-NMR metabolomic profiles among three unique genotypes of A. cervicornis: U25 (red; slow growth, high bleaching prevalence), U41 (green; rapid growth; moderate bleaching prevalence), and U44 (blue; moderate growth; low bleaching prevalence). The amount of variance explained is shown in parentheses on each axis.

Figure 3-2. Partial least squares discriminant analysis (PLS-DA) model comparing 1H- NMR metabolomic profiles among three unique genotypes of A. cervicornis: U25 (red; slow growth, high bleaching prevalence), U41 (green; rapid growth; moderate bleaching prevalence), and U44 (blue; moderate growth; low bleaching prevalence). The amount of variance explained is shown in parentheses on each axis.

Figure 3-3. Compounds driving separation of 1H-NMR metabolomic profiles among genotypes determined by partial least square discriminant analysis. Only compounds with a Variable Importance in Projection (VIP) score of 2 or higher are presented. Compounds are listed from highest VIP score (top) to lowest VIP score (bottom). For peak pattern, s = singlet, d = doublet, t = triplet, m = multiplet. Circle size illustrates the relative concentration of each compound compared to other genotypes (smallest circles = low, largest circles = high). Compounds with an asterisk (*) indicate less certainty is associated with these putative annotations.

Figure 3-4. Heat map showing differences in concentrations among genotypes for the top 50 most significant features identified via analysis of variance from LC-MS in the (A) positive and (B) negative ion modes. Each column in the heat map corresponds to an individual replicate. Replicates are labeled at the bottom of the heat map with genotype, colony sampled (A or B), and replicate number.

Figure 3-4. Continued.

Figure 3-5. Principal component analysis (PCA) model comparing LC-MS metabolomic profiles obtained in the (a) positive and (b) negative mode among three unique genotypes of A. cervicornis: U25 (red; slow growth, high bleaching prevalence), U41 (green; rapid growth; moderate bleaching prevalence), and U44 (blue; moderate growth; low bleaching prevalence). The amount of variance explained is shown in parentheses on each axis.

Figure 3-5. Continued.

Figure 3-6. Partial least squares discriminant analysis (PLS-DA) model comparing LC- MS metabolomic profiles obtained in the positive (left) and negative (right) mode among three unique genotypes of A. cervicornis: U25 (red; slow growth, high bleaching prevalence), U41 (green; rapid growth; moderate bleaching prevalence), and U44 (blue; moderate growth; low bleaching prevalence). The amount of variance explained is shown in parentheses on each axis.

Figure 3-6. Continued.

Discussion

Efforts to conserve and restore threatened corals require innovative approaches in the face of growing anthropogenic stressors (van Oppen et al., 2017, 2015).

Metabolomic profiling can provide valuable physiological information on corals used in restoration. The three genotypes investigated in the present study were previously found to possess unique phenotypes for linear extension, branching, calcification, and bleaching prevalence in a common garden setting (Lohr and Patterson, 2017).

Phenotype and metabolomic profile have been linked in previous studies of plant systems (Carreno-Quintero et al., 2012; Schauer et al., 2006). We hypothesized that, given the previously documented differences in phenotype among the three coral genotypes sampled, metabolomic profiles for each genotype would also differ. Our results generally support this hypothesis; we found distinct differences in metabolomic profiles derived using both 1H-NMR and LC-MS approaches paired with PLS-DA for the three genotypes sampled. However, within-group variability in the metabolome for genotype U41 contributed to varying degrees of overlap with metabolomic profiles for the other two genotypes in PCA models. Although within-genotype variability could complicate intraspecific metabolomic profiling, this study demonstrates that distinct metabolomic profiles can be resolved for some phenotypically-distinct coral genotypes of the same species in a common garden setting.

These results contrast somewhat with a previous study of other coral species

(Sogin et al., 2014). Comparisons among five colonies of Porites compressa revealed relatively close clustering of metabolomic profiles compared to those found in the present study (Sogin et al., 2014). However, these inter-colony samples were more dispersed relative to concurrent intra-colony and technical replicates (Sogin et al.,

2014). Furthermore, sampled colonies of P. compressa were not genotyped or known to have unique phenotypes (Sogin et al., 2014). Overlapping genotypes or similar phenotypes may have resulted in the relatively high similarity of metabolomes among all five colonies profiled.

Although genetic factors are highly important, the environment also plays a key role in coral physiology. The present study was conducted in a common garden to control for the effect of environment in order to isolate differences among genotypes.

Coral nurseries can be valuable tools for such studies (Kuffner et al., 2017; Lohr and

Patterson, 2017; O’Donnell et al., 2017). Additionally, a number of studies have examined genotype-environment interactions (Elias et al., 2016; Todd et al., 2004) during outplanting and found considerable phenotypic plasticity across varying sites

(Drury et al., 2017; Forrester et al., 2013). Although little information is available regarding genotype-environment effects on the coral metabolome, environmental conditions are known to play a key role in determining metabolomic profile in other organisms (Carreno-Quintero et al., 2013; Harrigan et al., 2007; Keurentjes, 2009).

Future studies of corals should consider plasticity in metabolite profiles among diverse sites. Because corals are sessile organisms, they could be useful for investigating metabolomic shifts following transplantation to differing sites from a common garden.

Such information could provide insight into site selection for restoration activities.

In addition to spatial variation, temporal changes in abiotic conditions within a site can occur and potentially affect the metabolome. Temperature data suggest low thermal stress in 2016 compared to 2015. Daily mean temperature exceeded 31°C for two 10- day periods in late summer 2015, inducing sub-lethal bleaching (Lohr and Patterson,

2017), but daily mean temperature did not exceed this threshold in 2016. Because samples in this study were not collected during a period of thermal stress, key metabolites involved in physiological responses to elevated temperature may not be present in metabolomic profiles reported here. Recent studies have demonstrated metabolomic shifts in corals in response to thermal stress (Hillyer et al., 2018, 2017) and combined thermal and chemical stress (Sogin et al., 2016). Thus, differences in metabolomes among genotypes could be more or less apparent depending on abiotic conditions at the time of sampling. To better capture the full range of metabolites associated with a particular genotype, time series designs could be incorporated into future studies. In particular, sampling points during periods of high and low thermal stress could be useful in identifying metabolites associated with thermotolerance.

Although we were able to resolve distinct differences in metabolite profiles for all three genotypes using PLS-DA, higher within-genotype variability in metabolite profile was found for genotype U41. When metabolomic profiles were examined using PCA, this variability resulted in overlap of profiles between U41 and one or both of the other two genotypes using LC-MS and 1H-NMR, respectively. A heat map of the top 50 metabolites driving separation among genotypes also indicated varying concentrations of key metabolites among U41 replicates, with no apparent pattern based on the specific colony sampled. Together, these results indicate intracolonial variability in the metabolome for genotype U41. Interestingly, U41 was found to have within-genotype differences in linear extension in our previous study (Lohr and Patterson, 2017). Within- colony differences in metabolic processes related to linear extension could explain variability in the metabolome for genotype U41. Our study design may have been

particularly effective in resolving metabolites involved in coral growth, as we sampled actively extending distal branch tips (Gladfelter, 1982b). A sampling design that incorporates both distal and proximal branch portions within the same colony could provide additional information on metabolites related to growth in A. cervicornis.

Together, observations of within-genotype differences in growth and intracolonial variability in metabolomic profile for genotype U41 could indicate intracolonial variation in genotype (Schweinsberg et al., 2015). Potential intracolonial variation in genotype in

A. cervicornis, and particularly in genotype U41, warrants further study.

Although a lack of metabolite databases for A. cervicornis and other coral species (Sogin et al., 2014) presents an ongoing challenge in metabolomic profiling for these taxa, we were able to identify a number of putative compounds that varied among genotypes. Threonine was the only putatively annotated metabolite common to the top

50 driving separation among genotypes for both 1H-NMR and LC-MS (positive and negative mode). This amino acid was also identified as a compound driving separation of metabolite profiles among four unique coral species in a previous study (Sogin et al.,

2014). Although threonine was the only putatively identified compound common to the top 50 for 1H-NMR and LC-MS, a number of significantly different putative metabolites were identified by both approaches, including homoserine, leucine, creatine, and isoleucine (see Supplementary Data). The differences in the number and identity of putative metabolites resolved by 1H-NMR and LC-MS may highlight the varying strengths of each method. For example, LC-MS has higher sensitivity compared to 1H-

NMR, and can therefore detect a higher number of mass features, including very rare compounds (Emwas, 2015). However, the types of features detected may be dependent

on the chromatography technique applied. 1H-NMR captures a smaller number of metabolites, but is less susceptible to matrix effects, produces quantitative estimates of concentration without standards curves, and is therefore highly reproducible across labs, making it a valuable tool for untargeted metabolomics (Emwas, 2015). Application of both methods in the present study therefore provides a more comprehensive metabolomic profile of A. cervicornis compared to the use of either technique alone.

1H-NMR identified 15 compounds highly important in driving separation among the three genotypes compared in this study (VIP >2; Table 1). These compounds are primarily classified as carbohydrates and aliphatic compounds, however our solvent likely extracted more polar entities compared to non-polar compounds, such as fatty acids. Future studies should consider a solvent system that optimizes extraction of both polar and non-polar entities (e.g. chloroform:methanol:water, see Klueter et al., 2015).

Regardless of this limitation, 1H-NMR was able to putatively identify trimethylamine N- oxide (TMAO) as the compound with the most variability among genotypes. In many marine organisms, TMAO is well-known as an important osmolyte that prevents the damaging effects of urea buildup (Seibel and Walsh, 2002; Yancey et al., 1982) and hydrostatic pressure (Yancey et al., 2001) on proteins. The function of TMAO in the coral holobiont is less clear, however synthesis of this compound has been linked to protection against hydrostatic protein damage in other cnidarians (Yancey et al., 2004).

Dimethylglycine, another putative metabolite that was highly important in driving differences among genotypes, has previously been reported from 1H-NMR profiling of acroporid corals (Westmoreland et al., 2017). In transcriptomic studies, upregulation of a gene that converts betaine to dimethylglycine has been linked to low-light stress

(DeSalvo et al., 2012) and hypo-osmotic stress (Aguilar et al., 2017) in corals.

Concentrations of both TMAO and dimethylglycine were highest in genotype U44

(moderate growth, low bleaching prevalence) and lowest in genotype U25 (low growth, high bleaching prevalence). A possible link between these metabolites and a physiological function driving these phenotypes requires further study. Metabolomic studies of other acroporids were also consistent with our putative identification of choline, malonate, and methylguanidine (Westmoreland et al., 2017) as well as homoserine (Hillyer et al., 2017) in A. cervicornis in the present study. Unfortunately, there is currently a lack of published information to link these or any other highly significant metabolites in our study with potential physiological roles in the A. cervicornis holobiont, particularly as they relate to phenotype.

A number of important mass features driving separation of profiles for the three genotypes examined were also resolved by LC-MS. In contrast to 1H-NMR results, very few carbohydrates were resolved by LC-MS. Carbohydrates are known to fragment readily using MS-based techniques, and can therefore be more difficult to resolve (Fenn and McLean, 2011). In the LC-MS positive mode, only two of the top 50 mass features driving separation among genotypes were putatively identified (L-homophenylalanine and proprionylcarnitine), and we could locate no previous studies linking these metabolites to corals or their physiology. A higher number of the top 50 mass features driving separation in the negative mode were putatively identified. One such metabolite was catechin, a well-studied flavonoid known for its antioxidant properties (Heim et al.,

2002). Interestingly, a recent study found that addition of catechin in the laboratory could prevent bleaching in thermally-stressed Porites astreoides (Marty-Rivera et al.,

2018). In the present study, catechin was most abundant in U25 (Fig. 3-4), the genotype most susceptible to bleaching. One possible explanation for this result is that genotype

U25 gained a more thermotolerant dominant symbiont following bleaching (i.e. adaptive bleaching (Buddemeier and Fautin, 1993; Silverstein et al., 2014)) and this new symbiont produced higher amounts of catechin. Alternatively, catechin production may have been upregulated in the host as a result of its bleaching history. Host gene expression can change following bleaching, and these changes can persist well after corals have visually recovered (Pinzon et al., 2015). Further study is required to confirm the presence of catechin in A. cervicornis, and subsequently improve understanding of its role in coral physiology and stress tolerance. A number of other organic compounds common in many metabolic processes also varied among genotypes, including guanosine and hexose. Additional studies could also reveal more about specific roles of these metabolites in coral physiology and shed light on why they vary among genotypes. Interestingly, compounds associated with human activity, such as coronene and flusilazole, were also identified by LC-MS. Although corals are known to absorb pollutants from the marine environment (Glynn et al., 1989; Hanna and Muir, 1990;

Scott, 1990), it is unclear why absorption of such compounds would vary among genotypes in a common garden. It is also possible these compounds were misidentified during LC-MS analysis or unintentionally introduced during sampling. This question could be resolved in the future by performing MS/MS to identify features of interest with a higher level of certainty compared to the present study.

A number of metabolic pathways were found to drive separation of metabolite profiles among genotypes in both the positive and negative mode for LC-MS. Both

datasets identified biosynthesis of aminoacyl-tRNA, a critical component of translation and protein synthesis (Ibba and Soll, 2000), as the most important pathway driving separation. Pathways for the synthesis and metabolism of a variety of amino acids as well as purine and pyrimidine metabolism varied based on genotype, potentially also suggesting differences in transcriptional and translational activity among the three coral genotypes. Future transcriptomic and proteomic analyses of these genotypes could aid in identifying specific gene activity and proteins driving separation among genotypes, as well as their physiological role within the holobiont.

Metabolomics is an emerging technology in coral reef science, and clearly linking individual metabolites to physiological processes in corals is an important next step in this field. High within-genotype variability in gene expression among A. cervicornis has complicated attempts to identify transcriptomic biomarkers related to stress tolerance

(Parkinson et al., 2018). The present study suggests that metabolomic profiles could be easier to distinguish among some A. cervicornis genotypes, and therefore it may be possible to identify metabolite biomarkers for traits of interest. Better understanding of metabolomic variation in corals can assist with future efforts to identify key physiological processes related to growth and stress tolerance, and also support selection of robust genotypes for restoration (van Oppen et al., 2017, 2015). Future research on the metabolome of A. cervicornis and other coral species can aid in building databases to improve metabolite identification. Additional work and improvements in coral bioinformatics are needed to begin linking specific metabolites to the physiological processes that underlie phenotypes. This study demonstrates that the metabolome can

vary among conspecific corals with unique phenotypes, a first step toward linking phenotype and metabolite profiles in corals.

CHAPTER 4 DIFFERENTIAL DISTURBANCE EFFECTS AND PHENOTYPIC PLASTICITY AMONG RESTORED ACROPORA CERVICORNIS AT PATCH AND FORE REEF SITES

Introduction

The global decline in live coral cover over the past several decades has resulted in the development of active solutions for restoring ecosystem function on reefs (e.g.

Rinkevich, 1995; van Oppen et al., 2017; Young et al., 2012). Healthy coral reefs are critical for protecting shorelines, building complex habitat, and supporting economies

(Moberg and Folke, 1999), and a loss of live coral cover threatens these services. Coral gardening has therefore become an increasingly popular strategy for rebuilding coral populations (Rinkevich, 2014; Young et al., 2012). Coral gardening generally refers to propagation of fragments from wild donor colonies within in situ nurseries, followed by transplantation of nursery-reared colonies back to degraded reefs (Johnson et al., 2011;

Schopmeyer et al., 2017). This strategy has been adopted for coral species globally

(Rinkevich, 2014), and is particularly popular in the Caribbean, where live coral cover

(Gardner et al., 2003), reef carbonate production (Perry et al., 2013), and reef complexity (Alvarez-Filip et al., 2009) have declined during the past several decades.

Staghorn coral (Acropora cervicornis) is the most common species used in

Caribbean coral reef restoration (Young et al., 2012). A. cervicornis alone has declined by up to 95% throughout its range since the early 1980s, primarily due to disease outbreaks (Aronson and Precht, 2001; Precht et al., 2002). Since then, disease has continued to affect A. cervicornis (Miller et al., 2014a), as have other factors such as thermal stress (Ladd et al., 2017) and altered benthic communities (van Woesik et al.,

2017). Following its precipitous decline, A. cervicornis was listed as threatened under

the US Endangered Species Act, with restoration noted as an important strategy for recovery of this species (National Marine Fisheries Service, 2015). Desire to replace lost ecosystem function, as well as its fast growth rate and branching morphology, make

A. cervicornis an ideal species for coral gardening, and nursery culture for this species has been well-developed (Johnson et al., 2011; Young et al., 2012). To date, tens of thousands of A. cervicornis colonies have been reared in nurseries and outplanted to reefs in Florida and Puerto Rico alone (Schopmeyer et al., 2017).

Although restoration of A. cervicornis has been widely implemented throughout the Caribbean, there is interest in improving the overall success of population enhancement efforts, including improving long-term survival of restored coral populations, increasing efficiency of restoration efforts, and ensuring restored populations are resilient to climate change impacts. One strategy to optimize performance of restored coral populations is increasing the abundance of genotypes with desirable traits (van Oppen et al., 2017, 2015). Variation in such traits, including growth, thermotolerance, and disease resistance, has been observed among A. cervicornis genotypes (Drury et al., 2017; Kuffner et al., 2017; Ladd et al., 2017; Lohr and Patterson, 2017; O’Donnell et al., 2017; Vollmer and Kline, 2008). Increasing the proportion of coral genotypes with desirable traits used in population enhancement

(while maintaining genetic diversity) could improve overall coral restoration outcomes.

For example, outplanting a higher number of thermotolerant genotypes could increase the ability of a restored population to withstand future bleaching events (van Oppen et al., 2017). Similarly, fast growth and high rates of new branch formation have been proposed as indicators of coral fitness (Drury et al., 2017; Lirman et al., 2014) and could

enhance habitat provision (Schopmeyer and Lirman, 2015), while corals with higher skeletal density could better withstand mechanical damage following outplanting

(Enochs et al., 2014; Kuffner et al., 2017; van Oppen et al., 2017). Despite increased understanding of the extent to which these key phenotypes vary in a common garden, it remains unclear whether a given genotype’s traits are conserved following transplantation to a variety of sites. Previous studies have suggested that A. cervicornis may display high phenotypic plasticity in traits such as linear extension (Drury et al.,

2017; O’Donnell et al., 2017), while others, such as calcification, are more conserved

(Kuffner et al., 2017). Improved understanding of how phenotype changes from nursery to outplant site is important for evaluating the utility of a trait-based system for A. cervicornis restoration, in which key phenotypes are systematically identified and strategically used in population enhancement activities (Hunt and Sharp, 2014).

In addition to information on phenotypic plasticity, better understanding of A. cervicornis outplant performance across sites can help to inform and improve restoration activities. Site selection plays an important role in coral restoration outcomes, and a number of studies have reported high variability in survival, growth, and/or condition of corals among sites (Drury et al., 2017; Goergen and Gilliam, 2018;

Pausch et al., 2018). In Florida, A. cervicornis restoration has generally focused on sites historically dominated by this species, specifically fore reef zones (Miller et al., 2008;

National Marine Fisheries Service, 2015). In contrast, patch reefs are not often prioritized in restoration due to less availability of consolidated substrate and shallow depth, as well as higher fluctuations in temperature, salinity, and water quality (Lirman and Fong, 2007). However, relatively high abundances of healthy wild corals, including

A. cervicornis, have been observed on patch reefs (Lirman and Fong, 2007; Miller et al.,

2008), suggesting these may also be suitable habitats for restoration. Better understanding of differences in outplant performance at fore reef and patch reef sites could aid in refining site selection for A. cervicornis restoration.

Regardless of site, restored coral populations are continually subject to large- scale stressors, including disease epizootics and climate change impacts. The effects of anthropogenic climate change and associated ocean warming, particularly mass coral bleaching, have been well-documented (Hoegh-Guldberg, 1999; Hoegh-Guldberg and

Bruno, 2010; Hughes et al., 2018, 2017b). However, climate change may also affect coral reefs by increasing tropical cyclone intensity (Holland and Bruyère, 2014; Knutson et al., 2010). Tropical cyclones can benefit corals by reducing sea temperature and potentially mitigating coral bleaching (Carrigan and Puotinen, 2014) and also by facilitating asexual propagation in healthy coral populations (Highsmith, 1982).

However, high-intensity storms can severely damage coral populations (Puotinen et al.,

2016; Woodley et al., 1981). This is a concern for restoration practitioners, as population enhancement is generally labor-intensive, can be costly, and is hindered by methodological bottlenecks that currently limit the number of corals outplanted per unit effort. Hurricane damage therefore has the potential to reduce overall efficiency of coral restoration activities and increase the cost-per-unit-effort for surviving corals.

To better understand the effects of genotype and the environment on restored corals, we conducted a year-long study to determine (1) whether phenotype observed in a nursery is consistent following outplanting and (2) the effect of site type (patch reef vs. fore reef) on outplant performance. To address our first objective, we outplanted 192 A.

cervicornis colonies collected from nursery broodstock previously evaluated for intraspecific variation in phenotype in a common garden (Lohr and Patterson, 2017), and monitored growth and condition at outplant sites. Colonies were outplanted to two patch reef and two fore reef sites in a balanced design with regard to genotype in order to address our second objective. The passage of Hurricane Irma during the study period provided important insight into the effects of hurricane disturbance across site types.

Methods

Outplanting Experiment

Intraspecific variation in phenotype was previously determined for ten unique genotypes of A. cervicornis in a common garden nursery maintained by the Coral

Restoration Foundation (CRF; Lohr and Patterson, 2017). A subset of six genotypes from Lohr and Patterson (2017) were selected for use in the present study: K2, K3, U25,

U41, U44, and U77. These genotypes were selected as they were observed to have a diverse range of growth and thermotolerance traits in the nursery that warranted further study in an outplant setting. Fragments for outplanting (n = 32 per genotype) were clipped from the same colonies reared and examined during the previous study.

Collected colonies were individually tagged for identification (genotype, donor colony, and replicate), then brought to the surface and measured before transplantation to outplant sites. Colony length, width, and height were measured in order to calculate ellipsoid volume (EV) following Kiel et al. (2012). Total linear extension (TLE) was also recorded for each colony. Mean TLE of experimental colonies at the time of outplanting was 24.6 ± 0.6 cm (mean ± SE).

Colonies collected from the nursery were outplanted to four sites in a balanced design with regard to genotype (n = 48 colonies per site). Study sites were selected

from a larger subset of reef locations at which CRF was permitted by the Florida Keys

National Marine Sanctuary (FKNMS) to conduct outplanting activities (permit #FKNMS-

2011-159-A4). Two of the four sites were located on the fore reef (Snapper Ledge, Little

Conch), and two were located on offshore patch reefs (U4 Patch Reef, U14 Patch Reef) in Tavernier, Florida. Study sites were generally paired such that each patch reef site was directly inshore of a fore reef site (Fig. 4-1), and mean depth among sites was 4.7 ±

0.7 m (range: 3.7 – 6.7 m). Depth varied among sites, but did not relate to site type

(patch reef versus fore reef). At each site, outplanting was constrained within a 15 x 15 m plot. Colonies were outplanted haphazardly to each plot in a stratified random design, such that corals were always placed on consolidated substrate, no closer than one meter from adjacent outplants. Each coral was considered an independent replicate.

Corals were affixed to the substrate at three points of contact using a two-part marine epoxy produced by CRF. GPS coordinates of each outplant site were recorded.

Outplanted colonies were located at each subsequent monitoring visit by navigating to GPS coordinates, then visually locating corals from the surface using snorkel equipment. In the event corals could not be located on snorkel, the absence of corals was confirmed by systematically searching the area for a minimum of 30 minutes on SCUBA. Colonies were monitored for EV and condition at t = 51 d post-outplant. A second monitoring date was planned at t = 131 d post-outplant to capture any potential effects of thermal stress, but monitoring was obstructed by the passage of Hurricane

Irma as a Category 4 storm at t = 126 d post-outplant. Hurricane damage to local infrastructure delayed further monitoring to t = 248 d post-outplant, at which point EV

and condition of live colonies were recorded. At the conclusion of the experiment at t =

365 d post-outplant, EV, TLE, branches, and condition were recorded.

Expanded Dataset

To better understand the wider impact of Hurricane Irma among restored corals at patch reef and fore reef sites in the Upper Florida Keys, we utilized existing monitoring data from CRF. The dataset included six sites that were monitored prior to and after the passage of Hurricane Irma (n = 3 patch reef sites, n = 3 fore reef sites). At each of the six sites considered, A. cervicornis colonies were outplanted 15 – 44 d prior to the passage of Hurricane Irma (n = 145 – 451 colonies per site). Each site was then monitored 33 – 52 d post-hurricane. During monitoring, two divers haphazardly laid a 30 m transect within the outplant plot. Each diver swam along one side of the transect tape and observed each A. cervicornis outplant within a 30 x 3 m area. Colonies displaying

<100% mortality were considered alive (see Schopmeyer et al., 2017). Divers visually estimated the percent of each live colony affected by new and old partial mortality.

Mortality was considered new if corallites were intact and exposed skeleton was not yet colonized by other organisms, indicating mortality occurred within days of the observation (Lang et al., 2010). Mortality was considered old if corallites were colonized by thick macroalgae or invertebrates, indicating mortality occurred within months to years of the observation (Lang et al., 2010). Recorded types of mortality were old mortality (including macroalgal overgrowth, damselfish damage, and uncategorized old mortality), abrasion, breakage, predation, and sediment burial.

Data Analysis

Statistical analysis of all data was performed using R statistical software version

3.5.0 (R Core Team, 2018) and all tests were conducted at a significance level of α =

0.05. Means are presented as mean ± SE. Data were assessed for normality and homogeneity of variance using the Shapiro-Wilk test and Levene’s test, respectively.

Ellipsoid volume (EV) was standardized for each colony using the equation:

퐸푉푓 − 퐸푉푖 푆푡푎푛푑푎푟푑푖푧푒푑⁡퐸푉 = ⁡ 퐸푉푖 (4-1)

Standardized EV data (pooled across genotypes) did not meet the assumption of normality. Therefore nonparametric Scheirer Ray Hare tests (Sokal and Rohlf, 1995) in the R package ‘rcompanion’ (Mangiafico, 2016) were used to compare standardized pre-hurricane (t = 51 d) and post-hurricane (t = 365 d) EV (both including and excluding breakage) against genotype and site, as well as genotype and site type (patch reef vs. fore reef). Breakage was defined as a loss in EV for a given colony compared to initial

EV measured at the time the colony was outplanted. The majority of experimental corals were affected by breakage between t = 51 d and t = 248 d, the time period encompassing the passage of Hurricane Irma. It is therefore likely that net growth from t

= 0 to t = 248 was broadly confounded by hurricane-induced breakage. To better understand genotype-specific differences in growth, we isolated net growth between t =

248 d and t = 365 d (the period following the passage of the hurricane, when breakage was less prevalent), and compared this value among genotypes. Any colonies that experienced breakage during that time period were excluded from this analysis.

Standardized EV between day 248 and day 365 was log transformed to meet the assumption of normality, then compared among genotypes using ANOVA. Standardized

TLE was calculated using the equation:

푇퐿퐸푓 − 푇퐿퐸푖 푆푡푎푛푑푎푟푑푖푧푒푑⁡푇퐿퐸 = ⁡ 푇퐿퐸푖 (4-2)

As TLE was measured only at t = 0 d and t = 365 d, only post-hurricane standardized

TLE could be compared among genotypes. Standardized TLE data (including breakage) were log transformed to meet the assumption of normality, then compared among genotypes using ANOVA.

Phenotypic plasticity in growth was assessed by comparing standardized EV for each genotype among sites. Pre-hurricane standardized EV values (t = 51) were used in order to exclude hurricane effects on growth (i.e. breakage), and any colonies displaying breakage were excluded from analysis. With the exception of U44, standardized EV values for each of the genotypes were log transformed to meet the assumption of normality, then standardized EV for each genotype was compared among sites using ANOVA.

Chi-square tests were used to determine whether frequency of colony breakage varied among sites, site types, and genotypes. Specific tests included breakage frequency (1) among sites (n = 4) pre-hurricane (t = 51 d) and (2) post-hurricane (t =

365 d), (3) among site types (patch reef and fore reef) pre-hurricane, and (4) among genotypes pre-hurricane and (5) post-hurricane. Survival patterns at the conclusion of the study (t = 365 d) were compared among sites (n = 4) and site types (patch reefs vs. fore reefs) using Kaplan-Meier Survival Analysis in the R package ‘survival’ (Therneau,

2015), with each coral considered as a replicate and site or site type considered treatments.

For the expanded dataset, post-hurricane survival was compared between fore reef and patch reef sites using a chi-square test. Percent old and new partial mortality per colony were compared between site types using Mann-Whitney-Wilcoxon tests.

Partial mortality was categorized based on whether it could be attributed to causes acting before the passage of Hurricane Irma (i.e. old mortality), during the passage of

Hurricane Irma (i.e. breakage, abrasion), or after the passage of Hurricane Irma (i.e. recent predation, disease, sediment burial). Prevalence of each category of partial mortality (before, during, and after Hurricane Irma) was compared between fore reef and patch reef sites using chi-square tests.

Results

At t = 51 d of the experiment, growth (in terms of standardized EV) did not vary among genotypes, sites, or site types, regardless of whether breakage was included or excluded. Differences in standardized EV among sites were found only for genotype

2 U41 (F3,14 = 3.79, p < 0.05; Fig. 4-2). Breakage varied among both sites (X = 32.81, p <

0.0001) and site types (X2 = 8.10, p < 0.01), and was highest at Little Conch and fore reef sites, respectively. Breakage did not vary among genotypes.

Following the passage of Hurricane Irma, no evidence of experimental outplants was located at either fore reef site (Little Conch and Snapper Ledge; Fig. 4-3) after both snorkel and SCUBA searches. At remaining sites, neither standardized EV nor standardized TLE at t = 365 d varied among genotypes, regardless of whether breakage was included or excluded. Standardized EV between t = 248 d and t = 365 d also did not vary among genotypes. Breakage did not vary among either sites or genotypes at t

= 365 d.

Survival patterns for experimental outplants varied among sites (Kaplan-Meier X2

= 45.4, p < 0.0001) at the conclusion of the experiment (Fig. 4-4). Survival also varied among site types (patch reef vs. fore reef; Kaplan-Meier X2 = 41.6, p < 0.0001), and was highest for patch reef sites (51.04 ± 9.38% compared to 0.00% survival at fore reef

sites). Post-hurricane survival for CRF monitoring sites was also significantly higher at patch reef sites compared to fore reef sites (patch reef = 43.06 ± 13.83%, fore reef =

17.06 ± 8.25%; X2 = 159.61, p < 0.00001). Percent partial old mortality per remaining live colony was also significantly higher on CRF-monitored fore reefs compared to patch reefs (fore reef = 15.06 ± 2.79%, patch reef = 5.22 ± 1.10%; W = 8048.5, p < 0.01), however percent partial new mortality did not vary between site types. The proportion of colonies affected by partial mortality attributed to causes acting before, during, and after

Hurricane Irma did not differ between site types.

Figure 4-1. Map displaying the location of experimental outplant sites and the nursery of origin. Land is shaded dark gray, and reefs are shaded light gray. Two sites were located on fore reefs, and are represented by circles (Little Conch, Snapper Ledge) and two sites were located on patch reefs, and are represented by squares (U4 Patch Reef, U14 Patch Reef). Study sites were paired such that each patch reef site was inshore of a fore reef site. Map layers were sourced from FWC (2001) and the University of Florida GeoPlan Center (2015).

Figure 4-2. Standardized change in ellipsoid volume (EV) for each genotype among sites at t = 51. Error bars represent SE. Differences in standardized EV among sites were found for genotype U41 (F3,14 = 3.789, p < 0.05). Letters denote differences among sites for genotype U41 as determined by Tukey HSD.

A B

Figure 4-3. Visual differences post-hurricane between (A) Snapper Ledge at t = 365 and (B) U14 Patch Reef at t = 248. No outplanted A. cervicornis could be located at either fore reef site, and reefs were generally dominated by macroalgae and bare substrate. In contrast, 51.04 ± 9.38% of outplanted A. cervicornis survived at patch reef sites post-hurricane, and less bare substrate was available. Photos courtesy of author.

Figure 4-4. Kaplan-Meier survival plot illustrating differences in survival proportion of outplanted colonies among sites over 365 days. Survival varied significantly among sites (Kaplan-Meier X2 = 45.4, df = 3, p < 0.0001) and site types (patch reef vs. fore reef; Kaplan-Meier X2 = 41.6, df = 1, p < 0.0001). Because no corals could be located at either fore reef site, lines overlap at t = 248 for Snapper Ledge and Little Conch.

Discussion

Coral restoration aims to assist in the recovery of degraded reefs, often via propagation and outplanting of coral colonies. However, a number of factors, including site selection, can affect the success of such efforts. The present study revealed a dramatic disparity in the ability of restored corals at fore reef versus patch reef sites to withstand hurricane disturbance. No coral outplants were located following the passage of Hurricane Irma at either fore reef site. In contrast, approximately half of the corals outplanted at patch reef sites were alive post-hurricane. A similar trend was found when our dataset was expanded to include additional fore reef and patch reef sites in the same region. Post-hurricane persistence of A. cervicornis on patch reefs in our study is consistent with distribution patterns for wild A. cervicornis in the Florida Keys in recent decades (Miller et al., 2008), and suggests that some fore reef sites may not be ideal candidates for restoration using current methods, despite historically high abundances of A. cervicornis at such sites.

Hurricane damage has negatively affected A. cervicornis populations for decades, even prior to the widespread decline of this species in the Caribbean (Gardner et al., 2005; Stoddart, 1962; Woodley et al., 1981).However, hurricane-affected populations were historically observed to recover rapidly following hurricane damage

(Shinn, 1976). Recovery was largely possible because the complex reef structure formed by benthic communities, particularly A. cervicornis thickets, facilitated retention of broken fragments within the reef system (Highsmith, 1982). The loss of complex reef structure (Alvarez-Filip et al., 2009) could therefore explain why no corals were retained at fore reef sites post-hurricane. Our results are also supported by a recent study, which found relatively low rates of survival for unattached A. cervicornis fragments (Mercado-

Molina et al., 2014), casting doubt on the ability of colonies dislodged by the hurricane to recover. In contrast, patch reef sites were located further inshore, and protected by a longer distance of reef structure compared to fore reef sites (Fig. 4-1), which may explain their relatively high survival rates. The reef crest, which alone dissipates 86% of wave energy (Ferrario et al., 2014), lies between fore and patch reefs, and therefore provides an added measure of protection from wave disturbance on patch reefs.

Together, these results indicate that restoration practitioners should consider increasing effort at patch reef sites when striving to restore A. cervicornis. However, given that this species was historically less abundant at patch reef sites compared to fore reef sites

(Miller et al., 2008), potential negative effects of competition with patch reef species should be carefully considered in restoration strategies.

Although our results suggest that restoration efforts on patch reefs could be expanded, they also indicate that practitioners, managers, and researchers should strive to improve resilience among restored populations at fore reef sites. With regard to hurricane impacts, increasing the scale of restoration could potentially enhance resilience. For example, outplanting larger A. cervicornis colonies at a higher density could allow hurricane-dislodged colonies to be retained at target reefs and recover (as described in Highsmith, 1982). Outplanting smaller fragments (~25 cm TLE) at a density of 1 m-2 in the present study was likely insufficient to facilitate reestablishment of broken fragments. The need to increase the scale of coral restoration to improve reef function is increasingly recognized (Edwards and Gomez, 2007; Hein et al., 2017; Lirman and

Schopmeyer, 2016; Rogers et al., 2015; Shaver and Silliman, 2017). However, labor- intensive outplanting practices currently limit the scale of A. cervicornis restoration. In

the present study, outplanting efficiency averaged ~11 outplants hour-1 person-1. A pneumatic drill-based method was shown to increase outplanting efficiency to 30 outplants hour-1 person-1 for the Indo-Pacific species Stylophora pistillata and

Pocillopora damicornis (Horoszowski-Fridman et al., 2015). Continued development of such innovative strategies is required to improve outplanting efficiency for nursery- reared corals in order to scale up restoration. In concert with improvements to outplant efficiency, a more holistic approach to coral restoration could increase resilience at target reefs. Although Caribbean coral restoration has largely focused on acroporids

(Young et al., 2012) and population enhancement for these species is an important goal, biodiversity is recognized as a key factor in ensuring high resilience of ecological systems (Epstein et al., 2003; Shaver and Silliman, 2017). Interest in and efforts to develop methods for restoring multiple coral species (Forsman et al., 2015; Lirman and

Schopmeyer, 2016; Shaver and Silliman, 2017) as well as other species key to reef function, such as the herbivore Diadema antillarum (Sharp et al., 2018), are increasing among coral restoration practitioners. An example from seagrass restoration demonstrates that survivorship increases with species richness among restored communities (Williams et al., 2017). Interspecies interactions can improve resilience of restored coral populations through a variety of mechanisms, including increased herbivory, mutualistic interactions, and facilitation, among others (Shaver and Silliman,

2017). Increasing focus on both scaling up restoration and increasing biodiversity in such efforts could therefore enhance the ability of reefs to recover from hurricane disturbance.

Prior to hurricane effects, patch reefs were characterized by lower rates of initial breakage compared to fore reef sites. This observation supports the idea that even in the absence of a tropical cyclone, less wave energy may have acted upon patch reef outplants due to protection by the reef crest, which has a key role in dissipating wave energy (Ferrario et al., 2014). However, the frequency of breakage among outplanted A. cervicornis also varied between fore reef sites, with higher frequency of breakage at

Little Conch compared to Snapper Ledge. This could be a result of structural differences or differences in depths between these sites. Maximum depth at Little Conch was 4.0 m, while Snapper Ledge was somewhat deeper at 6.7 m. Previous studies have found increased rates of colony breakage with decreased depth (Lohr et al., 2017; Tunnicliffe,

1981). In addition to breakage, the mean percent of each A. cervicornis colony affected by partial old mortality at CRF outplant sites was also lower at patch reef sites compared to fore reef sites. This is consistent with previous observations of lower partial mortality among a number of coral species at patch reef sites along the Florida Reef

Tract (Lirman and Fong, 2007). Although the percent of each colony affected by old mortality was lower at patch reef sites, the proportion of colonies per site affected by old mortality (pre-hurricane) did not vary between site types. Similarly, the prevalence of partial mortality attributed to causes acting during and after the passage of Hurricane

Irma did not differ between site types, suggesting that differences in hurricane survival between CRF-planted fore and patch reef sites likely resulted from complete removal of colonies, consistent with observations at experimental outplant plots in the present study. Other indicators of colony condition, such as predation, disease, or bleaching were not observed among experimental outplant plots, and it is unclear how these may

differ among sites over a longer study period. Observations of higher coral cover and faster growth on patch reefs compared to fore reefs along the Florida Reef Tract

(Lirman and Fong, 2007), as well as high bleaching resilience on Upper Florida Keys patch reefs (Gintert et al., 2018) support the idea that restored corals may generally have more success at patch reef sites. In contrast, Pausch et al. (2018) observed higher bleaching prevalence among restored A. palmata at Upper Florida Keys patch reef sites compared to fore reef sites. More frequent, long-term observations of restored A. cervicornis at both fore reef and patch reef sites could aid in further resolving differences in colony condition between site types. In particular, improved understanding of how bleaching patterns may differ among site types is critical to ongoing A. cervicornis restoration, as thermal stress events are expected to become more frequent over time (Hughes et al., 2018).

Although spatial variation in colony condition was observed, site did not have an effect on growth for live colonies after 51 or 365 days, nor did genotype. This contrasts starkly with a number of previous studies that have reported significant differences in growth among sites for restored A. cervicornis (Drury et al., 2017; Goergen and Gilliam,

2018; Lohr et al., 2017) as well as A. palmata (Forrester et al., 2013). Although physical characteristics (i.e. reef zone) varied among sites, it is possible that other factors, such as water quality, were similar across sites and had an effect on growth that outweighed site-specific factors. For example, nutrient enrichment (Renegar and Riegl, 2005) and turbidity (Kendall et al., 1985) have both been shown to alter growth rate in A. cervicornis. However, it is more likely that growth results were confounded by hurricane impacts, both through complete removal of colonies (and subsequent reductions in

sample size) and alteration of natural growth rates via high rates of breakage among remaining colonies. Hurricane effects are also likely an important factor in the observed lack of difference in growth among genotypes. The absence of an effect of genotype on growth is unusual compared to previous literature on A. cervicornis (Bowden-Kerby,

2008; Drury et al., 2017; Goergen and Gilliam, 2018; Ladd et al., 2017; Lirman et al.,

2014; O’Donnell et al., 2017), and is particularly surprising given that the same colonies grew at significantly different rates based on genotype when measured in the nursery prior to outplanting (Lohr and Patterson, 2017). However, at least one previous study reported no effect of genotype on growth rate of outplanted A. cervicornis (Ladd et al.,

2016). Ladd et al. (2016) found that outplant density was a more important factor in driving differences in growth compared to genotype. Although outplant density was consistent throughout the present study, it is possible that environmental factors among or across sites had a stronger effect on growth than genotype. For example, growth over the duration of the present study was likely confounded by hurricane effects.Additionally, pre-hurricane measurements (at t = 51 d) may have been taken too early to fully resolve differences among either sites or genotypes. Previous studies comparing differences in nursery growth among A. cervicornis genotypes over time illustrate that differences in intraspecific growth rate typically become more apparent over time (Lohr and Patterson, 2017; O’Donnell et al., 2017). Previous studies have found intraspecific variation in growth rate among restored A. cervicornis genotypes after up to 3 months of study (Drury et al., 2017) or more (Goergen and Gilliam, 2018;

Ladd et al., 2017).

Although genotype and site did not affect growth among outplants overall, when site-specific growth was compared individually for genotypes after 51 days, genotype

U41 displayed plasticity in growth among sites. Similar plasticity has been previously observed in A. cervicornis (Drury et al., 2017; Kuffner et al., 2017), as well as A. palmata (Pausch et al., 2018). Had genotypes grown over a longer study period in the absence of hurricane impacts, it is possible that other genotypes would also have exhibited plasticity in growth. Although plasticity in growth appears to be relatively common in acroporids, some genotypes are considered ‘generalists’ that have conserved growth phenotypes across a number of sites (Drury et al., 2017; Pausch et al., 2018). Identifying such ‘generalists’ versus ‘specialists’ characterized by site-specific performance could be beneficial for ongoing A. cervicornis restoration efforts, and suggests that a trait-based restoration strategy (Hunt and Sharp, 2014) could be useful for a subset of genotypes. However, such trait-based strategies must strive to consider traits beyond growth. Growth has been well-studied in A. cervicornis (e.g. Lirman et al.,

2014; O’Donnell et al., 2017; Goergen and Gilliam, 2018), as it is relatively simple to measure in laboratory, nursery, and outplant settings compared to other traits. Although growth has been proposed as an indicator of coral health (Lirman et al., 2014), studies of acroporids have also identified tradeoffs between growth and traits such as skeletal density (Lohr and Patterson, 2017; Kuffner et al. 2017) and thermotolerance (Jones and

Berkelmans, 2010). A similar relationship between growth and disease resistance could also be present among corals, as such a tradeoff has been found in other systems (e.g.

Huot et al., 2014). These findings demonstrate that a variety of traits must be

considered when aiming to restore A. cervicornis, and fast growers should not necessarily be broadly prioritized for restoration.

The effects of Hurricane Irma on experimentally outplanted A. cervicornis plots complicate our findings of consistency in phenotype across genotypes and sites. It is understandably difficult to predict environmental conditions during long-term field studies, however repeating this study in the absence of major disturbance could aid in determining whether and how phenotype may vary spatially and temporally throughout the restoration process. Although unexpected, the effect of Hurricane Irma on restored

A. cervicornis provides important new insight that could guide decision-making for restoration programs. The increased ability of A. cervicornis to withstand hurricane impacts at patch reef sites requires further research. In particular, future studies should aim to collect information on rugosity and wave energy at fore reef and patch reef restoration sites to better understand differences in outplant performance. Furthermore, additional research must address increasing resilience in coral restoration efforts, including scaling up outplanting capacity and enhancing biodiversity to ensure a wider range of ecological services are restored. Such improvements could increase the success of restoration at fore reef sites and also increase the ability of these sites to withstand and recover from future disturbance. Although the mechanism for differences in outplant performance among site types is not yet clear, our results suggest that at present, increased effort should be directed toward restoring patch reef sites within the broader scope of restoration programs. Overall, diversifying the reef types targeted for outplanting could increase the scope and impact of restoration while spreading the risk from disturbance across a wide array of habitats.

CHAPTER 5 ASSESSMENT OF WILD AND RESTORED ACROPORA CERVICORNIS ACROSS THREE REEF ZONES IN THE CAYMAN ISLANDS*

Introduction

Staghorn coral Acropora cervicornis is a fast-growing, branching species that historically occurred across a wide range of depths (i.e. 1-25 m, Aronson et al., 2008a) and reef types (i.e. fore reef, backreef, and lagoonal habitats, Aronson and Precht,

2001) in the Caribbean. A. cervicornis is critical for building three-dimensional structure on reefs, creating habitat for reef-dwelling species (Bellwood et al., 2004; Nyström,

2006; Young et al., 2012) and providing shoreline protection (Ferrario et al., 2014).

Though once spatially dominant in the Caribbean (Pandolfi and Jackson, 2006), this species has declined throughout its range and is now designated as critically endangered by the International Union for Conservation of Nature (Aronson et al.,

2008a). Combinations of stressors including disease (Gladfelter, 1982a), temperature stress (Lesser et al., 2007), pollution (Bak, 1987), and hurricanes (Woodley et al., 1981) have led to the decline of A. cervicornis. Remnant populations of A. cervicornis generally persist at low abundances throughout the Caribbean (Aronson et al., 2008a), but natural recovery has been limited due to low rates of sexual reproduction (Aronson and Precht, 2001).

To mitigate this decline, coral gardening has been widely adopted to grow new colonies via asexually propagating portions of healthy, wild colonies (Bowden-Kerby,

* This chapter is reprinted from Lohr, K.E., Cook McNab, A.A., Manfrino, C., Patterson, J.T., 2017. Assessment of wild and restored staghorn coral Acropora cervicornis across three reef zones in the Cayman Islands. Reg. Stud. Mar. Sci. 9,1–8.

2001; Rinkevich, 1995). The primary goal of coral gardening is to create a self- sustaining stock of maricultured coral within a nursery, which can later be transplanted to restore degraded reefs (Epstein et al., 2003, 2001; Rinkevich, 1995; Young et al.,

2012). Transplantation from nurseries to reefs, known as outplanting, immediately increases reef complexity and local abundance of A. cervicornis at the outplanted site.

In addition, A. cervicornis is typically outplanted in multi-genotype clusters, thereby creating opportunities for future sexual reproduction, which can increase genetic diversity and enhance natural recovery of the species (Johnson et al., 2011).

Thousands of A. cervicornis colonies from dozens of nurseries have been outplanted throughout the Caribbean (Young et al., 2012).

Best practices for outplanting nursery-reared A. cervicornis recommend selecting sites with factors including, but not limited to, suitable water quality, solid substrate, and the presence of wild A. cervicornis (Johnson et al., 2011). Following region-wide declines, A. cervicornis was completely extirpated from some reefs (Aronson and

Precht, 2001). Due to multiple stressors, some reefs have also transitioned from a coral- dominated state to a macroalgal-dominated state, which can inhibit coral settlement, growth, and survival (Fung et al., 2011; Hughes, 1994). For this reason, systems that have been severely degraded and transitioned to an alternative state may be unsuitable for restoration of formerly-present species (Suding et al., 2004). Therefore, the presence of remnant wild A. cervicornis in a target region provides evidence that local conditions can support at least some genotypes of this species, and therefore some potential for successful restoration exists (Johnson et al., 2011).

Recent efforts have integrated field observations, satellite data, and modelling to prioritize sites for Caribbean Acropora re-establishment on the scale of km2 (Wirt et al.,

2015). However, selection of outplanting sites within and among reefs with conditions suitable for restoration can still be subjective, and the fate of outplanted colonies may vary among sites. For example, in a study from Belize, outplant survivorship after one month ranged from 38–97% between plots at the same location (Bowden-Kerby and

Carne, 2012). Similar variability in survivorship among adjacent outplant plots has been observed in the Cayman Islands (Lohr, unpublished data). However, in Puerto Rico, no difference in survivorship was reported between two plots of outplanted A. cervicornis at similar sites several km apart, although differences in growth were found (Mercado-

Molina et al., 2015). In addition to understanding how outplant performance varies among sites, it is critical to characterize differences in restoration outcomes among diverse reef zones. Prior outplanting studies have investigated sites with depths ranging from 1 m (Bowden-Kerby and Carne, 2012) to 12 m (Hollarsmith et al., 2012). Shallow reefs in particular have been identified as important targets for coral restoration because of their ability to reduce wave energy (Ferrario et al., 2014). Although A. cervicornis occurs in shallow reef zones such as backreefs, it was historically dominant in deeper fore reef habitats (e.g. Goreau, 1959; Logan, 2013). It is therefore important to develop the ability to restore A. cervicornis in each zone where it occurs.

Despite the need to restore A. cervicornis across a diverse range of site types, the extent to which a single nursery may be used to conduct restoration among sites at varying reef zones has received little attention. Best practice guides recommend collecting, rearing, and outplanting coral at similar depths (Edwards and Gomez, 2007;

Johnson et al., 2011), however the authors are unaware of a study which has systematically investigated the outcomes of, and potential limitations for, conducting restoration among reef zones of varying depth. Previous outplanting studies of A. cervicornis have compared outplant performance among similar sites of depths within a range of no more than ~3m (e.g. Bowden-Kerby and Carne, 2012; Hollarsmith et al.,

2012; Mercado-Molina et al., 2015). However, corals are capable of acclimating to new environments, including a variety of temperature (Mayfield et al., 2012; Palumbi et al.,

2014) and light (Titlyanov et al., 2001) regimes, as well as high- and low-energy sites

(Smith et al., 2007). Therefore, it is possible that A. cervicornis outplants sourced from a single nursery location would be able to acclimate to multiple types of outplant sites.

Scientists, resource managers, and governments are interested in better understanding the outcome of A. cervicornis restoration efforts in diverse reef zones.

For example, the Cayman Islands government is in the process of expanding its coral nursery program to multiple locations nationwide, and information on outplant success among reef zones may be of interest. Reefs in the Cayman Islands consist of a deep wall system with well-defined zonation by depth (Fig. 5-1, Logan, 2013, 1994).

Developing the ability to effectively enhance A. cervicornis populations in each of these zones is a critical next step for restoration programs in the Cayman Islands, and could also have relevance to other Caribbean sites with distinct reef zonation.

As a first step in addressing this issue, we conducted a study to test growth and survivorship among outplants within three reef zones at Little Cayman, Cayman Islands: the shallow back reef (0-3 m), the intermediate spur-and-groove reef (8-15 m), and the deep reef terrace (>15 m) (Figs. 5-1, 5-2). Outplants in this study were sourced from an

existing A. cervicornis nursery at Little Cayman, which was established in 2012 at a depth ranging from 4 to 7 m (see Lohr et al., 2015). To determine the suitability of each zone for A. cervicornis restoration activities, we simultaneously surveyed and characterized the extant wild population in each zone. Recent surveys of Little Cayman reefs have not described the extant distribution of A. cervicornis (e.g. Manfrino et al.,

2013), however anecdotal evidence and previously published data suggest this species occurs in both the shallow back reef and deeper spur-and-groove reefs in the Cayman

Islands (Manfrino et al., 2003). We tested two hypotheses: (1) Wild A. cervicornis will be present in each of the three reef zones studied, and the condition (in terms of predation and disease prevalence) of each population will be similar, (2) Outplanted A. cervicornis will not differ in growth, survivorship, or condition (predation, disease prevalence, and post-outplant breakage) among reef zones. Information provided in this study could be used to inform future A. cervicornis restoration efforts in the Cayman Islands and the wider Caribbean.

Methods

Wild Population Surveys

Surveys were conducted throughout the western half of Little Cayman between

June and August 2015 to determine the extent and condition of wild Acropora cervicornis populations within the shallow backreef, intermediate spur-and-groove reef, and deep reef terrace. For the purposes of this study, these reef zones will be identified as shallow, intermediate, and deep, respectively. Given that A. cervicornis is sparse on

Little Cayman reefs (Manfrino et al., 2003), surveys were conducted using the roving diver method (modified from Schmitt et al., 2002). Shallow zone surveys were

completed on snorkel, while intermediate and deep zone surveys were done using

SCUBA. Approximately five hours of search time was applied in each reef zone.

Upon locating an A. cervicornis colony, site depth was recorded and the colony’s maximum diameter and height were measured using a PVC meter-stick marked with 10 cm increments. Occurrence of predation was determined for each colony by identifying and recording the presence or absence of predation scars following Miller et al. (2014a; the gastropod Coralliophila abbreviata), Miller et al. (2014b; the fireworm Hermodice carunculata), and Schopmeyer and Lirman (2015; farming damselfish, Stegastes spp.).

Occurrence of disease was determined for each colony by identifying and recording the presence or absence of disease signs following Miller et al. (2014a). Prevalence was then calculated for each type of predation and for disease as the percent of the total population affected by each condition.

Outplanting Experiment

A. cervicornis colonies grown in the Little Cayman nursery were outplanted in

May 2015. Outplanted colonies were clipped from established colonies that had been propagated within the nursery since 2012. A total of 60 outplants ranging in size from 11 to 33 cm total linear extension (TLE, Lirman et al., 2014; mean ± SE = 19.57 ± 0.75) were utilized for the study and were randomized across treatments. Four unique genotypes identified in the Little Cayman nursery in 2014 via genotyping-by-sequencing

(C. Drury, University of Miami/RSMAS, unpublished data) were used for this study. Of the 60 colonies clipped for outplanting, 24 colonies were sourced from genotype A, and

12 each were sourced from genotypes B, C, and D. Colonies were outplanted in a balanced design with respect to genotype.

100

Outplanting plots were selected at three distinct reef zones along a single longitudinal transect perpendicular to shore (Fig. 5-2). All plots were located within 200 m of the nursery site (Fig. 5-2), and were within the range of concurrent wild surveys.

Two plots of 10 colonies each were established in the shallow, intermediate, and deep reef zones at depths of 1, 10, and 16 m, respectively. Photosynthetically active radiation

(PAR; μmol m-2 s-1) was measured at solar noon on a clear day at the nursery and at each reef zone (n = 10 measurements per site) using a handheld PAR meter (Apogee

Instruments, USA). At each plot, masonry nails were installed into areas of bare substrate in a 1 x 4 m grid. Outplants were attached to nails using cable ties and secured to the substrate at the base of the nail using a two-part epoxy putty (Johnson et al., 2011; Lirman et al., 2014). Scaled photographs of outplanted colonies were taken at approximately 30-day intervals over the course of 85 days (n = 4 intervals) using a digital camera (Olympus, Japan). Condition of each colony was recorded, including signs of disease and predation, using the same methods as the wild A. cervicornis surveys. Broken colonies (i.e. colonies missing a portion of skeleton and tissue with an exposed skeletal lesion visible) were noted. Frequency of breakage was calculated as the percent of broken colonies per zone by the end of the experiment. The total number of branches (≥ 1 cm) was counted, and TLE of each colony was measured using

ImageJ 1.43 software (National Institutes of Health, USA). Temperature was recorded continuously in situ at one plot within each reef zone using HOBO Pendant® data loggers (Onset Computer Corporation, USA).

101

Statistical Analysis

Statistical analysis of experimental and survey data was conducted using R statistical software version 3.1.2 (R Core Team, 2014). All statistical tests were performed at a significance level of α = 0.05 and all means are presented as mean ±

SE. Normality and homogeneity of variance were assessed using the Shapiro-Wilk test and Levene’s test, respectively. Wild colony size data did not meet the assumptions of normality and homogeneity of variance, and was therefore compared among reef zones using a nonparametric Kruskal-Wallis test followed by Dunn’s post-hoc test in the R package ‘dunn.test’ (Dinno, 2015). Chi-square tests were conducted to determine whether prevalence of disease or any type of predation differed between populations of

A. cervicornis located in varying reef zones.

Initial outplant size data were log transformed to meet the assumptions of normality and homogeneity of variance. These data were compared among treatments using ANOVA to test for any differences in initial size of outplanted colonies among reef zones. Net growth and net change in number of branches were compared among reef zones using ANOVA with a Tukey HSD post-hoc test. Differences in colony survivorship among reef zones throughout the 85-day study were evaluated using Kaplan-Meier

Survival Analysis in the R package ‘survival’ (Therneau, 2015). Frequency of A. cervicornis colony breakage and predation was analyzed among reef zones using a chi- square test.

102

Results

Wild Population Surveys

Table 5-1 contains summary data from wild colony surveys. Extant populations of

A. cervicornis were observed in each reef zone surveyed in Little Cayman, and the abundance of wild colonies was highest in the deep zone. Maximum diameter of wild colonies did not differ among reef zones, but differences in colony height were detected

(H2 = 6.57, p < 0.05). Dunn’s post-hoc indicated that mean height of wild A. cervicornis was greater in the intermediate zone compared to the shallow and deep zones.

Prevalence of predation by H. carunculata and C. abbreviata did not differ among reef zones. However, predation by Stegastes spp. differed among reef zones (X2 = 6.20, p <

0.05) and was most frequent in the intermediate zone. Disease prevalence in wild A. cervicornis did not differ among reef zones.

Outplanting Experiment

Mean photosynthetically active radiation was 2,403.00 ± 120.48 μmol m-2 s-1 at the shallow site (1 m), 1,267.10 ± 45.64 μmol m-2 s-1 at the nursery (6 m), 964.20 ±

27.42 μmol m-2 s-1 at the intermediate site (10 m), and 757.22 ± 18.61 at the deep site

(16 m). Table 5-2 contains summary data from outplanted colony surveys. Mean initial size of outplanted A. cervicornis colonies did not differ significantly among reef zones.

Net total linear extension (TLE) differed among reef zones after 85 days (F2,48 = 20.46, p < 0.001; Fig. 5-3) and was lowest at the shallow back reef site, where colonies experienced a mean decrease in TLE as a result of breakage. Increases in mean net

TLE were observed at both the intermediate and deep sites, but no statistical difference was found between these sites. Net number of branches also differed among outplanted

103

colonies in varying reef zones (F2,48 = 13.58, p < 0.001), and was lowest at the shallow site compared to the intermediate and deep sites, which did not differ significantly.

Frequency of colony breakage varied among sites (X2 = 13.68, p = 0.001), and breakage was highest among outplanted colonies on the shallow back reef. No statistically significant difference existed in prevalence of predation by C. abbreviata, H. carunculata, or Stegastes spp. among reef zones.

Survival patterns of outplanted A. cervicornis differed between reef zones during the 85-day study period (Kaplan-Meier X2 = 14.7, p = 0.001), and survival was lowest at the deep site. Following the conclusion of the study period, total survivorship within the shallow plots dropped to 10% during a period of unusually high temperature in which mean daily sea temperature exceeded 30.5°C for 20 consecutive days. However, a population of wild A. cervicornis less than 5 m from one of the shallow outplant plots survived this period of elevated temperature.

104

Table 5-1. Characteristics of extant wild A. cervicornis in Little Cayman. Asterisks (*) in headings indicate that significant differences within that factor were present among reef zones. Letters on values denote differences among reef zones per Dunn’s post hoc test.

Reef Zone Number Mean Mean Max Mean Height H. C. Stegastes Disease of Depth Diameter (cm) ± SE carunculata abbreviata Predation prevalence colonies (m) ± SE (cm) ± SE (*) Predation Predation Prevalence located Prevalence Prevalence (*)

Shallow Back 16 0.74 ± 61.25 ± 6.73 31.13 ± 2.53a 25.00% 50.00% 6.25% 0% Reef 0.10 Intermediate 16 10.93 ± 89.38 ± 14.68 63.81 ± 25.00% 43.75% 43.75% 12.5% Spur-and- 0.26 13.95b Groove Reef Deep Reef 44 16.00 ± 74.73 ± 8.11 44.66 ± 5.67a 23.26% 27.91% 23.26% 11.36% Terrace 0.42

105

Table 5-2. Results for outplanted A. cervicornis survival, growth, and condition across three reef zones in Little Cayman. Asterisks (*) in headings indicate that significant differences within that factor were present among reef zones. Letters on values denote differences among reef zones per Tukey’s HSD.

Reef Zone Mean Net TLE Mean Net Breakage Survival at H. carunculata C. abbreviata Stegastes (cm) ± SE Number of Frequency t = 85 d Predation Predation Predation Branches ± SE (*) (*) Prevalence Prevalence Prevalence

Shallow Back -4.04 ± 1.98a 0.00 ± 0.53a 70% 100% 0% 0% 0% Reef Intermediate 11.90 ± 1.86b 4.16 ± 0.62b 30% 95% 15% 0% 0% Spur-and- Groove Reef Deep Reef 7.23 ± 1.66b 2.33 ± 0.48b 15% 60% 0% 5% 0% Terrace

106

Figure 5-1. Zonation by depths in Little Cayman, modified from Logan (1994)1. Three major coral-dominated reef zones are present, including (A) the shallow back reef (i.e. shallow), (B) the intermediate spur-and-groove reef (i.e. intermediate), and (C) the deep reef terrace (i.e. deep). Wild Acropora cervicornis surveys and the A. cervicornis outplanting experiment occurred in each of zones A, B, and C.

.1 This figure is reproduced with permission from Logan, A., 1994. Reefs and lagoons of Cayman Brac and Little Cayman. In: Brunt, M., Davies, J. (Eds.) The Cayman Islands: Natural History and Biogeography. Springer Netherlands, Dordrecht, pp. 105–124.

107

Figure 5-2. Outplant experiment location in context of the Caribbean region, with points indicating the location of the nursery and each study site. All outplant sites were within 200 m of the nursery location along an approximately longitudinal transect perpendicular to shore. Aberrations from this transect were due to variation in availability of suitable hard substrate for outplanting.

108

Figure 5-3. Total linear extension (TLE) of outplanted Acropora cervicornis in each reef zone over time. Error bars represent SE. No significant difference in initial size of outplanted A. cervicornis was detected among zones at the beginning of the experiment. Colonies at the shallow site experienced breakage, while TLE increased at both the deep and intermediate sites over time. Net TLE was significantly lower at the shallow site compared to both the intermediate and deep sites on day 85 (F2,48 = 20.46, p < 0.001).

109

Discussion

Regional assessments of population trends for A. cervicornis suggest that populations remain depleted following a sharp decline in abundance in the early 1980s

(Aronson et al., 2008a; Schutte et al., 2010). Our observations of small remnant populations of wild A. cervicornis in each coral-dominated reef zone at Little Cayman are consistent with this reported trend as well as with reports of sustained A. cervicornis cover at other locations in the Caribbean (Crabbe, 2014; Lidz and Zawada, 2013; Lucas and Weil, 2015). The persistence of remnant wild A. cervicornis populations in three major reef zones in the Cayman Islands warrants investigation of the potential for outplanting in each zone. Although the current study provides a snapshot of the extant wild A. cervicornis population in Little Cayman, continued monitoring of wild A. cervicornis is recommended to fully understand long-term, local demography of this species.

Among wild A. cervicornis, colony height and the prevalence of predation by

Stegastes spp. was found to differ among reef zones, and each of these was highest in the intermediate zone. Increases in Stegastes predation with A. cervicornis colony size have been documented in Florida (Schopmeyer and Lirman, 2015), suggesting that increased height among colonies in the intermediate zone could explain observed higher rates of Stegastes predation in that zone. The relationship between colony size and Stegastes predation also indicates that this type of predation could increase among outplanted colonies as they grow larger (Schopmeyer and Lirman, 2015).

Prevalence of both predation and disease among the outplanted A. cervicornis population was low and did not differ among reef zones, suggesting that predation is not an important factor in initial outplant success at Little Cayman. This finding is also

110

supported by Griffin et al. (2015), who observed minimal predation within three months following outplanting of A. cervicornis in the U.S. Virgin Islands. In the same study, mean A. cervicornis outplant growth after three months was 9.0 ± 0.76 cm (Griffin et al.,

2015), which is consistent with values observed at the intermediate and deep sites in the present study. Griffin et al. (2015) also reported 98% survivorship of outplants after three months, which is similar to our observations of high survivorship in both the intermediate and shallow sites during the study period.

In contrast, long-term studies report somewhat lower survivorship among outplanted A. cervicornis after one year (e.g. 61.2 – 65.9% in Puerto Rico, Mercado-

Molina et al., 2015; 89.5 ± 2.4% in Florida, Schopmeyer and Lirman, 2015), suggesting that time post-outplant could be a factor in survivorship. Our observations of high mortality in the shallow zone following the conclusion of the three-month study period suggest that long-term monitoring (i.e. ≥ 1 year) would be valuable to assess restoration outcomes. Ultimately, high mortality at the shallow plots immediately after the conclusion of the study as well as at the deep plots within the first month of study suggest that these zones may be unsuitable for outplanting using current restoration strategies. However, high rates of growth and branching at the intermediate site suggest that current A. cervicornis restoration methods in the Cayman Islands may be best suited to this zone. Importantly, high growth rates, branching, and survivorship among outplants in the intermediate zone could increase availability and complexity of habitat for other reef fauna relative to other zones. Increased cover of complex branching corals has been linked to increases in abundance and species richness of reef fishes

(Holbrook et al., 2008, 2002; Huntington et al., 2017).

111

Our data ultimately suggest that a single nursery may be ineffective for rearing outplants which can be used to restore the shallow and deep zones at Little Cayman.

Growth and branching patterns among outplanted A. cervicornis in the shallow zone were predominantly influenced by colony breakage. Avoidance of nursery and outplant sites characterized by high wave energy has been proposed as a strategy to avoid breakage (Johnson et al., 2011; Young et al., 2012), however A. cervicornis is known to be capable of persisting at sites with high wave energy (Bottjer, 1980), including at Little

Cayman as documented by this study. In addition, although no visible bleaching was observed during the study period, temperature stress was likely a factor in the shallow zone mortality event immediately following the conclusion of the study. Daily mean temperature exceeded 30.5°C, one degree greater than the maximum monthly mean temperature reported for Little Cayman (i.e. 29.5°C, van Hooidonk et al., 2012), for a period of nearly three weeks. Corals have been shown to experience considerable thermal stress when this threshold is exceeded (Glynn and D’Croz, 1990; Logan et al.,

2012). Additionally, temperatures tend to have higher ranges in shallow reef habitats compared to the outer reef (Craig et al., 2001), and branching corals are known to have relatively high sensitivity to warming events (Baird and Marshall, 2002; Brown and

Suharsono, 1990; Gates and Edmunds, 1999; Loya et al., 2001). It is therefore logical to conclude that A. cervicornis colonies in the shallow plots would have been negatively affected by extreme high temperatures after a relatively short acclimation period of ~3 months. Colony breakage may have compounded the negative effects of the warming event by allocating colony resources to repair damage, which can in some cases increase the likelihood of mortality (Hall, 2001). Unlike at the shallow site, all mortality at

112

the deep site occurred within the first month following outplanting. Reduced PAR is known to decrease calcification and linear extension (Chalker and Taylor, 1975; Enochs et al., 2014; Kendall et al., 1985) and can cause mortality of A. cervicornis (Rogers,

1979). This environmental variable could therefore be a factor in the poorer performance of colonies at the deep site. Ultimately, we lack the data necessary to fully understand the cause of mortality among deep outplants.

This work represents a case study of initial A. cervicornis outplant performance across three diverse reef zones where wild colonies persist, however we acknowledge limitations to our study as well as remaining questions that could be addressed in future investigations. First, our study was constrained by low sample size due to colony availability within the nursery, and therefore our results must be interpreted carefully.

We recommend repeating this study with a larger number of corals and over a longer time scale to better understand reef zone-specific differences in outplant performance.

Despite limitations, these data have value in guiding future outplant site selection, and will likely be particularly useful for practitioners in the Cayman Islands and other

Caribbean sites with similar reef zonation. Additionally, new strategies should be tested to attempt to improve outplant performance in the shallow and deep zones. In particular, the effect of nursery donor colony site of origin could be explored, as site of origin has been shown to influence colony performance in reciprocal transplantation experiments

(Raymundo, 2001). The effect of nursery location should also be considered and could be tested by establishing nurseries within each target outplanting zone and conducting reciprocal outplanting. Colony performance was overall best in the zone with characteristics most similar to the nursery (i.e. the intermediate zone; Fig. 5-1), a result

113

that tentatively supports the recommendations of Johnson et al. (2011), who suggest outplanting at depths similar to those of the nursery site. Finally, strategies to reduce breakage could be developed and tested to improve outplant performance at sites with high wave energy. For example, anecdotal evidence suggests that adding three points of contact between the colony and the substrate could increase stability of outplanted A. cervicornis (K. Ripple, Coral Restoration Foundation, personal communication) compared to the single point of contact method used in this study. The relative difference in colony breakage between outplant methods has yet to be directly evaluated.

This study reports characteristics of the extant wild A. cervicornis population at

Little Cayman, a site with minimal direct anthropogenic impact and healthy reefs relative to other Caribbean locations (Manfrino et al., 2013). In addition, a concurrent assessment of the effect of reef zonation by depth on growth and survivorship was performed among nursery-reared staghorn coral. These results suggest that outplanting efforts currently have the most initial success in the intermediate spur-and-groove reef zone. This information can be used to inform site selection for future outplanting activities in the Cayman Islands and potentially at sites with similar nursery practices and reef zonation. Additional studies should be conducted to better understand the effect of site selection on outplant performance. New techniques must also be developed to improve the outcomes of A. cervicornis restoration activities in the shallow back reef and deep reef terrace in order to extend the ecological services provided by coral restoration to these zones.

114

CHAPTER 6 RESOLVING THE DYNAMICS OF CORAL PHOTOACCLIMATION THROUGH COUPLED PHOTOPHYSIOLOGICAL AND METABOLOMIC PROFILING

Introduction

Light availability fundamentally regulates the ecological success of reef building corals over space and time (Muir et al., 2015). However, corals within shallow reef habitats are exposed to light intensities that continually change over both transient

(seconds to hours) and longer-term (days to weeks) time-scales (Anthony et al., 2004).

Consequently, corals have evolved many mechanisms to photo-protect or photo- enhance in order to optimize physiological performance of their algal endosymbionts

(see Roth, 2014). Such mechanisms include physiological, morphological, and behavioral adaptations of the coral host (e.g. Muscatine et al., 1984; Porter et al., 1984;

Gates & Edmunds, 1999; Enríquez et al., 2005; Lesser et al., 2010) needed to fine-tune light exposure to the algal endosymbionts, as well as continual photophysiological adjustments of the algal endosymbionts themselves (e.g. Iglesias-Prieto et al., 2004;

Frade et al., 2008).

Given the importance of light for sustaining coral productivity, it is unsurprising that many studies have investigated the ability of corals to acclimate to changes in light intensity, i.e. photoacclimation (e.g. Titlyanov et al., 2001; Anthony & Hoegh-Guldberg,

2003a; Hennige et al., 2008). Most of these studies have to date predominantly focused on “steady state” properties of photoacclimation (reviewed in Warner & Suggett, 2016), via photophysiological measurements from corals acclimated to long term exposure to different light intensities that occur naturally (Anthony & Hoegh-Guldberg, 2003b; Frade et al., 2008; Winters et al., 2009; Hennige et al., 2010) or are imposed experimentally

115

(Hennige et al., 2008; Schutter et al., 2011; Jeans et al., 2013; Langlois &

Hoogenboom, 2014; Cohen and Dubinsky, 2015). However, repeated photophysiological measurements during photoacclimation are needed to provide insight on the fine-scale physiological changes that occur following dynamic alterations in light availability. Comparatively few studies have actually analyzed fine-scale time- dependent changes to coral photophysiology that capture the dynamics inherent to photoacclimation rates and extents. Earlier studies (Anthony and Hoegh-Guldberg,

2003a) used repeated respirometry measurements to model changes in photophysiological parameters of Turbinaria mesenterina during acclimation to both increases and decreases in light. Actual photophysiological changes during short-term exposure of T. mesenterina to elevated light (days; Hoogenboom et al., 2006) as well as following longer term reciprocal transplants of Stylophora pistillata across two depths

(weeks to months; Cohen and Dubinsky, 2015) were subsequently characterized using repeated pulse-amplitude modulated (PAM) fluorometry. Relatively short term (days) photoacclimation of four coral species has similarly been assessed using PAM fluorometry in combination with other metrics, such as endosymbiont cell concentration and pigment content (Langlois and Hoogenboom, 2014). Such fluorometry approaches have proven extremely critical for retrieving highly resolved photophysiological parameterization of corals over space and time (e.g. Hennige et al., 2008; Suggett et al., 2012; Langlois and Hoogenboom, 2014; Warner et al., 2010), but tying these parameters to the underlying metabolic changes that regulate photo-acclimation remains challenging and largely unresolved (see Nitschke et al., 2018; Warner and

Suggett, 2016).

116

A metabolomic profile provides a snapshot of the metabolic physiological state of an organism at a given time, and can therefore provide insight into changes that underpin light acclimation (Obata and Fernie, 2012; Weckwerth, 2003). Numerous metabolic processes are likely involved in photoacclimation and a vast array of small chemical compounds (metabolites) are critical to the operation of (and signaling amongst) these processes. Changes in the presence or concentration of metabolites have been shown to provide new insight into the physiological processes activated in response to external stimuli, e.g. temperature, pH (Sogin et al., 2016; Hillyer et al.,

2017, 2018) and the presence of competitors (Quinn et al., 2016) of reef-building corals.

Some specific metabolites or targeted metabolite groups have also been assessed in corals during photoacclimation periods (e.g. mycosporine-like amino acids; Torres et al.,

2007). Increased light dramatically alters the metabolome in algae (Davis et al., 2013), cyanobacteria (Meissner et al., 2015), and terrestrial plants (Wulff-Zottele et al., 2010;

Obata and Fernie, 2012), and the response of the metabolome to both increased and decreased light has also been explored in the model plant species Arabidopsis thaliana

(Caldana et al., 2011). However, while shifts in metabolite profiles of cultured

Symbiodinium are known to follow changes in temperature and light (Klueter et al.,

2015), how the metabolome of the coral holobiont (coral host, endosymbiotic

Symbiodinium, and associated microorganisms) changes during photoacclimation remains unknown.

Characterizing the effects of changes in light availability on coral metabolism is critical to understand how corals respond to natural changes in light, but also how corals fundamentally cope with light-altering stressors attributed to human activity, such as

117

enhanced sedimentation (Bessell-Browne et al., 2017) and algal shading (Cetz-Navarro et al., 2015). Rapidly altered light conditions are also induced via increasingly popular interventional reef management strategies involving coral transplantation (e.g. Bruckner and Bruckner, 2001; Ross, 2014; Lohr et al., 2017; see also Cohen and Dubinsky,

2015). Many species of coral are clearly capable of acclimating to a wide range of light regimes (e.g. Anthony et al., 2003b, 2004; Langlois and Hoogenboom, 2014); thus a better understanding of the rate and extent of coral photoacclimation can aid in determining corals’ ability to withstand sedimentation and shading and also to tolerate transplantation to new light regimes during management interventions. We therefore conducted a fully reciprocal light exposure experiment to determine the nature and extent with which corals respond to low versus high light shifts. We assayed the time scale for photophysiological adjustment of the abundant reef flat coral Acropora muricata (Heron Island, Great Barrier Reef) using PAM fluorometry-derived acclimation metrics, and simultaneously used a metabolomics approach to evaluate whether and how metabolic changes were concurrent with photophysiological adjustment.

Methods

Experimental Design

A total of 58 Acropora muricata fragments at least 5 cm in length were collected from visually healthy source colonies on the reef crest and fore reef at Heron Island

(23.44°S, 151.91°E) in January 2017. The fore reef collection site was located approximately 10 m seaward of the reef crest collection site. At both collection sites, source colonies were part of continuous thickets and discrete genets could not be identified. Instead, fragments were haphazardly collected from thickets within an area of approximately 3 x 3 m at each collection site. Maximum photosynthetically active

118

radiation (PAR) was measured at each site at solar noon on a clear day using a PAR meter (4π spherical underwater quantum sensor, LI–193SA). Fragments collected from the reef crest (n = 29) were clipped from the tops of high light (HL) exposed thickets at

0.5 – 1 m (maximum PAR ≈ 2500-3000 μmol m-2 s-1). The remaining fragments (n = 29) were collected from the underside of thickets that were exposed to comparatively low light (LL) at a depth of 3.5 – 4 m (maximum PAR ≈ 500-1000 μmol m-2 s-1). Collected HL and LL fragments were transported to the Heron Island Research Station and placed in seawater-conditioned plastic test tube racks to ensure they remained upright.

Fragments were acclimated for three days in an outdoor direct flow-through system at sunlight intensities closely approximating their source environment. Irradiance levels approximating each source environment were achieved using shade cloth to approximate mid-day maximum light levels to corresponding values measured in situ at

HL and LL collection sites (HL ~2500-2600 μmol m-2 s-1 and LL ~450-600 μmol m-2 s-1 maximum daily PAR).

To generate initial metabolomic profiles associated with each light regime prior to the application of experimental treatments and to account for any tank effect following the acclimation period, 10 fragments (5 each, HL and LL) were snap frozen in liquid nitrogen (sample groups hereafter referred to as “initial HL (or LL) samples”). The remaining 48 fragments were distributed among two light treatments in six 68-L glass aquaria (n = 3 aquaria per light treatment). Fragments were distributed in a fully reciprocal design such that equal numbers of fragments sourced from HL and LL environments were exposed to each light treatment (Fig. 6-1). This resulted in two treatment groups (LL to HL and HL to LL) and two control groups (LL control, HL

119

control), each consisting of 12 colonies. Fragments from each source location were haphazardly assigned to each treatment tank, and were equally spaced in one test tube rack per tank. Each aquarium received 10 L min-1 flow-through seawater pumped directly from the reef flat where fragments were originally sourced. Cross-calibrated temperature loggers (HOBO Pendant® UA-002-64 or UA-001-64, Onset Corporation) were placed in each aquarium to monitor temperature throughout the experiment and ensure consistency across treatments.

Photophysiology

Endosymbiont photosystem II (PSII) photophysiology was assessed over a period of 21 days using a Pulse Amplitude Modulation fluorometer (Diving-PAM, Walz) as per Nitschke et al. (2018). Immediately after dawn on day 0, 1, 3, 4, 7, 11, 15, 20, and 21 of the experiment, test tube racks containing corals were transported to an adjacent laboratory in a seawater-filled container, and rapid light curves (RLC) with 8 actinic light steps were performed on each replicate fragment. Corals were kept under low light to remove potential artifacts from dark-induced plastoquinone reduction. Actinic light levels were calibrated with a PAR meter (4π spherical underwater quantum sensor,

LI–193SA) prior to each set of light curves. Irradiance steps were administered in 20 s intervals (as per Hennige et al. 2008, Nitschke et al. 2018). Briefly, diving-PAM settings were: actinic light factor = 1, light curve intensity = 3, saturation width = 0.8s, saturation intensity = 10, gain = 12, and signal damping = 3.

RLCs for each replicate fragment were fit to a least squares non-linear regression model that describes the light-dependent quantum efficiency of PSII

(Hennige et al. 2008),

120

퐹푞′/퐹푚′ = ⁡ [(퐹푞′/퐹푚′ 퐸 )(1 − exp⁡(−퐸/퐸 ))]/퐸 (푚푎푥) 푘 푘 (6-1)

-2 -1 where Ek is sub-saturation irradiance (μmol photons m s ), Fq’/Fm’(max) (dimensionless) estimates the maximum photochemical efficiency of PSII, and E is PAR. RLC data were also fit to a second least squares non-linear regression model describing light- dependent electron transport:

푟퐸푇푅 = ⁡ 푟퐸푇푅MAX ⁡× [1 − exp⁡(−훼 × 퐸/푟퐸푇푅MAX)] (6-2) where α (dimensionless) is light-dependent photosynthetic rate and rETRMAX (μmol electrons m-2 s-1) is maximum relative electron transport rate.

Metabolome Profiling

In addition to the 10 initial HL and LL samples collected immediately following the acclimation period, control and treatment groups were sampled from the six tanks after seven and 21 days for metabolome profiling. A total of n = 24 samples were collected at each time point, with each source x treatment group equally represented among samples. Samples were snap frozen in liquid nitrogen to halt metabolic activity, and stored at -80 °C prior to extraction. For metabolomics analysis, experimental data for HL and LL exposed corals was considered separately in order to better interpret how treated colonies changed with respect to relevant initial and control groups. This resulted in comparisons between (1) LL to HL treated colonies, LL controls, and initial

HL and LL samples, and (2) HL to LL treated colonies, HL controls, and initial HL and

LL samples. Both GC-MS and LC-MS analyses were performed in order to retrieve a comprehensive set of both primary and secondary metabolite responses.

121

GC-MS

GC-MS profiling was used in order to target primary metabolites (such as sugars, amino acids, and organic acids; Dias et al,. 2015). Metabolite extraction methods for

GC-MS profiling were modified from Hillyer et al. (2016, 2017). Metabolites were extracted using 750 µL of extraction solution (100 % methanol with three internal standards: 100 µM DL-Valine-d8, 60 µM Stearic acid-d3, 60 µM 5-α-Cholestane) per 50 mg of sample material (including tissue and skeleton). An additional extraction was performed using 750 µL of 50 % methanol. Samples were incubated for 20 min at 5 ºC and 1250 rpm in a thermomixer in respective solvents sequentially, then centrifuged for

3 min at 20,800 g. The resulting supernatant was collected and pooled. 150 µL of supernatant was transferred into a glass vial and evaporated to dryness in a vacuum evaporator for 3 h at room temperature. Samples were derivatized by adding 20 µL of

20 mg/mL methoxylamine in pyridine then incubating samples at 37 ºC for 2 h and 750 rpm in the agitator. 20 µL N-Methyl-N-(trimethylsilyl) trifluoroacetamide was then added and samples were incubated at 37 ºC for 30 min and 750 rpm in the agitator. Samples were subsequently incubated at room temperature for 1 h, and finally 1 µL was injected onto the GC-MS.

Samples were run on a GCMS-QP2020 (Shimadzu Corporation, Kyoto, Japan) equipped with an AOC-20is autosampler (Shimadzu Corporation). The column used was an SH-Rxi-5Sil MS fused silica capillary column (30.0 m x 0.25 mm x 0.25 μm) operating in electron impact mode at 70 eV. Helium was used as the carrier gas at a constant flow of 1.0 mL min-1 and an injection volume of 1 µL, with an injector temperature of 280 °C and an ion source temperature of 230 °C. The temperature gradient of the oven was 70 ºC for 1 min, then 7 ºC per minute to 325 ºC. The scan

122

range was m/z 50-600. Samples were run in a randomized order to account for any potential analytical drifts affecting experimental groups. A pooled QC sample was also injected periodically.

LC-MS

Metabolite extraction methods for LC-MS were modified from Gordon et al.

(2013). Frozen coral fragments were placed in 20 mL scintillation vials containing 10 mL of 100 % methanol (LC-MS Grade, B&J Brand, Honeywell, Shanghai, China) spiked with 0.005 mM aminoanthracene (Sigma-Aldrich, Castle Hill, Australia) as an internal standard, then stored at -20 °C overnight. Vials were sonicated in a chilled water bath for 15 min then vortexed for 30 s per sample. Resulting extracts were decanted into clean scintillation vials, and 3 mL of 70 % methanol was added to vials containing remaining coral nubbins to ensure extraction of hydrophilic metabolites. Samples in 70

% methanol were vortexed for another 30 s and the resulting extract was combined with the 100 % methanol extract. The combined extract was stored at -20 °C overnight. A syringe was then used to remove 1 mL of extract, which was passed through 0.22 μm

Hydraflavon Syringe filter (MicroAnalytix Pty Ltd, Taren Point, Australia) to remove any particulate matter and into clean 2 mL HPLC vials. Vials were stored at -20 °C overnight prior to processing.

Untargeted LC-MS metabolite profiling was conducted using a reverse phase

(C18) technique targeting semi-polar to hydrophobic secondary metabolites (De Vos et al., 2007). Samples were analyzed on 6550 iFunnel Q-TOF LC-MS (Agilent

Technologies, Santa Clara, CA, USA) equipped with Dual Automatic Jet Stream (AJS)

Electrospray Ionisation (ESI), coupled with an 1260 infinity HPLC system (Agilent

Technologies, Santa Clara, CA, USA). Separation was performed at 25 °C on an Agilent

123

Zorbax Eclipse XDB-C18 column (100 × 4.6 mm i.d., 1.8 μm). The HPLC program consisted of a linear gradient of milli-Q water (with 1 % formic acid) to 100 % acetonitrile

(with 1% formic acid) over 12 min, followed by isocratic elution at 100 % acetonitrile

(with 1% formic acid) at a flow rate of 1 mL min−1. Nitrogen was used as the nebulizing gas. Dual AJS ESI source was kept at a voltage of 3500 V in positive ion mode. Mass spectra were acquired with source conditions as follows: gas temperature 350 °C,

-1 drying gas 4 L min (N2), nebulizer pressure 35 psi (N2) and Vcap 3,500 V, fragmentor

160 V and skimmer 65 V. The mass range scanned was 70–1,700 m/z, at a scan rate of

2 spectra/s. Because analysis was untargeted, generic settings were applied to obtain as many compounds as possible. All samples were injected in a single batch and a randomized injection order was used to avoid sample mass. A quality control (QC) sample composed of replicates from each sample group was analyzed at the beginning, middle and end of the batch. External mass calibration was performed using a calibrating solution monitoring signals at m/z in positive polarity. Data were processed using the Mass Hunter Qualitative analysis software (version B.06.00 Agilent

Technologies). All solvents used were of High Purity Grade from Honeywell Burdick and

Jackson (Chem-Supply, Australia). LC-MS grade formic acid was obtained from Sigma

Aldrich (Castle Hill, Australia).

Data Analysis

Photochemical parameters retrieved from modelling PAM fluorescence light-

MAX response curves (Ek, Fq’/Fm’(max), and rETR ) were compared among treatments over time using linear mixed models with repeated measures. An ar(1) model was used to assess the correlation caused by repeated measures. Variance was allowed to vary by

124

treatment. All analyses of photochemical parameters were performed using SAS statistical software (version 9.4, SAS Institute).

Raw GC–MS data were transformed into CDF format using GCMSsolution software (v. 4.0, Shimadzu, Kyoto, Japan), and the converted files were subsequently imported into XCMS (v. 3.2, Galaxy Project Metabolomics, Roscoff, France). XCMS software, which is freely available under an open-source license at http://metlin.scripps.edu, incorporates nonlinear retention time alignment, matched filtration, peak detection and peak matching. For grouping, bandwidth was set to 10, the resulting peak list comprising features (ions, retention time, intensity) was further processed using Excel (Microsoft, Redmond, WA), where the area normalization was performed using the total peak area of internal standards and used for further statistical analysis. Statistical analysis of the normalized m/zRT features obtained from XCMS were performed using MetaboAnalyst 4.0 software (www.metaboanalyst.ca).

Normalized peak areas were log transformed and autoscaled (the mean area value of each feature throughout all samples was subtracted from each individual feature area and the result divided by the standard deviation) prior to statistical analysis. Features showing statistically significant differences among the groups were used to annotate peaks. Groups of features sharing the same retention time that were statistically significant and presented with a high degree of correlation were considered representative of a single metabolite (Escobar-Morreale et al., 2012). Metabolite profiling and tentative metabolite identification was performed using GCMSsolution software by combining mass spectra and database consultation (NIST17, match with library > 70 %). Further validation was also done through literature searches.

125

Raw LC–MS data were processed using the Batch Recursive Feature Extraction option in the Agilent Mass Hunter Profinder software (version B.06.00). Untargeted analysis of complex biological mixtures comprises tens of thousands of mass features, and thus allows profiling of a large number of molecules. Several preprocessing filters were applied to curate the large amount of data to an operational size. Using the feature extraction algorithm, a group of ions characterized by retention time, peak area and accurate mass was extracted in each sample as molecular features. This was performed by using a minimum absolute abundance threshold of 1000 counts with an m/z range of 100-1700. The charge state was set to 2 and minimum number of ions in the isotopic distribution was set to 2, following the isotope model of “common organic molecules”. This was followed by binning and alignment of molecular features as a function of retention time, fragmentation pattern and m/z value across the data matrix, using a tolerance window of ± 0.1 % + 0.2 min retention time and ± 10 ppm + 2 mDa

+ + + + mass window. Allowed ion species were H , Na , K , NH4 and neutral losses of H2O, and CO2. A further visual validation of the feature extraction results and manual editing of EIC peaks as required was performed using Profinder to reduce the number of false negatives and false positives in the dataset, thereby increasing the quality of the data exported for differential analysis. Data files were transformed into .CEF files containing extracted compounds, neutral mass, retention time, and abundance, and exported to

Mass Profiler Professional (MPP) software package (version B.14.9.1, Agilent

Technologies, Santa Clara, CA, USA).

Data were normalized with the internal standard and evaluated and filtered to remove low quality and inconsistent mass spectral features (only those appearing in 75

126

% of samples in at least one condition were considered). Thereafter, compound abundance values in each sample were baselined to the median of each compound in all samples. Resultant mass features were exported as .csv files and statistical comparisons including principal component analysis (PCA), partial least square discriminant analysis (PLS-DA) and hierarchical cluster analysis was performed using

MetaboAnalyst 4.0 software (www.metaboanalyst.ca). Statistically significant differences among sample groups were determined using one-way ANOVA and Tukey’s

HSD (α = 0.05) using the exported dataset.

Results

Photophysiology

Values of maximum PSII photochemical efficiency (Fq’/Fm’(max)) at time zero did not vary between LL and HL fragments. Furthermore, Fq’/Fm’(max) remained constant throughout experimentation, with no interaction between source location and time, and no clear pattern for differences among treatment groups over time (Fig. 6-2a,b). In

MAX contrast, changes in Ek and rETR demonstrated reciprocal photoacclimation patterns for A. muricata (Fig. 6-2c,d and 6-3). Specifically, mean Ek for HL-sourced corals

(252.25 ± 16.18 SE μmol photons m-2 s-1; Fig. 6-2d) was significantly greater than that of LL-sourced corals (121.44 ± 11.19 SE μmol photons m-2 s-1; Fig. 6-2c) at t = 0 (p <

0.0001). After three days of exposure to light treatments, there was no longer a significant difference in Ek based on source location, regardless of light treatment (p >

0.05). However, by the conclusion of the experiment (t = 21 days), light treatment had a significant effect on Ek, regardless of source location (p = 0.01; Ek = 230.08 ± 17.83 SE

μmol photons m-2 s-1 for LL to HL treated and HL control corals and 143.50 ± 8.99 SE

μmol photons m-2 s-1 for HL to LL treated and LL control corals), thereby demonstrating

127

clear emergent photoacclimation responses of treated coral fragments to their new light environments.

MAX Values for rETR followed a similar pattern to those for Ek (Fig. 6-3). At t = 0, mean rETRMAX for HL-sourced corals (94.43 ± 4.67 μmol electron m-2 s-1) was significantly greater than that of LL-sourced corals (47.70 ± 3.61 μmol electron m-2 s-1; p

< 0.0001). After one day of exposure to light treatments, rETRMAX no longer differed based on source location (p > 0.05), and differed significantly based on light treatment

(p = 0.01). Light treatment also had a significant effect on rETRMAX, regardless of source location at t = 7 (p = 0.01) and at the conclusion of the experiment (t = 21, p < 0.01). At t

= 21, mean rETRMAX was 90.88 ± 5.46 SE μmol electron m-2 s-1 for the LL to HL treated plus HL control colonies and 58.45 ± 3.07 SE μmol electron m-2 s-1 for the HL to LL treated plus LL control colonies.

Metabolomic Responses to Light Shifts

A total of 182 metabolites were resolved by GC-MS, 59 of which were annotated by comparison to the NIST2017 library. PCA and PLS-DA of all mass features suggest metabolic adjustment during the 21-day experimental period (Figs. 6-4, 6-5). Initial HL and LL coral samples clustered separately in both PCA and PLS-DA models.

Metabolomic profiles of LL to HL treated corals clustered separately compared to LL controls at both t = 7 and t = 21 (Figs. 6-4a, 6-5a). PCA and PLS-DA indicate that the majority of metabolic adjustment for LL to HL treated corals occurred during the first 7 days of treatment, i.e. a larger shift occurred between t = 0 and t = 7 than between t = 7 and t = 21. Separation of metabolomic profiles was less distinct for HL to LL treated colonies and HL controls (Figs. 6-4a, 6-5b). PCA demonstrates that HL to LL treated colonies cluster separately from HL controls at t = 7, however overlap between HL to LL

128

treated colonies and both initial HL and LL profiles is also apparent (Fig. 6-4b).

Additionally, a great deal of overlap with control and initial groups is apparent for HL to

LL treated colonies at t = 21. PLS-DA for the same treatment and control groups illustrates that HL to LL treated corals at t = 7 and t = 21 generally clustered closer to initial LL samples, however overlap with HL control colonies was apparent at both t = 7 and t = 21 (Fig. 6-5b). Consequently, metabolomic profiles for HL to LL treated corals do not appear to be as distinct from HL controls compared to those of LL to HL treated corals and LL controls.

A total of 33 metabolites differed significantly between LL controls and LL to HL treated corals. Of these metabolites, 21 were tentatively annotated using the NIST 2017 library, and included an array of amino acids, organic acids, fatty acids, and sterols.

Figure 6-6 illustrates differences in relative concentrations of significant metabolites among treatment and control groups. Heat maps indicated two major clusters when LL to HL treatments were compared against LL controls (Fig. 6-6a). LL to HL treated colonies at both t = 7 and t = 21 clustered with initial HL samples, while LL controls clustered with initial LL samples. Discrimination between these clusters was very effective, again suggesting that most metabolic adjustment occurred within 7 days of treatment application. LL to HL treated colonies and initial HL samples were characterized by relatively lower concentrations of the steroid campesterol and the alcohols 1-hexanol and glycerol compared to LL controls and initial LL samples. LL to

HL treated colonies and initial HL samples also had relatively higher concentrations of the fatty acids tetradecanoic acid (C14:0) and dodecanoic (C12:0) acid, the amino acid

L-cysteine, and the alcohol 1-hexadecanol compared to LL controls and initial LL

129

samples. Although a number of significant metabolites were also identified for HL to LL treated colonies compared to HL controls and HL and LL initial samples, there was no clear clustering pattern among treatment, control, and initial sample groups (Fig. 6-6b).

These results suggest that light exposure over the 21-day experimental period did not have a clear effect on metabolite concentrations among HL to LL treatment, HL control, and HL and LL initial sample groups.

Based on mass accuracy and retention time, 1175 molecular features were aligned across the LC-MS retrieved sample set, and this number was reduced to 1090 molecular features after filtering by frequency (see methods). A total of 28 molecular features were found to be statistically different between LL to HL treatments and LL controls. As with the GC-MS, PCA and PLS-DA of LC-MS retrieved molecular features generally demonstrated metabolic adjustment during the 21-day experimental period

(Figs. 6-7, 6-8). Initial HL and LL samples clustered separately in both PCA and PLS-

DA models. PCA indicates overlap in the metabolomes of LL to HL treated corals and

LL controls at t = 7, however treatment and control profiles are distinct by t = 21 (Fig. 6-

7a). PLS-DA shows distinct clustering of LL to HL treated corals and LL controls at all time points (Fig. 6-8a). In contrast to PCA, PLS-DA indicated that the majority of (LC-

MS) metabolic adjustment for LL to HL corals occurred during the first 7 days of treatment, with a larger shift between t = 0 and t = 7 than between t = 7 and t = 20 (Fig.

6-8a). Overlap between HL to LL treated colonies and HL controls was apparent at both time points in the PCA model (Fig. 6-7b). Separation of metabolomic profiles was also less distinct for HL to LL colonies and HL controls in the PLS-DA model (Fig. 6-8b).

While HL to LL treated corals at t = 7 and t = 21 generally clustered closer to initial LL

130

samples, overlap with HL controls was also apparent at each time point, indicating that metabolomic shifts resulting from the HL to LL treatment were less substantial than those associated with the LL to HL treatment. Heat maps resolved two major clusters when significant molecular features were compared for LL to HL treatments versus LL controls (Fig. 6-9). LL to HL treatments at both t = 7 and t = 21 clustered with initial HL samples, while LL controls and initial LL samples formed a separate cluster.

Discrimination between these clusters was very effective, again suggesting that most metabolomic changes occurred within 7 days of treatment. No molecular features were found to vary significantly between HL to LL treatment groups and HL controls.

131

Figure 6-1. A schematic illustrating experimental design for this study. Fragments sourced from high light (H) and low light (L) sites were equally distributed among high and low light treatments (open and shaded boxes, respectively). This schematic illustrates relative distribution of high and low light-sourced corals among light treatment tanks, but does not represent actual positions of treatment tanks and replicate colonies within tanks, which were haphazard. Photophysiological and metabolomic changes were assessed over a period of 21 days.

132

Figure 6-2. Maximum photochemical efficiency of PSII (Fq’/Fm’(max)) (A, B) and sub-saturation irradiance (Ek) (C, D) over time. Left panels (A, C) show data for corals sourced from low light (LL) and right panels (B, D) show data for corals sourced from high light (HL). Treatments had no clear effect on Fq’/Fm’(max). At t = 0, Ek varied significantly based on source location only (HL vs. LL). However, by t = 7 and at t = 21, treatment had a significant effect on Ek, illustrating photoacclimation over time.

133

Figure 6-3. Maximum relative electron transport rate (rETRMAX) over time for (A) corals sourced from low light (LL) and (B) corals sourced from high light (HL). At t = 0, rETRMAX varied significantly based on source location only (HL vs. LL). However, by t = 7 and at t = 20 and t = 21, treatment had a significant effect on rETRMAX, illustrating photoacclimation over time.

134

Figure 6-4. Principal component analysis (PCA) results illustrating shifts in the metabolome over time based on GC-MS data for corals sourced from (A) LL and (B) HL. Shading indicates 95% confidence intervals.

135

Figure 6-4. Continued.

136

Figure 6-5. Partial least squares discriminant analysis (PLS-DA) models based on GC- MS data illustrating shifts in the metabolome over time for corals sourced from (A) LL and (B) HL. Shading indicates 95% confidence intervals. Separation based on treatment light condition is apparent for LL to HL treated corals, with the metabolome of treated corals shifting closer to initial HL samples, while LL controls and initial LL samples cluster separately. While some metabolomic shifts occurred for HL to LL treated corals, separation between treatment groups and HL controls is less clear.

137

Figure 6-5. Continued.

138

Figure 6-6. Heat map comparing significant entities identified by GC-MS and ANOVA for corals sourced from (A) LL and (B) HL. Individual entities are presented using tentative annotations or mass/retention time for compounds that were not annotated. Based on relative concentrations of significant entities, LL to HL treated corals at both t = 7 and t = 21 cluster with initial HL samples, while LL controls cluster separately with initial LL samples. In contrast, no clear clustering based on significant entities was observed among HL to LL treated corals and HL controls.

139

Figure 6-6. Continued.

140

Figure 6-7. Principal component analysis (PCA) results illustrating shifts in the metabolome over time based on LC-MS data for corals sourced from (A) LL and (B) HL. Shading indicates 95% confidence intervals.

141

Figure 6-7. Continued.

142

Figure 6-8. Partial least squares discriminant analysis (PLS-DA) models based on LC- MS data illustrating shifts in the metabolome over time for corals sourced from (A) LL and (B) HL. Shading indicates 95% confidence intervals. Separation based on treatment light condition is apparent for LL-sourced corals exposed to HL, with the metabolome of treated corals shifting closer to that from corals from the treatment condition at t = 0. While metabolomic shifts also occurred for HL-sourced corals exposed to LL, separation among HL to LL treatment and controls is less clear.

143

Figure 6-8. Continued.

144

Figure 6-9. Heat map comparing significant entities identified by LC-MS and ANOVA for corals sourced from LL. Individual entities are presented as mass/retention time. The metabolome for LL to HL treated corals at both t = 7 and t = 21 clusters with initial HL samples, while LL controls and initial LL samples cluster separately. These results suggest the metabolome of treated corals reflects treatment, rather than source, light conditions.

145

Discussion

Coral photoacclimation properties and how they vary over space and time have become increasingly well studied (e.g. Anthony & Hoegh-Guldberg, 2003a; Frade et al.,

2008), particularly through the introduction of highly resolute bio-optical techniques such as PAM fluorometry (Hennige et al., 2008; Nitschke et al., 2018). However, associated metabolic changes that regulate cellular physiological processes inherent to photoacclimation in the coral holobiont have not been comprehensively explored. Our present study therefore applied coupled photophysiology and metabolomic measurements for the first time to generate new insight into the biology of coral photoacclimation. Consistent with previous studies, corals were able to acclimate to both increases and decreases in light availability (Falkowski & Dubinsky, 1981; Cohen and Dubinsky, 2015) during the time frame of our experiment. Changes in photophysiological parameters, notably the light intensity for saturated photosynthesis

MAX (Ek) and maximum electron transport rate (rETR ), but not maximum photochemical efficiency of PSII (Fq’/Fm’(max)), suggest photoacclimation in these shallow reef flat

Acropora muricata occurred primarily through changes in capacity for maximum photosynthesis rather than light harvesting (Hennige et al., 2008; Suggett et al., 2012).

Furthermore, metabolomic shifts were observed within 7 days of treatment, particularly following high light exposure, suggesting a link between metabolite profile and photoacclimation response.

Lack of change observed for Fq’/Fm’(max) contrasts with a number of previous studies, which report concurrent shifts in Fq’/Fm’(max) with shifts in light (Hennige et al.,

2008; Warner et al., 2010; Suggett et al., 2012; Nitschke et al., 2018). This result is

MAX particularly intriguing, as other metrics (i.e. Ek, rETR ) clearly indicate

146

photoacclimation occurred during the study period. Although these results appear conflicting, one possible explanation is that Fq’/Fm’(max) is modified by a complex interaction of active PSII reaction centers versus absorption characteristics. For example, opposing responses of enhanced photoprotection through more heat dissipation of absorbed light (reduced photochemical efficiency) coupled with parallel increases in the proportion of active PSII centers (enhanced photochemical efficiency) to drive photochemistry more efficiently per unit photon absorbed (Suggett et al., 2009) would yield a net outcome of relatively constant Fq’/Fm’(max). This notion is supported by previous findings of intracolonial variation in both light absorption (Kaniewska et al.,

2008; Wangpraseurt et al., 2012) and photosynthetic activity (Ralph et al., 2002), but whether a similar mechanism drives relatively constant Fq’/Fm’(max) in the present study cannot be currently resolved and clearly warrants more targeted investigation.

Other photophysiological adjustments following changes in light availability are more consistent with a number of findings from previous studies. Increases in Ek and rETRMAX with increasing light were also found for Porites lutea (Hennige et al., 2008) and various Brazilian coral species (Suggett et al., 2012) growing at different depths. In our dynamic light shift experiment, Ek did not vary significantly until day 21, consistent with previous reports of no change to Ek in A. muricata following nine days of exposure to both increased and decreased light (Langlois and Hoogenboom, 2014). In contrast,

Anthony & Hoegh-Guldberg (2003a) estimated that Ek could adjust and reach stable values within 5-10 days in Turbinaria mesenterina, far shorter than the time required for adjustment in A. muricata. This may be a result of species-specific differences in photoacclimation capacity or rate, which could relate to colony morphology (Nitschke et

147

al., 2018). Specifically, T. mesenterina has a plating growth form that results in continuous exposure of a high percentage of its polyps to altered light regimes; in contrast, many A. muricata polyps are oriented at angles up to 90° to incident irradiance, resulting in less consistent exposure of the colony to a given light regime, and potentially a lengthier acclimation period. Alternatively, the lack of change in Ek reported by Langlois & Hoogenboom (2014) may reflect combined effects of photoinhibition and acclimation over increased light levels (i.e. the capacity for photoacclimation in A. muricata was exceeded).

Intriguingly, Langlois & Hoogenboom (2014) also found no variation in rETRMAX among A. muricata during their nine-day acclimation period, which contrasts with our observations of changes in rETRMAX after one day of exposure to treatment light in the present study. Such contrasting findings may result from the fact that the highest and lowest light treatment in Langlois & Hoogenboom (2014) differed by only 905 μmol photon m-2 s-1, whereas the treatments in the present study differed by ~2000 μmol photon m-2 s-1. More substantial differences in light treatments in the present study may have elicited a greater photophysiological response in A. muricata. Contrasting results have been reported from studies of other coral species. Although increases in Ek for deep-water Stylophora pistillata transplanted to a shallow site agreed with the findings of our study, Ek was also found to increase for shallow-water colonies transplanted to deeper sites after 14 days, suggesting a lack of short-term acclimation to reduced light

(Cohen and Dubinsky, 2015). However, after six months, Ek for deep-water S. pistillata transplanted to a shallow site decreased to levels similar to Ek of deep-water control colonies (Cohen and Dubinsky, 2015). This indicates that susceptibility to

148

photoinhibition can increase with time in corals moved to high light environments

(Cohen and Dubinsky, 2015), and broadly suggests that observed photoacclimation may be temporary in some cases. Repeated photophysiological measurements at regular intervals over a longer study period (i.e. months) are therefore likely required to resolve the stability of adjustments to Ek in A. muricata.

Although photophysiological parameters changed within a relatively short time frame of 21 days, metabolomic shifts were resolved using PLS-DA within an even faster period of 7 days, particularly for LL to HL treated corals. Importantly, metabolomic shifts

MAX were apparent well before any significant differences in Ek and rETR were observed.

This indicates that metabolic reorganization occurs very shortly after a light shift, and that the emergent ‘signatures’ detected by PAM are an endpoint of these shifting metabolic processes, and hence only detected well into the photoacclimation process.

Paired metabolomic and photophysiological data therefore suggest that metabolites may be more precise biomarkers of fine-scale physiological changes during short-term photoacclimation. Although metabolomic shifts suggest a number of pathways are likely altered to achieve photoacclimation, it is unclear whether rapid adjustment is possible for all physiological responses in corals. For example, calcification rates for corals transplanted across depths remained significantly lower than those of control corals at the same depths for up to six months post-transplant (Cohen and Dubinsky, 2015).

Again, longer-term studies could better address the effects of rapid light shifts on A. muricata characteristics that may require a longer study period to exhibit a response, such as calcification and growth.

149

In order to fully characterize the effect of light shifts on the A. muricata metabolome, we paired GC-MS and LC-MS metabolomic profiling approaches. GC-MS is well-recognized as a useful method for identifying primary metabolites with low molecular weight, such as sugars, fatty acids, and amino acids (De Vos et al., 2007; Liu et al., 2017). An advantage of GC-MS is that it is highly reproducible, and a number of libraries have been developed to facilitate identification of primary metabolites (De Vos et al., 2007; Lee et al., 2013). In contrast, LC-MS is more beneficial for distinguishing secondary metabolites, such as terpenoids, alkaloids, and phenols (De Vos et al., 2007;

Liu et al., 2017). Because secondary metabolites are often species-specific (Lee et al.,

2013), their identification can be far more challenging, particularly in the absence of species-specific metabolite databases. Identification of molecular features resolved using LC-MS therefore typically requires subsequent application of targeted approaches such as tandem MS (Viant et al., 2017), which was beyond the scope of the present study. To facilitate future elucidation of the 28 primary targets of interest from the LC-

MS analysis, the raw data including composite spectra will be made publicly accessible via MetaboLights. Regardless, the use of GC-MS and LC-MS approaches in concert offers a more complete picture of metabolomic adjustment during photoacclimation than either approach could offer alone. The findings of this study are also strengthened by the fact that overall metabolomic trends were comparable between approaches.

GC-MS and LC-MS metabolomic profiling approaches also differ in the total number of metabolites they typically retrieve. In the present study, >1000 molecular features were found among samples using LC-MS profiling. The high number of features retrieved using LC-MS is generally consistent with previous studies of corals;

150

for example, Farag et al. (2016) reported >6000 mass signals among five species of the soft coral genus Sarcophyton. In contrast, GC-MS profiling detected 182 metabolites, 59 of which were annotated by comparison to the NIST17 library. This is generally comparable to previous GC-MS studies of corals and their symbionts; for example,

Klueter et al. (2015) detected a total of 188 metabolites among Symbiodinium samples,

110 of which were identified to at least the level of metabolite class. Hillyer et al. (2017) identified 66 unique metabolites among Acropora aspera host samples and 73 metabolites among isolated symbiont samples, and Matthews et al. (2017) identified 89 compounds in the host tissue of the cnidarian Exaiptasia pallida.

Both GC-MS and LC-MS approaches revealed overlap between metabolomic profiles of HL to LL treated corals and HL controls. This finding indicates that the metabolome of HL to LL treated corals did not shift substantially, particularly compared to LL to HL treated corals. This could suggest fewer metabolic pathways are altered in response to reduced light availability. Acclimation to high light generally requires rapid photoprotective responses to minimize oxidative stress and repair PSII damage (Lesser and Shick, 1989; Jeans et al., 2013), which could reasonably manifest in a larger and more rapid metabolic change compared to low light acclimation. In contrast, low light acclimation generally consists of increases in symbiont and/or photosynthetic pigment density (Falkowski and Dubinsky, 1981; Titylanov et al., 2001; Jeans et al., 2013;

Langlois and Hoogenboom, 2014). While changes in pigment density can occur rapidly, nearly six weeks were required to observe significant changes in symbiont density in

Stylophora pistillata (Titlyanov et al., 2001). Similarly, Falkowski and Dubinsky (1981) found that S. pistillata requires up to 8 weeks to acclimate to reduced light,

151

approximately twice the time required to acclimate to high light in the same study. The lack of metabolomic adjustment in A. muricata in the present study could therefore suggest that this species also requires longer than 21 days to complete metabolic adjustments required for low light acclimation. Limited capacity for rapid acclimation to low light could partially explain reports of high mortality following in situ transplantation of acroporid corals to lower light environments (Ross, 2014; Lohr et al., 2017).

Consequently, careful consideration should also be given to selection of deeper sites during coral transplantation activities. Additionally, techniques that could aid in facilitating acclimation, such as phased transfer to deeper depths, should be explored.

Future studies should also consider analyzing the host and symbiont metabolomes separately (see Hillyer et al., 2017) in order to elucidate any metabolomic responses of either partner during low light acclimation that could be confounded when exploring the holobiont metabolome.

In contrast to HL to LL treated corals, LL to HL treated corals underwent substantial metabolic adjustment, and their metabolomes were distinct from LL controls and similar to those of initial HL samples after both 7 and 21 days of treatment.

Although limited data on the effects of light on the coral metabolome are available, our findings of metabolomic variation between LL to HL treated colonies and LL controls are supported by existing literature on photosynthetic organisms. Broad differences in metabolomic profiles were reported for the model plant species Arabidopsis thaliana grown under three varying light conditions (Jänkänpää et al., 2012). Similarly, a study by Klueter et al. (2015) reported differences among metabolomes of four Symbiodinium species cultured under three light levels in the absence of a coral host. Glycerol was

152

found to be a key driver of metabolomic variation among Symbiodinium exposed to differing light levels (Klueter et al., 2015); this is consistent with our results, as glycerol was significantly more abundant in LL controls compared to LL to HL treated corals at both time points (Fig. 6-6a). Glycerol has long been recognized as a key photosynthetic product transferred from Symbiodinium to coral hosts (Muscatine, 1967; Muscatine and

Cernichiari, 1969; Suescún-Bolívar et al., 2012). Lower concentrations of glycerol in LL to HL treated corals compared to LL controls could therefore potentially indicate negative effects of HL exposure on photosynthetic processes. Hoogenboom et al.

(2006) noted reduced carbon fixation by high light-acclimated corals compared to colonies acclimated to lower light levels, primarily as a result of lower chlorophyll concentration combined with higher Ek and respiration. Acclimation to HL was associated with increased Ek in the present study (Fig. 6-2c), however chlorophyll concentration and respiration were not assessed. Such direct measurements of coral metabolism could help improve understanding of the potential effects of high light on carbon fixation and translocation. In addition to glycerol, campesterol was also less abundant among LL to HL treated samples compared to LL controls. Like glycerol, campesterol is produced by Symbiodinium and translocated to the coral host (Treignier et al., 2009; Crandall et al., 2016). Reduced synthesis of campesterol therefore also points to a potential adverse effect of high light on symbionts, and warrants further investigation.

In addition to a decrease in glycerol among LL to HL treated samples compared to LL controls, two saturated fatty acids, tetradecanoic acid (C14:0) and dodecanoic acid (C12:0), increased in these treated samples compared to controls. Hillyer et al.

153

(2017) reported increases in the same fatty acids for Symbiodinium exposed to heat stress compared to colonies kept in ambient conditions. Similarly, Matthews et al.

(2017) found differences in fatty acids between E. pallida colonized with two differing symbionts. Both Hillyer et al. (2017) and Matthews et al. (2017) suggest that accumulation of fatty acids could result from increased breakdown of lipid stores as an alternative source of energy. This could also be occurring among LL to HL treated corals in the present study, particularly given the concurrent reduction in the abundance

(and presumably translocation) of glycerol and campesterol for this treatment group.

Less information is available on the potential roles of additional metabolites that varied between LL to HL treatment and LL control colonies, including 1-hexanol, L-cysteine, and 1-hexadecanol. However, Shinzato et al. (2011) note that Acropora corals cannot synthesize their own cysteine, and are therefore reliant on their symbionts for this amino acid. The observed increase in L-cysteine in LL to HL treated corals compared to LL controls is therefore also likely a result of altered symbiont activity.

Metabolomic profiling, in conjunction with PAM fluorometry, provides new information on physiological changes that occur during the coral photoacclimation process. Although photophysiological datasets support findings of significant acclimation to high and low light within 21 days in A. muricata, paired metabolomic data suggests metabolic reorganization (particularly in LL to HL treated corals) begins well before acclimation is detected using fluorometry techniques. Interestingly, although Ek and rETRMAX changed significantly for corals exposed to both increased and decreased light, treatment did not have a clear effect on Fq’/Fm’(max), contrasting with a number of previous studies. The precise mechanisms maintaining relatively constant Fq’/Fm’(max)

154

values over time in high and low light regimes warrant further study. Despite changes in

MAX Ek and rETR for HL to LL treated colonies, metabolomes for this treatment group overlapped considerably with HL controls, suggesting limited alteration of metabolic pathways following LL exposure over 21 days. In contrast, distinct photophysiological and metabolomic differences were observed between LL to HL treated colonies and LL controls. Changes in the abundance of glycerol, campesterol, and two fatty acids suggest translocation of photosynthetic products from symbiont to host may be reduced following LL to HL shifts, despite apparently successful acclimation reflected by photophysiological parameters. These metabolites could therefore prove to be useful indicators to assess the effects of rapid changes in light history, and targeted metabolomic profiling approaches could improve our understanding of how these may change over time following changes in light availability. This study improves our understanding of coral photobiology through metabolomic profiling of rapid and fine- scale metabolic changes that may not be resolved using fluorometry-based approaches alone.

155

CHAPTER 7 CONCLUSIONS

Coral restoration is increasingly recognized as an integral part of larger management strategies aimed at conservation and recovery of threatened reef systems

(National Marine Fisheries Service 2015; van Oppen et al. 2017). Although capacity for and efficiency of coral restoration is generally increasing (Rinkevich 2014; Schopmeyer et al. 2017), variability in performance of restored corals can still be high due to genetic and environmental factors. This dissertation aimed to add to the growing body of knowledge regarding the effects of genotype and environmental factors on corals in the context of reef restoration, with a particular focus on acroporid corals, which are widely cultured for use in restoration (Young et al. 2012).

Previous studies have suggested that selecting genotypes for use in population enhancement based on the presence or absence of desirable traits could enhance overall restoration outcomes (Hunt and Sharp 2014; van Oppen et al. 2015, 2017).

Chapter 2 of this study demonstrates that key traits of interest, including total linear extension (TLE), buoyant weight (a measure of calcification), branch formation, and thermotolerance varied among ten unique Acropora cervicornis genotypes in a common garden. Genetic variation in the study population accounted for approximately a quarter of variability in growth traits. The ratio of buoyant weight to TLE, an approximation for skeletal density, also varied among genotypes. Furthermore, the ratio of buoyant weight to TLE decreased with increasing TLE, suggesting a tradeoff between skeletal density and linear growth in nursery-reared A. cervicornis. Finally, specific growth rate decreased following a sub-lethal bleaching event, indicating a negative effect of thermal stress on growth regardless of genotype. These results demonstrate the effect of

156

genotype on important traits in a common garden, supporting the rationale for a trait- based A. cervicornis restoration system. Rapid growth, high skeletal density, and low bleaching prevalence are considered desirable traits in A. cervicornis restoration, as these could aid in habitat provision, resistance to breakage, and climate change tolerance, respectively.

Chapter 3 built on findings of intraspecific variation in phenotype among nursery- reared A. cervicornis by exploring the extent to which metabolomic profiles differed among genetically and phenotypically unique corals. A subset of three genotypes previously characterized in the nursery and determined to differ in both growth and thermotolerance were subjected to both 1H-NMR and LC-MS metabolomic profiling.

Each genotype had a distinct metabolomic “fingerprint” when assessed using PLS-DA, regardless of the profiling method used. However, high within-genotype variability for genotype U41 resulted in overlap of metabolomic profiles with one or both of the other two genotypes when assessed using PCA. Putative metabolites driving separation in metabolomic profiles among genotypes included compounds with key osmotic and antioxidant function. A number of metabolic pathways related to protein synthesis likely drove separation among the three A. cervicornis genotypes studied. The presence of distinct metabolomic profiles among genotypes suggests that future work may be able to link individual metabolites and pathways to coral phenotype. Metabolomic profiling therefore has great potential to identify biomarkers that could aid in selection of robust genotypes for restoration, and could contribute to replacement of the time- and labor- intensive in situ measurements currently required to identify phenotype in A. cervicornis.

157

To fully assess the utility of a trait-based system for A. cervicornis restoration,

Chapter 4 aimed to study the effect of outplanting on phenotype for six genotypes previously characterized in a nursery setting (Chapter 2), while also assessing differences in outplanted coral performance between two reef zones (patch reef and fore reef). The passage of Hurricane Irma during this experiment also provided a key opportunity to explore differences in outplant response to hurricane disturbance among sites. Growth measured prior to the hurricane did not vary among genotypes, sites, or reef zones, however colony breakage was higher at fore reef sites compared to patch reef sites. Post-hurricane growth also did not vary among genotypes or sites. Stark differences in hurricane survival were apparent between reef zones: while approximately half of outplanted corals survived at patch reef sites, no outplants remained at either fore reef site. Outplant monitoring records obtained from a restoration practitioner in the same region also showed significantly higher survival and lower partial mortality on patch reef sites compared to fore reef sites following the passage of Hurricane Irma. The lack of effect of genotype or site on growth contrasts sharply with a number of studies on A. cervicornis, as well as with previous results for the same genotypes described in Chapter 2. The effects of genotype and site on growth were likely confounded by hurricane effects, and further investigation is required to fully assess how phenotype may change from a nursery to outplant setting. However, the effects of hurricane disturbance on survival and partial mortality suggest that restoration practitioners should consider shifting greater effort toward restoration of patch reef sites, while also striving to improve resilience among restored fore reef populations.

158

The effect of reef zone on performance of restored A. cervicornis in a location with different reef geology was explored in Chapter 5. Three reef zones were considered in this study: shallow back reef (0–3 m), intermediate spur-and-groove reef

(8–15 m), and deep reef terrace (>15 m). Roving diver surveys confirmed the presence of wild A. cervicornis in each reef zone prior to restoration, and determined that wild colony height and Stegastes spp. predation was highest on intermediate spur-and- groove reefs. When A. cervicornis colonies were planted to each of the three reef zones from a single nursery site (6 m), survival was lowest at deep reef terrace sites after an

85 day observation period. The cause of high initial mortality among outplants on the deep reef terrace is unclear, as signs of disease, predation, and physical damage were not observed. It is possible that reduced light availability resulted in negative physiological impacts on deep outplants. Post-outplant growth and branching also varied among reef zones, and was lowest among outplants in the shallow back reef due to high rates of colony breakage, likely a result of high wave energy in this zone.

Following the conclusion of the study period, pooled survivorship at the shallow back reef sites dropped to 10% during a period of elevated sea temperature. Overall, these results indicate that A. cervicornis restoration at the study site was most successful on the intermediate spur-and-groove reef, the zone most similar in depth to the nursery of origin. However, use of a single nursery could limit restoration across diverse reef zones. Establishing nurseries in multiple reef zones could aid in ensuring the full range of natural A. cervicornis habitat is effectively targeted in restoration activities.

The effect of rapid changes in light availability on acroporids was explored in greater detail in Chapter 6. Repeated photophysiological measurements using Pulse

159

Amplitude Modulation (PAM) fluorometry over a 21-day period demonstrated that the model acroporid A. muricata can generally acclimate to both increased and decreased light in a matter of days. Although no clear patterns in maximum photochemical efficiency of photosystem II were observed, shifts in sub-saturation irradiance and maximum relative electron transport rate indicated acclimation to both increased and decreased light levels by day 21. Concurrent LC-MS and GC-MS metabolomic profiling revealed clear metabolic adjustment within 7 days of treatment, well before acclimation was detected using PAM fluorometry. Metabolomic shifts were substantial for LL to HL treated corals, while the metabolomes of HL to LL treated colonies overlapped with those of HL controls throughout the study period. A number of metabolites drove separation between LL to HL treated colonies and LL controls, including amino acids, organic acids, fatty acids, and sterols. Decreased glycerol and campesterol suggest translocation of photosynthetic products from symbiont to host was reduced in LL to HL treated corals, while concurrent increases in fatty acid abundance may indicate reliance on lipid stores for energy. These results indicate that while acclimation is possible for acroporid corals moved between light environments during restoration, signs of stress that are not apparent visually could be present. Additionally, acclimation to low light conditions could require more time for metabolic adjustment. Careful consideration should therefore be given to light conditions during site selection for acroporid restoration.

Specific recommendations for coral restoration practice include focusing greater effort on restoring A. cervicornis at Florida Keys patch reef sites, as well as establishing nurseries across a diverse range of reef zones inhabited by target species. This

160

dissertation also illustrates useful applications of physiological techniques, including metabolomic profiling, for coral restoration. While these results generally support the rationale for a trait-based A. cervicornis restoration system, they also demonstrate that additional studies are required to fully understand how phenotype may change from nursery to outplant sites. Together, these results provide actionable information for restoration practitioners while informing future research efforts for acroporid corals, particularly A. cervicornis.

161

APPENDIX REPRINT PERMISSIONS

162

LIST OF REFERENCES

Adey, W.H., 2000. Coral reef ecosystems and human health: Biodiversity counts! Ecosyst. Heal. 6, 227–236. doi:10.1046/j.1526-0992.2000.006004227.x

Aguilar, C., Raina, J.B., Motti, C.A., Fôret, S., Hayward, D.C., Lapeyre, B., Bourne, D.G., Miller, D.J., 2017. Transcriptomic analysis of the response of Acropora millepora to hypo-osmotic stress provides insights into DMSP biosynthesis by corals. BMC Genomics 18, 1–14. doi:10.1186/s12864-017-3959-0

Alvarez-Filip, L., Dulvy, N.K., Gill, J.A., Côté, I.M., Watkinson, A.R., 2009. Flattening of Caribbean coral reefs: Region-wide declines in architectural complexity. Proc. R. Soc. London B Biol. Sci. 276, 3019–3025. doi: 10.1098/rspb.2009.0339

Anderson, K., Pratchett, M., Baird, A., 2012. Summer growth rates of corals at Lord Howe Island, Australia, in: Riegl, B., Dodge, R.E. (Eds.), Proceedings of the 11th International Coral Reef Symposium. Nova Southeastern University National Coral Reef Institute, Davie, pp. 9–13.

Anthony, K.R.N., Hoegh-Guldberg, O., 2003a. Kinetics of photoacclimation in corals. Oecologia 134, 23–31. doi:10.1007/s00442-002-1095-1

Anthony, K.R.N., Hoegh-Guldberg, O., 2003b. Variation in coral photosynthesis, respiration and growth characteristics in contrasting light microhabitats: An analogue to plants in forest gaps and understoreys? Funct. Ecol. 17, 246–259. doi:10.1046/j.1365-2435.2003.00731.x

Anthony, K.R.N., Ridd, P. V., Orpin, A.R., Larcombe, P., Lough, J., 2004. Temporal variation of light availability in coastal benthic habitats: Effects of clouds, turbidity, and tides. Limnol. Oceanogr. 49, 2201–2211. doi:10.4319/lo.2004.49.6.2201

Aronson, R., Bruckner, A., Moore, J., Precht, B., Weil, E., 2008a. Acropora cervicornis. The IUCN Red List of Threatened Species: e.T133381A3716457. doi:10.2305/IUCN.UK.2008.RLTS.T133381A3716457.en

Aronson, R., Bruckner, A., Moore, J., Precht, B., Weil, E., 2008b. Acropora palmata. The IUCN Red List of Threatened Species: e.T133006A3536699. doi:http://dx.doi.org/10.2305/IUCN.UK.2008.RLTS.T133006A3536699.en

Aronson, R.B., Precht, W.F., 2001. White-band disease and the changing face of Caribbean coral reefs. Hydrobiologia 460, 25–38. doi:10.1023/A:1013103928980

Bailey, M., Sumaila, U.R., 2015. Destructive fishing and fisheries enforcement in eastern Indonesia. Mar. Ecol. Prog. Ser. 530, 195–211. doi:10.3354/meps11352

Baird, A., Marshall, P., 2002. Mortality, growth and reproduction in scleractinian corals following bleaching on the Great Barrier Reef. Mar. Ecol. Prog. Ser. 237, 133– 141. doi:10.3354/meps237133

163

Bak, R.P.M., 1987. Effects of chronic oil pollution on a Caribbean coral reef. Mar. Pollut. Bull. 18, 534–539. doi:10.1016/0025-326X(87)90537-6

Baird, A.H., Bhagooli, R., Ralph, P.J., Takahashi, S., 2009. Coral bleaching: The role of the host. Trends Ecol. Evol. 24, 16–20. doi:10.1016/j.tree.2008.09.005

Baums, I.B., 2008. A restoration genetics guide for coral reef conservation. Mol. Ecol. 17, 2796–2811. doi:10.1111/j.1365-294X.2008.03787.x

Baums, I.B., Devlin-Durante, M.K., Pinzón, J.H., 2009. Nine novel, polymorphic microsatellite markers for the study of threatened Caribbean acroporid corals. Mol. Ecol. Resour. 9, 1155–1158. doi:10.1111/j.1755-0998.2009.02581.x

Baums, I.B., Johnson, M.E., Devlin-Durante, M.K., Miller, M.W., 2010. Host population genetic structure and zooxanthellae diversity of two reef-building coral species along the Florida Reef Tract and wider Caribbean. Coral Reefs 29, 835–842. doi:10.1007/s00338-010-0645-y

Bellwood, D.R., Hughes, T.P., Folke, C., Nyström, M., 2004. Confronting the coral reef crisis. Nature 429, 827–833. doi:10.1038/nature02691

Bender, M.A., Knutson, T.R., Tuleya, R.E., Sirutis, J.J., Vecchi, G.A., Garner, S.T., Held, I.M., 2010. Modeled impact of anthropogenic warming on the frequency of intense Atlantic hurricanes. Science 327, 454–458. doi:10.1126/science.1180568

Bessell-Browne, P., Negri, A.P., Fisher, R., Clode, P.L., Duckworth, A., Jones, R., 2017. Impacts of turbidity on corals: The relative importance of light limitation and suspended sediments. Mar. Pollut. Bull. 117, 161–170. doi:10.1016/j.marpolbul.2017.01.050

Birkeland, C., 2015. Coral Reefs in the Anthropocene, in: Birkeland, C. (Ed.), Coral Reefs in the Anthropocene. Springer Science & Business Media, Dordrecht, pp. 1–15.

Blackall, L.L., Wilson, B., Van Oppen, M.J.H., 2015. Coral-the world’s most diverse symbiotic ecosystem. Mol. Ecol. 24, 5330–5347. doi:10.1111/mec.13400

Bosch, T.C.G., Miller, D.J., 2016. Bleaching as an obvious dysbiosis in corals, in: The Holobiont Imperative. Springer Vienna, Vienna, pp. 113–125. doi:10.1007/978-3- 7091-1896-2_9

Bottjer, D.J., 1980. Branching morphology of the reef coral Acropora cervicornis in different hydraulic regimes. J. Paleontol. 54, 1102–1107.

Bowden-Kerby, A., 2001. Low-tech coral reef restoration methods modeled after natural fragmentation processes. Bull. Mar. Sci. 69, 915–931.

164

Bowden-Kerby, A., 2008. Restoration of threatened Acropora cervicornis corals: Intraspecific variation as a factor in mortality, growth, and self-attachment, in: Riegl, B., Dodge, R.E. (Eds.), Proceedings of the 11th International Coral Reef Symposium. Nova Southeastern University National Coral Reef Institute, Davie, pp. 7–11.

Bowden-Kerby, A., Carne, L., 2012. Thermal tolerance as a factor in Caribbean Acropora restoration, in: Yellowlees, D., Hughes, T.P. (Eds.), Proceedings of the 12th International Coral Reef Symposium. ARC Centre of Excellence for Coral Reef Studies, Townsville, pp. 1–4.

Brown, B.E., Suharsono, 1990. Damage and recovery of coral reefs affected by El Niño related seawater warming in the Thousand Islands, Indonesia. Coral Reefs 8, 163–170. doi:10.1007/BF00265007

Brown, B.E., 1997. Coral bleaching: causes and consequences. Coral Reefs 16, S129– S138. doi:10.1007/s003380050249

Bruckner, A., Bruckner, R., 2001. Condition of restored Acropora palmata fragments off Mona Island, Puerto Rico, 2 years after the Fortuna Reefer ship grounding. Coral Reefs 20, 235–243. doi:10.1007/s003380100164

Bruckner, A.W., Aronson, R.B., Bruckner, R.J., 2002. Endangered acroporid corals of the Caribbean. Coral Reefs 21, 41–42. doi:10.1007/s00338-001-0209-2

Buddemeier, R.W., Fautin, D.G., 1993. Coral bleaching as an adaptative mechanism: A testable hypothesis. Bioscience 43, 320–326. doi:10.2307/1312064

Caldana, C., Degenkolbe, T., Cuadros-Inostroza, A., Klie, S., Sulpice, R., Leisse, A., Steinhauser, D., Fernie, A.R., Willmitzer, L., Hannah, M.A., 2011. High-density kinetic analysis of the metabolomic and transcriptomic response of Arabidopsis to eight environmental conditions. Plant J. 67, 869–884. doi:10.1111/j.1365- 313X.2011.04640.x

Canelas, A.B., Ras, C., ten Pierick, A., van Dam, J.C., Heijnen, J.J., van Gulik, W.M., 2008. Leakage-free rapid quenching technique for yeast metabolomics. Metabolomics 4, 226–239. doi:10.1007/s11306-008-0116-4

Carreno-Quintero, N., Acharjee, A., Maliepaard, C., Bachem, C.W.B., Mumm, R., Bouwmeester, H., Visser, R.G.F., Keurentjes, J.J.B., 2012. Untargeted metabolic quantitative trait loci analyses reveal a relationship between primary metabolism and potato tuber quality. Plant Physiol. 158, 1306–18. doi:10.1104/pp.111.188441

Carreno-Quintero, N., Bouwmeester, H.J., Keurentjes, J.J.B., 2013. Genetic analysis of metabolome-phenotype interactions: From model to crop species. Trends Genet. 29, 41–50. doi:10.1016/j.tig.2012.09.006

165

Carrigan, A.D., Puotinen, M., 2014. Tropical cyclone cooling combats region-wide coral bleaching. Glob. Chang. Biol. 20, 1604–1613. doi:10.1111/gcb.12541

Cetz-Navarro, N.P., Quan-Young, L.I., Espinoza-Avalos, J., 2015. Morphological and community changes of turf algae in competition with corals. Sci. Rep. 5, 1–12. doi:10.1038/srep12814

Chalker, B., Taylor, D., 1975. Light-enhanced calcification, and the role of oxidative phosphorylation in calcification of the coral Acropora cervicornis. Proc. R. Soc. London B Biol. Sci. 190, 323–331.

Chow, M.H., Tsang, R.H.L., Lam, E.K.Y., Ang, P., 2016. Quantifying the degree of coral bleaching using digital photographic technique. J. Exp. Mar. Bio. Ecol. 479, 60– 68. doi:10.1016/j.jembe.2016.03.003

Cinner, J., 2014. Coral reef livelihoods. Curr. Opin. Environ. Sustain. 7, 65–71. doi:10.1016/j.cosust.2013.11.025

Cohen, I., Dubinsky, Z., 2015. Long term photoacclimation responses of the coral Stylophora pistillata to reciprocal deep to shallow transplantation: photosynthesis and calcification. Front. Mar. Sci. 2, 1–13. doi:10.3389/fmars.2015.00045

Connell, J.H., 1978. Diversity in tropical rain forests and coral reefs. Science 199, 1302– 1310. doi:10.1126/science.199.4335.1302

Crabbe, M.J.C., 2014. Evidence of initial coral community recovery at Discovery Bay on Jamaica’s North Coast. Rev. Biol. Trop. 62, 137–140.

Craig, P., Birkeland, C., Belliveau, S., 2001. High temperatures tolerated by a diverse assemblage of shallow-water corals in American Samoa. Coral Reefs 20, 185– 189. doi:10.1007/s003380100159

Crandall, J.B., Teece, M.A., Estes, B.A., Manfrino, C., Ciesla, J.H., 2016. Nutrient acquisition strategies in mesophotic hard corals using compound specific stable isotope analysis of sterols. J. Exp. Mar. Bio. Ecol. 474, 133–141. doi:10.1016/j.jembe.2015.10.010

Császár, N.B.M., Ralph, P.J., Frankham, R., Berkelmans, R., van Oppen, M.J.H., 2010. Estimating the potential for adaptation of corals to climate warming. PLoS One 5. doi:10.1371/journal.pone.0009751

Davis, M.C., Fiehn, O., Durnford, D.G., 2013. Metabolic acclimation to excess light intensity in Chlamydomonas reinhardtii. Plant, Cell Environ. 36, 1391–1405. doi:10.1111/pce.12071

De’ath, G., Fabricius, K., 2010. Water quality as a regional driver of coral biodiversity and macroalgae on the Great Barrier Reef. Ecol. Appl. 20, 840–850. doi:Doi 10.1890/08-2023.1

166

DeSalvo, M.K., Sunagawa, S., Voolstra, C.R., Medina, M., 2010. Transcriptomic responses to heat stress and bleaching in the elkhorn coral Acropora palmata. Mar. Ecol. Prog. Ser. 402, 97–113. doi:10.3354/meps08372

DeSalvo, M.K., Estrada, A., Sunagawa, S., Medina, M., 2012. Transcriptomic responses to darkness stress point to common coral bleaching mechanisms. Coral Reefs 31, 215–228. doi:10.1007/s00338-011-0833-4

De Vos, R.C., Moco, S., Lommen, A., Keurentjes, J.J., Bino, R.J., Hall, R.D., 2007. Untargeted large-scale plant metabolomics using liquid chromatography coupled to mass spectrometry. Nat. Protoc. 2, 778

Dias, D.A., Hill, C.B., Jayasinghe, N.S., Atieno, J., Sutton, T., Roessner, U., 2015. Quantitative profiling of polar primary metabolites of two chickpea cultivars with contrasting responses to salinity. J. Chromatogr. B 1000, 1–13

Dieterle, F., Ross, A., Schlotterback, G., Senn, H., 2006. Probabilistic quotient normalization as robust method to account for dilution of complex biological mixtures. Application in 1H-NMR metabonomics. Anal. Chem. 78, 4281–4290.

Dinno, A., 2015. dunn.test: Dunn’s test of multiple comparisons using rank sums. R package version 1.3.1.

Dixon, G.B., Davies, S.W., Aglyamova, G.V., Meyer, E., Bay, L.K., Matz, M.V., 2015. Genomic determinants of coral heat tolerance across latitudes. Science 348, 1460–1462. doi:10.1126/science.1261224

Drury, C., Dale, K.E., Panlilio, J.M., Miller, S. V., Lirman, D., Larson, E.A., Bartels, E., Crawford, D.L., Oleksiak, M.F., 2016. Genomic variation among populations of threatened coral: Acropora cervicornis. BMC Genomics 17, 286. doi:10.1186/s12864-016-2583-8

Drury, C., Manzello, D., Lirman, D., 2017. Genotype and local environment dynamically influence growth, disturbance response and survivorship in the threatened coral, Acropora cervicornis. PLoS One 12, e0174000. doi:10.1371/journal.pone.0174000

Eakin, C.M., Lough, J.M., Heron, S.F., 2009. Climate variability and change: Monitoring data and evidence for increased coral bleaching stress, in: van Oppen, M.J.H., Lough, J. (Eds.), Coral Bleaching: Patterns, processes, causes and consequences. Springer, Berlin, pp. 41–67. doi:10.1007/978-3-540-69775-6_4

Edmunds, P.J., 1994. Evidence that reef-wide patterns of coral bleaching may be the result of the distribution of bleaching-susceptible clones. Mar. Biol. 121, 137–142. doi:10.1007/BF00349482

167

Edwards, A., Gomez, E., 2007. Reef Restoration Concepts and Guidelines: Making Sensible Management Choices in the Face of Uncertainty. Coral Reef Targeted Research and Capacity Building for Management Programme, St. Lucia.

Elias, A.A., Robbins, K.R., Doerge, R.W., Tuinstra, M.R., 2016. Half a century of studying genotype × environment interactions in plant breeding experiments. Crop Sci. 56, 2090–2105. doi:10.2135/cropsci2015.01.0061

Emanuel, K., 2005. Increasing destructiveness of tropical cyclones over the past 30 years. Nature 436, 686–688. doi:10.1038/nature03906

Emanuel, K.A., 2013. Downscaling CMIP5 climate models shows increased tropical cyclone activity over the 21st century 110, 12219–12224. doi:10.1073/pnas.1301293110

Emwas, A.-H.M., 2015. The strengths and weaknesses of NMR spectroscopy and mass spectrometry with particular focus on metabolomics research, in: Bjerrum, J. (Ed.), Metabonomics. Humana Press, New York, NY, pp. 161–193. doi:10.1007/978-1-4939-2377-9_13

Enochs, I.C., Manzello, D.P., Carlton, R., Schopmeyer, S., van Hooidonk, R., Lirman, D., 2014. Effects of light and elevated pCO2 on the growth and photochemical efficiency of Acropora cervicornis. Coral Reefs 33, 477–485. doi:10.1007/s00338-014-1132-7

Enríquez, S., Méndez, E.R., Iglesias-Prieto, R., 2005. Multiple scattering on coral skeletons enhances light absoprtion by symbiotic algae. Limnol. Ocean. 50, 1025–1032. doi:10.4319/lo.2005.50.4.1025

Epstein, N., Bak, R.P.M., Rinkevich, B., 2003. Applying forest restoration principles to coral reef rehabilitation. Aquat. Conserv. Mar. Freshw. Ecosyst. 13, 387–395. doi:10.1002/aqc.558

Epstein, N., Bak, R.P.M., Rinkevich, B., 2001. Strategies for gardening denuded coral reef areas: The applicability of using different types of coral material for reef restoration. Restor. Ecol. 9, 432–442. doi:10.1046/j.1526-100X.2001.94012.x

Eriksson, L., Johansson, E., Kettaneh-Wold, N., Trygg, J., Wikström, C., Wold, S., 2006. Multi- and Megavariate Data Analysis: Basic Principles and Applications, Third. ed. MKS Umetrics AB, Malmö, Sweden.

Escobar-Morreale, H.F., Samino, S., Insenser, M., Vinaixa, M., Luque-Ramírez, M., Lasunción, M.A., Correig, X., 2012. Metabolic heterogeneity in polycystic ovary syndrome is determined by obesity: Plasma metabolomic approach using GC- MS. Clin. Chem. 58, 999–1009. doi:10.1373/clinchem.2011.176396

168

Fabricius, K.E., 2005. Effects of terrestrial runoff on the ecology of corals and coral reefs: review and synthesis. Mar. Pollut. Bull. 50, 125–146. doi:10.1016/j.marpolbul.2004.11.028

Falkowski, P.G., Dubinsky, Z., 1981. Light-shade adaptation of Stylophora pistillata, a hermatypic coral from the Gulf of Eilat. Nature 289, 172–174. doi:10.1038/289172a0

Fenn, L.S., McLean, J.A., 2011. Structural resolution of carbohydrate positional and structural isomers based on gas-phase ion mobility-mass spectrometry. Phys. Chem. Chem. Phys. 13, 2196–2205. doi:10.1039/C0CP01414A

Ferrario, F., Beck, M.W., Storlazzi, C.D., Micheli, F., Shepard, C.C., Airoldi, L., 2014. The effectiveness of coral reefs for coastal hazard risk reduction and adaptation. Nat. Commun. 5, 1–9. doi:10.1038/ncomms4794

Fiehn, O., 2002. Metabolomics – the link between genotypes and phenotypes. Plant Mol. Biol. 48, 155–171.

Forrester, G.E., Taylor, K., Schofield, S., Maynard, A., 2013. Colony growth of corals transplanted for restoration depends on their site of origin and environmental factors. Mar. Ecol. 34, 186–192. doi:10.1111/maec.12000

Forsman, Z.H., Page, C.A., Toonen, R.J., Vaughan, D., 2015. Growing coral larger and faster: micro-colony-fusion as a strategy for accelerating coral cover. PeerJ 3, e1313. doi:10.7717/peerj.1313

Frade, P.R., Bongaerts, P., Winkelhagen, A.J.S., Tonk, L., Bak, R.P.M., 2008. In situ photobiology of corals over large depth ranges: A multivariate analysis on the roles of environment, host, and algal symbiont. Limnol. Oceanogr. 53, 2711– 2723. doi:10.4319/lo.2008.53.6.2711

Fung, T., Seymour, R.M., Johnson, C.R., 2011. Alternative stable states and phase shifts in coral reefs under anthropogenic stress. Ecology 92, 967–982.

FWC (Florida Fish and Wildlife Conservation Commission), Fish and Wildlife Research Institute, 2001. South Florida Benthic Habitats. Florida Geographic Data Library, University of Florida, Gainesville, FL. http://www.fgdl.org (accessed 28 June 2018).

Gardner, T.A., Côté, I.M., Gill, J.A., Grant, A., Watkinson, A.R., 2003. Long-term region- wide declines in Caribbean corals. Science 301, 958–960. doi:10.1126/science.1086050

Gardner, T.A., Côté, I.M., Gill, J.A., Grant, A., Watkinson, A.R., 2005. Hurricanes and Caribbean coral reefs: Impacts, recovery patterns, and role in long-term decline. Ecology 86, 174–184. doi:10.1890/04-0141

169

Gates, R.D., Edmunds, P., 1999. The physiological mechanisms of acclimatization in tropical reef corals. Am. Zool. 39, 30–43. doi:10.1093/icb/39.1.30

Gilmore, D.M., Hall, B.R., 1976. Life history, growth habits, and constructional roles of Acropora cervicornis in the patch reef environment. J. Sediment. Petrol. 46, 519– 522.

Gintert, B.E., Manzello, D.P., Enochs, I.C., Kolodziej, G., Carlton, R., Gleason, A.C.R., Gracias, N., 2018. Marked annual coral bleaching resilience of an inshore patch reef in the Florida Keys: A nugget of hope, aberrance, or last man standing? Coral Reefs 1–15. doi:10.1007/s00338-018-1678-x

Gladfelter, E.H., Monahan, R.K., Gladfelter, W.B., 1978. Growth rates of five reef- building corals in the northeastern Caribbean. Bull. Mar. Sci. 28, 728–734.

Gladfelter, W.B., 1982a. White-band disease in Acropora palmata: Implications for the structure and growth of shallow reefs. Bull. Mar. Sci. 32, 639–643.

Gladfelter, E.H., 1982b. Skeletal development in Acropora cervicornis: I. Patterns of calcium carbonate accretion in the axial corallite. Coral Reefs 1, 45–51.

Glynn, P.W., Szmant, A.M., Corcoran, E.F., Cofer-Shabica, S. V., 1989. Condition of coral reef cnidarians from the northern Florida reef tract: Pesticides, heavy metals, and histopathological examination. Mar. Pollut. Bull. 20, 568–576. doi:10.1016/0025-326X(89)90359-7

Glynn, P.W., D’Croz, L., 1990. Experimental evidence for high temperature stress as the cause of El Niño-coincident coral mortality. Coral Reefs 8, 181–191. doi:10.1007/BF00265009

Goergen, E.A., Gilliam, D.S., 2018. Outplanting technique, host genotype, and site affect the initial success of outplanted Acropora cervicornis. PeerJ 6, e4433. doi:10.7717/peerj.4433

Gordon, B.R., Leggat, W., 2010. Symbiodinium - invertebrate symbioses and the role of metabolomics. Mar. Drugs 8, 2546–2568. doi:10.3390/md8102546

Gordon, B.R., Leggat, W., Motti, C.A., 2013. Extraction protocol for nontargeted NMR and LC-MS metabolomics-based analysis of hard coral and their algal symbionts, in: Roessner, U., Dias, D.A. (Eds.), Metabolomics Tools for Natural Product Discovery: Methods and Protocols, Methods in Molecular Biology. Springer Science & Business Media, pp. 129–147. doi:10.1007/978-1-62703-577-4

Goreau, T.F., 1959. The ecology of Jamaican coral reefs I. Species composition and zonation. Ecology 40, 67–90.

170

Goulitquer, S., Potin, P., Tonon, T., 2012. Mass spectrometry-based metabolomics to elucidate functions in marine organisms and ecosystems. Mar. Drugs 10, 849– 880. doi:10.3390/md10040849

Greenstein, B.J., Curran, H.A., Pandolfi, J.M., 1998. Shifting ecological baselines and the demise of Acropora cervicornis in the western North Atlantic and Caribbean Province: A Pleistocene perspective. Coral Reefs 17, 249–261. doi:10.1007/s003380050125

Greff, S., Aires, T., Serrão, E.A., Engelen, A.H., Thomas, O.P., Pérez, T., 2017. The interaction between the proliferating macroalga Asparagopsis taxiformis and the coral Astroides calycularis induces changes in microbiome and metabolomic fingerprints. Sci. Rep. 7, 42625. doi:10.1038/srep42625

Griffin, J.N., Schrack, E.C., Lewis, K.-A., Baums, I.B., Soomdat, N., Silliman, B.R., 2015. Density-dependent effects on initial growth of a branching coral under restoration. Restor. Ecol. 23, 197–200. doi:10.1111/rec.12173

Hall, V.R., 2001. The response of Acropora hyacinthus and Montipora tuberculosa to three different types of colony damage: Scraping injury, tissue mortality and breakage. J. Exp. Mar. Bio. Ecol. 264, 209–223. doi:10.1016/S0022- 0981(01)00318-5

Hanna, R.G., Muir, G.L., 1990. Red sea corals as biomonitors of trace metal pollution. Environ. Monit. Assess. 14, 211–222. doi:10.1007/BF00677917

Harrigan, G.G., Stork, L.G., Riordan, S.G., Reynolds, T.L., Ridley, W.P., Masucci, J.D., MacIsaac, S., Halls, S.C., Orth, R., Smith, R.G., Wen, L., Brown, W.E., Welsch, M., Riley, R., McFarland, D., Pandravada, A., Glenn, K.C., 2007. Impact of genetics and environment on nutritional and metabolite components of maize grain. J. Agric. Food Chem. 55, 6177–6185. doi:10.1021/jf070494k

Harvell, C.D., Mitchell, C.E., Ward, J.R., Altizer, S., Dobson, A.P., Ostfeld, R.S., Samuel, M.D., 2002. Climate warming and disease risks for terrestrial and marine biota. Science 296, 2158–2162. doi:10.1126/science.1063699

Heim, K.E., Tagliaferro, A.R., Bobilya, D.J., 2002. Flavonoid antioxidants: chemistry, metabolism and structure-activity relationships. J. Nutr. Biochem. 13, 572–584. doi:10.1016/S0955-2863(02)00208-5

Hein, M.Y., Willis, B.L., Beeden, R., Birtles, A., 2017. The need for broader ecological and socioeconomic tools to evaluate the effectiveness of coral restoration programs. Restor. Ecol. 25, 873–883. doi:10.1111/rec.12580

Hennige, S., Smith, D.J., Walsh, S.-J., McGinley, M.P., Warner, M.E., Suggett, D.J., 2010. Acclimation and adaptation of scleractinian coral communities along environmental gradients within an Indonesian reef system. J. Exp. Mar. Bio. Ecol. 391, 143–152. doi:10.1016/j.jembe.2010.06.019

171

Hennige, S.J., Smith, D.J., Perkins, R., Consalvey, M., Paterson, D.M., Suggett, D.J., 2008. Photoacclimation, growth and distribution of massive coral species in clear and turbid waters. Mar. Ecol. Prog. Ser. 369, 77–88. doi:10.3354/meps07612

Herlan, J., Lirman, D., 2008. Development of a coral nursery program for the threatened coral Acropora cervicornis in Florida, in: Riegl, B., Dodge, R.E. (Eds.), Proceedings of the 11th International Coral Reef Symposium. Nova Southeastern University National Coral Reef Institute, Davie, pp. 1244–1247.

Hicks, C.C., Graham, N.A.J., Cinner, J.E., 2013. Synergies and tradeoffs in how managers, scientists, and fishers value coral reef ecosystem services. Glob. Environ. Chang. 23, 1444–1453. doi:10.1016/j.gloenvcha.2013.07.028

Highsmith, R.C., 1982. Reproduction by fragmentation in corals. Mar. Ecol. Prog. Ser. 7, 207–226. doi:10.3354/meps007207

Hillyer, K.E., Dias, D., Lutz, A., Roessner, U., Davy, S.K., 2018. 13C metabolomics reveals widespread change in carbon fate during coral bleaching. Metabolomics 14, 1–9. doi:10.1007/s11306-017-1306-8

Hillyer, K.E., Dias, D.A., Lutz, A., Wilkinson, S.P., Roessner, U., Davy, S.K., 2017. Metabolite profiling of symbiont and host during thermal stress and bleaching in the coral Acropora aspera. Coral Reefs 36, 105–118. doi:10.1007/s00338-016- 1508-y

Hillyer, K.E., Tumanov, S., Villas-Bôas, S., Davy, S.K., 2016. Metabolite profiling of symbiont and host during thermal stress and bleaching in a model cnidarian – dinoflagellate symbiosis. J. Exp. Biol. 219, 516–527. doi:10.1242/jeb.128660

Hoegh-Guldberg, O., 1999. Climate change, coral bleaching and the future of the world’s coral reefs. Mar. Freshw. Res. 50, 839–866.

Hoegh-Guldberg, O., Bruno, J.F., 2010. The impact of climate change on the world’s marine ecosystems. Science 328, 1523–1528. doi:10.1126/science.1189930

Hoffmann, A., Sgrò, C., 2011. Climate change and evolutionary adaptation. Nature 470, 479–485. doi:10.1038/nature09670

Holbrook, S.J., Brooks, A.J., Schmitt, R.J., 2002. Predictability of fish assemblages on coral patch reefs. Mar. Freshw. Res. 53, 181–188. doi:10.1071/MF01137

Holbrook, S.J., Brooks, A.J., Schmitt, R.J., 2002. Variation in structural attributes of patch-forming corals and in patterns of abundance of associated fishes. Mar. Freshw. Res. 53, 1045–1053.

Holbrook, S.J., Schmitt, R.J., Brooks, A.J., 2008. Resistance and resilience of a coral reef fish community to changes in coral cover. Mar. Ecol. Prog. Ser. 371, 263– 271. doi:10.3354/meps07690

172

Holland, J.B., Nyquist, W.E., Cervantes-Martínez, C.T., 2003. Estimating and interpreting heritability for plant breeding: an update, in: Janick, J. (Ed.) Plant Breeding Reviews, Volume 22. John Wiley & Sons, Inc., Oxford, pp. 9-112.

Holland, G., Bruyère, C.L., 2014. Recent intense hurricane response to global climate change. Clim. Dyn. 42, 617–627. doi:10.1007/s00382-013-1713-0

Hollarsmith, J.A., Griffin, S.P., Moore, T.D., 2012. Success of outplanted Acropora cervicornis colonies in reef restoration, in: Yellowlees, D., Hughes, T.P. (Eds.), Proceedings of the 12th International Coral Reef Symposium. ARC Centre of Excellence for Coral Reef Studies, Townsville.

Hollywood, K., Brison, D.R., Goodacre, R., 2006. Metabolomics: Current technologies and future trends. Proteomics 6, 4716–4723. doi:10.1002/pmic.200600106

Hoogenboom, M.O., Anthony, K.R.N., Connolly, S.R., 2006. Energetic cost of photoinhibition in corals. Mar. Ecol. Prog. Ser. 313, 1–12. doi:10.3354/meps313001

Horoszowski-Fridman, Y.B., Brêthes, J.C., Rahmani, N., Rinkevich, B., 2015. Marine silviculture: Incorporating ecosystem engineering properties into reef restoration acts. Ecol. Eng. 82, 201–213. doi:10.1016/j.ecoleng.2015.04.104

Hughes, T.P., 1994. Catastrophes, phase shifts and large-scale degradation of a Caribbean coral reef. Science 265, 1547–1551.

Hughes, T.P., Connell, J.H., 1997. Multiple stressors on coral reefs: a long-term perspective. Limnol. Oceanogr. 44, 932–940. doi:10.4319/lo.1999.44.3_part_2.0932

Hughes, A.R., Stachowicz, J.J., 2004. Genetic diversity enhances the resistance of a seagrass ecosystem to disturbance. Proc. Natl. Acad. Sci. 101, 8998–9002. doi:10.1073/pnas.0402642101

Hughes, T.P., Barnes, M.L., Bellwood, D.R., Cinner, J.E., Cumming, G.S., Jackson, J.B.C., Kleypas, J., van de Leemput, I.A., Lough, J.M., Morrison, T.H., Palumbi, S.R., van Nes, E.H., Scheffer, M., 2017a. Coral reefs in the Anthropocene. Nature 546, 82–90. doi:10.1038/nature22901

Hughes, T.P., Kerry, J.T., Álvarez-Noriega, M., Álvarez-Romero, J.G., Anderson, K.D., Baird, A.H., Babcock, R.C., Beger, M., Bellwood, D.R., Berkelmans, R., Bridge, T.C., Butler, I.R., Byrne, M., Cantin, N.E., Comeau, S., Connolly, S.R., Cumming, G.S., Dalton, S.J., Diaz-Pulido, G., Eakin, C.M., Figueira, W.F., Gilmour, J.P., Harrison, H.B., Heron, S.F., Hoey, A.S., Hobbs, J.-P.A., Hoogenboom, M.O., Kennedy, E. V., Kuo, C., Lough, J.M., Lowe, R.J., Liu, G., McCulloch, M.T., Malcolm, H.A., McWilliam, M.J., Pandolfi, J.M., Pears, R.J., Pratchett, M.S., Schoepf, V., Simpson, T., Skirving, W.J., Sommer, B., Torda, G., Wachenfeld,

173

D.R., Willis, B.L., Wilson, S.K., 2017b. Global warming and recurrent mass bleaching of corals. Nature 543, 373–377. doi:10.1038/nature21707

Hughes, T.P., Kerry, J.T., Baird, A.H., Connolly, S.R., Dietzel, A., Eakin, C.M., Heron, S.F., Hoey, A.S., Hoogenboom, M.O., Liu, G., McWilliam, M.J., Pears, R.J., Pratchett, M.S., Skirving, W.J., Stella, J.S., Torda, G., 2018. Global warming transforms coral reef assemblages. Nature 1–5. doi:10.1038/s41586-018-0041-2

Hunt, J., Sharp, W., 2014. Developing a Comprehensive Strategy for Coral Restoration for Florida. State Wildlife Grant Award T-32-R 1169: Final Report. Florida Fish and Wildlife Conservation Commission, Tallhassee.

Huntington, B.E., Miller M.W., Pausch, R., Richter, L., 2017. Facilitation in Caribbean coral reefs: High densities of staghorn coral foster greater coral condition and reef fish composition. Oecologica 184, 247–257. doi:10.1007/s00442-017-3859-7

Huot, B., Yao, J., Montgomery, B.L., He, S.Y., 2014. Growth-defense tradeoffs in plants: A balancing act to optimize fitness. Mol. Plant 7, 1267–1287. doi:10.1093/mp/ssu049

Ibba, M., Soll, D., 2000. Aminoacyl-tRNA synthesis. Annu. Rev. Biochem. 69, 617–150. doi:10.1146/annurev.biochem.69.1.617

Iglesias-Prieto, R., Beltran, V.H., LaJeunesse, T.C., Reyes-Bonilla, H., Thome, P.E., 2004. Different algal symbionts explain the vertical distribution of dominant reef corals in the eastern Pacific. Proc. R. Soc. B Biol. Sci. 271, 1757–1763. doi:10.1098/rspb.2004.2757

Jackson, J.B.C., Kirby, M.X., Berger, W.H., Bjorndal, K.A., Botsford, L.W., Bourque, B.J., Bradbury, R.H., Cooke, R., Erlandson, J., Estes, J.A., Hughes, T.P., Kidwell, S., Lange, C.B., Lenihan, H.S., Pandolfi, M., Peterson, C.H., Steneck, R.S., Tegner, M.J., Warner, R.R., Pandolfi, J.M., 2001. Historical overfishing and the recent collapse of coastal ecosystems. Science 293, 629–638. doi:10.1126/science.1059199

Jaenisch, R., Bird, A., 2003. Epigenetic regulation of gene expression: How the genome integrates intrinsic and environmental signals. Nat. Genet. 33, 245–254. doi:10.1038/ng1089

Jänkänpää, H.J., Mishra, Y., Schröder, W.P., Jansson, S., 2012. Metabolic profiling reveals metabolic shifts in Arabidopsis plants grown under different light conditions. Plant, Cell Environ. 35, 1824–1836. doi:10.1111/j.1365- 3040.2012.02519.x

Jeans, J., Campbell, D.A., Hoogenboom, M.O., 2013. Increased reliance upon photosystem II repair following acclimation to high-light by coral-dinoflagellate symbioses. Photosynth Res 118, 219–229. doi:10.1007/s11120-013-9918-y

174

Johnson, M., Lustic, C., Bartels, E., Baums, I., Gilliam, D., Larson, L., Lirman, D., Miller, M., Nedimyer, K., Schopmeyer, S., 2011a. Caribbean Acropora Restoration Guide: Best Practices for Propagation and Population Enhancement. The Nature Conservancy, Arlington.

Jokiel, P., Maragos, J., Franzisket, L., 1978. Coral growth : Buoyant weight technique, in: Stoddart, D., Johannes, R. (Eds.), Coral Reefs: Research Methods. UNESCO, Paris, pp. 529-541.

Jones, A., Berkelmans, R., 2010. Potential costs of acclimatization to a warmer climate: Growth of a reef coral with heat tolerant vs. sensitive symbiont types. PLoS One 5, e10437. doi:10.1371/journal.pone.0010437

Kaniewska, P., Anthony, K.R.N., Hoegh-Guldberg, O., 2008. Variation in colony genometry modulates internal light levels in branching corals, Acropora humilis and Stylophora pistillata. Mar Biol 155, 649–660

Kashman, Y., Groweiss, A., Carmely, S., Kinamoni, Z., Czarkie, D., Rotem, M., 1982. Recent research in marine natural products from the Red Sea. Pure Appl. Chem. 54, 1995–2010. doi:10.1351/pac198254101995

Kendall, J.J., Powell, E.N., Connor, S.J., Bright, T.J., Zastrow, C.E., 1985. Effects of turbidity on calcification rate, protein concentration and the free amino acid pool of the coral Acropora cervicornis. Mar. Biol. 87, 33–46. doi:10.1007/BF00397003

Kenkel, C.D., Goodbody-Gringley, G., Caillaud, D., Davies, S.W., Bartels, E., Matz, M. V., 2013. Evidence for a host role in thermotolerance divergence between populations of the mustard hill coral (Porites astreoides) from different reef environments. Mol. Ecol. 22, 4335–4348. doi:10.1111/mec.12391

Keurentjes, J.J., 2009. Genetical metabolomics: Closing in on phenotypes. Curr. Opin. Plant Biol. 12, 223–230. doi:10.1016/j.pbi.2008.12.003

Kiel, C., Huntington, B.E., Miller, M.W., 2012. Tractable field metrics for restoration and recovery monitoring of staghorn coral Acropora cervicornis. Endanger. Species Res. 19, 171–176. doi:10.3354/esr00474

Klueter, A., Crandall, J.B., Archer, F.I., Teece, M.A., Coffroth, M.A., 2015. Taxonomic and environmental variation of metabolite profiles in marine dinoflagellates of the genus Symbiodinium. Metabolites 5, 74–99. doi:10.3390/metabo5010074

Knowlton, N., Lang, J.C., Christine Rooney, M., Clifford, P., 1981. Evidence for delayed mortality in hurricane-damaged Jamaican staghorn corals. Nature 294, 251–252. doi:10.1038/294251a0

Knowlton, N., Lang, J.C., Keller, B.D., 1990. Case study of natural population collapse: post-hurricane predation on Jamaican staghorn corals. Smithson. Contrib. Mar. Sci. 31, 1–25. doi:10.5479/si.01960768.31.1

175

Knowlton, N., Jackson, J.B.C., 2008. Shifting baselines, local impacts, and global change on coral reefs. PLoS Biol. 6, e54. doi:10.1371/journal.pbio.0060054

Knutson, T.R., McBride, J.L., Chan, J., Emanuel, K., Holland, G., Landsea, C., Held, I., Kossin, J.P., Srivastava, A.K., Sugi, M., 2010. Tropical cyclones and climate change. Nat. Geosci 3, 157–163. doi:10.1038/ngeo779

Kroon, F.J., Kuhnert, P.M., Henderson, B.L., Wilkinson, S.N., Kinsey-Henderson, A., Abbott, B., Brodie, J.E., Turner, R.D.R., 2012. River loads of suspended solids, nitrogen, phosphorus and herbicides delivered to the Great Barrier Reef lagoon. Mar. Pollut. Bull. 65, 167–181. doi:10.1016/j.marpolbul.2011.10.018

Kuffner, I.B., Bartels, E., Stathakopoulos, A., Enochs, I.C., Kolodziej, G., Toth, L.T., Manzello, D.P., 2017. Plasticity in skeletal characteristics of nursery-raised staghorn coral, Acropora cervicornis. Coral Reefs 36, 679–684. doi:10.1007/s00338-017-1560-2

Ladd, M., Shantz, A., Bartels, E., Burkepile, D., 2017. Thermal stress reveals a genotype-specific tradeoff between growth and tissue loss in restored Acropora cervicornis. Mar. Ecol. Prog. Ser. 572, 129–139. doi:10.3354/meps12169

Ladd, M.C., Shantz, A.A., Nedimyer, K., Burkepile, D.E., 2016. Density dependence drives habitat production and survivorship of Acropora cervicornis used for restoration on a Caribbean coral reef. Front. Mar. Sci. 3, 261. doi:10.3389/fmars.2016.00261

LaJeunesse, T.C., Parkinson, J.E., Gabrielson, P.W., Jeong, H.J., Reimer, J.D., Voolstra, C.R., Santos, S.R., 2018. Systematic revision of Symbiodiniaceae highlights the antiquity and diversity of coral endosymbionts. Curr. Biol. 28, 2570–2580. doi:10.1016/j.cub.2018.07.008

Lang, J.C., Marks, K.W., Kramer, P.A., Kramer, P.R., Ginsburg, R.N., 2010. Atlantic and Gulf Rapid Reef Assessment (AGRRA) Protocols Version 5.4. AGRRA, Miami.

Langlois, L.A., Hoogenboom, M.O., 2014. Capacity for short-term physiological acclimation to light does not control the lower depth distributions of branching corals. Mar. Ecol. Prog. Ser. 508, 149–162.

Lee, D.-K., Yoon, M.H., Kang, Y.P., Yu, J., Park, J.H., Lee, J., Kwon, S.W., 2013. Comparison of primary and secondary metabolites for suitability to discriminate the origins of Schisandra chinensis by GC/MS and LC/MS. Food Chem. 141, 3931–3937. doi:10.1016/J.FOODCHEM.2013.06.064

Lehnert, E.M., Mouchka, M.E., Burriesci, M.S., Gallo, N.D., Schwarz, J.A., Pringle, J.R., 2014. Extensive differences in gene expression between symbiotic and aposymbiotic cnidarians. G3 Genes, Genomes, Genet. 4, 277–295. doi:10.1534/g3.113.009084

176

Lesser, M.P., Shick, J.M., 1989. Effects of irradiance and ultraviolet radiation on photoadaptation in the zooxanthellae of Aiptasia pallida: primary production, photoinhibition, and enzymic defenses against oxygen toxicity. Mar. Biol. 102, 243–255. doi:10.1007/BF00428286

Lesser, M.P., Bythell, J.C., Gates, R.D., Johnstone, R.W., Hoegh-Guldberg, O., 2007. Are infectious diseases really killing corals? Alternative interpretations of the experimental and ecological data. J. Exp. Mar. Bio. Ecol. 346, 36–44. doi:10.1016/j.jembe.2007.02.015

Lesser, M.P., Marc, S., Michael, S., Michiko, O., Gates, R.D., Andrea, G., 2010. Photoacclimatization by the coral Montastraea cavernosa in the mesophotic zone: Light, food, and genetics. Ecology 91, 990–1003. doi:10.1890/09-0313.1

Lessios, H., 1988. Mass mortality of Diadema antillarum in the Caribbean: What have we learned? Annu. Rev. Ecol. Syst. 19, 371–393.

Lessios, H., Robertson, D., Cubit, J., 1984. Spread of Diadema mass mortality through the Caribbean. Science 226, 335–337.

Li, S., Park, Y., Duraisingham, S., Strobel, F.H., Khan, N., Soltow, Q.A., Jones, D.P., Pulendran, B., 2013. Predicting network activity from high throughput metabolomics. PLoS Comput. Biol. 9, e1003123. doi:10.1371/journal.pcbi.1003123

Libro, S., Vollmer, S. V., 2016. Genetic signature of resistance to white band disease in the Caribbean staghorn coral Acropora cervicornis. PLoS One 11, e0146636. doi:10.1371/journal.pone.0146636

Lidz, B.H., Zawada, D.G., 2013. Possible return of Acropora cervicornis at Pulaski Shoal, Dry Tortugas National Park, Florida. J. Coast. Res. 287, 256–271. doi:10.2112/JCOASTRES-D-12-00078.1

Lirman, D., 1999. Reef fish communities associated with Acropora palmata: Relationships to benthic attributes. Bull. Mar. Sci. 65, 235–252.

Lirman, D., Fong, P., 2007. Is proximity to land-based sources of coral stressors an appropriate measure of risk to coral reefs? An example from the Florida Reef Tract. Mar. Pollut. Bull. 54, 779–791. doi:10.1016/J.MARPOLBUL.2006.12.014

Lirman, D., Bowden-Kerby, A., Schopmeyer, S., Huntington, B., Thyberg, T., Gough, M., Gough, T., Gough, R., Gough, Y., 2010. A window to the past: Documenting the status of one of the last remaining “megapopulations” of the threatened staghorn coral Acropora cervicornis in the Dominican Republic. Aquat. Conserv. Mar. Freshw. Ecosyst. 20, 773–781. doi:10.1002/aqc.1146

Lirman, D., Schopmeyer, S., Galvan, V., Drury, C., Baker, A.C., Baums, I.B., 2014. Growth dynamics of the threatened Caribbean staghorn coral Acropora

177

cervicornis: Influence of host genotype, symbiont identity, colony size, and environmental setting. PLoS One 9, e107253. doi:10.1371/journal.pone.0107253

Lirman, D., Schopmeyer, S., 2016. Ecological solutions to reef degradation: Optimizing coral reef restoration in the Caribbean and Western Atlantic. PeerJ 4, e2597. doi:10.7717/peerj.2597

Liu, J., Liu, Y., Wang, Y., Abozeid, A., Zu, Y.-G., Tang, Z.-H., 2017. The integration of GC–MS and LC–MS to assay the metabolomics profiling in Panax ginseng and Panax quinquefolius reveals a tissue- and species-specific connectivity of primary metabolites and ginsenosides accumulation. J. Pharm. Biomed. Anal. 135, 176–185. doi:10.1016/J.JPBA.2016.12.026

Logan, A., 2013. Coral reefs of the Cayman Islands, in: Sheppard, C. (Ed.), Coral Reefs of the United Kingdom Overseas Territories. Springer Netherlands, Dordrecht, pp. 61-68. doi:10.1007/978-94-007-5965-7

Logan, A., 1994. Reefs and lagoons of Cayman Brac and Little Cayman, in: Brunt, M.A., Davies, J.E. (Eds.), The Cayman Islands: Natural History and Biogeography. Springer Netherlands, Dordrecht, pp. 105–124. doi:10.1007/978-94-011-0904-8

Logan, C.A., Dunne, J.P., Eakin, C.M., Donner, S.D., 2012. A framework for comparing coral bleaching thresholds, in: Yellowlees, D., Hughes, T.P. (Eds.), Proceedings of the 12th International Coral Reef Symposium. ARC Centre of Excellence for Coral Reef Studies, Townsville.

Lohr, K., Bejarano, S., Lirman, D., Schopmeyer, S., Manfrino, C., 2015. Optimizing the productivity of a coral nursery focused on staghorn coral Acropora cervicornis. Endanger. Species Res. 27, 243–250. doi:10.3354/esr00667

Lohr, K.E., Cook McNab, A.A., Manfrino, C., Patterson, J.T., 2017. Assessment of wild and restored Staghorn coral Acropora cervicornis in three reef zones at Little Cayman, Cayman Islands. Reg. Stud. Mar. Sci. 9, 1–8. doi:10.1016/j.rsma.2016.11.003

Lohr, K.E., Patterson, J.T., 2017. Intraspecific variation in phenotype among nursery- reared staghorn coral Acropora cervicornis (Lamarck, 1816). J. Exp. Mar. Bio. Ecol. 486, 87–92. doi:10.1016/j.jembe.2016.10.005

Loya, Y., Sakai, K., Yamazato, K., Nakano, Y., Sambali, H., van Woesik, R., 2001. Coral bleaching: The winners and the losers. Ecol. Lett. 4, 122–131. doi:10.1046/j.1461-0248.2001.00203.x

Lucas, M.Q., Weil, E., 2015. Recent recovery in Acropora cervicornis and abundance of A. prolifera off La Parguera, Puerto Rico. Mar. Biodivers., 1–2. doi:10.1007/s12526-015-0399-4

178

Mangiafico, S.S., 2016. Summary and analysis of extension program evaluation in R, version 1.15.0.

Manfrino, C., Riegl, B., Hall, J.L., Graifman, R., 2003. Status of Coral Reefs of Little Cayman, Grand Cayman, and Cayman Brac, British West Indies, in 1999 and 2000 (Part 1: Stony Corals and Algae), in: Lang, J.C. (Ed.), Status of Coral Reefs in the Western Atlantic: Results of Initial Surveys, Atlantic & Gulf Rapid Reef Assessment (AGRRA). Atoll Research Bulletin 496, 204–225. doi:10.5479/si.00775630.496-12.226

Manfrino, C., Jacoby, C.A., Camp, E., Frazer, T.K., 2013. A positive trajectory for corals at Little Cayman Island. PLoS One 8, e75432. doi:10.1371/journal.pone.0075432

Markley, J.L., Brüschweiler, R., Edison, A.S., Eghbalnia, H.R., Powers, R., Raftery, D., Wishart, D.S., 2017. The future of NMR-based metabolomics. Curr. Opin. Biotechnol. 43, 34–40. doi:10.1016/j.copbio.2016.08.001

Marsh, A.G., Hoadley, K.D., Warner, M.E., 2016. Distribution of CpG motifs in upstream gene domains in a reef coral and sea anemone: Implications for epigenetics in cnidarians. PLoS One 11, e0150840. doi:10.1371/journal.pone.0150840

Marty-Rivera, M., Roberson, L.M., Yudowski, G.A., 2018. Mitigation of coral bleaching by antioxidants. Biology Preprint (non-peer-reviewed).

Matsuda, F., Okazaki, Y., Oikawa, A., Kusano, M., Nakabayashi, R., Kikuchi, J., Yonemaru, J.I., Ebana, K., Yano, M., Saito, K., 2012. Dissection of genotype- phenotype associations in rice grains using metabolome quantitative trait loci analysis. Plant J. 70, 624–636. doi:10.1111/j.1365-313X.2012.04903.x

Matthews, J.L., Crowder, C.M., Oakley, C.A., Lutz, A., Roessner, U., Meyer, E., Grossman, A.R., Weis, V.M., Davy, S.K., 2017. Optimal nutrient exchange and immune responses operate in partner specificity in the cnidarian-dinoflagellate symbiosis. Proc. Natl. Acad. Sci. U. S. A. 114, 13194–13199. doi:10.1073/pnas.1710733114

Mayfield, A.B., Chan, P.-H., Putnam, H.M., Chen, C.-S., Fan, T.-Y., 2012. The effects of a variable temperature regime on the physiology of the reef-building coral Seriatopora hystrix: results from a laboratory-based reciprocal transplant. J. Exp. Biol. 215, 4183–4195. doi:10.1242/jeb.071688

Maynard, J.A., Anthony, K.R.N., Marshall, P.A., Masiri, I., 2008. Major bleaching events can lead to increased thermal tolerance in corals. Mar. Biol. 155, 173–182. doi:10.1007/s00227-008-1015-y

Maynard, J., van Hooidonk, R., Eakin, C.M., Puotinen, M., Garren, M., Williams, G., Heron, S.F., Lamb, J., Weil, E., Willis, B., Harvell, C.D., 2015. Projections of climate conditions that increase coral disease susceptibility and pathogen

179

abundance and virulence. Nat. Clim. Chang. 5, 688–694. doi:10.1038/nclimate2625

McManus, J.W., 1997. Tropical marine fisheries and the future of coral reefs: A brief review with emphasis on Southeast Asia. Coral Reefs 16, S121–S127. doi:10.1007/s003380050248

Meesters, E.H.W.G., Smith, S.R., Becking, L.E., 2015. A review of coral restoration techniques. Report number C028/14. Institute for Marine Resources and Ecosystem Studies, Wageningen University and Research Centre, Wageningen.

Meissner, S., Steinhauser, D., Dittmann, E., 2015. Metabolomic analysis indicates a pivotal role of the hepatotoxin microcystin in high light adaptation of Microcystis. Environ. Microbiol. 17, 1497–1509. doi:10.1111/1462-2920.12565

Mercado-Molina, A.E., Ruiz-Diaz, C.P., Sabat, A.M., 2014. Survival, growth, and branch production of unattached fragments of the threatened hermatypic coral Acropora cervicornis. J. Exp. Mar. Bio. Ecol. 457, 215–219. doi:10.1016/j.jembe.2014.04.017

Mercado-Molina, A.E., Ruiz-Diaz, C.P., Sabat, A.M., 2015. Demographics and dynamics of two restored populations of the threatened reef-building coral Acropora cervicornis. J. Nat. Conserv. 24, 17–23. doi:10.1016/j.jnc.2015.01.001

Mercado-Molina, A.E., Ruiz-Diaz, C.P., Sabat, A.M., 2016. Branching dynamics of transplanted colonies of the threatened coral Acropora cervicornis: Morphogenesis, complexity, and modeling. J. Exp. Mar. Bio. Ecol. 482, 134–141. doi:10.1016/j.jembe.2016.05.004

Meyer, E., Davies, S., Wang, S., Willis, B.L., Abrego, D., Juenger, T.E., Matz, M. V., 2009. Genetic variation in responses to a settlement cue and elevated temperature in the reef-building coral Acropora millepora. Mar. Ecol. Prog. Ser. 392, 81–92. doi:10.3354/meps08208

Miller, S.L., Chiappone, M., Rutten, L.., Swanson, D.W., 2008. Population status of Acropora corals in the Florida Keys, in: Riegl, B., Dodge, R.E. (Eds.), Proceedings of the 11th International Coral Reef Symposium. Nova Southeastern University National Coral Reef Institute, Davie, pp. 775–779.

Miller, M.W., Bourque, A.S., Bohnsack, J.A., 2002. An analysis of the loss of acroporid corals at Looe Key, Florida, USA: 1983-2000. Coral Reefs 21, 179–182. doi:10.1007/s00338-002-0228-7

Miller, M.W., Lohr, K.E., Cameron, C.M., Williams, D.E., Peters, E.C., 2014a. Disease dynamics and potential mitigation among restored and wild staghorn coral, Acropora cervicornis. PeerJ 2, e541. doi:10.7717/peerj.541

180

Miller, M.W., Marmet, C., Cameron, C.M., Williams, D.E., 2014b. Prevalence, consequences, and mitigation of fireworm predation on endangered staghorn coral. Mar. Ecol. Prog. Ser. 516, 187–194. doi:10.3354/meps10996

Miller, M.W., Kerr, K., Williams, D.E., 2016. Reef-scale trends in Florida Acropora spp. abundance and the effects of population enhancement. PeerJ 4, e2523. doi:10.7717/peerj.2523

Moberg, F., Folke, C., 1999. Ecological goods and services of coral reef ecosystems. Ecol. Econ. 29, 215–233. doi:10.1016/S0921-8009(99)00009-9

Muir, P.R., Wallace, C.C., Done, T., Aguirre, J.D., 2015. Limited scope for latitudinal extension of reef corals. Science 348, 1135–1138.

Muller, E.M., Rogers, C.S., Spitzack, A.S., van Woesik, R., 2008. Bleaching increases likelihood of disease on Acropora palmata (Lamarck) in Hawksnest Bay, St John, US Virgin Islands. Coral Reefs 27, 191–195. doi:10.1007/s00338-007-0310-2

Muscatine, L., 1967. Glycerol excretion by symbiotic algae from corals and Tridacna and its control by the host. Science 156, 516–519. doi:10.1126/science.142.3594.956

Muscatine, L., Cernichiari, E., 1969. Assimilation of photosynthetic products of zooxanthellae by a reef coral. Biol. Bull. 137, 506–523. doi:10.2307/1540172

Muscatine, L., Porter, J.W., 1977. Reef corals: Mutualistic symbioses adapted to nutrient-poor environments. Bioscience 27, 454–460.

Muscatine, L., Falkowski, P.G., Porter, J.W., Dubinsky, Z., 1984. Fate of photosynthetic fixed carbon in light- and shade-adapted colonies of the symbiotic coral Stylophora pistillata. Proc. R. Soc. London B Biol. Sci. 222, 181–202.

National Marine Fisheries Service, 2006. Endangered and threatened species: final listing determinations for elkhorn coral and staghorn coral. Fed. Regist. 71, 26852–26861.

National Marine Fisheries Service, 2015. Recovery Plan: Elkhorn Coral (Acropora palmata) and Staghorn Coral (A. cervicornis). United States Department of Commerce, National Oceanographic and Atmospheric Administration, Saint Petersburg.

Nedimyer, K., Gaines, K., Roach, S., 2011. Coral tree nursery©: An innovative approach to growing corals in an ocean-based field nursery. AACL Bioflux 4, 442–446.

Nitschke, M.R., Gardner, S.G., Goyen, S., Fujise, L., Camp, E.F., Ralph, P.J., Suggett, D.J., 2018. Utility of photochemical traits as diagnostics of thermal tolerance

181

amongst Great Barrier Reef corals. Front. Mar. Sci. 5, 45. doi:10.3389/fmars.2018.00045

Nyström, M., 2006. Redundancy and response diversity of functional groups: implications for the resilience of coral reefs. Ambio 35, 30–35. doi:10.1579/0044- 7447-35.1.30

Obata, T., Fernie, A.R., 2012. The use of metabolomics to dissect plant responses to abiotic stresses. Cell Mol. Life Sci. 69, 3225–3243. doi:10.1007/s00018-012- 1091-5

O’Donnell, K.E., Lohr, K.E., Bartels, E., Patterson, J.T., 2017. Evaluation of staghorn coral (Acropora cervicornis, Lamarck 1816) production techniques in an ocean- based nursery with consideration of coral genotype. J. Exp. Mar. Bio. Ecol. 487, 53–58. doi:10.1016/j.jembe.2016.11.013

Palumbi, S.R., Barshis, D.J., Traylor-Knowles, N., Bay, R.A., 2014. Mechanisms of reef coral resistance to future climate change. Science 344, 895–898. doi:10.1126/science.1251336

Pandolfi, J.M., Jackson, J.B.C., 2006. Ecological persistence interrupted in Caribbean coral reefs. Ecol. Lett. 9, 818–826. doi:10.1111/j.1461-0248.2006.00933.x

Parkinson, J.E., Bartels, E., Devlin-Durante, M.K., Lustic, C., Nedimyer, K., Schopmeyer, S., Lirman, D., LaJeunesse, T.C., Baums, I.B., 2018. Extensive transcriptional variation poses a challenge to thermal stress biomarker development for endangered corals. Mol. Ecol. 27, 1103–1119. doi:10.1111/mec.14517

Patti, G.J., Yanes, O., Siuzdak, G., 2012. Metabolomics: the apogee of the omics trilogy. Nat. Rev. Mol. Cell Biol. 13, 263–9. doi:10.1038/nrm3314

Pausch, R., Williams, D., Miller, M., 2018. Impacts of fragment genotype, habitat, and size on outplanted elkhorn coral success under thermal stress. Mar. Ecol. Prog. Ser. 592, 109–117. doi:10.3354/meps12488

Perry, C.T., Murphy, G.N., Kench, P.S., Smithers, S.G., Edinger, E.N., Steneck, R.S., Mumby, P.J., 2013. Caribbean-wide decline in carbonate production threatens coral reef growth. Nat. Commun. 4, 1402. doi:10.1038/ncomms2409

Pinzon, J.H., Kamel, B., Burge, C.A., Harvell, C.D., Medina, M., Weil, E., Mydlarz, L.D., 2015. Whole transcriptome analysis reveals changes in expression of immune- related genes during and after bleaching in a reef-building coral. R. Soc. Open Sci. 2, 140214–140214. doi:10.1098/rsos.140214

Porter, J.W., Muscatine, L., Dubinsky, Z., Falkowski, P.G., 1984. Primary production and photoadaptation in light- and shade-adapted colonies of the symbiotic coral,

182

Stylophora pistillata. Proc. R. Soc. London B Biol. Sci. 222, 161–180. doi:doi:10.1086/303379

Precht, W.F., Bruckner, A.W., Aronson, R.B., Bruckner, R.J., 2002. Endangered acroporid corals of the Caribbean. Coral Reefs 21, 41–42. doi:10.1007/s00338- 006-0086-9

Puotinen, M., Maynard, J.A., Beeden, R., Radford, B., Williams, G.J., 2016. A robust operational model for predicting where tropical cyclone waves damage coral reefs. Sci. Rep. 6, 26009. doi:10.1038/srep26009

Quinn, N.J., Kojis, B.L., 2008. The recent collapse of a rapid phase-shift reversal on a Jamaican north coast coral reef after the 2005 bleaching event. Rev. Biol. Trop. 56, 149–159. doi:10.15517/rbt.v56i0.5584

Quinn, R.A., Vermeij, M.J.A., Hartmann, A.C., Galtier d’Auriac, I., Benler, S., Haas, A., Quistad, S.D., Lim, Y.W., Little, M., Sandin, S., Smith, J.E., Dorrestein, P.C., Rohwer, F., 2016. Metabolomics of reef benthic interactions reveals a bioactive lipid involved in coral defence. Proc. Biol. Sci. 283, 20160469-. doi:10.1098/rspb.2016.0469

R Core Team, 2014. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna.

Ralph, P., Gademann, R., Larkum, A., Kühl, M., 2002. Spatial heterogeneity in active chlorophyll fluorescence and PSII activity of coral tissues. Mar. Biol. 141, 639– 646. doi:10.1007/s00227-002-0866-x

Randall, C.J., van Woesik, R., 2015. Contemporary white-band disease in Caribbean corals driven by climate change. Nat. Clim. Chang. 5, 375–379. doi:10.1038/nclimate2530

Ravanbakhsh, S., Liu, P., Bjordahl, T.C., Mandal, R., Grant, J.R., Wilson, M., Eisner, R., Sinelnikov, I., Hu, X., Luchinat, C., Greiner, R., Wishart, D.S., 2015. Accurate, fully-automated NMR spectral profiling for metabolomics. PLoS One 10, e0124219. doi:10.1371/journal.pone.0124219

Raymundo, L.J., 2001. Mediation of growth by conspecific neighbors and the effect of site in transplanted fragments of the coral Porites attenuata Nemenzo in the Central Philippines. Coral Reefs 20, 263–272. doi:10.1007/s003380100170

Reaka-Kudla, M.L., 2005. Biodiversity of Caribbean Coral Reefs, in: Miloslavich, P., Klein, E. (Eds.), Caribbean Marine Biodiversity: The Known and the Unknown. DEStech Publications, Inc., Lancaster, pp. 259–276.

Renegar, D., Riegl, B., 2005. Effect of nutrient enrichment and elevated CO2 partial pressure on growth rate of Atlantic scleractinian coral Acropora cervicornis. Mar. Ecol. Prog. Ser. 293, 69–76. doi:10.3354/meps293069

183

Richards, Z.T., Delbeek, J.T., Lovell, E.R., Bass, D., Aeby, G., Reboton, C., 2014. Acropora formosa. IUCN Red List of Threatened Species. doi:10.2305/IUCN.UK.2014-1.RLTS.T133644A54299282.en

Richards, Z.T., Delbeek, J.T., Lovell, E.R., Bass, D., Aeby, G., Reboton, C., 2008. Acropora cytherea [WWW Document]. IUCN Red List Threat. Species. doi:10.2305/IUCN.UK.2008.RLTS.T133313A3684008.en

Riedelsheimer, C., Czedik-Eysenberg, A., Grieder, C., Lisec, J., Technow, F., Sulpice, R., Altmann, T., Stitt, M., Willmitzer, L., Melchinger, A.E., 2012. Genomic and metabolic prediction of complex heterotic traits in hybrid maize. Nat. Genet. 44, 217–220. doi:10.1038/ng.1033

Rinkevich, B., 1995. Restoration strategies for coral reefs damaged by recreational activities: The use of sexual and asexual recruits. Restor. Ecol. 3, 241–251. doi:10.1111/j.1526-100X.1995.tb00091.x

Rinkevich, B., 2014. Rebuilding coral reefs: Does active reef restoration lead to sustainable reefs? Curr. Opin. Environ. Sustain. 7, 28–36. doi:10.1016/j.cosust.2013.11.018

Rinkevich, B., 2015. Climate Change and Active Reef Restoration—Ways of Constructing the “Reefs of Tomorrow.” J. Mar. Sci. Eng. 3, 111–127. doi:10.3390/jmse3010111

Roff, G., Mumby, P.J., 2012. Global disparity in the resilience of coral reefs. Trends Ecol. Evol. 27, 404–413. doi:10.1016/j.tree.2012.04.007

Rogers, C.S., 1979. The effect of shading on coral reef structure and function. J. Exp. Mar. Bio. Ecol. 41, 269–288.

Rogers, C.S., 1990. Responses of coral reefs and reef organisms to sedimentation. Mar. Ecol. Prog. Ser. 62, 185–202. doi:10.3354/meps062185

Rogers, A., Harborne, A.R., Brown, C.J., Bozec, Y., Castro, C., Chollett, I., Hock, K., Knowland, C.A., Marshell, A., Ortiz, J.C., Razak, T., Roff, G., Samper-Villarreal, J., Saunders, M.I., Wolf, N.H., Mumby, P.J., 2015. Anticipative management for coral reef ecosystem services in the 21st century. Glob. Chang. Biol. 21, 504– 514. doi:10.1111/gcb.12725

Rohwer, F., Seguritan, V., Azam, F., Knowlton, N., 2002. Diversity and distribution of coral-associated bacteria. Mar. Ecol. Prog. Ser. 243, 1–10. doi:10.3354/meps243001

Rosenberg, E., Koren, O., Reshef, L., Efrony, R., Zilber-Rosenberg, I., 2007. The role of microorganisms in coral health, disease and evolution. Nat. Rev. Microbiol. 5, 355–362. doi:10.1038/nrmicro1635

184

Ross, A.M., 2014. Genet and reef position effects in out-planting of nursery-grown Acropora cervicornis (Scleractinia: Acroporidae) in Montego Bay, Jamaica. Rev. Biol. Trop. 62, 95–106.

Roth, M.S., Deheyn, D.D., 2013. Effects of cold stress and heat stress on coral fluorescence in reef-building corals. Sci. Rep. 3, 1–8. doi:10.1038/srep01421

Roth, M.S., 2014. The engine of the reef: photobiology of the coral-algal symbiosis. Front. Microbiol. 5, 1–22. doi:10.3389/fmicb.2014.00422

Schauer, N., Semel, Y., Roessner, U., Gur, A., Balbo, I., Carrari, F., Pleban, T., Perez- Melis, A., Bruedigam, C., Kopka, J., Willmitzer, L., Zamir, D., Fernie, A.R., 2006. Comprehensive metabolic profiling and phenotyping of interspecific introgression lines for tomato improvement. Nat. Biotechnol. 24, 447–454. doi:10.1038/nbt1192

Schmitt, F., Sluka, D., Sullivan, M., 2002. Evaluating the use of roving diver and transect surveys to assess the coral reef fish assemblage off southeastern Hispaniola. Coral Reefs 21, 216–223. doi:10.1007/s00338-002-0216-y

Schopmeyer, S.A., Lirman, D., 2015. Occupation dynamics and impacts of damselfish territoriality on recovering populations of the threatened staghorn coral, Acropora cervicornis. PLoS One 10, e0141302. doi:10.1371/journal.pone.0141302

Schopmeyer, S.A., Lirman, D., Bartels, E., Gilliam, D.S., Goergen, E.A., Griffin, S.P., Johnson, M.E., Lustic, C., Maxwell, K., Walter, C.S., 2017. Regional restoration benchmarks for Acropora cervicornis. Coral Reefs 36, 1047–1057.

Schutte, V.G.W., Selig, E.R., Bruno, J.F., 2010. Regional spatio-temporal trends in Caribbean coral reef benthic communities. Mar. Ecol. Prog. Ser. 402, 115–122. doi:10.3354/meps08438

Schutter, M., van der Ven, R.M., Janse, M., Verreth, J. a. J., Wijffels, R.H., Osinga, R., 2011. Light intensity, photoperiod duration, daily light flux and coral growth of Galaxea fascicularis in an aquarium setting: a matter of photons? J. Mar. Biol. Assoc. United Kingdom 92, 703–712. doi:10.1017/S0025315411000920

Schweinsberg, M., Weiss, L.C., Striewski, S., Tollrian, R., Lampert, K.P., 2015. More than one genotype: how common is intracolonial genetic variability in scleractinian corals? Mol. Ecol. 24, 2673–2685. doi:10.1111/mec.13200

Schymanski, E.L., Jeon, J., Gulde, R., Fenner, K., Ruff, M., Singer, H.P., Hollender, J., 2014. Identifying small molecules via high resolution mass spectrometry: Communicating confidence. Environ. Sci. Technol. 48, 2097–2098. doi:10.1021/es5002105

Scott, P.J.B., 1990. Chronic pollution recorded in coral skeletons in Hong Kong. J. Exp. Mar. Bio. Ecol. 139, 51–64. doi:10.1016/0022-0981(90)90038-E

185

Seibel, B.A., Walsh, P.J., 2002. Trimethylamine oxide accumulation in marine animals: relationship to acylglycerol storage. J. Exp. Biol. 205, 297–306. doi:10.1016/0305-0491(95)00108-5

Sha, W., da Costa, K.-A., Fischer, L.M., Milburn, M. V, Lawton, K. a, Berger, A., Jia, W., Zeisel, S.H., 2010. Metabolomic profiling can predict which humans will develop liver dysfunction when deprived of dietary choline. FASEB J. 24, 2962–2975. doi:10.1096/fj.09-154054

Sharp, W.C., Delgado, G.A., Hart, J.E., Hunt, J.H., 2018. Comparing the behavior and morphology of wild-collected and hatchery-propagated long-spined urchins (Diadema antillarum): Implications for coral reef ecosystem restoration. Bull. Mar. Sci. 94, 103–122. doi:10.5343/bms.2017.1068

Shaver, E.C., Silliman, B.R., 2017. Time to cash in on positive interactions for coral restoration. PeerJ 5, e3499. doi:10.7717/peerj.3499

Shearer, T.L., Porto, I., Zubillaga, A.L., 2009. Restoration of coral populations in light of genetic diversity estimates. Coral Reefs 28, 727–733. doi:10.1007/s00338-009- 0520-x

Shinn, E.A., 1976. Coral reef recovery in Florida and the Persian Gulf. Environ. Geol. 1, 241–254.

Shinzato, C., Shoguchi, E., Kawashima, T., Hamada, M., Hisata, K., Tanaka, M., Fujie, M., Fujiwara, M., Koyanagi, R., Ikuta, T., Fujiyama, A., Miller, D.J., Satoh, N., 2011. Using the Acropora digitifera genome to understand coral responses to environmental change. Nature 476, 320–323. doi:10.1038/nature10249

Silverstein, R.N., Cunning, R., Baker, A.C., 2014. Change in algal symbiont communities after bleaching, not prior heat exposure, increases heat tolerance of reef corals. Glob. Chang. Biol. 21, 236–249. doi:10.1111/gcb.12706

Smith, L.W., Barshis, D., Birkeland, C., 2007. Phenotypic plasticity for skeletal growth, density and calcification of Porites lobata in response to habitat type. Coral Reefs 26, 559–567. doi:10.1007/s00338-007-0216-z

Sogin, E.M., Anderson, P., Williams, P., Chen, C.S., Gates, R.D., 2014. Application of 1H-NMR metabolomic profiling for reef-building corals. PLoS One 9, e111274. doi:10.1371/journal.pone.0111274

Sogin, E.M., Putnam, H.M., Anderson, P.E., Gates, R.D., 2016. Metabolomic signatures of increases in temperature and ocean acidification from the reef-building coral, Pocillopora damicornis. Metabolomics 12, 1–12. doi:10.1007/s11306-016-0987-8

Sogin, E.M., Putnam, H.M., Nelson, C.E., Anderson, P., Gates, R.D., 2017. Correspondence of coral holobiont metabolome with symbiotic bacteria, archaea

186

and Symbiodinium communities. Environ. Microbiol. Rep. 9, 310–315. doi:10.1111/1758-2229.12541

Sokal, R.R., Rohlf, F.J., 1995. Biometry, 3rd ed. W.H. Freeman, New York

Steinfath, M., Strehmel, N., Peters, R., Schauer, N., Groth, D., Hummel, J., Steup, M., Selbig, J., Kopka, J., Geigenberger, P., van Dongen, J.T., 2010. Discovering plant metabolic biomarkers for phenotype prediction using an untargeted approach. Plant Biotechnol. J. 8, 900–911. doi:10.1111/j.1467- 7652.2010.00516.x

Stewart, D., McDougall, G.J., Sungurtas, J., Verrall, S., Graham, J., Martinussen, I., 2007. Metabolomic approach to identifying bioactive compounds in berries: Advances toward fruit nutritional enhancement. Mol. Nutr. Food Res. 51, 645– 651. doi:10.1002/mnfr.200700056

Stoddart, D.R., 1962. Catastrophic storm effects on the British Honduras reefs and cays. Nature 196, 512–515.

Suding, K.N., Gross, K.L., Houseman, G.R., 2004. Alternative states and positive feedbacks in restoration ecology Trends Ecol. Evol. 19, 46–53. doi:10.1016/j.tree.2003.10.005

Suescún-Bolívar, L.P., Iglesias-Prieto, R., Thomé, P.E., 2012. Induction of glycerol synthesis and release in cultured Symbiodinium. PLoS One 7, e47182. doi:10.1371/journal.pone.0047182

Suggett, D.J., Hall-Spencer, J.M., Rodolfo-Metalpa, R., Boatman, T.G., Payton, R., Tye Pettay, D., Johnson, V.R., Warner, M.E., Lawson, T., 2012. Sea anemones may thrive in a high CO2 world. Glob. Chang. Biol. 18, 3015–3025. doi:10.1111/j.1365-2486.2012.02767.x

Sutherland, K.P., Shaban, S., Joyner, J.L., Porter, J.W., Lipp, E.K., 2011. Human pathogen shown to cause disease in the threatened eklhorn coral Acropora palmata. PLoS One 6, e23468. doi:10.1371/journal.pone.0023468

Sweet, M., Ramsey, A., Bulling, M., 2017. Designer reefs and coral probiotics: Great concepts but are they good practice? Biodiversity 18, 19–22. doi:10.1080/14888386.2017.1307786

Teng, Q., Huant, W., Collette, T.W., Ekman, D.R., Tan, C., 2009. A direct cell quenching method for cell-culture based metabolomics. Metabolomics 5, 199–208. doi:10.1007/s11306-008-0137-z

Therneau, T.M., 2015. survival: Survival analysis. R package version 2.38-3.

187

Titlyanov, E.A., Titlyanova, T.V., Yamazato, K., Van Woesik, R., 2001. Photo- acclimation dynamics of the coral Stylophora pistillata to low and extremely low light. J. Exp. Mar. Bio. Ecol. 263, 211–225. doi:10.1016/S0022-0981(01)00309-4

Todd, P.A., Ladle, R.J., Lewin-Koh, N.J.I., Chou, L.M., 2004. Genotype x environment interactions in transplanted clones of the massive corals Favia speciosa and Diploastrea heliopora. Mar. Ecol. Prog. Ser. 271, 167–182. doi:10.3354/meps271167

Toker, C., 2004. Estimates of broad-sense heritability for seed yield and yield criteria in faba bean (Vicia faba L.). Hereditas 140, 222-225. doi:10.1111/j.1601- 5223.2004.01780.x

Torres, J.L., Armstrong, R.A., Corredor, J.E., Gilbes, F., 2007. Physiological responses of Acropora cervicornis to increased solar irradiance. Photochem. Photobiol. 83, 839–850. doi:10.1562/2006-09-01-ra-1025

Treignier, C., Tolosa, I., Grover, R., Reynaud, S., sa, C.F.-P., 2009. Carbon isotope composition of fatty acids and sterols in the scleractinian coral Turbinaria reniformis: Effect of light and feeding. Limnol. Oceanogr. 54, 1933–1940. doi:10.4319/lo.2009.54.6.1933

Tunnicliffe, V., 1981. Breakage and propagation of the stony coral Acropora cervicornis. Proc. Natl. Acad. Sci. 78, 2427–2431. doi:10.1073/pnas.78.4.2427

University of Florida GeoPlan Center, 2015. Florida County Boundaries with Generalized Shorelines – September 2015. Florida Geographic Data Library, University of Florida, Gainesville, FL. http://www.fgdl.org (accessed 28 June 2018). van Hooidonk, R.J., Manzello, D.P., Moye, J., Brandt, M.E., Hendee, J.C., McCoy, C., Manfrino, C., 2012. Coral bleaching at Little Cayman, Cayman Islands 2009. Estuar. Coast. Shelf Sci. 106, 80–84. doi:10.1016/j.ecss.2012.04.021 van Oppen, M.J.H., Oliver, J.K., Putnam, H.M., Gates, R.D., 2015. Building coral reef resilience through assisted evolution. Proc. Natl. Acad. Sci. 112, 2307-2313. doi:10.1073/pnas.1422301112 van Oppen, M.J.H., Gates, R.D., Blackall, L.L., Cantin, N., Chakravarti, L.J., Chan, W.Y., Cormick, C., Crean, A., Damjanovic, K., Epstein, H., Harrison, P.L., Jones, T.A., Miller, M., Pears, R.J., Peplow, L.M., Raftos, D.A., Schaffelke, B., Stewart, K., Torda, G., Wachenfeld, D., Weeks, A.R., Putnam, H.M., 2017. Shifting paradigms in restoration of the world’s coral reefs. Glob. Chang. Biol. 1–12. doi:10.1111/gcb.13647 van Woesik, R., Ripple, K., Miller, S.L., 2017. Macroalgae reduces survival of nursery- reared Acropora corals in the Florida reef tract. Restor. Ecol. 26, 563–569. doi:10.1111/rec.12590

188

Vega Thurber, R.L., Burkepile, D.E., Fuchs, C., Shantz, A.A., Mcminds, R., Zaneveld, J.R., 2014. Chronic nutrient enrichment increases prevalence and severity of coral disease and bleaching. Glob. Chang. Biol. 20, 544–554. doi:10.1111/gcb.12450

Veron, J.E.N., 2000. Corals of the world. Australian Institute of Marine Science, Townsville, Australia.

Viant, M.R., Kurland, I.J., Jones, M.R., Dunn, W.B., 2017. How close are we to complete annotation of metabolomes? Curr. Opin. Chem. Biol. 36, 64–69. doi:10.1016/J.CBPA.2017.01.001

Vollmer, S. V., Kline, D.I., 2008. Natural disease resistance in threatened staghorn corals. PLoS One 3, e3718. doi:10.1371/journal.pone.0003718

Vollmer, S. V, Palumbi, S.R., 2002. Hybridization and the evolution of reef coral diversity. Science 296, 2023–2025. doi:10.1126/science.1069524

Wangpraseurt, D., Larkum, A.W.D., Ralph, P.J., Kühl, M., 2012. Light gradients and optical microniches in coral tissues. Front. Microbio. 3, 316

Want, E., Masson, P. Processing and analysis of GC/LC-MS-based metabolomics data, in: Metz, T.O. (Ed.), Metabolic Profiling. Humana Press, pp. 277–298.

Ware, M., 2015. Assessment of nursery-raised Acropora cervicornis transplants in the upper Florida Keys. MS Thesis, Nova Southeastern University, Fort Lauderdale.

Warner, M.E., Lesser, M.P., Ralph, P.J., 2010. Chlorophyll fluorescence in reef building corals, in: Suggett, D.J., Prášil, O., Borowitzka, M.A. (Eds.), Chlorophyll a Fluorescence in Aquatic Sciences: Methods and Applications. Springer Netherlands, pp. 209–222. doi:10.1007/978-90-481-9268-7

Warner, M.E., Suggett, D.J., 2016. The photobiology of Symbiodinium spp.: Linking physiological diversity to the implications of stress and resilience. In: Goffredo S, Dubinsky Z (eds) The Cnidaria, Past, Present and Future. Springer Science & Business Media, p 489–509

Weckwerth, W., 2003. Metabolomics in systems biology. Annu. Rev. Plant Biol. 54, 669–689. doi:10.1146/annurev.arplant.54.031902.135014

Westmoreland, L.S.H., Niemuth, J.N., Gracz, H.S., Stoskopf, M.K., 2017. Altered acrylic acid concentrations in hard and soft corals exposed to deteriorating water conditions. Facets 2, 531–544. doi:10.1139/facets-2016-0064

White, A.T., Vogt, H.P., Arin, T., 2000. Philippine coral reefs under threat: The economic losses caused by reef destruction. Mar. Pollut. Bull. 40, 598–605. doi:10.1016/S0025-326X(00)00022-9

189

Wiedenmann, J., D’Angelo, C., Smith, E.G., Hunt, A.N., Legiret, F.E., Postle, A.D., Achterberg, E.P., 2013. Nutrient enrichment can increase the susceptibility of reef corals to bleaching. Nat. Clim. Chang. 3, 160–164. doi:10.1038/nclimate1661

Williams, S.L., Ambo-Rappe, R., Sur, C., Abbott, J.M., Limbong, S.R., 2017. Species richness accelerates marine ecosystem restoration in the Coral Triangle. Proc. Natl. Acad. Sci. 114, 11986–11991. doi:10.1073/pnas.1707962114

Williams, D.E., Miller, M.W., Bright, A.J., Pausch, R.E., Valdivia, A., 2017. Thermal stress exposure, bleaching response, and mortality in the threatened coral Acropora palmata. Mar. Pollut. Bull. 124, 189–197. doi:10.1016/j.marpolbul.2017.07.001

Williams, D.E., Miller, M.W., 2005. Coral disease outbreak: Pattern, prevalence and transmission in Acropora cervicornis. Mar. Ecol. Prog. Ser. 301, 119–128. doi:10.3354/meps301119

Winters, G., Beer, S., Ben Zvi, B., Brickner, I., Loya, Y., 2009. Spatial and temporal photoacclimation of Stylophora pistillata: Zooxanthella size, pigmentation, location and clade. Mar. Ecol. Prog. Ser. 384, 107–119. doi:10.3354/meps08036

Wirt, K.E., Hallock, P., Palandro, D., Lunz, K.S., 2015. Potential habitat of Acropora spp. on reefs of Florida, Puerto Rico, and the US Virgin Islands. Glob. Ecol. Conserv. 3, 242–255. doi:10.1016/j.gecco.2014.12.001

Woodley, J.D., Chornesky, E.A., Clifford, P.A., Jackson, J.B., Kaufman, L.S., Knowlton, N., Lang, J.C., Pearson, M.P., Porter, J.W., Rooney, M.C., Rylaarsdam, K.W., Tunnicliffe, V.J., Wahle, C.M., Wulff, J.L., Curtis, A.S., Dallmeyer, M.D., Jupp, B.P., Koehl, M.A., Neigel, J., Sides, E.M., 1981. Hurricane Allen’s Impact on Jamaican Coral Reefs. Science 214, 749–755. doi:10.1126/science.214.4522.749

Wulff-Zottele, C., Gatzke, N., Kopka, J., Orellana, A., Hoefgen, R., Fisahn, J., Hesse, H., 2010. Photosynthesis and metabolism interact during acclimation of Arabidopsis thaliana to high irradiance and sulphur depletion. Plant, Cell Environ. 33, 1974–1988. doi:10.1111/j.1365-3040.2010.02199.x

Xia, J., Sinelnikov, I. V., Han, B., Wishart, D.S., 2015. MetaboAnalyst 3.0—making metabolomics more meaningful. Nucleic Acids Res. 43, W251–W257. doi:10.1093/nar/gkv380

Yancey, P.H., Clark, M.E., Hand, S.C., Bowlus, R.D., Somero, G.N., 1982. Living with water stress: Evolution of osmolyte systems. Science 217, 1214–1222.

Yancey, P.H., Fyfe-Johnson, A.L., Kelly, R.H., Walker, V.P., Aun, M.T., 2001. Trimethylamine oxide counteracts effects of hydrostatic pressure on proteins of

190

deep-sea teleosts. J. Exp. Zool. 289, 172–176. doi:10.1002/1097- 010X(20010215)289:3<172::AID-JEZ3>3.0.CO;2-J

Yancey, P.H., Rhea, M.D., Kemp, K.M., Bailey, D.M., 2004. Trimethylamine oxide, betaine and other osmolytes in deep-sea animals: depth trends and effects on enzymes under hydrostatic pressure. Cell. Mol. Biol. 50, 371–376. doi:10.1170/T526

Yonge, C.M., 1973. The nature of reef-building (hermatypic) corals. Bull. Mar. Sci. 23, 1–15.

Young, C.N., Schopmeyer, S.A., Lirman, D., 2012. A review of reef restoration and coral propagation using the threatened genus Acropora in the Caribbean and western Atlantic. Bull. Mar. Sci. 88, 1075–1098. doi:10.5343/bms.2011.1143

191

BIOGRAPHICAL SKETCH

Kathryn Elaine Lohr was born and raised in Bloomfield Hills, Michigan. She first became interested in aquatic science while spending time exploring the Great Lakes and snorkeling with her parents and grandparents during trips to the Florida Keys. She graduated with a Bachelor of Science in biology from the University of Michigan in 2009, and went on to complete a Master of Professional Science degree in marine biology and fisheries at the University of Miami’s Rosenstiel School of Marine and Atmospheric

Science in 2011.

After completing her master’s degree, Kathryn went on to work for the Central

Caribbean Marine Institute in Little Cayman, Cayman Islands. While there, she had the opportunity to set up and run Cayman’s first coral nursery in conjunction with the

Cayman Islands Department of Environment. She also gained valuable experience in other aspects of coral reef science from invasive species management to long-term benthic and fish monitoring.

In January 2015, she left the Caribbean to accept a PhD position in the School of

Forest Resources and Conservation at the University of Florida. Kathryn has spent the past four years working from the Florida Conservation and Technology Center in Apollo

Beach, FL, and has had the opportunity to complete projects for her dissertation in the

Florida Keys, Little Cayman, and Heron Island, Australia.

Kathryn will spend the year following the completion of her PhD as a Sea Grant

John A. Knauss Marine Policy Fellow within NOAA’s Office of National Marine

Sanctuaries, where she hopes she will be able to apply her knowledge and experience in marine science and restoration to broader resource management issues.

192