Long-term variation of octocoral populations in St. John, US Virgin Islands

REPORT

Long-term variation of octocoral populations in St. John, USVirgin Islands

Elizabeth A. Lenz1,2 • Lorenzo Bramanti1,3 • Howard R. Lasker4 • Peter J. Edmunds1

Received: 17 August 2014 / Accepted: 29 May 2015

� Springer-Verlag Berlin Heidelberg 2015

Abstract The decline in abundance of scleractinian cor-

als over the past three decades in the Caribbean has raised

the possibility that other important benthic taxa, such as

octocorals, are also changing in abundance. We used

photoquadrats taken over 20 yr from reefs (7–9 m depth) at

six sites on the south coast of St. John, US Virgin Islands,

to test the hypothesis that octocorals have changed in

abundance since 1992. Octocorals were counted in 0.25 m2

photoquadrats at 2- to 3-yr intervals and identified to genus

or family. Overall, there was variation over time in popu-

lation density of octocorals (pooled among taxa, and also

separately for Antillogorgia spp., Gorgonia spp., and

plexaurids) at each site, and densities remained unchanged

or increased over 20 yr; where increases in density

occurred, the effects were accentuated after 2002. The

local-scale analysis was expanded to the Caribbean

(including the Florida Keys) by compiling data for octo-

coral densities from 31 studies for reefs at B25 m depth

between 1968 and 2013. At this scale, analyses were lim-

ited by the paucity of historical data, and despite a weak

trend of higher octocoral densities in recent decades, sta-

tistically, there was no change in octocoral abundance over

time. Together with data from the whole Caribbean, the

present analysis suggests that octocorals have not experi-

enced a decadal-scale decline in population density, which

has occurred for many scleractinian corals.

Keywords Gorgonians � Coral reefs � Octocorals �Caribbean

Introduction

In the Caribbean, large disturbances impacting coral reefs

have attracted widespread attention since the 1960s

(Hughes and Connell 1999; Cote et al. 2005; Jackson et al.

2014) with these events including hurricanes (Woodley

et al. 1981), the die-off of the echinoid Diadema antillarum

(Lessios et al. 1984), overfishing (Jackson et al. 2001),

disease outbreaks (Aronson and Precht 2001; Weil and

Rogers 2011), and large-scale bleaching due to increased

seawater temperature (Glynn 1993; Lesser 2011; Thornhill

et al. 2011). Overall, the percent cover of scleractinians has

declined *70 % throughout the region from 1970 to the

present, with a region-wide mean cover of only 17 % in

2012 (Jackson et al. 2014). The decline in cover of scler-

actinians has been associated with a reduction in topo-

graphic complexity of the benthos (Alvarez-Filip et al.

2009), which has negative implications for fish and inver-

tebrates using coral reefs as habitat (Lirman 1999; Idjadi

and Edmunds 2006).

Communicated by Ecology Editor Dr. Alastair Harborne

Electronic supplementary material The online version of thisarticle (doi:10.1007/s00338-015-1315-x) contains supplementarymaterial, which is available to authorized users.

& Elizabeth A. Lenz

[email protected]

1 Department of Biology, California State University,

18111 Nordhoff Street, Northridge, CA 91330-8301, USA

2 Hawai‘i Institute of Marine Biology, Universtity of Hawai‘i,

PO Box 1346, Kaneohe, HI 96744, USA

3 LECOB-UPMC-CNRS, Universite Pierre et Marie Curie,

UMR8222, Observatoire Oceanologique Banyuls sur mer,

18 Avenue du Fontaule, 66650 Banyuls sur mer, France

4 Department of Geology and Graduate Program in Evolution,

Ecology and Behavior, University at Buffalo, Buffalo,

NY 14260-1300, USA

123

Coral Reefs

DOI 10.1007/s00338-015-1315-x

Author's personal copy

In addition to the Caribbean-wide decline in cover of

scleractinians (Jackson et al. 2014), there is evidence from

shallow reefs (i.e., B25 m depth) that scleractinian com-

munities have changed in species composition over several

decades (Green et al. 2008; Edmunds 2013), and similar

changes have been recorded in the Pacific (Loya et al.

2001). These changes have important implications because

they suggest that the goods and services provided by coral

reef ecosystems (Moberg and Folke 1999) may also shift in

form and magnitude due to a reduction in scleractinian

species diversity (Worm et al. 2006). For example, the

replacement of massive corals like Orbicella spp. and Di-

ploria spp. in the Caribbean with weedy corals (sensu

Knowlton 2001) like Agaricia spp. and Porites spp. (Green

et al. 2008; Darling et al. 2012) could favor the formation

of reefs with a reduced ability to resist the damaging effects

of storms and an impaired ability to form structurally

complex communities (Alvarez-Filip et al. 2013).

As the global decline in cover of scleractinian corals has

intensified, many Caribbean reefs have undergone a tran-

sition favoring macroalgae over scleractinians (Hughes

1994; Rogers and Miller 2006; Roff and Mumby 2012).

However, a shift in community structure favoring

macroalgae is only one of the several possible outcomes

following a decline in cover of scleractinians, and in a few

cases, sponges, ascidians, and non-scleractinian anthozoans

have come to dominate the benthos (Norstrom et al. 2009).

For example, several studies in the Caribbean have

described increases in the abundance of sponges and

octocorals on some reefs that have experienced declines in

scleractinian cover (Norstrom et al. 2009; Ruzicka et al.

2013; Loh and Pawlik 2014). Recent changes in abundance

of octocorals on some Caribbean reefs may represent one

of the ways by which coral reef communities respond to

disturbances that have generated declines in the abundance

of scleractinians (e.g., Ruzicka et al. 2013).

While octocorals are susceptible to stressors such as

outbreaks of the disease aspergillosis (Smith et al. 1998;

Alker et al. 2001; Bruno et al. 2011), and extreme thermal

stress (Lasker 2005), in comparison with scleractinians,

they are less severely damaged by hurricanes of moderate

strength (Yoshioka and Yoshioka 1987, 1989; Witman

1992), mild increases in seawater temperature (Lasker et al.

1984; Kirk et al. 2005; Ruzicka et al. 2013), and possibly

ocean acidification (Inoue et al. 2013; Gabay et al. 2014;

Gomez et al. 2014). Moreover, like scleractinians (High-

smith 1982), at least one species of octocoral exploits

storm-mediated fragmentation for asexual reproduction

(Lasker 1984), while others can rapidly preempt vacant

space on benthic surfaces through high recruitment or rapid

linear extension (Witman 1992; Lasker et al. 2003; Ruz-

icka et al. 2013). These characteristics probably played

important roles in facilitating the threefold increase in

octocoral cover that occurred on shallow fore-reef habitats

at 3–7 m depth in the Florida Keys between 1999 and 2009

(Ruzicka et al. 2013). On these reefs, Ruzicka et al. (2013)

found that percent cover of octocorals increased 138 %

11 yr after the 1998 El Nino, while the mean percent cover

of scleractinians declined 22 %, without signs of recovery.

Against a backdrop of consistently low percent cover of

scleractinians (i.e., *4.5 %) on shallow, nearshore fring-

ing reefs in St. John, US Virgin Islands (Edmunds 2013),

the present study tested the null hypothesis that octocorals

have not changed in abundance on the same reefs over the

last 20 yr. We first assessed changes in the abundance

(colonies m-2) of octocorals on a local scale (B20 km;

Mittelbach et al. 2001) in St. John, using photoquadrats

(0.25 m2) recorded from 1992 to 2012 on shallow reefs

along *5 km of the south shore. At these sites, sclerac-

tinian cover remained *4.5 % from 1992 to 2012.

Although scleractinian cover exhibited statistically signif-

icant differences between some years, there were no

directional trends (Edmunds 2002, 2013). We then

expanded the spatial scale of the analysis to evaluate the

abundance of octocorals throughout the Caribbean [i.e., on

a regional scale of 200–4000 km; (Mittelbach et al. 2001)]

over 45 yr using data compiled from the present study as

well as from peer-reviewed and selected gray literature.

The compiled data were used to test for changes over time

in abundance of octocorals at a regional scale. Given the

long-standing, regional-scale decline in cover of sclerac-

tinians that has occurred since at least the 1970s throughout

the Caribbean (Jackson et al. 2014), even detection of

population stasis over time (i.e., an outcome other than a

reduction in population size) for octocorals would be

notable.

Materials and methods

Local scale

For the local-scale (*5 km) analysis, octocoral densities

were measured along the south shore of St. John, in the

Virgin Islands National Park (VINP) and Biosphere

Reserve (Rogers and Teytaud 1988; Rogers et al. 2008).

Within this area, coral reef community structure has been

studied since the 1950s (Randall 1961; Collette and Earle

1972) and in a systematic manner since 1987 (Rogers et al.

1991; Rogers and Beets 2001; Miller et al. 2006; Edmunds

2013). Six sites on shallow reefs (7–9 m depth) were

selected between Cabritte Horn and White Point in 1992

(Fig. 2a, with sites restricted to hard substrata but other-

wise identified based on randomly selected coordinates.

These sites have been censused annually to measure the

percent cover of macroalgae, scleractinians, and a

Coral Reefs

123

Author's personal copy

combined category of crustose coralline algae, algal turf,

and bare space (CTB; Edmunds 2013).

At each site, photoquadrats (0.25 m2) were recorded at

random positions along a fixed transect each year, with

20-m transects and 17–20 photoquadrats site-1 between

1992 and 1999, and 40-m transects and 40 photoquadrats

site-1 from 2000 to present (Edmunds 2002, 2013). Before

2000, photoquadrats were recorded with a NikonosTM V

camera (fitted with a 28-mm lens, two Nikonos SB 105

strobes, and Kodachrome 64 film) mounted on a quadrapod

that held the camera perpendicular to the seafloor (Ed-

munds 2002, 2013). In 2000, the method was upgraded to

digital images using first a 3.3 megapixel camera

(2000–2006, Nikon Coolpix 990) and then a 6.1 megapixel

camera (2007–2012, Nikon D70). The camera framer

remained unchanged throughout the sampling, and the

images allowed objects C10 mm in diameter to be

resolved. While photoquadrats from the six sites (archived

at http://mcr.lternet.edu/vinp) have been recorded annually,

photoquadrats in the present analysis were sampled at 2- to

3-yr intervals: 1992, 1994, 1997, 1999, 2002, 2004, 2007,

2009, and 2012 (n = 1630 images). Logistical constraints

precluded analyzing the photoquadrats every year for

octocoral abundance, and by subsampling every 2–3 yr, it

was possible to effectively capture temporal trends in

octocoral abundance. With this temporal resolution, we

completed a coarse-grained analysis of changes over time

in octocoral abundance at each site.

For each photoquadrat, octocoral colonies were counted

based on the presence of their holdfasts within the framer.

In some cases, other organisms, including large octocorals,

obscured the holdfasts, or holdfasts were hidden in cre-

vices. Most octocorals could not be distinguished to species

in the photographs, in part because taxonomic identifica-

tion requires inspection of sclerites, and therefore, analyses

were constrained to genera [Antillogorgia spp. (formerly

Pseudopterogorgia; Williams and Chen 2012), Briareum

spp., Eunicea spp., Erythropodium spp., Gorgonia spp.,

Muricea spp., Muriceopsis spp., Plexaura spp., Plexaurella

spp., Pseudoplexaura spp., and Pterogorgia spp.].

Encrusting Erythropodium caribaeorum and the encrusting

form of Briareum asbestinum were excluded from the

analysis, which instead focused on arborescent octocoral

(hereafter referenced as octocorals) that dominate octocoral

communities in St. John. Of the eleven octocoral genera

found on these shallow reefs, Eunicea spp., Plexaurella

spp., Pseudoplexaura spp., and Plexaura spp. could not be

distinguished from each other in the images when colonies

were small (\12 cm tall), and therefore, members of these

genera were categorized by their family (i.e., Plexauridae).

This retrospective analysis was augmented with in situ

surveys in July and August 2013 that were used to quantify

the accuracy and precision of the octocoral population

census conducted using photoquadrats. Given the chal-

lenges of quantifying arborescent colonies in planar ima-

ges, we did not expect perfect concordance between

methods, but expected to detect a strong correlation

between the approaches and quantify the underestimation

associated with the photographic technique. To compare

these census methods, octocorals at the six sites were

censused with 0.25 m2 quadrats placed along the same

40-m transect as was used for the photoquadrats

(n = 40 site-1). Photoquadrats were recorded, and in situ

counts completed at each site within 5 d of each other.

Quadrats for counts were placed at the same location where

photoquadrats were recorded, but underwater logistics

prevented perfect concordance between sampling areas.

Densities of octocorals from the in situ counts and photo-

quadrats were tested for association and concordance (de-

scribed below) using site-specific means (n = 6).

Octocorals were also surveyed in 2012 and 2013 at

Booby Rock, 1.3 km east of Cabritte Horn (Fig. 2a) where

octocorals were very abundant (Fig. 1). Data from this site

provided insight into the upper range of octocoral densities

that occur along the south shore of St. John. At this loca-

tion, octocorals were censused in situ using 20 quadrats

(1.0 9 1.0 m) placed at random points along a 40-m

transect running along the 7–9 m depth contour. As there

were no historic data for Booby Rock, octocoral densities

were not used in the contrast of octocoral abundances over

time in St. John, although they were included in the

regional-scale assessment.

Fig. 1 Representative photograph of a shallow (8 m depth) fringing

reef at Booby Rock, St. John (N18�18.1330, W64�42.6000), where

some of the highest population densities of octocorals were found in

August 2012. In this location, mean (±SE) octocoral density across

all sites was 7.97 ± 0.51 colonies m-2 in 2012, and the octocoral

fauna was represented mostly by Antillogorgia (18 % of all octoco-

rals), Gorgonia (17 %), and plexaurids (57 %). Together, these

octocorals formed a canopy *1–2 m high. For scale, the central

Dendrogyra cylindrus is *1 m tall

Coral Reefs

123

Author's personal copy

Regional scale

Octocoral densities from 1968 to 2013 were examined at a

broad spatial scale throughout the Caribbean, including the

Florida Keys (Fig. 2b) using the present surveys in St. John

in combination with densities of octocorals obtained from

multiple sources. We used 29 peer-reviewed articles and

two online sources, the Coral Reef Evaluation and Moni-

toring Project (CREMP) of the Florida Fish and Wildlife

Commission/Fish and Wildlife Research Institute (http://

ocean.floridamarine.org/FKNMS_WQPP/pages/cremp.html)

and the United Nations Educational, Scientific and Cultural

Organization (Table 15 in UNESCO Caribbean Coastal

Marine Productivity Program Data Report 1994–1995;

http://www.unesco.org/csi/pub/papers/Table15.htm). These

sources were identified from standard bibliographic search

techniques, references in key publications, and expert

referrals and are summarized in Electronic Supplementary

Materials, ESM, Table S3. The analysis was restricted to

results reporting the abundance of colonies per area on reefs

at B25 m depth. This restriction resulted in the exclusion of

several studies that have used percentage cover of octocorals

to record abundances (i.e., Ruzicka et al. 2013). E. carib-

aeorum and the encrusting form of Briareum spp. were

excluded to match the taxonomic scope of the local-scale

study.

Statistical analysis

The association between densities of octocorals obtained

from photoquadrats and in situ counts was tested with a

Pearson correlation, and the relationship between the two

methods was described using model II regression (Sokal

and Rohlf 2012). The resulting equation quantified the

75° W 70° W 65° W

25° N

60° W

20° N

15° N

10° N

Atlantic Ocean

Caribbean Sea

Gulf of Mexico

0

500 km

80° W85° W

Caribbean

St. John, USVI

St. John1 km

WhitePoint

Europa

Cabritte Horn

WestTektite

N

EastTektite

Booby Rock

WLL

300 m

GreatLameshur

Bay

18o 19.002

’

N 64o 43.453

’

W

(a)

(b)

Fig. 2 Map showing a the

study sites in St. John, US

Virgin Islands, and b the sites

for the regional-scale

assessment across the

Caribbean. The local-scale

analysis from St. John was

based on surveys at six sites at

7–9 m depth: Cabritte

Horn = N18�18.4040,W064�43.3080, west

Tektite = N18�18.7600,W064�43.3830, east

Tektite = N18�18.6820,W064�43.3860, west Little

Lameshur

(WLL) = N18�19.0280,W064�43.6670,Europa = N18�18.9760,W064�43.784, and White

Point = N18�18.9100

Coral Reefs

123

Author's personal copy

efficacy of photoquadrats for the assessment of the density

of octocorals, and it was used to evaluate the bias associ-

ated with using planar images to quantify octocorals. This

bias was not used to correct estimates of octocoral densities

obtained from photographs, because the effect was small at

low densities and inconsistent among colonies differing in

size (described below).

In the local-scale analysis, the densities of octocorals

(i.e., pooled among taxa), Antillogorgia spp., Gorgonia

spp., and plexuarids were compared among sites and years

with a mixed-model two-way ANOVA. Comparisons

among years at each site were conducted using one-way

ANOVA, first for octocorals pooled among taxa and then

for the most abundant groups, which were Antillogorgia

spp., Gorgonia spp., and plexaurids (see ESM Table S1 for

densities of octocorals and Table S2 for statistical results).

Where a significant effect of time was detected, post hoc

analyses by Tukey’s HSD tests were used to determine

which pairs of years statistically differed. Analyses of

octocoral densities at a local scale were performed with

SYSTAT version 13 (Systat Software Inc., Chicago, Illi-

nois, USA), and the statistical assumptions of normality

and homoscedasticity were tested through graphical anal-

ysis of residuals.

For the regional-scale analysis, sources of data were

heterogeneously distributed across Caribbean regions,

years, habitats, and sites. Some studies reported densities of

octocorals from multiple sites, in some cases extending

over a few kilometers and in others between more distant

localities. Moreover, different reef habitats were sampled

inconsistently and unequally among studies, with some

studies combining results from back-reef and fore-reef

habitats and others reporting results individually from

multiple habitats. Similarly, sampling intensity varied

greatly among decades, with more data from recent dec-

ades (1990–2010) than earlier decades (1960–1990).

Among regions within the Caribbean, some were heavily

sampled (e.g., the Florida Keys), while other regions were

virtually ignored (e.g., the eastern Caribbean). To assess

changes over time at the regional scale, the compiled data

were divided into two temporal groups: studies conducted

in 1998 or earlier and those conducted after 1998. This

improved the balance of the analytic design by creating two

groups with approximately equal replication. The use of

1998 as a point of demarcation also separates data collected

after the extensive bleaching that occurred in the Caribbean

in 1998 (McWilliams et al. 2005) from the earlier surveys.

The data were further classified by site (the location

reported in the study) and geographic region (generally, the

country in which the study was completed) in order to

separate the effects of other sources of variance in the data.

Given the complex and unbalanced sampling of the

wider Caribbean that is inherent in a retrospective data

compilation, the data were analyzed using a generalized

linear mixed model (GLMM, SPSS V22.0, Armonk, New

York, USA). To further resolve temporal effects in octo-

coral densities, the analyses were repeated first restricting

the data set to those regions for which there were data from

both periods (in or before 1998 versus after 1998) and then

in an additional analysis on a subset of the data restricted to

sites at which surveys were conducted in both periods. All

analyses were conducted using period (‘‘time’’ in the

accompanying tables, which corresponded to a contrast of

1998 and earlier versus after 1998), geographic region

(region) and time–region interactions as fixed effects, and

sites within regions as a random effect. Analyses in which

region was considered a random effect were conducted, but

these results are not reported as they generated identical

conclusions.

Results

Local scale

Octocoral densities recorded in situ and from photo-

quadrats were correlated for all octocorals (i.e., pooled

among taxa), as well as for Antillogorgia spp., Gorgonia

spp., and plexaurids (r C 0.82, df = 4, P B 0.044; Fig. 3).

The model II regressions describing these relationships

(0.98 C r2 C 0.68) show that in situ densities of octocorals

can be estimated from densities determined in photo-

quadrats. For all octocorals (pooled among taxa), the two

methods provided virtually identical results at low densities

(\1 colonies 0.25 m-2), at intermediate densities

(1–3 colonies 0.25 m-2) photoquadrats underestimated

actual densities by 22 %, and at high densities (3–6 colo-

nies 0.25 m-2) photoquadrats underestimated densities by

28 % (Fig. 3a). The downward bias in density estimates

obtained from photoquadrats versus in situ counts differed

among Antillogorgia spp., Gorgonia spp., and plexaurids,

and at intermediate densities for these taxa (0.18–4.00

colonies 0.25 m-2), photoquadrats yielded densities that

were *40, 4, and 21 % lower, respectively, than densities

obtained in situ.

The photoquadrats provided 102–248 records yr-1 for

the 9 yr analyzed between 1992 and 2012, and 62 % of the

photoquadrats (n = 1630) contained C1 octocoral.

Between 1992 and 2007, octocoral holdfasts were found in

51–69 % of the photoquadrats (n = 1144 photoquadrats),

but in 2009 and 2012 this proportion increased to 71–74 %

(n = 486 photoquadrats). Mean (±SE) densities of all

octocorals (i.e., pooled among taxa) between 1992 and

2013 varied from 0.71 ± 0.38 colonies m-2 at West Tek-

tite in 1997, to 18.22 ± 1.31 colonies m-2 at Cabritte

Horn in 1994 (Fig. 4a). In general, mean densities at

Coral Reefs

123

Author's personal copy

Cabritte Horn were 2–4 times greater than the others sites

in all years (P\ 0.001).

Densities of pooled octocorals changed over time at five

sites (F8,260-167 C 3.707, P\ 0.001; Fig. 4a), but not at

White Point where mean density (±SE) averaged among

years was 4.71 ± 0.28 colonies m-2 (F8,260 = 1.871,

P = 0.065). At West Little Lameshur, mean densities

(±SE) declined from 6.25 ± 1.32 colonies m-2 in 1992 to

1.68 ± 0.64 colonies m-2 (P = 0.004) in 1994 and

remained low thereafter (P C 0.726). At Cabritte Horn,

East Tektite, Europa, and West Tektite, mean densities

varied among years with large decreases between 1997 and

2002, but increases thereafter. For example, mean (±SE)

densities at Cabritte Horn were stable between 1992 and

1999 (13.68 ± 0.95 colonies m-2 averaged among years;

n = 69, P C 0.116), but decreased (5.46 ± 0.89

colonies m-2) in 2002. In 2007, mean densities increased

twofold to 13.00 ± 1.42 colonies m-2 and remained stable

between 2007 and 2012 (at 13.74 ± 0.77 colonies m-2

averaged among years; n = 124, P C 0.168). At Europa,

mean densities decreased 63 % between 1994 and 2002

(P = 0.046), then increased to 7.40 ± 1.30 colonies m-2

in 2009 (P = 0.013), and remained unchanged in 2012

(P = 0.980). At East Tektite, mean densities increased

fivefold between 1997 (1.88 ± 0.70 colonies m-2) and

2012 (10.44 ± 1.45 colonies m-2; P\ 0.001). At West

Tektite, mean densities were lowest in 1999 (0.71 ±

0.38 colonies m-2), but increased eightfold to 6.00 ±

1.06 colonies m-2 by 2009 (P = 0.001). When all six sites

were combined, mean densities increased from 5.18 ±

0.56 colonies m-2 in 1992 to 7.97 ± 0.51 colonies m-2 in

2012 (F8,1576 = 17.015, P\ 0.001; Fig. 4a red line).

When the octocoral fauna was separated by the most

common taxa in 2012 (Antillogorgia spp.,Gorgonia spp., and

plexaurids that accounted for 18, 17, and 57 % of colonies,

respectively; n = 4202), densities differed among years in

ways that varied among sites forGorgonia spp. and plexaurids

(F20,1576 C 1.018, P\ 0.001), but not for Antillogorgia spp.

(F40,1576 = 1.018, P = 0.440). These trends were consis-

tent when sites were pooled for Antillogorgia spp.

(F8,1576 = 1.123,P = 0.345),Gorgonia spp. (F8,1576 = 6.33,

P\0.001), and plexuarids (F8,1576 = 12.381, P\ 0.001;

0

1

2

3

4

5

6

7

0 1 2 3 4 5 6 7

In situ densities(colonies 0.25 m-2)

Pho

toqu

adra

t den

sitie

s(c

olon

ies

0.25

m-2)

y = 0.73x + 0.3r = 0.99P < 0.001

0

1

2

3

4

5

0 1 2 3 4 5

y = 0.64x - 0.05r = 0.82P = 0.044

y = 0.75x + 0.14r = 0.99P = 0.001

y = 0.94x + 0.09r = 0.92P = 0.011

0

0.5

1

1.5

2

0 0.5 1 1.5 2

0

0.5

1

1.5

2

0 0.5 1 1.5 2

Pooled Octocorals Antillogorgia spp.

Gorgonia spp. Plexaurids

(a) (b)

(d)(c)

Fig. 3 Scatterplot displaying

the relationships between

octocoral densities recorded

in situ versus from photographs

in 2013 on shallow reefs in St.

John for a all gorgonians

(pooled among taxa minus

Erythropodium caribaeorum

and encrusting Briareum spp.),

b Antillogorgia spp.,

c Gorgonia spp., and

d plexaurids. Values show

mean ± SE for the six sites

(Fig. 2 with n = 40 quadrats

site-1 (colonies m-2). Densities

obtained by the two methods

were significantly and positively

correlated (r C 0.82, df = 4,

P B 0.011); lines display model

II linear regression

Coral Reefs

123

Author's personal copy

Fig. 4b–d red line). Densities of Antillogorgia spp. did not

change over time at any site (Fig. 4b; F8,259-267 C 0.551,

P = 0.065), with members of this genus absent at West

Tektite in 1999, but amounting to 2.67 ± 0.97 colonies m-2

at Cabritte Horn in 1994 (Fig. 4b). Densities of Gorgonia

spp. remained unchanged over time at five sites

(F8,259-267 B 4.957,P C 0.072), but they increased from zero

in 1992 to 3.42 ± 0.38 colonies m-2 in 2012 at East Tektite

(Fig. 4c; F8,264 = 4.957, P\ 0.001). Densities of plexaurids

differed over time at all sites (F8,261-271 C 2.390,P B 0.017),

but the causes of these effects involved fluctuations in abun-

dance that differed among sites in timing, extent, and direction

of change. Plexaurid densities increased between 1992 and

2012 at Cabritte Horn (67 %), East Tektite (50 %), Europa

(166 %), and West Tektite (91), but decreased at West Little

Lameshur (36 %) and White Point (12 %).

Regional scale

Caribbean-wide data were obtained from 31 studies and

described densities of octocorals from 359 surveys at 249

sites. These data covered 16 geographic regions over 45 yr

(1968–2013), and all came from shallow (B25 m depth),

nearshore reefs (Fig. 5; ESM Table S3). The critical result

of the GLMM analysis of the entire data is that densities of

all octocorals did not differ between periods (on and before

1998 versus after 1998; P = 0.426; ESM Table S4 and

Fig. S1), while densities differed among regions

(P\ 0.001). Analyses using subsets of the data in which

there was a more equitable balance of sampling effort over

time generated identical results. For example, there was no

difference between periods when only regions with data

from both periods were analyzed (P = 0.426, ESM

Table S5). Likewise, when the analysis was restricted to

the three studies (Florida Keys, Roatan, and St. John) in

which individual sites were surveyed in both periods, there

was also no difference between periods (P = 0.374; ESM

Table S6 and Fig. S2). Regions and sites within regions had

significant effects on density of octocorals in all analyses,

but there were no interactions between region and time

(ESM Table S4, S5, S6).

Discussion

More than 60 yr of measurements of percent cover of

scleractinians on Caribbean reefs provides a compelling

summary of coral mortality caused by anthropogenic and

1

2

3

4

0

5

10

15

20

0

1

2

3

4

0

2

4

6

8

10

12

Pooled Octocorals Antillogorgia

Gorgonia Plexuarids

Year

Col

onie

s m

-2

Cabritte HornEast Tektite

EuropaWest Tektite

West Little LameshurWhite Point

1992 1997 2002 2007 2012

All Sites

H H HB B

1992 1997 2002 2007 2012

H H HB B

(a) (b)

(d)(c)

Fig. 4 Mean densities of

octocorals from 1992 to 2012 at

the six survey sites (±SE,

n = 102–248 quadrats yr-1)

used for the local-scale

assessment. Densities of

octocorals are based on analyses

of photoquadrats subsampled

from a larger sampling scheme

at 2- to 3-yr intervals with

results shown for a all

octocorals (pooled among taxa),

b Antillogorgia spp.,

c Gorgonia spp., and

d plexaurids. The red line

indicates densities of octocorals

when sites were pooled

together. Years when hurricanes

(H) and bleaching event

(B) affected the reefs in St. John

are highlighted in gray bars

Coral Reefs

123

Author's personal copy

natural disturbances (Hughes 1994; Gardner et al. 2003;

Schutte et al. 2010; Jackson et al. 2014). These losses

highlight the importance of long-term studies in detecting

changes in coral reef communities (Connell et al. 1997;

Gardner et al. 2003; Jackson et al. 2014), but to date most

studies have focused on scleractinians and macroalgae,

with little attention to changes in other benthic taxa despite

their inclusion in sampling methods (e.g., when recorded in

photoquadrats or video transects). There are exceptions to

this trend, however, with examples of abundance changing

over time for other sessile taxa including octocorals (Las-

ker et al. 1984; Yoshioka and Yoshioka 1987; Ruzicka

et al. 2013), the hydrocoral Millepora spp. (Lewis 2006;

Brown and Edmunds 2013), and sponges (Loh and Pawlik

2014; Loh et al. 2015). Overall, it is remarkable how much

of the recent history of coral reefs has been viewed through

the large monochromatic lens of ‘‘coral cover.’’ The pre-

sent study provides evidence from both local and regional

scales that octocoral populations have not declined in

abundance over the last several decades as has occurred

with scleractinians in many locations (Jackson et al. 2014).

Arguably, evidence from this study, and elsewhere (Ruz-

icka et al. 2013), suggests Caribbean octocorals are at least

maintaining their abundances, while the coral reefs around

them undergo changes in community structure, usually

resulting in declining abundance of scleractinians.

While the abundances of octocorals on St. John varied

over time and displayed site- and taxon-specific patterns, a

steady decline in population size was not recorded at any

site, regardless of taxon. When sites were pooled, there was

a trend for population densities of all octocorals (pooled

among taxa), Gorgonia spp., and plexaurids to increase

over time, at least following 2002 (Fig. 4). In St. John, the

long-term differences in variation of octocoral densities

among sites require further research to identify the mech-

anisms driving site-specific population dynamics. In com-

parison with the regional trend for declining abundances of

scleractinians in the Caribbean (Schutte et al. 2010; Jack-

son et al. 2014), it is striking that the present analyses did

not reveal declines in population densities for octocorals on

shallow reefs (7–9 m in depth) in St. John. Rather, popu-

lation densities appear to have increased following 2002,

which is similar to the trend reported for octocoral cover in

the Florida Keys (Ruzicka et al. 2013).

Changes in abundance of octocorals in the regional-

scale data compilation were not statistically discernible in

the data that we were able to compile. Given the small

number of studies, especially in the early decades of the

1968–2013 period, and the heterogeneous distribution of

these studies over time and space, the ability to statistically

test for changes over time was limited. Thus, while average

densities of octocorals were greater after than before 1998

(Fig. 5), this trend was not statistically significant. How-

ever, the absence of significant changes in the octocoral

population density (i.e., population stability) at the same

time that scleractinian abundances have declined (i.e.,

Gardner et al. 2003; Jackson et al. 2014) leads to an

increase in their relative abundance in the community.

Evidence of population stability for octocorals on con-

temporary reefs may have important implications for the

goods and services typically provided by these ecosystems.

For instance, while octocorals supply habitat provisioning

and trophic functions similar to scleractinians (Gili and

Coma 1998), they generally do not create wave-resistant

platforms like scleractinians (Lugo et al. 2000) and thus

cannot deliver equivalent capacity for coastal protection

and ecosystem engineering.

The population dynamics we report for octocorals on the

shallow reefs of St. John have occurred against a backdrop

of low scleractinian cover and increasing macroalgal cover

(Edmunds 2013). In St. John, octocoral abundance varied

between 1992 and 2012 with the greatest densities in 2012,

yet over this period the cover of scleractinians at these sites

has remained B4.5 % (Edmunds 2013). Although the cover

of scleractinians changed significantly at the present study

sites, there was no consistent directional trend (i.e., neither

up nor down) between 1992 and 2012 (Edmunds 2013). In

contrast, the mean cover of macroalgae at these sites

increased from 14.4 ± 1.7 % (±SE, n = 102) in 1992 to

27.9 ± 1.1 % (n = 240) in 2013 (Edmunds 2013; PJ

Edmunds unpublished data), which is equivalent to an

annualized linear increase of 1.1 % yr-1. Much of the

macroalgae was Dictyota spp., which can overgrow and

inhibit the growth of juvenile scleractinians (Lirman 2001;

Box and Mumby 2007), yet this increase in cover appears

not to have affected the population density of octocorals.

One feature potentially of importance in allowing octocoral

0

10

20

30

40

50

1960 1970 1980 1990 2000 2010 2020Year

Col

onie

s m

-2

Fig. 5 Scatterplot used to illustrate no significant changes in the

densities of octocorals (n = 351 surveys) from the compilation of

studies conducted throughout the Caribbean from 1968 to 2013. Gray

shading indicates densities of octocorals in or after 1998

Coral Reefs

123

Author's personal copy

populations to remain unchanged in size (or to increase)

during a period when conditions generally were unfavor-

able for scleractinians, is their fast linear growth, which can

be used to escape the constraints of a two-dimensional

benthic environment. For example, octocorals extend their

branches at 4–18 cm yr-1 (Brazeau and Lasker 1992;

Lasker et al. 2003), whereas massive scleractinians like

Orbicella annularis extend 0.6–1.1 cm yr-1 (Hudson

1981), and branching scleractinians like Acropora palmata

extend 4.7–9.9 cm yr-1 (Gladfelter et al. 1978).

In St. John, octocoral recruits (i.e., colonies B4 cm tall)

were found at all sites, and for these colonies, fast linear

growth may represent a means to reduce competition with

macroalgae, allowing them to quickly escape from benthic

surfaces rather than compete for space (as occurs with

juvenile scleractinians; Box and Mumby 2007). While

there are some cases where the potential for linear exten-

sion of scleractinians and octocorals is similar, generally

the growth of scleractinians is more detrimentally affected

by high sea surface temperature (leading to bleaching;

Thornhill et al. 2011) and macroalgal competition (Tanner

1995; Box and Mumby 2007) than in octocorals (Maida

et al.1995; Lasker et al. 2003). As these effects (i.e., warm

temperatures and high macroalgal abundance) are expected

to intensify in the coming century (Pandolfi et al. 2003), it

is possible that octocoral populations will increase in size.

Since small octocorals (B4 cm tall) were common (e.g.,

9.30 colonies m-2 at East Cabritte; n = 40) in St. John in

July and August 2013 when the present surveys were

completed, and moreover represented 19 % of octocorals at

the six sites, it is likely that recruitment was supporting

population growth. This interpretation is consistent with

the detection of even smaller octocorals (*1 cm tall) in

some of our surveys. Assuming Caribbean octocorals grow

linearly at ca. 3.90 cm yr-1 (Lasker 1990), all the small

octocoral colonies found in St. John in 2013 settled in the

previous year, and for the 1-cm-tall colonies, within the

4 months prior to the census. The location of the source

population(s) supplying the larvae to support this recruit-

ment is unknown, and indeed, the location of these source

populations is likely to vary among octocoral taxa

depending on the dispersal abilities of their larvae (Lasker

and Kim 1996; Gutierrez-Rodrıguez and Lasker 2004). The

high density of octocorals (e.g., Fig. 1) in locations

upstream from Great Lameshur Bay creates intriguing

possibilities for source populations capable of supplying

octocoral larvae to locations between Cabritte Horn and

White Point, and it might be valuable to test this possibility

through a population genetics approach (Kim et al. 2004).

Although photoquadrats used in the present study have

analytical limitations for quantifying three-dimensional

organisms, they are a reliable source in determining past

octocoral populations (Fig. 3). Discrepancies arise from

multiple sources that include limitations in detecting

holdfasts of octocorals obscured by other octocoral colo-

nies and detecting small colonies (i.e., B2 cm tall).

Moreover, the resolution of the photographs limited the

identifications of octocorals to genera, and in one case, it

was necessary to pool genera into a single family (i.e.,

plexaurids). Notwithstanding these limitations, the relia-

bility of photoquadrats in quantifying octocorals was sup-

ported by the significant correlation between octocoral

densities determined by in situ counts and from photo-

quadrats at the same locations. Photoquadrats provided

estimates of octocoral densities that were downwardly

biased depending on density, and at low densities, the bias

was small. Assuming the biases in estimating octocoral

densities from photoquadrats are consistent among years,

the limitation of this technique does not alter our principle

conclusions regarding changes in density over time.

Identifying the mechanisms driving the abundance of

octocorals in St. John is beyond the scope of the present

study, but our results suggest that a number of research

directions may be productive. First, further research is

required to enhance the resolution of studies on octocoral

abundance on both individual and percentage cover scales.

Greater resolution is required at spatial (i.e., more locations

in the Caribbean), temporal (i.e., annual increments and

greater longevity), and taxonomic (i.e., to species) scales to

detect ecological transitions favoring octocoral dominance,

the mechanisms driving such changes (and how these

might differ among species), and the implications of such

transitions. Second, the present study, as well as that of

Ruzicka et al. (2013), indirectly draws attention to two

mechanisms favoring increased octocoral abundance. One

possibility focuses on the capacity of octocoral recruits to

preempt available space on the benthos (Ruzicka et al.

2013), particularly during large, episodic recruitment

events involving octocorals (Yoshioka 1996). As the pre-

sent results suggest, this possibility is not simply preemp-

tion of space conceded by dying scleractinians, as increases

in octocoral abundance were recorded at sites in St. John

where the cover of scleractinians has remained stable and

low for at least two decades. The second possibility that

could account for increases in abundance of octocorals on

Caribbean reefs focuses on the ability of octocorals to

better compete with macroalgae, at least in comparison

with scleractinians (Tanner 1995; Lirman 2001), and even

outcompete other benthic taxa (Sebens and Miles 1988)

including scleractinian recruits (Maida et al. 1995, 2001).

Octocorals can avoid spatial competition with macroalgae

through rapid vertical growth (Sanchez et al. 2004), and the

exploitation of tree-like morphologies (sensu Jackson

1979) facilitates growth into the seawater column with

minimal resources committed to benthic attachment on

hard substrate. Overall, it is clear that there is more to the

Coral Reefs

123

Author's personal copy

changing community ecology of Caribbean reefs than what

has been depicted by analyses of cover of scleractinian

corals and macroalgae alone, and the present study draws

attention to the roles played by octocorals in these changes.

Acknowledgments This project was funded by the US National

Science Foundation grants DEB 03-43570, DEB 08-41441, DEB

13-50146, OCE 13-32915, and OCE 13-34052 and was conducted

under permits issued by the Virgin Islands National Park (most

recently VIIS-2013-SCI-0015). This research was submitted in partial

fulfillment of the MS degree to E.A.L. We are grateful to R. Brown,

R. Fish, and the staff of the Virgin Islands Ecological Resource

Station for making our visits productive and enjoyable. Fieldwork

would have been impossible without the assistance of many students,

and in 2012 and 2013, these included A. Yarid, and K. Privitera-

Johnson, with assistance from C. Didden. We are also grateful to N.

Evensen for assistance in organizing the data compilation. We would

like to thank the three anonymous reviewers and A. Harborne for

comments that improved an earlier draft of this manuscript. This is

contribution number 229 of the marine biology program of California

State University, Northridge.

References

Alker AP, Smith GW, Kim K (2001) Characterization of Aspergillus

sydowii (Thom et Church), a fungal pathogen of Caribbean sea

fan corals. Hydrobiol 460:105–111

Alvarez-Filip L, Carricart-Ganivet JP, Horta-Puga G, Iglesias-Prieto

R (2013) Shifts in coral-assemblage composition do not ensure

persistence of reef functionality. Sci Rep 3:3486

Alvarez-Filip L, Dulvy NK, Gill JA, Cote IM, Watkinson AR (2009)

Flattening of Caribbean coral reefs: region-wide declines in

architectural complexity. Proc R Soc Lond B Biol Sci

276:3019–3025

Aronson RB, Precht WF (2001) White-band disease and the changing

face of Caribbean coral reefs. Hydrobiol 159:25–38

Box SJ, Mumby PJ (2007) Effect of macroalgal competition on

growth and survival of juvenile Caribbean corals. Mar Ecol Prog

Ser 342:139–149

Brazeau DA, Lasker HR (1992) Reproductive success in the Caribbean

octocoral Briareum asbestinum. Mar Biol 114:157–163

Brown DJ, Edmunds PJ (2013) Long-term changes in the population

dynamics of the Caribbean hydrocoral Millepora spp. J Exp Mar

Bio Ecol 441:62–70

Bruno JF, Ellner SP, Vu I, Kim K, Harvell CD (2011) Impacts of

aspergillosis on sea fan coral demography: modeling a moving

target. Ecol Monogr 81:123–139

Collette BB, Earle SA (1972) Results of the Tektite Program: ecology

of coral-reef fishes. Los Angeles County Natural History

Museum Science Bulletin 14:180p

Connell JH, Hughes TP, Wallace CC (1997) A 30-year study of coral

abundance, recruitment, and disturbance at several scales in

space and time. Ecol Monogr 67:461–488

Cote IM, Gill JA, Gardner TA, Watkinson AR (2005) Measuring

coral reef decline through meta-analyses. Philos Trans R Soc B

Biol Sci 360:385–395

Darling ES, Alvarez-Filip L, Oliver TA, McClanahan TR, Cote IM

(2012) Evaluating life-history strategies of reef corals from

species traits. Ecol Lett 15:1378–1386

Edmunds PJ (2002) Long-term dynamics of coral reefs in St. John,

US Virgin Islands. Coral Reefs 21:357–367

Edmunds PJ (2013) Decadal-scale changes in the community

structure of coral reefs of St. John, US Virgin Islands. Mar Ecol

Prog Ser 489:107–123

Gabay Y, Fine M, Barkay Z, Benayahu Y (2014) Octocoral tissue

provides protection from declining oceanic pH. PLoS One

9:e91553

Gardner T, Cote IM, Gill JA, Grant A, Watkinson AR (2003) Long-

term region-wide declines in Caribbean corals. Science

301:958–960

Gili JM, Coma R (1998) Benthic suspension feeders: their paramount

role in littoral marine food webs. Trends Ecol Evol 13:316–321

Gladfelter EH, Monahan RK, Gladfelter WB (1978) Growth rates of

five reef-building corals in the Northeastern Caribbean. Bull Mar

Sci 28:728–734

Glynn PW (1993) Coral reef bleaching: ecological perspectives. Coral

Reefs 12:1–17

Gomez CE, Paul VJ, Ritson-Williams R, Muehllehner N, Langdon C,

Sanchez JA (2014) Responses of the tropical gorgonian coral

Eunicea fusca to ocean acidification conditions. Coral Reefs

34:451–460

Green DH, Edmunds PJ, Carpenter RC (2008) Increasing relative

abundance of Porites astreoides on Caribbean reefs mediated by

an overall decline in coral cover. Mar Ecol Prog Ser 359:1–10

Gutierrez-Rodrıguez C, Lasker HR (2004) Reproductive biology,

development, and planula behavior in the Caribbean gorgonian

Pseudopterogorgia elisabethae. Invert Biol 123:54–67

Highsmith RC (1982) Reproduction by fragmentation in corals. Mar

Ecol Prog Ser 7:207–226

Hudson HJ (1981) Growth rates in Montastraea annularis: a record of

environmental change in Key Largo Coral Reef Sanctuary,

Florida. Bull Mar Sci 31:444–459

Hughes TP (1994) Catastrophes, phase shifts, and large-scale

degradation of a Caribbean coral reef. Science 265:1547–1551

Hughes TP, Connell JH (1999) Multiple stressors on coral reefs: a

long-term perspective. Limnol Oceanogr 44:932–940

Idjadi JA, Edmunds PJ (2006) Scleractinian corals as facilitators for

other invertebrates on a Caribbean reef. Mar Ecol Prog Ser

319:117–127

Inoue S, Hayanne H, Yamamoto S, Kurihara H (2013) Spatial

community shift from hard to soft corals in acidified water. Nat

Clim Chang 3:683–687

Jackson JBC (1979) Morphological strategies of sessile animals. In:

Larwood G, Rosen BR (eds) Biology and systematics of colonial

organisms. Academic Press, London, pp 499–555

Jackson JBC, Kirby MX, Berger WH, Bjorndal KA (2001) Historical

overfishing and recent collapse of coastal ecosystems. Science

293:629–637

Jackson JBC, Donovan MK, Cramer K, Lam V (2014) Status and

trends of Caribbean coral reefs: 1970-2012. Global Coral Reef

Monitoring Network, IUCN, Gland, Switzerland

Kim E, Lasker HR, Coffroth MA, Kim K (2004) Morphological and

genetic variation across reef habitats in a broadcast-spawning

octocoral. Hydrobiol 530:423–432

Kirk NL, Ward JR, Coffroth MA (2005) Stable Symbiodinium

composition in the sea fan Gorgonia ventalina during temper-

ature and disease stress. Biol Bull 209:227–234

Knowlton N (2001) The future of coral reefs. Proc Natl Acad Sci

98:5419–5425

Lasker HR (1984) Asexual reproduction, fragmentation, and skeletal

morphology of a plexaurid gorgonian. Mar Ecol Prog Ser

19:261–268

Lasker HR (1990) Clonal propagation and population dynamics of a

gorgonian coral. Ecology 71:1578–1589

Lasker HR (2005) Gorgonian mortality during a thermal event in the

Bahamas. Bull Mar Sci 76:155–162

Coral Reefs

123

Author's personal copy

Lasker HR, Kim K (1996) Larval development and settlement

behavior of the gorgonian coral Plexaura kuna (Lasker, Kim,

and Coffroth). J Exp Mar Bio Ecol 207:161–175

Lasker HR, Peters EC, Coffroth MA (1984) Bleaching of reef

coelenterates in the San Blas Islands, Panama. Coral Reefs

3:183–190

Lasker HR, Boller ML, Castanaro J, Sanchez JA (2003) Determinate

growth and modularity in a gorgonian octocoral. Biol Bull

205:319–330

Lesser MP (2011) Coral bleaching: causes and mechanisms. In:

Dubinsky Z, Stambler N (eds) Coral reefs: an ecosystem in

transition. Springer pp 405–419

Lessios HA, Robertson DR, Cubit JD (1984) Spread of Diadema mass

mortality through the Caribbean. Science 226:335–337

Lewis CL (2006) Biology and ecology of the hydrocoral Millepora on

coral reefs. Adv Mar Biol Rev 50:1–55

Lirman D (1999) Reef fish communities associated with Acropora

palmat relationships to benthic attribute. Bull Mar Sci

65:235–252

Lirman D (2001) Competition between macroalgae an corals: effects

of herbivore exclusion and increased algal biomass on coral

survivorship. Coral Reefs 19:392–399

Loh TL, Pawlik JR (2014) Chemical defenses and resource trade-offs

structure sponge communities on Caribbean coral reefs. Proc

Natl Acad Sci USA 111:4151–4156

Loh TL, McMurray SE, Henkel TP, Vicente R, Pawlik JR (2015)

Indirect effects of overfishing on Caribbean reefs: sponges

overgrow reef-building corals. PeerJ 3:e901

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

Lugo AE, Rogers CS, Nixon SW (2000) Hurricanes, coral reefs, and

rainforests: resistance, ruin and recovery in the Caribbean.

AMBIO: A Journal of the Human. Environment 29:106–114

Maida M, Sammarco PW, Coll JC (1995) Effects of soft corals on

scleractinian coral recruitment. I: Directional allelopathy and

inhibition of settlement. Mar Ecol Prog Ser 121:191–202

Maida M, Sammarco PW, Coll JC (2001) Effects of soft corals on

scleractinian coral recruitment. II: Allelopathy, spat survivorship

and reef community structure. Mar Ecol 22:397–414

McWilliams JP, Cote IM, Gill JA, Sutherland WJ, Watkinson AR

(2005) Accelerating impacts of temperature-induced coral

bleaching in the Caribbean. Ecology 86:2055–2060

Miller J, Waara R, Muller E, Rogers C (2006) Coral bleaching and

disease combine to cause extensive mortality of reefs in US

Virgin Islands. Coral Reefs 25:418

Mittelbach GG, Steiner CF, Scheiner SM, Gross KL, Reynolds HL,

Waide RB, Willig MR, Dodson SI, Gough L (2001) What is the

observed relationship between species richness and productivity?

Ecology 82:2381–2396

Moberg F, Folke C (1999) Ecological good and services of coral reef

ecosystems. Ecol Econ 29:215–233

Norstrom AV, Nystrom M, Lokrantz J, Folke C (2009) Alternative

states on coral reefs: beyond coral-macroalgal phase shifts. Mar

Ecol Prog Ser 376:295–306

Pandolfi JM, Bradbury RH, Sala E, Hughes TP, Bjorndal KA, Cooke

RG, McArdle D, McClenachan L, Newman MJG, Paredes G,

Warner RR, Jackson JBC (2003) Global trajectories of the long-

term decline of coral reef ecosystems. Science 301:955–958

Randall JE (1961) Overgrazing of algae by herbivorous marine fishes.

Ecology 42:812

Roff G, Mumby PJ (2012) Global disparity in the resilience of coral

reefs. Trends Ecol Evol 27:404–413

Rogers CS, Teytaud R (1988) Marine and terrestrial ecosystems of the

Virgin Islands National Park and Biosphere Reserve. Biosphere

Reserve Rep 29:1–112

Rogers CS, Beets J (2001) Degradation of marine ecosystems and

decline of fishery resources in marine protected areas in the US

Virgin Islands. Environ Conserv 28:312–322

Rogers CS, Miller J (2006) Permanent ‘phase shifts’ or reversible

declines in coral cover? Lack of recovery of two coral reefs in St.

John, US Virgin Islands. Mar Ecol Prog Ser 306:103–114

Rogers CS, McLain LN, Tobias CR (1991) Effects of Hurricane Hugo

(1989) on a coral reef in St. John, USVI. Mar Ecol Prog Ser

78:189–199

Rogers CS, Miller J, Muller EM, Edmunds PJ (2008) Ecology of coral

reefs in the US Virgin Islands. In: Riegle BML, Dodge RE (eds)

Coral reefs of the USA. Springer, Dordrecht, pp 303–373

Ruzicka RR, Colella MA, Porter JW (2013) Temporal changes in

benthic assemblages of Florida Keys reefs 11 years after the

1997/1998 El Nino. Mar Ecol Prog Ser 489:125–141

Sanchez JA, Lasker HR, Nepomuceno EG, Sanchez JD, Woldenberg

MJ (2004) Branching and self-organization in marine modular

organisms: a model. Am Nat 163:24–39

Schutte VGW, Selig ER, Bruno JF (2010) Regional spatio-temporaltrends in Caribbean coral reef benthic communities. Mar Ecol

Prog Ser 402:115–122

Sebens KP, Miles JS (1988) Sweeper tentacles in a gorgonian

octocoral: morphological modifications for interference compe-

tition. Biol Bull 175:378–387

Smith GW, Harvel CD, Kiho K (1998) Response of sea fans to

infection with Aspergillus sp. (Fungi). Rev Biol Trop 5:205–208

Sokal RR, Rohlf FJ (2012) Biometry, 4th edn. Freeman, New York

Tanner JE (1995) Competition between scleractinian corals and

macroalgae: an experimental investigation of coral growth,

survival and reproduction. J Exp Mar Bio Ecol 190:151–168

Thornhill DJ, Rotijan RD, Todd BD, Chilcoat GC, Iglesias-Prieto R,

Kemp DW, LaJeunesse TC, Reynolds JM, Schmidt GW,

Shannon T, Warner ME, Fitt WK (2011) A connection between

colony biomass and death in Caribbean reef-building corals.

PLoS One 6:e29535

Weil E, Rogers CS (2011) Coral reef diseases in the Atlantic-

Caribbean. In: Dubinsky Z, Stambler N (eds) Coral reefs: an

ecosystem in transition. Springer pp 465–491

Williams GC, Chen JY (2012) Resurrection of the octocorallian genus

Antillogorgia for Caribbean species previously assigned to

Pseudopterogorgia and a taxonomic assessment of the relation-

ship of these genera with Leptogorgia (Cnidara, Anthozoa,

Gorgoniidae). Zootaxa 3505:39–52

Witman JD (1992) Physical disturbance and community structure of

exposed and protected reefs: a case study from St. John, US

Virgin Islands. Am Zool 32:641–654

Woodley JD, Chornesky EA, Cliffo PA, Jackson JBC, Kaufman LS,

Knowlton N, Lang JC, Pearson MP, Porter JW, Rooney MC,

Rylaarsdam KW, Tunnicliffe VJ, Wahle CM, Wulff JL, Curtis

ASG, Dallmeyer MD, Jupp BP, Koehl MAR, Neigel J, Sides EM

(1981) Hurrican Allen’s impact on Jamaican coral reefs. Science

214:749–755

Worm B, Barbier EB, Beaumont N, Duffy JE, Folke C, Halpern BS,

Jackson JBC, Lotze HK, Micheli F, Palumbi SR, Sala E, Selkoe

KA, Stachowicz JJ, Watson R (2006) Impacts of biodiversity

loss on ocean ecosystem services. Science 314:787–790

Yoshioka PM (1996) Variable recruitment and its effects on the

population and community structure of shallow-water gorgoni-

ans. Bull Mar Sci 59:433–443

Yoshioka PM, Yoshioka BB (1987) Variable effects of hurricane

David on the shallow water gorgonians of Puerto Rico. Bull Mar

Sci 40:132–144

Yoshioka PM, Yoshioka BB (1989) Effects of wave energy,

topographic relief and sediment transport on the distribution of

shallow-water gorgonians in Puerto Rico. Coral Reefs 8:145–152

Coral Reefs

123

Author's personal copy