Content uploaded by Elizabeth R Selig
Author content
All content in this area was uploaded by Elizabeth R Selig on Jan 17, 2014
Content may be subject to copyright.
MARINE ECOLOGY PROGRESS SERIES
Mar Ecol Prog Ser
Vol. 402: 115–122, 2010
doi: 10.3354/meps08438
Published March 8
INTRODUCTION
Based on a recently published global map of the
magnitude and geographic extent of 17 anthropogenic
threats, Halpern et al. (2008) argued that coral reefs
are one of the world’s most heavily impacted marine
ecosystems. This finding is consistent with a vast body
of literature documenting the global degradation of
tropical reefs over the last several decades (Wilkinson
1992, Grigg 1994, Gardner et al. 2003, Bruno & Selig
2007, Edmunds & Elahi 2007). Coral loss can lead to
reductions in fish abundance and diversity (Jones et al.
2004, Pratchett et al. 2008) and declines in topographi-
cal complexity (Alvarez-Filip et al. 2009). Coral loss is
also associated with compensatory increases in the
abundance of several other taxa, including sponges,
© Inter-Research 2010 · www.int-res.com*Email: vschutte@uga.edu
Regional spatio-temporal trends in Caribbean
coral reef benthic communities
Virginia G. W. Schutte
1, 3,
*
, Elizabeth R. Selig
1, 2, 4
, John F. Bruno
1
1
Department of Marine Sciences, The University of North Carolina at Chapel Hill, 340 Chapman Hall CB# 3300, Chapel Hill,
North Carolina 27599-3300, USA
2
Curriculum for the Environment and Ecology, 207 Coastes Building CB# 3275, The University of North Carolina at Chapel
Hill, Chapel Hill, North Carolina 27599-3275, USA
3
Present address: Odum School of Ecology, The University of Georgia, 140 E. Green St., Athens, Georgia 30602-2202, USA
4
Present address: Center for Applied Biodiversity Science, Conservation International, 2011 Crystal Drive, Suite 500,
Arlington, Virginia 22202, USA
ABSTRACT: Coral cover has declined on reefs worldwide with particularly acute losses in the
Caribbean. Despite our awareness of the broad-scale patterns and timing of Caribbean coral loss,
there is little published information on: (1) finer-scale, subregional patterns over the last 35 yr,
(2) regional-scale trends since 2001, and (3) macroalgal cover changes. We analyzed the spatio-
temporal trends of benthic coral reef communities in the Caribbean using quantitative data from 3777
coral cover surveys of 1962 reefs from 1971 to 2006 and 2247 macroalgal cover surveys of 875 reefs
from 1977 to 2006. A subset of 376 reefs was surveyed more than once (monitored). The largest 1 yr
decline in coral cover occurred from 1980 to 1981, corresponding with the beginning of the
Caribbean-wide Acropora spp. white band disease outbreak. Our results suggest that, regionally,
coral cover has been relatively stable since this event (i.e. from 1982 to 2006). The largest increase in
macroalgal cover was in 1986, 3 yr after the regional die-off of the urchin grazer Diadema antillarum
began. Subsequently, macroalgal cover declined in 1987 and has been stable since then. Regional
mean (±1 SE) macroalgal cover from 2001 to 2005 was 15.3 ± 0.4% (n = 1821 surveys). Caribbean-
wide mean (±1 SE) coral cover was 16.0 ± 0.4% (n = 1547) for this same time period. Both macroalgal
and coral cover varied significantly among subregions from 2001 to 2005, with the lowest coral cover
in the Florida Keys and the highest coral cover in the Gulf of Mexico. Spatio-temporal patterns from
the subset of monitored reefs are concordant with the conclusions drawn from the entire database.
Our results suggest that coral and macroalgal cover on Caribbean reef benthic communities has
changed relatively little since the mid-1980s.
KEY WORDS: Coral cover · Macroalgae · Coral disease · Coral bleaching
Resale or republication not permitted without written consent of the publisher
OPENPEN
ACCESSCCESS
Mar Ecol Prog Ser 402: 115–122, 2010
gorgonians, and macroalgae (Hughes 1994, Aronson et
al. 2002, Norström et al. 2009). A wide variety of causes
and human activities have been implicated in driving
these changes, including overfishing,
increasing ocean temperatures, coral
disease and predator outbreaks, and
poor land-use practices that lead to sed-
imentation and nutrient pollution
(Endean 1977, Jackson et al. 2001,
Hughes et al. 2003, Fabricius 2005,
Hoegh-Guldberg et al. 2007).
Although the key drivers of coral loss
have been identified, there is no con-
sensus on their relative importance and
how this varies in space and time
(Grigg & Dollar 2005, Precht et al.
2005). Quantifying the spatio-temporal
changes in coral reef benthic communi-
ties at regional and decadal scales can
lead to a broader understanding of the
patterns and causes of reef degradation
and provide information which will
result in more effective management
actions (Côté et al. 2005).
The purpose of the present study was
to quantify the regional-scale trends in
coral and macroalgal cover on Carib-
bean coral reefs over the last 35 yr in
each of 7 subregions (Fig. 1A). Our
study builds on previous analyses of
benthic changes on particular reefs
(Edmunds & Elahi 2007), whole islands
(Hughes 1994), and a meta-analysis of
the entire region (Gardner et al. 2003).
Although the broad-scale patterns and
timing of Caribbean coral loss were dis-
cussed in Gardner et al. (2003), there is
no published information on regional-
scale trends for macroalgal cover or for
coral cover since 2001. Our analysis
was based on data from 3777 surveys of
1962 reefs performed between 1971
and 2006 — the largest compilation of
quantitative Caribbean reef surveys to
date.
MATERIALS AND METHODS
All analyses were based on quantita-
tive in situ surveys of subtidal coral
reefs in the Caribbean basin that mea-
sured the percentage of the bottom cov-
ered by living scleractinian coral tissue.
We grouped these survey sites into 7
subregions, which were similar to previous delineation
schemes and were based on areas of similar biodiver-
sity and biogeography, major bathymetric changes, or
116
Fig. 1. (A) Locations of survey sites (purple dots; n = 1667) and delineations of the 7
subregions used in our analyses. The 295 sites for which we could not obtain ex-
act coordinates are not shown. (B) Number of thermal stress anomalies during
2005. Sea surface temperature data were derived from the Advanced Very High
Resolution Radiometer sensor and were processed to a resolution of approxi-
mately 4.6 km at the equator. An anomaly is defined as 1°C more than the typical
maximum climatological week defined for each grid cell (see also Text S3 in the
supplement, www.int-res.com/articles/suppl/m402p115_app. pdf)
Schutte et al.: Caribbean reef trends
management regimes (Fig. 1A; Spalding et al. 2007).
The coral and macroalgal cover survey database was
compiled by searching the scientific literature using a
variety of search engines (e.g. ISI Web of Science and
Google Scholar) using terms including ‘Caribbean’
and ‘coral cover’. We also examined every issue of
Coral Reefs, Atoll Research Bulletin, the Proceedings
of the Colloquium on Global Aspects of Coral Reefs
(Ginsburg 1993), and the Proceedings of the Interna-
tional Coral Reef Symposiums. When cover data were
presented in graphical form, we used ImageJ (Rasband
2006) to extract the raw data (the percent cover). In
addition, we obtained survey data directly from sev-
eral monitoring programs (Text S1, Fig. S1 & Table S1
in the supplement, www.int-res.com/articles/suppl/
m402p115_app.pdf), including the Florida Coral Reef
Evaluation and Monitoring Project (CREMP; data com-
prise 35.6% of surveys and 13.1% of sites), the Atlantic
and Gulf Rapid Reef Assessment (AGRRA) Program
(17.1% of surveys and 30.3% of sites), and Reef Check
(16.5% of surveys and 22.8% of sites).
The database included 3777 coral cover surveys from
1962 sites surveyed between 1971 and 2006 (Text S1 &
Table S2 in the supplement, www.int-res.com/articles/
suppl/m402p115_app.pdf). We pooled observations
across depths (1 to 30 m depth, 8.2 ± 0.08 m, mean ±
1 SE; Fig. S2 in the supplement, www.int-res.com/
articles/suppl/m402p115_app.pdf) and reef zones for
our analyses because there was little variation in sur-
vey depth among subregions or years and there was
no relationship between depth and cover (Text S2 &
Fig. S3 in the supplement, www.int-res.com/articles/
suppl/m402p115_app.pdf). The cover of Millepora spp.
(hydrozoans) was included in the total live coral cover
value for 181 surveys (4.1% of all surveys). Macroalgal
cover, including both fleshy and calcifying macroalgae
(sensu Steneck 1988), was measured in 2247 surveys
on 875 reefs (58.4 and 44.6% of the total, respectively).
We did not include cover from turf or coralline encrust-
ing algae. In 64.6% of surveys, benthic cover was mea-
sured in situ using related techniques such as the
AGRRA Program methodology (Lang 2003) and point
intercept cover estimates (Hodgson et al. 2006, Lam et
al. 2006). For the other 35.4%, the substrate was re-
corded with video or still photographs (Edmunds &
Bruno 1996), which were analyzed in the laboratory
using image analysis software or the point count tech-
nique (Aronson et al. 2002, Idjadi & Edmunds 2006,
Lam et al. 2006, Rogers & Miller 2006).
Annual sample sizes for the whole region (~200 to
400 surveys; Fig. 2) and for many subregions (Table S1)
over the last decade are quite large, and most survey
sites were more or less randomly or haphazardly se-
lected (particularly with respect to the hypotheses being
tested in this analysis). In other words, we used survey
data regardless of the original research purpose. There-
fore, our analyses are likely representative of the broad-
scale trends on benthic reef communities, particularly
since the mid-1990s. Yet meta-analyses of field surveys
such as this have a number of limitations, and several
caveats need to be considered when interpreting the re-
sults (Bruno & Selig 2007). First, year-to-year variations
in which sites are sampled can cause apparent short-
term temporal trends that are likely sampling artifacts
and not real phenomena. As in any long-term analysis,
short-term and temporary fluctuations are expected,
may not be real, and should not be over-interpreted.
Second, for a variety of reasons, some sites and sub-
regions have been sampled far more intensively than
others. We used a repeated measures linear regression
analysis to test the null hypothesis that there was
no relationship between percent coral or macroalgal
cover and time (using Stata Version 9.1, STATA). This
test was based on the individual trajectories of the 7
subregions rather than pooled data for each year. Per-
forming this analysis on yearly subregional averages
equalizes the influence of each subregion. However, it
does not remove the influence of highly sampled sites
on subregional values. Additionally, some important
locations were sampled comparatively sparsely rela-
tive to their size, e.g. Belize. Therefore, the regional
and subregional trends may not be representative of
all parts of the Caribbean. Our general trends also
might differ from the true, but unknowable, population
values. The population-level trends we quantified are
not meant to represent the trajectory of any individual
reef. Additionally, trends from the early years of the
database should be interpreted with great caution
given their relatively small annual sample sizes. We
still know very little about the natural baselines of
Caribbean reefs and even less about their state even in
the near-past, e.g. the 1960s. Thus, putting our results
into a broader historical context is very difficult. It is
possible that Caribbean reefs began to degrade long
before we began surveying them (Pandolfi et al. 2003),
in which case the actual degree of degradation is much
greater than we can currently quantify.
The database largely consisted of reefs surveyed only
once, but included 376 reefs surveyed in 2 or more
years. We analyzed this subset of monitored reefs sepa-
rately from the entire database, restricting the analysis
to surveys performed between 1996 and 2006 (n = 331
reefs; Table S3 in the supplement, www.int-res.com/
journals/articles/m402p115_app.pdf), because few
reefs were monitored before this time period. Re-
peated-measures regression analysis was used to de-
termine whether coral or macroalgal cover changed
significantly from 1996 to 2006. We performed separate
analyses for each subregion and also an analysis of all
the monitoring data combined.
117
Mar Ecol Prog Ser 402: 115–122, 2010
RESULTS AND DISCUSSION
Regional temporal trends
Numerous studies indicate that Caribbean reef ben-
thic communities changed dramatically in the 1980s
(Hughes et al. 1985, Carpenter 1990, Liddell & Ohl-
horst 1992, Hughes 1994, Gardner et al. 2003). Despite
the general perception that Caribbean reefs have con-
tinued to degrade, the broad-scale cover of hard corals
and macroalgae appears to have changed very little
since at least the mid-1980s. Our analysis suggests that
the regional coral cover average has not changed sig-
nificantly since the 1981 Acropora spp. mass mortality
(i.e. from 1982 to 2006, p = 0.99, based on repeated-
measures regression of annual subregional means;
Fig. 2A). This pattern of regional stasis does not contra-
dict studies that have documented recent, i.e. post-
1980s, coral loss on individual reefs (e.g. Aronson et
al. 2002, Edmunds 2007) and in some subregions such
as the Florida Keys (Fig. S4 in the supplement, www.
int-res.com/articles/suppl/m402p115_app.pdf; Porter &
Meier 1992, Maliao et al. 2008). Instead it suggests that
local losses on some reefs have apparently been
roughly balanced by local coral recovery on other
reefs. In other words, the observed pattern of regional
stasis is probably a dynamic equilibrium, masking
greater spatio-temporal variance in benthic commu-
nity structure at finer spatial scales. Additionally,
because we were only able to analyze broad-scale pat-
terns in total coral and macroalgal cover, we could not
evaluate other reef characteristics such as trophic com-
plexity (Paddack et al. 2009), reef rugosity (Alvarez-
Filip et al. 2009), or changes in species composition
(e.g. Aronson et al. 2004), which may indeed have
changed during the study period and are also impor-
tant indicators of coral reef health.
Disturbances before the mid-1980s caused substan-
tial regional losses in coral cover. Caribbean-wide
coral cover averages were highest from 1971 to 1980
(Fig. 2A). Annual regional means during this period
ranged from ~25 to 40% (n = 43 surveys), which is
somewhat lower than generally assumed (Gardner et
al. 2003). Our results suggest that regional absolute
coral cover declined by ~18% between 1972, the year
with the highest yearly coral cover mean (38.3%, n = 1
118
40
50
60
300
400
500
40
50
60
400
500
0
10
20
30
0
Q
Q
100
200
0
10
20
30
40
1970 1975 1980 1985 1990 1995 2000 2005 1970 1975 1980 1985 1990 1995 2000 2005
0
100
200
300
A) Coral B) Macroalgae
D) FLK coralC) Coral-no FLK
Year
Cover (%)
No. of studies
Fig. 2. Annual cover values (±1 SE, closed circles, left y-axis) and site sample sizes (open circles, right y-axis) for (A) mean coral
cover for all sites in the Caribbean basin (n = 1962; star: 1980, the year in which Hurricane Allen struck and white band disease
outbreaks began); (B) mean macroalgal cover for all sites for which data were available (n = 875; star: 1983, the year in which the
Diadema antillarum die-off began); (C) mean coral cover for all sites in the greater Caribbean except those in the Florida Keys
(FLK; n = 1515); and (D) mean coral cover for all sites in the FLK subregion (n = 447)
Schutte et al.: Caribbean reef trends
survey), and 1982 (20.8 ± 4.2%, n = 7 surveys), when
the current period of regional stasis began (Fig. 2A).
However, the annual coral cover means during this
period are based on a small number of surveys
(Table S2) and it is possible that the historical regional
baseline was higher. The largest 1 yr decline in coral
cover (24.9 ± 3.2% absolute coverage) took place from
1980 to 1981, coincident with and presumably due to
the regional Acropora spp. die-off from the white band
disease epizootic (Aronson & Precht 2001, 2006) and
the passage of Hurricane Allen through Jamaica
(Woodley et al. 1981), where 9 of the 21 surveys con-
ducted in 1980 and 1981 were performed.
Just over half of the coral cover surveys had accom-
panying macroalgal cover values; our database inclu-
ded 2247 macroalgal surveys conducted between 1977
and 2006. Macroalgal cover did not change between
1987 and 2006 (p = 0.13; Fig. 2B), although the small
number of quantitative surveys in the 1980s and early
1990s makes estimates of subregion-specific macroal-
gal trends during this period less reliable (Table S2). It
is possible that some other rarely measured attribute of
macroalgae changed during the period of stasis, e.g.
biomass, height, or composition, although macroalgal
cover is generally a good predictor of macroalgal bio-
mass (Miller et al. 2003).
The regional mean coverage of macroalgae increa-
sed dramatically in 1986 (26.0 ± 13.3%; Fig. 2B) follow-
ing the Caribbean-wide die-off of the sea urchin Dia-
dema antillarum in 1983 and 1984 (Hughes et al. 1985,
Lessios 1988). Mean macroalgal cover before the D.
antillarum die-off (based on 37 surveys performed
from 1977 to 1983) was 8.0 ± 1.6%. In 1986, macroalgal
cover increased to 38.1 ± 13.3% (n = 8 surveys). How-
ever, it declined again to 14.5% in 1987 (± 5.7, n = 13
surveys) and the regional mean remained below 20%
for most years between 1987 and 2006. The drop in
macroalgal cover following the spike in 1986 could be
due to compensatory population increases or behav-
ioral responses by other fish and urchin grazers to the
loss of the once dominant herbivore, D. antillarum
(Aronson et al. 2000, Haley & Solandt 2001). Popula-
tions of D. antillarum have since recovered on some
Caribbean reefs (e.g. Carpenter & Edmunds 2006,
Myhre & Acevedo-Gutiérrez 2007), which could ex-
plain the general absence of macroalgal cover changes
since 1987. Conversely, the 1986 spike in macroalgal
cover could be an artifact of non-random site selection;
many macroalgal cover studies conducted in the mid-
1980s focused on reefs that experienced significant
losses in coral cover or were designed to document the
indirect effects of the D. antillarum die-off (e.g.
Hughes 1994).
The combined multi-decade, regional patterns of
changes in coral and macroalgal cover indicate that,
although the region has experienced substantial coral
losses, there has not been a concomitant increase in
macroalgal cover (Fig. 3). Almost half (48.9%) of the
2247 macroalgal surveys documented a higher percen-
tage of macroalgal cover than coral cover, but macroal-
gal cover has rarely exceeded 50% (just 5.2% of sur-
veys). The observed regional increase in macroalgal
cover in 1986 occurred 5 yr after the collapse of coral
cover in 1981, supporting the argument that coral loss
in the 1980s was not caused by an increase in macro-
algal cover (Aronson & Precht 2006, Bruno et al. 2009).
The relationship between coral and macroalgal
cover on reefs over time can be divided into 3 temporal
categories (Fig. 3): (1) the late 1970s, the baseline for
the present study; (2) the 1980s and early 1990s, after
the Acropora spp. and Diadema antillarum disease
outbreaks; and (3) from 1993 to 2006, a period of post-
disease stability. We pooled the annual means within
each temporal period in order to compare the 3 peri-
ods, and the greatest coral cover loss occurred from
Time Period 1 to Time Period 2. There was not a long-
term increase in macroalgal cover proportional to the
coral cover loss in the 1980s.
Recent spatial patterns
Recent (2001 to 2005) macroalgal cover varied signi-
ficantly among subregions (ANOVA, p < 0.0001; Fig. 4)
and ranged from 6.2 ± 4.0% (n = 16 surveys) in the Gulf
of Mexico to 22.8 ± 1.6% (n = 100 surveys) in the south-
119
Fig. 3. Average annual coral and macroalgal cover values
from 1977 to 2006. Horizontal and vertical lines represent
1 SE. The 3 points from the 1981 to 1992 group that are
clumped with the 1993 to 2006 values are from 1981, 1987,
and 1988
Mar Ecol Prog Ser 402: 115–122, 2010
western Caribbean. The region-wide mean cover of
macroalgae from 2001 to 2005 was 15.3 ± 0.4% (n =
1821 surveys), while mean coral cover was 16.0 ± 0.4%
(n = 1547 surveys). Recent coral cover also varied sig-
nificantly among subregions (ANOVA, p < 0.0001;
Fig. 4; see also Fig. S4). Subregional coral cover dif-
ferences are concordant with a previous analysis of the
Caribbean (Gardner et al. 2003). Assuming the histori-
cal coral cover baseline was similar across the region,
this pattern of current spatial variability could be inter-
preted as evidence of variable rates of coral loss.
Recent coral cover was highest in the Gulf of Mexico
(58.1 ± 3.5% from 2001 to 2005, n = 10 surveys). The
high cover in this subregion was likely due to the
absence of Acropora spp. host populations in the
Flower Garden Banks (until recently; Precht & Aron-
son 2004), where most of the surveys from the Gulf of
Mexico were conducted, which precluded white band
disease from reducing Acropora spp. coral cover there
(Aronson et al. 2005). The Flower Garden Banks reefs
are also atypical sites because survey depths there
were from 20 to 30 m (e.g. Dokken et al. 2003), more
than twice the mean depth of most surveys in the
database.
Recent coral cover was lowest in the Florida Keys
(FLK; 8.6 ± 0.4% from 2001 to 2005, n = 747 surveys).
This subregion was extensively sampled (50.6% of all
surveys performed from 1996 to 2006 were conducted
in the FLK; Table S2) and this overrepresentation
could have unduly influenced the Caribbean-wide
analyses. Therefore, we also analyzed the trends from
1996 to 2005 in regional coral and macroalgal cover
without the FLK data. This resulted in annual regional
coral cover means as much as 13.1% higher than when
the FLK data were included (Fig. 2C,D), but did not
noticeably influence macroalgal cover values. Mean
regional Caribbean coral cover without the FLK data
from 1996 to 2005 was 21.8 ± 0.3%.
The southeastern Caribbean experienced a severe
warming event in late 2005, with levels of thermal
stress exceeding standard bleaching thresholds
(Fig. 1B; Donner et al. 2007, Wilkinson & Souter 2008).
This led to widespread coral bleaching, subsequent
coral disease outbreaks, and moderate to severe local
coral mortality and loss in some locations, including
the US Virgin Islands and the Lesser Antilles (Miller
et al. 2006, Wilkinson & Souter 2008). There was a mi-
nor regional reduction in coral cover in 2006 (Fig. 2C),
which could have been caused in part by coral mortal-
ity in several subregions, particularly those that expe-
rienced the most severe temperature anomalies (Fig. 1;
see also Fig. S4), e.g. the Northern Caribbean (–6.0%
absolute cover from 2005 to 2006) and the Lesser An-
tilles (–3.8% absolute cover). However, we did not
have enough post-bleaching event data to reliably es-
timate subregional declines since 2005 (Table S2).
Monitored sites
Our analysis was primarily based on reefs that were
randomly selected and surveyed only once. The results
of such randomized population sampling can be gener-
alized to a greater degree than those from longitudinal
monitoring studies. However, randomized sampling
has some drawbacks, including sensitivity to sample
composition. Minor, short-term fluctuations in coral
and macroalgal cover (Fig. 2; see also Fig S4) could be
due to the subpopulation of reefs that were surveyed in
a given year, rather than to real year-to-year changes
in community state. Although non-random initial site
selection can cause similar biases, monitoring studies
are generally less sensitive to sample composi-
tion and have several advantages, including
greater power to detect small changes in com-
munity state. Unfortunately, there are still rela-
tively few quantitative reef monitoring pro-
grams and most are focused on well-studied
reefs within Marine Protected Areas.
Spatio-temporal patterns from the subset of
376 monitored reefs are very similar to those
from the entire database (Fig. S5 in the sup-
plement, www.int-res.com/articles/suppl/m402
p115_app.pdf) and are concordant with our
general conclusions. The regional trends from
the sites monitored from 1996 to 2006 indicate
that, since 1996, there has been very little tem-
poral variation in both coral (n = 331) and
macroalgal cover (n = 215) and the trends rein-
force the finding that reefs in the Caribbean
entered a period of general regional stasis in
120
Fig. 4. Recent (2001 to 2005) macroalgal and coral cover in the 7
Caribbean subregions (Fig. 1). Values are means (+1 SE) and sample
sizes (no. of sites) are shown to the right of the bars. Mesoam Reef:
Mesoamerican Reef; N Caribbean: Northern Caribbean; SW Carib-
bean: Southwestern Caribbean
Schutte et al.: Caribbean reef trends
the mid-1990s, at least in terms of coral and macroalgal
cover. Linear repeated-measures regression analyses
indicate that there was no significant change in coral
cover (p = 0.32, n = 331 reefs) or macroalgal cover
(p = 0.109, n = 215 reefs) from 1996 to 2006 across the
region (Table S4 in the supplement, www.int-res.com/
articles/suppl/m402p115_app.pdf). We also analyzed
monitoring data from each subregion independently
from 1996 to 2006; in most cases, there was no sig-
nificant change in cover (Tables S5 & S6 in the supple-
ment, www.int-res.com/articles/suppl/m402p115_app.
pdf. There was a statistically, although perhaps not
ecologically, significant decrease in both coral and
macroalgal cover in the intensively monitored Florida
Keys. Coral cover slightly increased in the Northern
Caribbean and decreased in the Lesser Antilles and
the Southwestern Caribbean. Otherwise, our regres-
sion analyses of the monitoring data suggest that there
have been few other signs of temporal change within
subregions since 1996.
CONCLUSIONS
The first meta-analysis of Caribbean hard coral cover
documented a reduction from ~50% in 1977 to ~10%
in 2001 (Gardner et al. 2003). Our study built on that
work by adding surveys from an additional 1699 sites
and macroalgal data and by expanding the analysis to
2006. The timing of coral loss we documented is consis-
tent with the hypothesis that the acroporid white band
epizootic was the primary cause of hard coral cover
loss in the Caribbean (Aronson & Precht 2006). Like-
wise, the regional macroalgal bloom that occurred sev-
eral years after the disease-induced Diadema antil-
larum die-off supports the argument that a reduction
in herbivory, rather than increased nutrient availabil-
ity, caused the observed increases in macroalgal cover
(Hughes et al. 1999). Therefore, disease, whether nat-
ural or exacerbated by human activities, appears to
have been the primary driver of regional-scale
changes in Caribbean reef benthic communities over
the last 35 yr. Our results indicate that, since these 2
major disturbances, regional coral and macroalgal
benthic coverage has been relatively stable. However,
the future of Caribbean reefs is uncertain. There are
many factors besides disease that could cause fur-
ther changes in reef benthic community structure,
including climate change, ocean acidification, and
direct anthropogenic stressors like overfishing and
nutrient pollution (Hughes et al. 2003). Although our
results could be interpreted as relatively good news,
the observed regional pattern could also be a tempo-
rary plateau preceding a potential collapse in coral
cover.
Acknowledgements. We thank K. France, R. Katz, L. Ladwig,
S. C. Lee, M. I. O’Connor, C. Shields, G. Smelick, I. Vu, and
A. M. Melendy for assistance with this project, and J. C. Lang,
G. Hodgson, W. F. Precht, and W. K. Fitt for very helpful com-
ments on this paper. We are especially grateful to everyone
who shared data with us, including Reef Check, AGRRA,
Florida CREMP, the scientists and data managers who com-
municated with us personally, and all the volunteers and
researchers who collected the data. This project was funded
in part by the National Science Foundation, an Environmen-
tal Protection Agency STAR fellowship to E.R.S., and the Uni-
versity of North Carolina at Chapel Hill.
LITERATURE CITED
Alvarez-Filip L, Dulvy NK, Gill JA, Côté IM, Watkinson AR
(2009) Flattening of Caribbean coral reefs: region-wide
declines in architectural complexity. Proc Biol Sci 276:
3019–3025
Aronson RB, Precht WF (2001) White-band disease and the
changing face of Caribbean coral reefs. Hydrobiologia
460:25–38
Aronson RB, Precht WF (2006) Conservation, precaution, and
Caribbean reefs. Coral Reefs 25:441–450
Aronson RB, Precht WF, Macintyre IG, Murdoch TJT (2000)
Ecosystems—coral bleach-out in Belize. Nature 405:36
Aronson RB, Precht WF, Toscano MA, Koltes KH (2002) The
1998 bleaching event and its aftermath on a coral reef in
Belize. Mar Biol 141:435–447
Aronson RB, Macintyre IG, Wapnick CM, O’Neill MW (2004)
Phase shifts, alternative states, and the unprecedented
convergence of two reef systems. Ecology 85:1876–1891
Aronson RB, Precht WF, Murdoch TJT, Robbart ML (2005)
Long-term persistence of coral assemblages on the Flower
Garden Banks, northwestern Gulf of Mexico: implications
for science and management. Gulf Mex Sci 23:84–94
Bruno JF, Selig ER (2007) Regional decline of coral cover in the
Indo-Pacific: timing, extent, and subregional comparisons.
PLoS ONE 2(8):e711. doi: 10.1371/journal.pone.0000711
Bruno JF, Sweatman H, Precht WF, Selig ER, Schutte VGW
(2009) Assessing evidence of phase shifts from coral to
macroalgal dominance on coral reefs. Ecology 90:
1478–1484
Carpenter RC (1990) Mass mortality of Diadema antillarum.
Mar Biol 104:67–77
Carpenter RC, Edmunds PJ (2006) Local and regional scale
recovery of Diadema promotes recruitment of scleractin-
ian corals. Ecol Lett 9:271–280
Côté IM, Gill JA, Gardner TA, Watkinson AR (2005) Measur-
ing coral reef decline through meta-analyses. Philos Trans
R Soc Lond B 360:385–395
Dokken QR, MacDonald IR, Tunnell JW Jr, Wade T and oth-
ers (2003) Long-term monitoring at the East and West
Flower Garden Banks, 1998–2001. Final report, U.S.
Department of the Interior, Minerals Management Ser-
vice, Gulf of Mexico OCS Region, New Orleans, LA
Donner SD, Knutson TR, Oppenheimer M (2007) Model-
based assessment of the role of human-induced climate
change in the 2005 Caribbean coral bleaching event. Proc
Natl Acad Sci USA 104:5483–5488
Edmunds PJ (2007) Evidence for a decadal-scale decline in
the growth rates of juvenile scleractinian corals. Mar Ecol
Prog Ser 341:1–13
Edmunds PJ, Bruno JF (1996) The importance of sampling
scale in ecology: kilometer-wide variation in coral reef
communities. Mar Ecol Prog Ser 143:165–171
121
➤➤
➤
➤
➤➤
➤
➤
➤
➤➤
➤
➤➤
➤
➤➤
➤
➤
Mar Ecol Prog Ser 402: 115–122, 2010
Edmunds PJ, Elahi R (2007) The demographics of a 15-year
decline in cover of the Caribbean reef coral Montastrea
annularis. Ecol Monogr 77:3–18
Endean R (1977) Acanthaster planci infestations of reefs of the
Great Barrier Reef. In: Taylor DL (ed) Proceedings of the
3rd international coral reef symposium. University of
Miami, Miami, FL, p 185–191
Fabricius KE (2005) Effects of terrestrial runoff on the ecology
of corals and coral reefs: review and synthesis. Mar Pollut
Bull 50:125–146
Gardner TA, Côté IM, Gill JA, Grant A, Watkinson AR (2003)
Long-term region-wide declines in Caribbean corals. Sci-
ence 301:958–960
Ginsburg RN (ed) (1993) Global aspects of coral reefs: health,
hazards and history. University of Miami, Miami, FL
Grigg RW (1994) Effects of sewage discharge, fishing pres-
sure and habitat complexity on coral ecosystems and reef
fishes in Hawaii. Mar Ecol Prog Ser 103:25–34
Grigg RW, Dollar SJ (2005) Reassessing U.S. coral reefs. Sci-
ence 308:1740–1742
Haley MP, Solandt JL (2001) Population fluctuations of the sea
urchin Diadema antillarum and Tripneustes ventricosus at
Discovery Bay, Jamaica: a case of biological succession?
Caribb J Sci 37:239–245
Halpern BS, Walbridge S, Selkoe KA, Kappel CV and others
(2008) A global map of human impact on marine ecosys-
tems. Science 319:948–952
Hodgson G, Hill J, Kiene W, Maun L and others (2006)
Instruction manual: a guide to coral reef monitoring. Reef
Check Foundation, Pacific Palisades, CA
Hoegh-Guldberg O, Mumby PJ, Hooten AJ, Steneck RS and
others (2007) Coral reefs under rapid climate change and
ocean acidification. Science 318:1737–1742
Hughes TP (1994) Catastrophes, phase shifts, and large-scale
degradation of a Caribbean coral reef. Science 265:
1547–1551
Hughes TP, Keller BD, Jackson JCB, Boyle MJ (1985) Mass
mortality of the echinoid Diadema antillarum Philippi in
Jamaica. Bull Mar Sci 36:377–384
Hughes TP, Szmant AM, Steneck R, Carpenter RC, Miller SL
(1999) Algal blooms on coral reefs: What are the causes?
Limnol Oceanogr 44:1583–1586
Hughes TP, Baird AH, Bellwood DR, Card M and others
(2003) Climate change, human impacts, and the resilience
of coral reefs. Science 301:929–933
Idjadi JA, Edmunds PJ (2006) Scleractinian corals as facilita-
tors for other invertebrates on a Caribbean reef. Mar Ecol
Prog Ser 319:117–127
Jackson JBC, Kirby MX, Berger WH, Bjorndal KA and others
(2001) Historical overfishing and the recent collapse of
coastal ecosystems. Science 293:629–637
Jones GP, McCormick MI, Srinivasan M, Eagle JV (2004)
Coral decline threatens fish biodiversity in marine
reserves. Proc Natl Acad Sci USA 101:8251–8253
Lam K, Shin PKS, Bradbeer R, Randall D, Ku KKK, Hodgson P,
Cheung SG (2006) A comparison of video and point inter-
cept transect methods for monitoring subtropical coral
communities. J Exp Mar Biol Ecol 333: 115–128
Lang JL (2003) Status of coral reefs in the western Atlantic: re-
sults of initial surveys, Atlantic and Gulf Rapid Reef Assess-
ment (AGRRA) Program. Atoll Res Bull 496:1–651
Lessios HA (1988) Mass mortality of Diadema antillarum in
the Caribbean: What have we learned? Annu Rev Ecol
Syst 19:371–393
Liddell WD, Ohlhorst SL (1992) Ten years of disturbance and
change on a Jamaican fringing reef. In: Richmond RH (ed)
Proceedings of the 7th international coral reef symposium.
University of Guam Press, Guam, p 149–155
Maliao RJ, Turingan RG, Lin J (2008) Phase-shift in coral reef
communities in the Florida Keys National Marine Sanctu-
ary (FKNMS), USA. Mar Biol 154:841–853
Miller MW, Aronson RB, Murdoch TJT (2003) Monitoring
coral reef macroalgae: different pictures from different
methods. Bull Mar Sci 72:199–206
Miller J, Waara R, Muller E, Rogers C (2006) Coral bleaching
and disease combine to cause extensive mortality on reefs
in US Virgin Islands. Coral Reefs 25:418
Myhre S, Acevedo-Gutiérrez A (2007) Recovery of sea urchin
Diadema antillarum populations is correlated to increased
coral and reduced macroalgal cover. Mar Ecol Prog Ser
329:205–210
Norström AV, Nyström M, Lokrantz J, Folke C (2009) Alterna-
tive states on coral reefs: beyond coral–macroalgal phase
shifts. Mar Ecol Prog Ser 376:295–306
Paddack M, Reynolds J, Aguilar C, Appledoorn R and others
(2009) Recent region-wide declines in Caribbean reef fish
abundance. Curr Biol 19:590–595
Pandolfi JM, Bradbury RH, Sala E, Hughes TP and others
(2003) Global trajectories of the long-term decline of coral
reef ecosystems. Science 301:955–958
Porter JW, Meier OW (1992) Quantification of loss and
change in Floridian reef coral populations. Am Zool 32:
625–640
Pratchett MS, Munday PL, Wilson SK, Graham NAJ and oth-
ers (2008) Effects of climate-induced coral bleaching on
coral-reef fishes — ecological and economic consequen-
ces. Oceanogr Mar Biol Annu Rev 46:251–296
Precht WF, Aronson RB (2004) Climate flickers and range
shifts of reef corals. Front Ecol Environ 2:307–314
Precht WF, Miller SL, Aronson RB, Bruno JF, Kaufman L
(2005) Reassessing US coral reefs. Science 308:
1740–1742
Rasband WS (2006) ImageJ. US National Institutes of Health,
Bethesda, MD
Rogers CS, Miller J (2006) Permanent ‘phase shifts’ or rever-
sible declines in coral cover? Lack of recovery of two coral
reefs in St. John, US Virgin Islands. Mar Ecol Prog Ser
306:103–114
Spalding MD, Fox HE, Allen GR, Davidson N and others
(2007) Marine ecoregions of the world: a bioregionaliza-
tion of coastal and shelf areas. Bioscience 57:573–583
Steneck RS (1988) Herbivory on coral reefs: a synthesis. In:
Choat JH, Barnes D, Borowitzka MA, Coll JC and others
(eds) Proceedings of the 6th international coral reef sym-
posium. 6th International Coral Reef Symposium Execu-
tive Committee, Townsville, p 37–49
Wilkinson CR (1992) Coral reefs of the world are facing wide-
spread devastation: Can we prevent this through sustain-
able management practices? In: Richmond RH (ed) Pro-
ceedings of the 7th international coral reef symposium.
University of Guam Press, Guam, p 11–21
Wilkinson CR, Souter D (2008) Status of Caribbean coral reefs
after bleaching and hurricanes in 2005. Global Coral Reef
Monitoring Network and Reef and Rainforest Research
Centre, Townsville
Woodley JD, Chornesky EA, Clifford PA, Jackson JBC and
others (1981) Hurricane Allen’s impact on Jamaican coral
reefs. Science 214:749–755
122
Editorial responsibility: Tim McClanahan,
Mombasa, Kenya
Submitted: May 18, 2009; Accepted: November 23, 2009
Proofs received from author(s): February 18, 2010
➤
➤➤
➤➤
➤
➤➤
➤➤
➤➤
➤➤
➤➤
➤
➤➤
➤➤
➤
➤
➤
➤
➤➤
➤➤
➤
➤
➤
➤➤
