COBI cobi12171 Dispatch: September 20, 2013 CE: AFL

Journal MSP No. No. of pages: 10 PE: XXXXX1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

UNCORRECTEDPROOF

Special Section

Incorporating Climate and Ocean Change intoExtinction Risk Assessments for 82 Coral SpeciesRUSSELL E. BRAINARD,∗ MARISKA WEIJERMAN,† C. MARK EAKIN,‡ PAUL MCELHANY,§MARGARET W. MILLER,∗∗ MATT PATTERSON,†† GREGORY A. PINIAK,‡‡MATTHEW J. DUNLAP,† AND CHARLES BIRKELAND§§∗Pacific Islands Fisheries Science Center, National Marine Fisheries Service, 1125B Ala Moana Boulevard, Honolulu, HI 96814, U.S.A.email rusty.brainard@noaa.gov.†Joint Institute for Marine and Atmospheric Research, University of Hawaii at Manoa, Honolulu, HI 96822, U.S.A.‡Center for Satellite Applications and Research, NOAA Coral Reef Watch, 5830 University Research Court, College Park, MD 20737,U.S.A.§Northwest Fisheries Science Center, National Marine Fisheries Service, 2725 Montlake Boulevard East, Seattle, WA 98112, U.S.A.∗∗Southeast Fisheries Science Center, National Marine Fisheries Service, 75 Virginia Beach Drive, Miami, FL 33149–1099, U.S.A.††South Florida Inventory & Monitoring Network, National Park Service, 18001 Old Cutler Road Suite 419, Miami, FL 33157, U.S.A.‡‡National Centers for Coastal Ocean Science, National Ocean Service, 1305 East-West Highway, Silver Spring, MD 20910, U.S.A.§§Department of Biology, University of Hawaii at Manoa, University Ave, Honolulu, HI 96822, U.S.A.

Abstract: Many marine invertebrate species facing potential extinction have uncertain taxonomies and

Q1

poorly known demographic and ecological traits. Uncertainties are compounded when potential extinctiondrivers are climate and ocean changes whose effects on even widespread and abundant species are onlypartially understood. The U.S. Endangered Species Act mandates conservation management decisions foundedon the extinction risk to species based on the best available science at the time of consideration—requiringprompt action rather than awaiting better information. We developed an expert-opinion threat-based ap-proach that entails a structured voting system to assess extinction risk from climate and ocean changes andother threats to 82 coral species for which population status and threat response information was limited.Such methods are urgently needed because constrained budgets and manpower will continue to hinderthe availability of desired data for many potentially vulnerable marine species. Significant species-specificinformation gaps and uncertainties precluded quantitative assessments of habitat loss or population declinesand necessitated increased reliance on demographic characteristics and threat vulnerabilities at genus orfamily levels. Adapting some methods (e.g., a structured voting system) used during other assessments anddeveloping some new approaches (e.g., integrated assessment of threats and demographic characteristics),we rated the importance of threats contributing to coral extinction risk and assessed those threats againstpopulation status and trend information to evaluate each species’ extinction risk over the 21st century. Thisqualitative assessment resulted in a ranking with an uncertainty range for each species according to theirestimated likelihood of extinction. We offer guidance on approaches for future biological extinction riskassessments, especially in cases of data-limited species likely to be affected by global-scale threats.

Q2 Keywords: Climate and ocean change, coral, data-limited species, extinction risk, ESA, Risk assessment

Incorporacion del Cambio Climatico y Oceanico en Estudios de Riesgo de Extincion para 82 Especies de Coral

Resumen: Muchas especies de invertebrados marinos que enfrentan extincion potencial tienen taxonomıasinciertas y caracterısticas demograficas y ecologicas poco conocidas. Las incertidumbres estan compuestascuando los conductores potenciales de extincion son los cambios climaticos y oceanicos, cuyos efectos, inclusosobre especies abundantes y con distribucion extensa, son parcialmente entendidos. El Acta Estadunidensede Especies en Peligro dicta decisiones de manejo de conservacion basadas en el riesgo de extincion paralas especies basado en la mejor ciencia disponible en el momento de consideracion, lo que requiere una

∗email mariska.weijerman@noaa.govPaper submitted July 26, 2012; revised manuscript accepted June 29, 2013.

1Conservation Biology, Volume 00, No. 0, 1–10C© 2013 Society for Conservation BiologyDOI: 10.1111/cobi.12171

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

UNCORRECTEDPROOF

2 Incorporating Climate and Ocean Change

pronta accion en lugar de esperar mejor informacion. Desarrollamos un acercamiento basado en amenazasy opiniones de expertos que involucra un sistema de votacion estructurado para estudiar el riesgo de extinciona partir de los cambios climaticos y oceanicos y otras amenazas para 82 especies de coral, para las cualesla informacion sobre el estado de la poblacion y la respuesta a la amenaza era limitada. Tales metodos sonuna necesidad urgente porque los presupuestos restringidos y la voluntad de la mano de obra continuarandificultando la disponibilidad de la informacion deseada para muchas especies marinas potencialmentevulnerables. Vacıos de informacion, significativos y especıficos de especie e incertidumbres impidieron estudioscuantitativos de la perdida de habitat o disminuciones poblacionales y requirieron dependencia incrementadade los caracteres demograficos y las debilidades de amenaza en los niveles de genero o familia. Al adaptaralgunos metodos (p. ej.: un sistema estructurado de votacion) usados durante otros estudios y desarrollandoalgunos acercamientos nuevos (p. ej.: estudios integrados de amenazas y caracterısticas demograficas), cali-ficamos la importancia de las amenazas que contribuyen al riesgo de extincion de los corales y estudiamosesas amenazas contra el estado de la poblacion y la tendencia de la informacion para evaluar el riesgo deextincion de cada especie en el siglo 21. Este estudio cualitativo resulto en una clasificacion con un rango deincertidumbre para cada especie de acuerdo a su probabilidad estimada de extincion. Ofrecemos orientacionen los acercamientos para estudios futuros de riesgo de extincion biologica, especialmente en casos de especiescon informacion limitada y con probabilidad de ser afectadas por amenazas a escala global.

Introduction

Many species extinction risk analyses require an account-

Q3

ing of both local and global threats (i.e., climate andocean change), but it is challenging to develop meth-ods to make scientifically sound and reliable extinction-risk assessments of species for which little is known.Traditional analyses rely on demographic patterns, butthe use of trait-specific, threat-based ecological-risk as-sessments has been increasing. For even well-studiedspecies, comprehensive information on emerging threatssuch as climate and ocean changes is often not available.Recent expansions of International Union for Conserva-tion of Nature (IUCN) Red List assessments for sea tur-tles (Wallace et al. 2011) and pinnipeds (Kovacs et al.2012) specifically considered potential climate effects.The sea turtle assessment divided species into regionalmanagement units (RMUs) but omitted climate changefrom the overall threat assessment when they found38 of the 58 RMUs were data deficient with respectto climate-change effects. The pinniped assessment in-cluded an extensive summary of climate effects, butin evaluating individual species and subspecies it sim-ply listed climate effects as a primary or impendingsecondary threat without detailing the species-specificeffects. Climate effects were a primary threat for 12species and an impending secondary threat for 28 more.Eight species did not have climate effects as an explicitconcern.

In a more comprehensive approach, Chin et al. (2010)devised an integrated risk assessment for climate changethat blends 3 components of species vulnerability traits—exposure, sensitivity, and adaptive capacity—with tradi-tional fishery assessments to evaluate sharks and rays onthe Great Barrier Reef. Little information was available oneffects to individual species, so all species within habitatswere considered together to determine relative vulnera-bility to a suite of direct (temperature, acidification, and

freshwater input) and indirect climate effects for each

Q4

habitat. Thirty of the 133 species evaluated were assessedto be highly or moderately vulnerable to climate effects.The authors identified potential actions to reduce risk,but the assessment method itself was not linked to anyparticular conservation or extinction framework. Theseexamples advanced the assessments of their respectivespecies, but generally efforts to incorporate considera-tions of climate and ocean change into extinction-riskevaluations are in early developmental stages relative totraditional demographic techniques.

In October 2009, the Center for Biological Diversitypetitioned the National Marine Fisheries Service (NMFS)to list 83 coral species under the U.S. Endangered SpeciesAct (ESA), citing climate and ocean change-related effectsas primary contributing factors in extinction risk. TheNMFS determined further assessment was warranted for82 of the petitioned species and convened a biologicalreview team (BRT) of 7 subject-matter experts (all ofwhom are authors on this paper) charged with assessingthe status of and extinction risk to each species. TheESA requires a comprehensive threat evaluation, so wedeveloped a methodology that was transferable across asuite of threats, both local and global. However, giventhe large number of species petitioned, significant datagaps were likely. Also, the complex evolutionary his-tory, hybridization, and inconsistencies between classicalmorphologically based taxonomy and newer genetic ap-proaches (e.g., Veron 1995; Forsman et al. 2010; Pinzon& LaJeunesse 2011) even made defining coral speciesproblematic. Because corals are colonial organisms thatincur partial mortality and most coral reef surveys identifycorals to the genus level, defining individual abundanceand population trends was similarly difficult. Our objec-tive here is to present the methods developed for review-ing the status of and extinction risk for the 82 candidatecoral species in the climate-based ESA petition and toprovide lessons learned for incorporating considerations

Conservation BiologyVolume 00, No. 0, 2013

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

UNCORRECTEDPROOF

Brainard et al. 3

of climate and ocean change that might be applicable toextinction-risk assessments for other data-limited species.

General Approach

We reviewed species risks at 2 levels—considering keythreats to corals in general and evaluating available infor-mation on population status and susceptibility to threatsthat was either available for or applicable to each species.We then used a structured expert-opinion voting systemto evaluate the overall risk of extinction for each of the82 species across its entire range.

Absolute extinction is reached when there are zeroindividuals of a particular species alive. Prior to that, aspecies may become functionally extinct when extinc-tion is inevitable, although some individuals may still bealive. For more practical conservation management deci-sion making under ESA, we defined the term critical-riskthreshold to indicate a condition in which the speciesis of such low abundance, so spatially fragmented, or atsuch reduced genetic or genotypic diversity that extinc-tion is extremely likely (conceptually similar to extinc-tion thresholds for habitat fragmentation; Ovaskainen &Hanski 2003). Critical-risk threshold was not defined as asingle abundance number, density, spatial distribution,or trend value, but rather as a qualitative descriptionencompassing such characteristics where quantitative in-formation is lacking. In assessing the critical-risk thresh-old, we considered depensatory processes, environmen-tal stochasticity, catastrophic events, and known localand global anthropogenic stressors, including climate andocean change, to allow for evaluation despite insufficientdata to compute quantitative thresholds for the species.

Assessment Time Horizon

Lacking a formal ESA definition for foreseeable future,agency policy recommends linking risk evaluation to thetime frame over which it is possible to scientifically pre-dict threat effects (U.S. Department of Interior 2009).The analyses and synthesis of information developed forthe Intergovernmental Panel on Climate Change (IPCC)Fourth Assessment Report (AR4) (IPCC 2007) were themost thoroughly documented and reviewed assessmentsof future climate and ocean change ever issued and repre-sented the best scientific information on potential futurechanges in Earth’s climate system available at the time thespecies’ status was reviewed (Brainard et al. 2011). Therange of CO2 emission scenarios and collection of globalclimate models provided projections with adequate con-fidence to the year 2100 for us to reasonably use this timeframe for evaluation of critical-risk thresholds for the can-didate coral species. Much of the scientific informationavailable on the potential effects of ocean acidification

on corals has likewise been based on IPCC CO2 emissionscenarios and model projections.

Components of Extinction Risk

The general trends of coral decline and the importanceof various global change and local threats are reasonablywell documented for corals, but information was sorelylacking on the specifics of how these threats would af-fect most of the individual candidate species. Therefore,the focus was placed on exposure to threats and sus-ceptibility of the species to them. Here, we evaluatedextinction risk for each species by considering 2 sepa-rate, but related, types of information: global and localthreats, including their intensity and trends (Crawford& Rumsey 2009) and traits, dynamics, and responsesof the species to these threats and distribution of thespecies themselves. In most cases, there was no quanti-tative species-level information regarding abundance andtrends or threat susceptibility. There was also a lack ofgeographically refined predictions regarding exposure,especially to global threats.

Identifying Important Threats

We considered status and trends of human populationand consumption of natural resources to be the rootdrivers of the important environmental threats, whichfall into 2 main categories: global climate- and ocean-change threats (atmospheric CO2, ocean warming, oceanacidification, sea-level rise, changing ocean circulation,changing storm characteristics, atmospheric dust, andchanges in insolation) and local threats (land-based pol-lution, disease, predation, fishing and trophic cascades,destructive fishing practices, ornamental trade, naturalphysical damage, human-induced physical damage, andinvasive species). Furthermore, we explicitly consideredinteractions between threats during the assessment ofrisks.

Evaluating the severity and likelihood of effects basedon documented responses to natural changes and exper-iments from the threats outlined above across the globalrange of reef corals, we concluded that the 3 most im-portant threats in terms of extinction risk of coral speciesbetween now and the year 2100 are ocean warming,disease, and ocean acidification (Table 1). Local threatswith widespread effect, such as sedimentation, eutroph-ication, and fishing, were considered medium extinctionrisks. Many other threats (e.g., physical damage, invasivespecies, collection, and trade) can cause extirpation atlocal scales but pose significant extinction risk only forspecies with restricted geographic ranges or species thathave already undergone precipitous population declines.To date, there is little evidence that corals can adapt oracclimate to the projected magnitude and rate of change

Conservation BiologyVolume 00, No. 0, 2013

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

UNCORRECTEDPROOF

4 Incorporating Climate and Ocean Change

Table 1. Summary of proximate threats considered in assessing ex-tinction risks to the 82 coral species petitioned for listing under theU.S. Endangered Species Act, ordered on the basis of the Coral Biolog-ical Review Team’s estimate of the threat’s importance for extinctionrisk.

Scale Proximate threat Importance

Global ocean warming highLocal disease highGlobal ocean acidification medium-highLocal reef fishing—trophic

effectsmedium

Local sedimentation low-mediumLocal nutrients low-mediumGlobal sea-level rise low-mediumLocal toxins lowGlobal changing ocean

circulationlow

Global changing storm tracksand intensities

low

Local predation lowLocal reef fishing—habitat

effects, destructivefishing practices

low

Local ornamental trade lowLocal natural physical damage lowLocal human-induced physical

damagenegligible-low

Local aquatic invasive species negligible-lowLocal salinity negligibleLocal African/Asian dust negligibleGlobal changes in insolation probably negligible

in environmental conditions (Baskett et al. 2009a, b;Rodolfo-Metalpa et al. 2011), let alone to multiple stres-sors simultaneously and interactively.

Specific mechanisms of climate- and ocean-changethreats were examined individually (Table 1) (seeBrainard et al. 2011 for a detailed assessment). The 2greatest global threats to extinction to all or most coralspecies over the next century were ocean warming andacidification, which will continue to increase as atmo-spheric CO2 continues to rise. Globally, thermal stress hasand will continue to affect corals throughout all or most oftheir ranges (Hoegh-Guldberg et al. 2007). An associationof mass coral bleaching events with rising temperaturesbegan on a large scale in the early 1980s (Glynn 1991;Fitt et al. 2001). The return frequency of mass bleachingevents has already begun to exceed the ability of manyreefs and coral species to recover (Baker et al. 2008; Eakinet al. 2010), and thermal stress may have already causedthe first modern coral extinction (Glynn et al. 2001). Al-though, it is theoretically possible that future technologi-cal or policy changes could reduce these climate threats,management cannot rely on assumptions about futurepolicy changes or technological advances. Moreover, thethreat posed by even the most optimistic scenarios of21st century CO2 emissions used in the IPCC AR4 (IPCC2007)—or simply the threat posed by committed climate

change (Hare & Meinshausen 2006)—represents a plau-sible extinction risk to many coral species. Fortunately,some combination of field and lab studies on bleachingeffects was available for many species being evaluated ortheir congeners.

The magnitude of the threat of ocean acidification tocoral reefs has been accepted only within the last decade.The atmospheric CO2 concentration recently surpassed400 ppm and continues to accelerate, already exceed-ing the worst-case emission trajectories used in the AR4(IPCC 2007; Allison et al. 2009). Rising atmospheric CO2

will continue to induce ocean acidification, but thereremain many unknowns in understanding of how oceanacidification will affect the diversity and interdependenceof marine species. Based on current knowledge and pro-jections for the future, we ranked ocean acidification asa medium-high extinction risk to corals. The severity ofthis threat to the growth and, particularly, recruitmentof corals will make it more difficult for many of themto recover from other effects (reviewed in Brainard etal. 2011). Some corals are already experiencing repro-ductive impairment and recruitment failure (Hughes &Tanner 2000; Albright et al. 2010; Polato et al. 2010). Un-fortunately, ocean acidification effects have been quanti-fied for only a handful of coral species, so we had to as-sume effects of congeneric or confamilial species wouldapply to the species being evaluated.

It is clear that multiple stressors affect organisms si-multaneously and interactively, although it is not cer-tain whether the effects will be additive, multiplicative,or negative. There are numerous examples of climate-related interactive effects on corals. Growing evidencefrom both the Caribbean and Indo-Pacific suggests coralsexperiencing physiological stress from poor-water qual-ity (Carilli et al. 2009; Wooldridge 2009) or ocean acidifi-cation (Anthony et al. 2008; Crawley et al. 2010) are moresensitive to high temperatures. Increased seawater tem-perature also may act synergistically with coral diseasesto reduce coral health and survival (Bruno et al. 2007).Although, we first considered the overall importance ofseparate threats, each species also was assessed on theavailable information regarding its responses to the iden-tified threats acting interactively and collectively.

Species Demography, Ecology, and Traits

Ideally, ESA biological reviews should include informa-tion on each species’ physiological response and threatsusceptibility (van Woesik et al. 2012; Foden et al. 2013),as well as a demographic assessment—but there was ageneral lack of most of these data for the coral specieswe considered. The vulnerabilities of different coralsto stressors are fundamentally related to natural-historytraits that are often common to genera or even families.Although, we were cautious regarding presumed eco-logical similarities among the species within a genus,

Conservation BiologyVolume 00, No. 0, 2013

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

UNCORRECTEDPROOF

Brainard et al. 5

some reliable intrageneric traits were used as indica-tors of likely species-level vulnerabilities. For example,members of the genus Acropora have high-growth ratesand porous skeletons. However, their thin superficial tis-sues provide low-energy reserves, and they consistentlydisplay high sensitivity to a range of stressors (e.g., in-creased temperatures, sedimentation, disease, and pre-dation). Hence, Acropora species are usually among thefirst corals to show substantial losses under stressful con-ditions, putting that genus at relatively high risk to someof the climate-related and disease threats. That said, it isrecognized that such traits may also provide opportuni-ties for potentially rapid recovery.

Lacking much of the desired information on species-specific susceptibility, the modern known geographicand habitat distributions of each species were used asproxies for species’ threat exposures and responses. Inthe case of invertebrates, the ESA requires that evaluationand protection be applied across a species’ entire range.Many corals have vast ranges spanning much of the globaltropics. Population trends were available for only a few(<5) of the 82 species; these species generally showedsignificant range-wide population declines. Range mapswere the only information available for all 82 species,but inadequacies in taxonomy, field identification, surveyeffort, and anecdotal reporting meant even these werenot always reliable (Brainard et al. 2011). Because it wasstill the only information consistently available, a species’distribution—by range and habitat type—was used to in-dicate both threat exposure and tolerance. Species in therelatively confined Caribbean basin, the eastern Pacific,or with extremely localized distribution within the Indo-Pacific were considered more susceptible to extinctionfrom stochastic or local disturbances, whereas those in-habiting a wide range of latitudes and longitudes mightexperience extirpation in some areas but could still per-sist across at least a portion of their ranges. Similarly,occupation of a wide range of habitat types or harshenvironments (shallow or intertidal areas) was presumedto indicate a greater degree of physiological tolerance tokey stressors, such as warm temperatures or increasedwave energy. Occupancy of potential refugia, such asrelatively deep mesophotic reefs (∼30–150 m), was alsotaken to indicate less exposure to surface-driven threatsand thus a potentially lower overall threat exposure.

The geologic record indicates that corals have re-sponded, sometimes dramatically, to changes over geo-logic time. Geographic distributions have shifted withocean temperature and sea-level stands (Precht &Aronson 2004; Greenstein & Pandolfi 2008; Yamanoet al. 2011), although current and future human activi-ties (Buddemeier et al. 2004) and environmental factorslike carbonate saturation state and light availability maylimit poleward range expansion (Kleypas et al. 1999).Unlike many other ESA-assessed species, reef corals relyextensively on symbionts to meet their energetic needs,

and changing that relation offers increased potential forphysiological tolerance. Although scientists have startedto model the potential for coral symbiont diversity to al-low corals to adapt to warming and acidification (Baskettet al. 2009a), many coral species would likely experiencesignificant mortality (e.g., Hoeke et al. 2011), as demon-strated by mass coral extinctions over geologic scales.Applying such concepts within an ESA context remainsproblematic because there is little predictive informationon any coral species’ potential for bleaching acclimati-zation, range expansion, or other mechanisms (geneticevolution, etc.) to cope with climate and ocean change.

Voting System for Risk Assessment

Extinction-risk evaluations were based on a compilationof the best available information on the biology, ecology,and the known vulnerabilities of the candidate speciesto the overall coral extinction threats (Table 1). In theabsence of quantitative demographic status and trendinformation, each expert independently evaluated thelikelihood that a species’ status will fall below the critical-risk threshold by the year 2100. Votes were cast basedon the best available science, following extensive litera-ture review and group discussion about the natural andanthropogenic threats and the susceptibility of each can-didate species. The voting process was similar to thatof some other ESA status evaluations (e.g., Good et al.2005), whereby each member anonymously assigned 10points to 8 risk-likelihood categories linked to qualitativeestimates of the probability of reaching the critical-riskthreshold. These categories ranged from exceptionallyunlikely that a species would fall below the critical-riskthreshold to virtually certain it would fall below thecritical-risk threshold. In a new approach for an ESAevaluation, we applied the risk-likelihood categories de-veloped by the IPCC to assess confidence (IPCC 2007)(example in Fig. 1). The spread of votes each ex-pert cast expressed their assessment of uncertainty—concentrating many votes in a single category reflectedhigh confidence, whereas a widespread allocation re-flected uncertainty. After the initial round of voting, theBRT reviewed the overall distribution of votes, with eachexpert’s votes listed anonymously for each of the 82species. Following a discussion about the specific factorsthat each expert considered as they voted on each indi-vidual species, a second round of anonymous voting wasperformed and results were tallied. A summation of thevotes from all experts produced the likelihood of eachspecies status falling below the critical-risk threshold, ex-pressed as a histogram of the percentage of likelihoodpoints for each risk category (Fig. 1). The mean and me-dian were used to indicate the overall likelihood, andthe distribution of votes reflected the uncertainty in theestimate across all experts.

Conservation BiologyVolume 00, No. 0, 2013

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

UNCORRECTEDPROOF

6 Incorporating Climate and Ocean Change

0

10

20

30

40

50

estimated likelihood of species status falling below CRT by 2100 (%)

po

ints

(%

)

66 1 99 33 50 9010

<1% 1-10% 33-50% 50-66% 66-90% 90-99%10-33% >99%

exceptionally unlikely

very unlikely

unlikely less likelythan not

more likely than not likely

very likely

virtually certain

mean = 53%

Figure 1. Estimates of the likelihood that the status of Pavona diffluens will fall below the critical-risk threshold(CRT) (i.e., the point at which the species is of such low abundance, so spatially fragmented, or at such reduceddiversity that extinction is extremely likely) by 2100.

The voting process yielded a ranked list of the critical-risk threshold likelihoods for each of the 82 candidatecoral species (Fig. 2 & Supporting Information). Boththe mean risk-likelihood scores and the uncertaintiesshould be considered in the application of these esti-mates. Scores indicated only a qualitative ranking anddid not support fine parsing among species whose meanscores differed by only a few points. The mean estimatedlikelihood of falling below the critical-risk threshold by2100 ranged from 78% (likely) to 19% (unlikely) amongthe 82 candidate coral species. The mean and medianlikelihood scores across all 82 candidate coral speciesboth exceeded 50%, thereby falling into the more-likely-than-not category.

The status review report was peer-reviewed by theCenter for Independent Experts (http://cufer.rsmas.miami.edu/wordpress/wp-content/uploads/2007/05/p41.pdf) to ensure that the overall approach wasscientifically valid and the risk assessments werereasonable. The results of the biological review (Brainardet al. 2011), together with the evaluation of otherfactors such as adequacy of existing regulations andconservation initiatives (NMFS 2012a) and additionalinformation collected on the species (NMFS 2012b),resulted in 12 coral species being proposed asendangered (5 Caribbean and 7 Indo-Pacific), 54 coralspecies being proposed as threatened (2 Caribbean and

52 Indo-Pacific), and the determination that 16 coralspecies (all Indo-Pacific) did not warrant listing underthe ESA (NMFS 2012c). Final decisions on the remaining66 coral species proposed for listing are scheduled to bereleased by NMFS in December 2013.

Comparisons with Previous Coral Assessments

In its first foray into the application of ESA to corals, theNMFS listed the Caribbean species Acropora cervicornisand A. palmata as threatened (NMFS 2006). Considerabledata documenting significant population declines wereavailable for those species. The Acropora BRT rankedthreats (bleaching was an important but secondary threatand ocean acidification received little attention; Acrop-ora Biological Review Team 2005) but did not adopta structured voting process to assess risk. More recentobservations of climate conditions (IPCC 2007; Allisonet al. 2009) indicate the best climate models used by theAcropora BRT were overly optimistic.

The IUCN’s comprehensive red-list assessment for 845reef corals (Carpenter et al. 2008) was conducted bya large (39 member) expert panel. They were similarlyhamstrung by a lack of abundance and trend information.Instead they used a surrogate of estimated habitat loss toinfer population reduction, concluding that losses will

Conservation BiologyVolume 00, No. 0, 2013

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

UNCORRECTEDPROOF

Brainard et al. 7

Montastraea annularisMontastraea faveolata

Acropora lokaniAcropora jacquelineae

Dendrogyra cylindrusMontastraea franksi

Montipora patula (verrilli)

Mycetophyllia feroxAcropora rudis

Acropora doneiAcropora pharaonisEuphyllia paradivisaMillepora foveolataMillepora tuberosa

Euphyllia paraancoraIsopora cuneata

Euphyllia cristataMontipora dilatata/flabellata (turgescens)

Agaricia lamarckiAnacropora spinosa

Dichocoenia stokesiAcropora microclados

Montipora lobulataAcropora striataAcropora listeri

Acropora globicepsPachyseris rugosa

Alveopora fenestrata

Alveopora verrilliana

Alveopora allingi

Acanthastrea brevisAcanthastrea hemprichii

Pavona bipartita

Pavona venosaPavona decussata

Pavona cactusAcanthastrea regularis

Galaxea astreataTurbinaria stellulataCyphastrea ocellinaCyphastrea agassizi

Turbinaria mesenterinaTurbinaria peltata

Turbinaria reniformisHeliopora coerulea

Porites (Clade 1 variety pukoensis)

Leptoseris yabeiLeptoseris incrustansPsammacora stellata

Acanthastrea ishigakiensis

Montipora australiensis

Isopora crateriformisMontipora caliculata

Acropora speciosaAcropora verweyi

Acropora retusa

Acropora aculeusAcropora acuminata

Acropora asperaPectinia alcicornis

Caulastrea echinulataSeriatopora aculeata

Montipora calcarea

Montipora angulata

Anacropora puertogalerae

Pocillopora danaePavona diffluens

Acropora paniculataAcropora polystomaAcropora vaughani

Acropora palmeraeAcropora horrida

Astreopora cucullata

Physogyra lichtensteiniPorites horizontalata

Pocillopora elegans (W Pacific)Porites napopora

Porites nigrescens

Barabattoia laddi

Acropora tenella

Acropora dendrumPocillopora elegans (E Pacific)

<1% 1-10% 33-50% 50-66% 66-90% 90-99%10-33% >99%

exceptionally unlikely

very unlikely

unlikely less likelythan not

more likelythan not likely

very likely

virtually certain

Figure 2. Summary of votes tallied across critical-risk threshold (i.e., point at which the species is of such lowabundance, so spatially fragmented, or at such reduced diversity that extinction is extremely likely) likelihoodcategories for all candidate Indo-Pacific and Caribbean (in bold) coral species ranked by mean likelihood offalling below the critical-risk threshold (the darker the shade, the more votes in a risk category).

Conservation BiologyVolume 00, No. 0, 2013

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

UNCORRECTEDPROOF

8 Incorporating Climate and Ocean Change

be driven largely by climate and ocean changes. Theyused a consensus-based approach to review the scienceand estimate loss and then equated relatively arbitrarythresholds of habitat loss to red-list categories. Subse-quent to publication of the BRT’s report, a consensus-based approach was used to assess extinction risk basedon trait-based predictions of coral resilience to climatechange (van Woesik et al. 2012). That work was reportedat the genus level and showed discrepancies in a few gen-era relative to the IUCN results (Carpenter et al. 2008);consequently, they recommended extinction risk inte-grate both population reduction and species traits. Thatis precisely the approach required by ESA and used in ourreview (Brainard et al. 2011), although several of our con-clusions differed from those of van Woesik et al. (2012).Most prominently, we considered the Montastrea annu-laris complex at considerably more risk than estimatedby van Woesik et al. (2012), although this conclusionwas heavily driven by documented recruitment failure(e.g., population decline) more so than species traits onwhich van Woesik et al. (2012) relied. We also consideredgeographic and habitat distributions a potential mitigatorof risk, whereas they explicitly noted that species withwidespread distributions could still go extinct.

Alternate Approaches and Species Evaluations

Many commonly applied endangerment criteria arepurely demographic. For example, the criteria used forthe IUCN Red List of Threatened Species addresses spe-cific levels of reduction in population size, geographicrange, subpopulation structure, etc. Many previous NMFSBRTs, particularly those for vertebrates, were able torely on population-level modeling to assess populationstatus (e.g., Good et al. 2005). In addition to the 2Caribbean Acropora and the 66 proposed corals, NMFSpreviously listed 3 benthic species: white abalone (Hali-otis sorenseni), black abalone (Haliotis cracherodii),and Johnson’s seagrass (Halophila johnsonii) (NMFS2013). The abalone listings had the advantage of fishery-dependent abundance data, whereas Johnson’s seagrasswas known to occur rarely. Those listings were all donebefore climate and ocean change gained widespreadrecognition as a potential extinction driver. However, thecritical habitat designation for black abalone cited oceanacidification as factor in abalone mortality, and the rulespecifically solicited public comments on acidificationto help address the knowledge gap. Similarly, the 5-yearreview update for H. johnsonii explicitly commented onthe lack of climate effects in the original status reviewand listing and suggested newer information is availableand the issue should be addressed.

For an ESA application, our approach offers severaladvantages. By focusing on a critical-risk threshold ratherthan an explicit population threshold, our approach can

apply to species with abundance information (e.g., fish-eries data for abalone) or with qualitative distributioninformation (e.g., seagrass). Climate effects can be ex-plicitly considered, but within a comprehensive threatassessment as the ESA requires; this is also useful foriconic or charismatic species with available populationestimates that still can be data poor with respect to threats(e.g., Wallace et al. 2011). In evaluating risk threshold rel-ative to IPCC categories, we were able to more directlylink our assessment to a climate framework rather thanextinction categories (e.g., IUCN Red List criteria) or ahigh–medium–low scheme. More importantly, aggregat-ing the assessment of experts with varying perceptions ofrisk to the species into a numerical score placed multiplespecies in context to one another and explicitly portrayedthe uncertainty in the outcomes. This approach is animportant advantage in an ESA context because it allowsa BRT to articulate what the science suggests the risk iswhile giving the regulatory agency managers discretion tochoose their level of acceptable extinction risk in makingtheir listing decisions. Incorporating uncertainty in riskassessment has proven valuable in other managementframeworks, such as ecosystem-based fisheries manage-ment (Hobday 2011).

Although the Center for Biological Diversity coral pe-tition explicitly emphasized climate- and ocean-changeeffects, we were required by ESA to consider the fullsuite of threats. This necessitated an approach that wastractable across multiple types of threats that varied inthe extent of available information at the species level,and we defaulted to a simple categorical threat evalua-tion. An alternative approach could have been to use ahierarchical framework that considers, but screens out,lower level threats and thus focuses on higher risk factors,as has recently been done for fisheries risk assessments(Hobday 2011). Although that example focuses on effectsof a single threat, it incorporates uncertainty and can beapplied to data-poor species. For applications in which anexpanded treatment of climate as a standalone threat isdesired, the integrated risk assessment for climate changefor sharks and rays on the Great Barrier Reef (Chin et al.2010) is somewhat analogous to our approach in thatit considers vulnerability to multiple climate- and ocean-change effects (temperature, acidification, freshwater in-put, circulation, sea-level rise, storms, and light) and canaccommodate data-poor species.

Future ESA Assessments

The ESA mandates timely decisions based on the bestavailable information. To meet these prescribed dead-lines, novel approaches are needed when informationavailability precludes standard approaches. We assessedthe largest single list of ESA candidate species ever pe-titioned to NMFS using a mix of previously applied and

Conservation BiologyVolume 00, No. 0, 2013

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

UNCORRECTEDPROOF

Brainard et al. 9

novel approaches to assess the extinction-risk likelihoodof 82 species explicitly at risk from threats driven by cli-mate and ocean change. With accelerating atmosphericCO2, warming ocean and air temperatures, ocean acid-ification, and rising sea level, we expect the need forsimilar assessments for other data-limited marine specieswill continue to increase. We hope the approaches wedeveloped will help guide future assessments and aid inconservation of vulnerable species.

Acknowledgments

Many subject-matter experts graciously offered their timeand expertise when called on by the BRT, particularlyearly in the process of initiating the status review of the82 candidate coral species. Unfortunately, the list is toolong to include all who contributed to the writing ofthe status review report. We acknowledge M. McClure,R. Griffis, E. Seney, R. Waples, and their National Ma-rine Fisheries Service team for coordinating this specialissue and the anonymous reviewers whose commentssubstantially improved the manuscript. The contents inthis manuscript are solely the opinions of the authorsand do not constitute a statement of policy, decision, orposition on behalf of NOAA or the U.S. Government.

Supporting Information

A table with the results of the BRT voting process talliedper risk-likelihood category (Appendix S1) is availableonline. The authors are solely responsible for the con-tent and functionality of these materials. Queries otherthan absence of the material should be directed to thecorresponding author.

Literature Cited

Acropora Biological Review Team. 2005. Atlantic Acropora Status Re-view Document. Report to National Marine Fisheries Service, South-east Regional Office. 152 pp.

Albright, R. A., B. Mason, M. Miller, and C. Langdon. 2010. Oceanacidification compromises recruitment success of the threatenedCaribbean coral, Acropora palmata. Proceedings of the NationalAcademy of Sciences 107:20400–20404.

Allison, et al. 2009. The copenhagen diagnosis: updating the world onthe latest climate science. Available from http://www.fcrn.org.uk/sites/default/files/Copenhagen_Diagnosis_2009.pdf (accessed June2012).Q5

Anthony, K. R. N., D. I. Kline, G. Diaz-Pulido, S. Dove, and O. Hoegh-Guldberg. 2008. Ocean acidification causes bleaching and pro-ductivity loss in coral reef builders. Proceedings of the NationalAcademy of Sciences 105:17442–17446.

Baker, A. C., P. W. Glynn, B. Riegl. 2008. Climate change and coral reefbleaching: An ecological assessment of long-term impacts, recov-ery trends and future outlook. Estuarine, Coastal and Shelf Science80:435–471.

Baskett, M. L., S. D. Gaines, and R. M. Nisbet. 2009a. Symbiont diversitymay help coral reefs survive moderate climate change. EcologicalApplications 19:3–17.

Baskett, M. L., R. M. Nisbet, C. V. Kappel, P. J. Mumby, and S. D.Gaines. 2009b. Conservation management approaches to protectingthe capacity for corals to respond to climate change: a theoreticalcomparison. Global Change Biology 16:29–1246.

Brainard, R. E., C. Birkeland, C. M. Eakin, P. McElhany, M. W. Miller, M.Patterson, and G. A. Piniak. 2011. Status review report of 82 can-didate coral species petitioned under the U.S. Endangered SpeciesAct. U.S. Dep. Commer., NOAA Technical Memorandum, NOAA-TM-NMFS-PIFSC-27, Honolulu, HI. 530 p.

Bruno, J. F., E. R. Selig, K. S. Casey, C. A. Page, B. L. Willis, C. D.Harvell, H. Sweatman, and A. M. Melendy. 2007. Thermal stress andcoral cover as drivers of coral disease outbreaks. PLoS Biology 5:e124.

Buddemeier, R. W., J. A. Kleypas, and R. B. Aronson. 2004. Coral reefs& global climate change: potential contributions of climate changeto stresses on coral reef ecosystems. Pew Center on Global ClimateChange, Arlington,VA.

Carilli, J. E., R. D. Norris, B. A. Black, S. M. Walsh, and M. McField.2009. Local stressors reduce coral resilience to bleaching. PLoS ONE4:e6324.

Carpenter, K. E., et al. 2008. One-third of reef-building corals face ele-vated extinction risk from climate change and local impacts. Science321:560–563.

Chin, A., P. M. Kyne, T. I. Walker, and R. B. McAuley. 2010. An integratedrisk assessment for climate change: analysing the vulnerability ofsharks and rays on Australia’s Great Barrier Reef. Global ChangeBiology 16:1936–1953.

Crawford, B., and S. Rumsey. 2009. Guidance for monitoring re-covery of Pacific Northwest salmon and steelhead. NOAA Na-tional Marine Fisheries Service, Northwest Region, Seattle. Availablefrom http://www.nwr.noaa.gov/publications/recovery_planning/salmon_steelhead/domains/rme-guidance.pdf (accessed December2012).

Crawley, A., D. I. Kline, S. Dunn, K. Anthony, and S. Dove. 2010.The effect of ocean acidification on symbiont photorespirationand productivity in Acropora formosa. Global Change Biology 16:851–863.

Eakin, C.M., et al. 2010. Caribbean corals in crisis: record thermal stress,bleaching, and mortality in 2005. PloS ONE 5:e13969.

Fitt, W.K., B. E. Brown, M. E. Warner, and R. P. Dunne. 2001. Coralbleaching: interpretation of thermal tolerance limits and thermalthresholds in tropical corals. Coral Reefs 20:51–65.

Foden, W. B., et al. 2013. Identifying the world’s most climate changevulnerable species: a systematic trait-based assessment of all birds,amphibians and corals. PLoS ONE 8. Q6

Forsman, Z. H., G. T. Concepcion, R. D. Haverkort, R. W. Shaw, J. E.Maragos, and R. J. Toonen. 2010. Ecomorph or endangered coral?DNA and microstructure reveal Hawaiian species complexes: Mon-tipora dilitata/flabellata/turgescens & M. patula/verrilli. PLoSONE 5:e15021.

Glynn, P. W. 1991 Coral reef bleaching in the 1980s and possibleconnections with global warming. Trends in Ecology & Evolution6:175–179.

Glynn, P. W., J. L. Mate, A. C. Baker, and M. O. Calderon. 2001. Coralbleaching and mortality in Panama and Ecuador during the 1997–1998 El Nino-Southern Oscillation Event: spatial/temporal patternsand comparisons with the 1982–1983 event. Bulletin of Marine Sci-ence 69:79–109.

Good, T. P., R. S. Waples, P. Adams (editors). 2005. Updated status offederally listed ESUs of West Coast salmon and steelhead. U.S. Dept.Commer., NOAA Tech. Memo. NMFS-NWFSC-66.

Greenstein, B. J., and J. M. Pandolfi. 2008. Escaping the heat: rangeshifts of reef coral taxa in coastal Western Australia. Global ChangeBiology 14:513–528.

Conservation BiologyVolume 00, No. 0, 2013

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

UNCORRECTEDPROOF

10 Incorporating Climate and Ocean Change

Hare, B., and M. Meinshausen. 2006. How much warming are we com-mitted to and how much can be avoided? Climatic Change 75:111–149.

Hobday, A. J. 2011. Sliding baselines and shuffling species: implicationsof climate change for marine conservation. Marine Ecology 32:392–403.

Hoegh-Guldberg, O., et al. 2007. Coral reefs under rapid climate changeand ocean acidification. Science 318:1737–1742.

Hoeke, R. K., P. L. Jokiel, R. W. Buddemeier, and R. E. Brainard. 2011.Projected changes to growth and mortality of Hawaiian corals overthe next 100 years. PLoS ONE 6.Q7

Hughes, T. P., and J. E. Tanner. 2000. Recruitment failure, life histories,and long-term decline of Caribbean corals. Ecology 81:2250–2263.

IPCC. 2007. Climate Change 2007: Synthesis Report. Contribution ofWorking Groups I, II and III to the Fourth Assessment Report ofthe Intergovernmental Panel on Climate Change. IntergovernmentalPanel on Climate Change, Geneva, Switzerland.

Kleypas, J. A., J. W. McManus, and L. A. B. Menez. 1999. Environmen-tal limits to coral reef development: where do we draw the line?American Zoologist 39:146–159.

Kovacs, K. M., et al. 2012. Global threats to pinnipeds. Marine MammalScience 28(2):414–436.

National Marine Fisheries Service (NMFS). 2006. Endangered and threat-ened Species: final listing determinations for elkhorn coral andstaghorn coral. Federal Register 71:26852–26861.

National Marine Fisheries Service (NMFS). 2012a. Management reportfor 82 coral status review under the Endangered Species Act: exist-ing regulatory mechanisms and conservation efforts. Available fromhttp://www.fpir.noaa.gov/Library/PRD/Coral/Final%20Corals%20Mgmt%20Report%2011--29--12.pdf (accessed December 2012).

National Marine Fisheries Service (NMFS). 2012b. Supplementalinformation report on status review report and draft manage-ment report for 82 coral candidate species. Available fromhttp://www.fpir.noaa.gov/Library/PRD/Coral/Final%20Corals%20Supplemental%20Information%20Report%2011--28--12.pdf(accessed December 2012).

National Marine Fisheries Service (NMFS). 2012c. Endangered andthreatened wildlife and plants: Proposed listing determinationsfor 82 reef-building coral species; Proposed reclassification of

Acropora palmate and Acropora cervicornis from threatened toendangered. Federal Register 77:773220–73262. Available fromhttp://www.gpo.gov/fdsys/pkg/FR-2012--12--07/pdf/2012--29350.pdf (accessed December 2012).

National Marine Fisheries Service (NMFS). 2013. Endangeredand threatened species under NMFS’ jurisdiction. Availablefrom http://www.nmfs.noaa.gov/pr/pdfs/species/esa_table.pdf(accessed June 2013).

Ovaskainen, O., and I. Hanski. 2003. Extinction threshold in metapop-ulation models Annales Zoologici Fennici 40:81–97.

Pinzon, J. H., and T. LaJeunesse. 2011. Species delimitation of com-mon reef corals in the genus Pocillopora using nucleotide sequencephylogenies, population genetics, and symbiosis ecology. MolecularEcology 20:311–325.

Polato, N., C. R. Voolstra, J. Schnetzer, M. K. DeSalvo, C. J. Randall,A. M. Szmant, M. Medina, and I. B. Baums. 2010. Location-specificresponses to thermal stress in larvae of the reef-building coral Mon-tastraea faveolata. PLoS ONE 5:e11221.

Precht, W. F., and R. B. Aronson. 2004. Climate flickers and range shiftsof reef corals. Frontiers in Ecology and the Environment 2:307–314.

Rodolfo-Metalpa, et al. 2011. Coral and mollusc resistance to oceanacidification adversely affected by warming. Nature Climate Change1:308–312.

U.S. Department of Interior. 2009. The meaning of “Foreseeable Fu-ture”. In: Section 3(20) of the Endangered Species Act. Dept. of theInterior Solicitor’s Memo M-37021, January 16, 2009.

van Woesik, R., E. C. Franklin, J. O’Leary, T. R. McClanahan, J. S. Klaus,and A. F. Budd. 2012. Hosts of the Plio-Pleistocene past reflectmodern-day coral vulnerability. Proceedings of the Royal Societyof London Series B, DOI:10.1098/rspb.2011.2621. Q8

Veron, J. E. N. 1995. Corals in space and time. Cornell University Press. Q9

Wallace B. P., et al. 2011. Global conservation priorities for marineturtles. PLoS ONE 6. Q10

Wooldridge, S. A. 2009. Water quality and coral bleaching thresholds:formalising the linkage for the inshore reefs of the Great BarrierReef, Australia. Marine Pollution Bulletin 58:745–751.

Yamano, H., K. Sugihara, and K. Nomura. 2011. Rapid poleward rangeexpansion of tropical reef corals in response to rising sea surfacetemperatures. Geophysical Research Letters 38:L04601.

Conservation BiologyVolume 00, No. 0, 2013

Queries

Q1: Author: Please check authors’ field and their affiliations as typeset.Q2: Author: Please check keywords as typeset for correctness.Q3: Author: Please provide Palabras Clave section for the article.Q4: Author: A running head short title was not supplied; please check if this one is suitable and, if not, please supply a

short title that can be used instead.Q5: Author: Please check reference Allison et al. (2009) as typeset.Q6: Author: Please provide page range or DOI code for reference Foden et al. (2013).Q7: Author: Please provide page range or DOI code for reference Hoeke et al. (2011).Q8: Author: Please provide volume number for reference van Woesik et al. (2012).Q9: Author: Please provide publisher location for reference Veron (1995).

Q10: Author: Please provide page range or DOI code for reference Wallace et al. (2011).