ArticlePDF Available

The reduction of harmful algae on Caribbean coral reefs through the reintroduction of a keystone herbivore, the long spined sea urchin, Diadema antillarum

  • Institute for Socio-Ecological Research (ISER), Coastal Survey Solutions LLC (CSS)

Abstract and Figures

Herbivores play an essential role in the health and recovery of a coral reef ecosystem. The lack of recovery of the keystone herbivore, Diadema antillarum, has had long-lasting effects, evidenced with many reefs persisting in algal dominance. This study restocked 756 lab-reared D. antillarum to four coral reefs on the east and south coast of Puerto Rico. Sea urchins were placed in experimental plots (“corrals”) for two months, and the change in benthic composition was measured. Significant changes in the benthic structure were observed during the first week after the restocking. Significant reductions of fleshy macroalgae (Dictyota spp.) and thick turf algal/sediment mats (TAS), both unsuitable substrates (e.g., coral settlement), contributed to this change. Also, restocked D. antillarum significantly reduced the cover of encrusting red algae, Ramicrusta spp. By the end of the study, the abundance of fleshy macroalgae decreased by a mean of 77% (max of 100%) and Ramicrusta and TAS by 53% (max 71%) and 56% (max 100%), respectively. Clean substrate (“pavement”), crustose coralline algae (CCA), and filamentous turf algae increased between one to two orders of magnitude. The restoration of native sea urchins is a non-invasive and useful approach to aid in the mitigation of algae, especially potentially dangerous alga like Ramicrusta. The results of this study highlight the importance of herbivores in improving the conditions on coral reefs. This article is protected by copyright. All rights reserved.
Content may be subject to copyright.
The reduction of harmful algae on Caribbean coral reefs
through the reintroduction of a keystone herbivore,
the long-spined sea urchin Diadema antillarum
Stacey M. Williams
Herbivores play an essential role in the health and recovery of a coral reef ecosystem. The lack of recovery of the keystone her-
bivore Diadema antillarum has had long-lasting effects, evidenced with many reefs persisting in algal dominance. This study
restocked 756 lab-reared D. antillarum to four coral reefs on the east and south coast of Puerto Rico. Sea urchins were placed
in experimental plots (corrals) for 2 months, and the change in benthic composition was measured. Signicant changes in the
benthic structure were observed during the rst week after the restocking. Signicant reductions of eshy macroalgae (Dictyota
spp.) and thick turf algal/sediment mats (TAS), both unsuitable substrates (e.g. coral settlement), contributed to this change.
Also, restocked D. antillarum signicantly reduced the cover of encrusting red algae, Ramicrusta spp. By the end of the study,
the abundance of eshy macroalgae decreased by a mean of 77% (max of 100%) and Ramicrusta and TAS by 53% (max 71%)
and 56% (max 100%), respectively. Clean substrate (pavement), crustose coralline algae (CCA), and lamentous turf algae
increased between one to two orders of magnitude. The restoration of native sea urchins is a non-invasive and useful approach
to aid in the mitigation of algae, especially potentially dangerous alga like Ramicrusta. The results of this study highlight the
importance of herbivores in improving the conditions on coral reefs.
Key words: Caribbean, coral reef, Diadema antillarum, herbivore, restoration, sea urchin
Implications for Practice
Coral restoration has primarily focused on restocking
scleractinian corals. The reef substrate housing the coral
fragments must be manually cleaned by divers to remove
algae before and after outplanting. As seen in this study,
restocked Diadema antillarum effectively reduces ben-
thic algae. D. antillarum can be restocked simultaneously
with corals to aid in restoration success.
D. antillarum is the only known consumer of harmful
peyssonnelids, like Ramicrusta.D. antillarum can be
restocked to coral reefs experiencing a high abundance
of Ramicrusta.
Post-larval capture and culture is an alternative technique
to produce individuals in the laboratory for restoration
Further studies are needed to understand the habitat pref-
erences of D. antillarum to enhance the retention of
restocked individuals.
Over the past four decades, coral reefs in the Caribbean have
dramatically changed (Hughes 1994; Wilkinson 2008; Jackson
et al. 2014). The abundance of reef-associated organisms,
especially corals, has suffered a massive decline due to cumula-
tive factors, such as hurricanes, disease outbreaks, bleaching,
pollution, and overshing (Bythell & Sheppard 1993; Littler
et al. 1993; Hughes 1994). One of the most dramatic shifts in
community structure occurred after the massive die-off of Dia-
dema antillarum, a keystone herbivore. The 19831984 mass
mortality of D. antillarum occurred throughout the Caribbean
basin and was one of the most extensive and severe die-offs ever
recorded for a marine invertebrate (Lessios 1995).
Before 1983, D. antillarum was common (1318 ind m
coral reefs in Puerto Rico (Bauer 1980; Vicente & Goe-
naga 1984). They played an important role in structuring coral
reef communities by controlling algal abundance (Odgen
et al. 1973; Carpenter 1981; Sammarco 1982) and productivity
(Williams & Carpenter 1990). Given their high numbers at the
time, D. antillarum was one of the principal agents of bioerosion
on reefs (Scofn et al. 1980; Bak et al. 1984; Lidz & Hal-
lock 2000). After the massive die-off, populations were
Author contributions: SMW conceived and designed the research, performed the
experiments, analyzed the data, and edited the manuscript.
Institute for Socio-Ecological Research, PO Box 3151, Lajas, 00667, Puerto Rico
Coastal Survey Solutions LLC, PO Box 1362, Lajas, 00667, Puerto Rico
Address correspondence to S. M. Williams, email
© 2021 Society for Ecological Restoration.
doi: 10.1111/rec.13475
Supporting information at:
Restoration Ecology 1of11
drastically reduced by 95100% in many Caribbean locations
(Lessios 1995), and at the same time, eshy macroalgal cover
increased between 100% and 250% (Phinney et al. 2001). The
long absence of D. antillarum has not only inuenced the ben-
thic algal productivity of coral reef communities, it also has
inuenced other reef processes (i.e. coral recruitment) (Hughes
et al. 1987).
Presently, the recovery of D. antillarum has been slow and
even absent at many locations in the Caribbean (Lessios 2016).
In Puerto Rico, there has beena modest recovery in the population
of D. antillarum (Mercado-Molina et al. 2015; Tuohy et al. 2020);
nevertheless, densities are still far below pre-mass mortality num-
bers (Lessios 2016). In La Parguera, Tuohy et al. (2020) observed
no signicant increase in D. antillarum populations from 2001
surveys, and populations were dominated by medium to large
(59 cm test diameter) individuals that were concentrated at
shallower (<5 m), more complex reefs. Larval mortality and/or
post-recruitment mortality processes could be the bottlenecks reg-
ulating the recovery of D. antillarum to coral reefs (Karlson &
Levin 1990; Williams et al. 2010). In Puerto Rico, the larval sup-
ply and survival do not seem to be inhibiting the recovery of these
populations (Williams et al. 2010). Consequently, recruitment-
limited processes, such as post-settler and/or juvenile mortality,
may be inhibiting the recovery of D. antillarum in Puerto Rico
(Williams et al. 2011).
Many reefs in the world, and in particular in the Caribbean,
have lost their capacity to recover from recurrent disturbances
and have undergone long-term phase shifts (Hughes et al.
2003). One of the most spoken about phase-shifts in scientic
literature is the coral to eshy macroalgal shift (Hughes 1994;
Rogers & Miller 2006). Reefs characterized in permanent states
of algal dominance usually signify a loss of resiliency (Hughes
et al. 2007) because macroalgal assemblages limit coral settle-
ment, affect sediment deposition, and alter chemical properties
close to the benthos (Birrell et al. 2008). Reef degradation in
Puerto Rico is occurring at a rapid pace (Ballantine et al. 2008;
Weil et al. 2009). Reefs are now characterized by a high abun-
dance of eshy macroalgae and turf algae that have monopo-
lized substrate previously covered by corals and other sessile
reef invertebrates.
There is a new threat to the coral reefs in Puerto Rico, Rami-
crusta spp. Ramicrusta is an encrusting red alga from the Peys-
sonneliaceae family (Rhodophyta). Currently, there are three
species of Ramicrusta identied in the Caribbean; textilis,
bonairensis, and monensis (Ballantine et al. 2011). Ramicrusta
forms a thin, crustose layer that spreads over the substrate and
can grow over the living tissues of other organisms
(Pueschel & Saunders 2009; Eckrich & Engel 2013). Not much
is known of the origins of this alga, and the reasons for its
increase around the Caribbean are unknown. The overgrowth
rate is high for this alga, ranging between 0.06 mm d
0.08 mm d
(Eckrich & Engel 2013). The spread of Rami-
crusta has been occurring throughout Puerto Rico (Williams &
García-Sais 2020). In Puerto Rico, the abundance of Ramicrusta
varies between coasts, with the east coast having a high cover,
ranging between 4575% (Williams & García-Sais 2020). The
rapid growth of Ramicrusta makes this alga not only a threat
to slow-growing sessile-benthic organisms but has the potential
to reduce the area of suitable substratum for coral settlement.
Coral species listed under the Endangered Species Act (ESA),
such as Acropora and Orbicella, have been negatively impacted
by the expansion of Ramicrusta (Ruiz 2015; Williams & García-
Sais 2020).
D. antillarum has been witnessed eating Ramicrusta in the
laboratory (S. M. Williams 2018, Institute for Socio-Ecological
Research, personal observation), and some studies have con-
cluded these sea urchins might be eating these algae in the eld
(Ruiz 2015; Williams & García-Sais 2020). In this article, I
describe the changes in the benthic assemblages after increasing
the densities of D. antillarum to four coral reefs in Puerto Rico.
Coral reefs targeted for this project were characterized by a high
abundance of nuisance algae, such as Ramicrusta,eshy macro-
algae (Dictyota spp.), and thick turf mats with sediment. Also,
this study aimed to conrm that D. antillarum consumes Rami-
crusta in the eld.
The sea urchins restored to the coral reefs for this project were
from settlers collected in the eld and grown in the laboratory
(Williams 2016). The supply of Diadema antillarum settlers
was collected at a shelf-edge reef in La Parguera, Puerto Rico.
The methodology of settler collection is outlined in Williams
et al. (2010, 2011) studies. Settlers were brought back to the lab-
oratory and grown in the wet-lab facility at the Department of
Marine Science at the University of Puerto Rico, Mayagüez.
The settlers (<1 mm in test diameter) were grown in semi-closed
raceways and fed a mix of algae such as Dictyota spp., Ulva sp.,
Acanthophora sp., and Padina sp. Settlers were reared in the
tanks from 10 to 12 months, until they reached young adult
stage (34 cm in test diameter).
Lab-reared D. antillarum were reintroduced to four reefs in
Puerto Rico. In August 2018, D. antillarum juveniles were trans-
ferred from the laboratory to the backreefs of Cayo Diablo
(N1821037.37, W6531059.7) and Los Lobos (N1822026.58,
W6534012.9) in Fajardo (Fig. 1) and in August 2019, to the back-
reefs of El Coral (N1756055.1, W67104.73) and at Mario
(N1757011.23, W673022.82) in La Parguera (Fig. 1). These
coral reefs were chosen for herbivore restoration because they
were dominated by algae, specically Ramicrusta spp. (hereafter
Ramicrusta), eshy macroalgae (Dictyota spp.), and turf algal/
sediment mats (hereafter TAS). TAS substrate is created when
turf algae grow in thick mats that trap sediment. Also, the natural
populations of D. antillarum on all reefs were low (<0.01 ind
). All reefs were 57 m deep and were characterized by
mostly dead O. annularis colonies surrounded by sandy habitats.
Before restoration, six corrals were installed at each reef and
held into place with rebar. Corrals were necessary to measure
the changes in benthic composition because the urchins disperse
if released freely. Sometimes, they move far (>30 m) from the
area of reintroduction (S. M. Williams, personal observation).
The diameter of each corral was approximately 2.4 m (4.5 m
area). Corrals were made of galvanized chicken wire with a
1 in. diameter mesh size. The plastic chicken netting was
Restoration Ecology2of11
Grazing effectiveness by Diadema antillarum
attached to the bottom of the corral to mold to the reef (Fig. 2).
Corrals were placed in the sand around isolated O. annularis
colonies. Tops were temporarily placed on corrals for the rst
2 weeks after restocking. Tops were removed after 2 weeks.
The reason for the tops was to reduce the number of urchins
from escaping. Past restoration activities have shown that sea
Figure 1. A map of the restoration sites, Cayo Diablo and Los Lobos in Fajardo and El Coral and Mario in La Parguera, Puerto Rico.
Figure 2. Photograph of (A) a corral at Los lobos, (B) restocked Diadema antillarum inside a corral at Cayo Diablo that was dominant in Dictyota spp. and
Ramicrusta spp., (C) top view of a corral at Los Lobos, Fajardo, and (D) Diadema antillarum 1 month after restocking event in Fajardo.
Restoration Ecology 3of11
Grazing effectiveness by Diadema antillarum
urchins start displaying homing behavior after 2 weeks of rein-
troduction (S. M. Williams, personal observation). All corrals
and rebars were removed from the reef after 2 months of
On 22 August 2018, 480 lab-reared D. antillarum were trans-
ferred to Fajardo. Twenty-ve D. antillarum were placed in each
corral (5.5 ind m
), and the rest of the sea urchins were released
freely on the reef. In 2019, a total of 276 sea urchins were trans-
ferred to El Coral and Mario on 31 August 2018. Twenty-three
sea urchins were placed in each of the corrals (5.1 ind m
To monitor benthic change through time, I photographed three
xed and three random 25 cm 25 cm quadrats in each corral.
Fixed quadrats were placed in areas with high algal cover, espe-
cially Ramicrusta. Masonry nails were installed in the substrate
and were used to mark the position of xed quadrats, allowing
for the estimation of change. In addition, change in benthic com-
position was also monitored outside the corrals (control) with
three random and xed quadrats. One limitation of this study
is that there were no procedural controls (corrals without sea
urchins) installed. However, past research in Puerto Rico dem-
onstrates that there are no caging (corral) effects on the abun-
dance of algae in a 6-month period, even when the corrals were
fully enclosed (Olmeda-Saldaña et al. 2021; Williams &
Olmeda-Saldaña 2021). Corrals were placed between 5 and
10 m apart, and the controls were placed 2030 m away from
corrals. The percentage cover of algae was discriminated to the
lowest possible taxonomic level. The photographs were exam-
ined in the laboratory, and the relative percentage cover of ses-
sile organisms was estimated using Coral Point Count with
Excel extensions (CPCe). In CPCe, 50 points were placed in a
uniform grid for the xed photo-quadrats, and 50 points were
randomly placed for the haphazard photo-quadrats. Photographs
were taken before restocking the sea urchins and then 1 week,
2 weeks, 1 month, and 2 months after the reintroduction. Photo-
graphs were not taken at 2 weeks at Fajardo sites and after
1 month at Los Lobos because of a camera malfunction.
A three-way permutational multivariate analysis of variance
(PERMANOVA) test (Anderson 2001) was conducted for each
location (Fajardo and La Parguera). For the three-way PERMA-
NOVAs, the changes in benthic taxonomic structure were exam-
ined between sites (Fajardo: Cayo Diablo and Los Lobos, La
Parguera: Mario and El Coral), plots (six corrals with
D. antillarum and control without D. antillarum), and sampling
time. Plots were nested within sites (random). For analyses, ran-
dom quadrats were considered non-independent through time
because of the limited sample area inside each corral (higher
chance of sampling the same area through time). Therefore, per-
manent and random quadrats were pooled for the analyses and
the data were square-root transformed to reduce the inuence
on the dominant taxa (Anderson et al. 2008). Sea urchin abun-
dance inside each corral during each sampling time was used
as a covariate in the analyses. PERMANOVA procedures were
based on BrayCurtis similarity measures, and pvalues were
obtained using 9,999 permutations of the residuals under a
reduced model and Monte Carlo simulations were included. A
three-way PERMANOVA was used to assess the differences
in Ramicrusta cover between the sites, plots (nested in sites),
and sampling time. This analysis was only conducted for
Fajardo because Ramicrusta was absent in the plots in La Par-
guera. For the analyses of Ramicrusta cover (univariate), the
similarity matrix was based on Euclidean distances. Euclidean
distance measures for univariate PERMANOVA analyses pro-
duce sums-of-squares estimates equivalent to parametric
ANOVA (Anderson 2001). All data were square-root trans-
formed and sea urchin abundance was used as a covariate in
the analyses. Pair-wise comparison tests were performed to
identify the variation in benthic cover inside the control plots
through time in each of the models. SIMPER tests were run to
identify the contribution of benthic categories to the overall dif-
ferences between the variables. The multivariate statistical tests
were carried out with the PRIMER v.6 with PERMANOVA
add-on software (Anderson et al. 2008; Clarke & Gorley 2015).
Retention of Restocked Individuals
The retention of restocked Diadema antillarum inside cor-
rals decreased through time, independent of location or site.
As seen in Figure 3, most of the sea urchins escaped the
corrals after 1 week of restocking. The mean retention of
sea urchins inside the corrals was similar between La Par-
guera (27%) and in Fajardo (26%); however, it varied
between sites with El Coral showing the greatest densities
by month 2 (43% retention).
Benthic Change
Fajardo. Before the reintroduction of D. antillarum, the ben-
thic composition between the corrals and controls was similar:
both treatments were mainly characterized by the high abun-
dance of Ramicrusta at both Cayo Diablo (Fig. 4) and Los Lobos
(Fig. 4), with means (SE) ranging from 30 9.9% at Cayo
Diablo to 80 4.8% at Los Lobos. Filamentous turf algae
(hereafter turf algae) and Dictyota spp. were also abundant
outside and inside corrals. Inside the corrals, the cover of other
benthic organisms was low, especially that of coral cover, which
ranged in cover between 0% to 14.7 7.5% at Cayo Diablo to
0% to 13.7 8.9% at Los Lobos.
In the control plot, benthic composition did not show any
signicant changes through time at Cayo Diablo and Los
Lobos. By the end of the study, the benthic composition
was distinctly different inside the experimental plots com-
pared to the controls (Fig. 4). In the experimental plots, the
changes in benthic composition were dependent on the num-
ber of D. antillarum inside the plots (Table S1). The grazing
effects by the sea urchins varied signicantly between sites,
plots, and sampling time (Table S1).
The benthic composition in the control plots did not signi-
cantly differ after 1 week of restocking (three-way PERMA-
NOVA, P[MC] =0.17). However, benthic composition did
signicantly change within 1 week after the reintroduction of
sea urchins in the experimental plots. Benthic substrate was
characterized by more turf algae and clean substrate (hereafter
Restoration Ecology4of11
Grazing effectiveness by Diadema antillarum
pavement) at both sites after 1 week. Ramicrusta was reduced
by 44% at Cayo Diablo and 62.6% at Los Lobos (Fig. 5A, B).
There was also a lower abundance of eshy macroalgae,
specically Dictyota spp. 14.8% at Cayo Diablo (Fig. 5A).
There was a more marked shift in benthic community structure
after 2 months of restocking, and this was independent of sites.
Figure 3. Mean abundance of Diadema antillarum inside the corrals through the monitoring at sites in Fajardo (Cayo Diablo and Los Lobos) and La Parguera
(El Coral and Mario), Puerto Rico. The bars denote SE.
Figure 4. Principal coordinate analysis (PCO) of the benthic composition between control and experimental plots (corrals) before (left) and 2 months (right) after
the reintroduction of Diadema antillarum in Fajardo, at Cayo Diablo (top) and Los Lobos (bottom), Puerto Rico. Plots 16 were the experimental corrals with
D. antillarum, and plot 7 was the control. CCA =crustose coralline algae.
Restoration Ecology 5of11
Grazing effectiveness by Diadema antillarum
The benthic substrates that contributed to the change in composi-
tion over time were Ramicrusta, pavement, Dictyota spp., CCA,
and turf algae (SIMPER). The cover of Ramicrusta inside the
control plots did not signicantly differ through the monitoring
(Table S2). For the experimental corrals, the differences in ben-
thic assemblage after 1 and 2 months was due to the change in
Ramicrusta abundance (Fig. 5A, B). D. antillarum signicantly
reduced Ramicrusta (Table S2), which was replaced by pavement
and turf algae (Fig. 6). The rate at which D. antillarum reduced
Ramcrusta cover varied through the monitoring period (Table
S2). Ramicrusta was reduced on average by 46% at Cayo Diablo
and 51% at Los Lobos by the end of the study. Turf algae and
pavement increased by 191% and 669% (respectively) at Cayo
Diablo and by 114% and 182% at Los Lobos.
La Parguera. Similar to Fajardo, the benthic composition
was not signicantly different between the controls
and experimental plots before the reintroduction of
D. antillarum (Fig. 7). Before the restocking, the benthic
substrate at El Coral was characterized by a high mean cover
(SE) of Dictyota spp. of 36.9 4.2% and turf mats,
23.1 3.8%. Other eshy macroalgae, such as Padina
spp., and articulate calcarous algae, Halimeda spp., were
common. Crustose coralline algae (CCA) and pavement
were low at both sites. At Mario, the dominant substrate in
the quadrats before the restocking was turf algae with a mean
coverof29.64.0%, Dictyota spp. (16.1 2.9%), and
unidentied eshy macroalgae (14.5 3.1). There was no
CCA and pavement present in the corrals before restocking.
The benthic composition in the control plots did not differ through
the monitoring (Table S3). By the end of the study, the benthic com-
position was distinctly different inside the experimental plots com-
pared to the controls (Fig. 7). In La Parguera sites, the changes in
benthic composition of the experimental plots were dependent on
Figure 5. Mean cover of the benthic substrate inside the corrals with Diadema antillarum through the sampling time at (A) Cayo Diablo (n=164) and (B) Los
Lobos (n=126) in Fajardo, Puerto Rico. The bars denote SE.
Figure 6. Photographs of a permanent quadrat before (left) and 2 months (right) after the restocking of Diadema antillarum in corral 5 at Cayo Diablo, Fajardo.
Red circles indicate Ramicrusta.
Restoration Ecology6of11
Grazing effectiveness by Diadema antillarum
the number of D. antillarum (Table S3). The grazing effects by the
sea urchins varied signicantly between sites, plots, and sampling
At both El Coral and Mario, the grazing effects of
D. antillarum signicantly changed the benthic composition
inside the corrals. These changes were more pronounced
through time at El Coral, when compared to Mario. Even after
1 week of restocking, the grazing effects of D. antillarum were
evident at El Coral and Mario (Fig. 8A,B), as Dictyota cover
was reduced by 49% at El Coral, and 48% at Mario. At El Coral,
turf mats were signicantly reduced by 43% during the rst
week. Other eshy macroalgae were signicantly reduced by
88% at Mario during the rst week.
The reef substrate was characterized by turf algae, CCA, and
pavement (Fig. 8A,B) at El Coral by the end of the study. Turf
algae, CCA, and pavement increased by two and one orders of
magnitude, respectively. By the second month after restocking,
most of the D. antillarum escaped the plots, and therefore the
grazing effects at Mario were not as pronounced as at El Coral.
By the end of the study, Dictyota spp. and eshy macroalgal cover
was reduced by 87%, respectively. The substrate at Mario was
characterized by more turf algae, CCA, and pavement (Fig. 8B).
Figure 7. Principal coordinate analysis (PCO) of the benthic composition between control and experimental plots (corrals) before (left) and 2 months (right) after
the reintroduction of Diadema antillarum in La Parguera, at El Coral (top) and Mario (bottom), Puerto Rico. Plots 16 were the experimental corrals with
D. antillarum, and plot 7 was the control. TWS =turf with sediment and CCA =crustose coralline algae.
Restoration Ecology 7of11
Grazing effectiveness by Diadema antillarum
Modern Caribbean coral reefs have been overwhelmed with ben-
thic algae, partly due to the lack of recovery of Diadema antil-
larum (Lessios 2016) and the absence of other invertebrate
herbivores (Francis et al. 2019; Spadaro & Butler 2021) and larger
parrotsh (rainbow, midnight, etc.). As seen in this project, the
reintroduction of D. antillarum is an effective mitigation tool to
reduce the benthic algae on coral reefs. The reduction of algae
was independent of benthic composition and/or location of
restocking. In just 2 months after reintroduction, eshy macroal-
gae were reduced by 88% in cover. These grazing rates of eshy
macroalgae were higher than reported for other herbivore inverte-
brates (Butler & Mojica 2012; Butler IV & Kintzing 2016; Spa-
daro & Butler 2021). The largest variation in benthic cover and
algal reduction occurred with a D. antillarum density of 5 ind
(Olmeda-Saldaña et al. 2021). However, a signicant reduc-
tion of algae continued with two individuals per square meter. The
restocking of D. antillarum did not signicantly impact the coral
cover, yet three small coral recruits were observed in one corral at
El Coral and Cayo Diablo by the end of the study. Given the
recruit size and duration of this study (2 months), these recruits
were most likely established before the experiment and covered
by macroalgae. The likelihood of these coral recruits surviving
has increased as they do not have to compete for space with eshy
The negative relationship between D. antillarum abundance
and eshy macroalgal cover is not novel. Past studies have
found D. antillarum recovery, whether natural or human-
induced via restoration, greatly reduces the eshy macroalgal
abundance and positively impacts coral recruitment on a coral
reef (Edmunds & Carpenter 2001; Nedimyer & Moe 2003;
Macia et al. 2007). However, eshy macroalgae is not the dom-
inant algal type on some modern Caribbean coral reefs. Coral
reefs once characterized by turf and eshy macroalgae are now
dominated by these peyssonnelid algal crusts (PAC, Edmunds
et al. 2019), more specically, Ramicrusta in Puerto Rico (Wil-
liams & García-Sais 2020). There is limited information about
the distribution and ecology of most peyssonnelids, especially
Ramicrusta. Currently in Puerto Rico, Ramicrusta cover is
remarkably high on east coast reefs, areas like Fajardo.
However, it has been progressively spreading throughout Puerto
Rico (Williams & García-Sais 2020). Peyssonnelids, like Rami-
crusta, are known as detractorsas they do not provide a suit-
able surface for coral settlement and might be chemically
defended against the predation of herbivorous shes (AGRRA
2020), as scrape or scar marks or predation from herbivorous
shes have not been observed in the eld (Williams & García-
Sais 2020). However, scar marks on the Ramicrusta were
observed in corrals with D. antillarum. Actively restoring
D. antillarum populations can reduce Ramicrusta cover by as
much as 71% in 2 months. This study is the rst report of a
marine organism effectively grazing and controlling the abun-
dance of Ramicrusta. These results are promising since Rami-
crusta poses a real threat to corals and possibly other benthic
organisms in Puerto Rico and other places in the Caribbean
(Edmunds et al. 2019; Williams & García-Sais 2020).
All the sites in this study were characterized by a substrate
cover not optimal for the settlement and coral growth before
the reintroduction of D. antillarum. The abundance of Rami-
crusta was exceptionally high at the Fajardo sites, and in La Par-
guera, sites were dominated by mostly Dictyota spp. and thick
turf algal mats with sediments. Fleshy macroalgae, especially
Dictyota spp., can retard growth rates of juvenile corals by out-
shading and abrasion (Box & Mumby 2007). The allochemicals
produced by brown, eshy macroalgae can also introduce dan-
gerous bacteria and inhibit coral larvae from settling (Morrow
et al. 2017). Turf, which is short, lamentous, and without sed-
iment, will not affect the settlement and survivorship of corals
(OBrien & Scheibling 2018). However, quite often, algal turf
will accumulate sediments when there is a terrestrial source
close by, or there is constant resuspension of sediments, as
observed at the La Parguera sites. Birrell et al. (2005) found that
coral settlement and juveniles were signicantly reduced when
algal turf accumulated sediments. In the Florida Keys, turf with
sediment signicantly inhibits coral from settling by 10- to
13-fold, compared to turf algae alone (Speare et al. 2019). By
the end of this study, the substrates inside the corrals were char-
acterized by a cleaner substrate with signicantly less eshy
macroalgae, thick turf mats with sediments, and Ramicrusta.
Pavement, crustose coralline algae, and lamentous turf algae
Figure 8. Mean cover of the benthic substrate inside the corrals with Diadema antillarum through the sampling time at (A) El Coral (n=197) and (B) Mario
(n=189) in La Parguera, Puerto Rico. The bars denote SE. TAS, Turf algae with sediment.
Restoration Ecology8of11
Grazing effectiveness by Diadema antillarum
increased from one to two orders of magnitude in the corrals at
all sites. Enhancing herbivory (Francis et al. 2019), in this case,
D. antillarum populations, resulted in more optimal reef sub-
strate and persistence of coral reefs.
Even though D. antillarum were restocked to similar reef
habitats (mostly dead O. annularis colonies), the retention of
the sea urchins between the corrals and sites varied. Trying to
maintain D. antillarum densities in the experimental plots was
challenging in this study and other D. antillarum restoration pro-
jects (Nedimyer & Moe 2003; Miller et al. 2006). The sites with
the highest retention of D. antillarum were at El Coral in La Par-
guera and Cayo Diablo in Fajardo. Two possible reasons for the
higher retention at these sites could be food availability and
the absence of conspecic aggressors (Sammarco & Wil-
liams 1982). The two corrals that had the most D. antillarum
at the end of the experiment had a higher eshy algal abundance
at the beginning of the study. Sea urchins may have escaped the
corrals with less eshy algae to search for food at nearby coral
heads. Williams (1979) recorded threespot damselsh to be more
aggressive to D. antillarum than to other sea urchins, and directly
affecting the distribution of sea urchins on the reef. There were less
damselsh on coral heads at El Coral and Cayo Diablo than at the
other two sites. There were at least two to four threespot damselsh,
Stegastes planifrons,ineachcorralatMario.S. planifrons are one
of the most aggressive damselsh in the Caribbean, and are known
to be more aggressive to conspecics than to predators (Kapeta-
naki 2008). Damselsh were noticed chasing and pecking at the
spines of the reintroduced D. antillarum. Pieces of spines were
commonly observed inside the corrals at Mario. Further studies
are needed to better understand the habitat preferences of
D. antillarum, as this may further the understanding of their distri-
bution and improve restocking efforts.
Since 2014, D. antillarum settlers have been collected and lab-
reared in Puerto Rico. The rst major restocking of D. antillarum
took place in 2016 on the backreef of Media Luna in La Parguera
(Williams 2016). Three hundred and forty-three lab-reared
D. antillarum were reintroduced at Media Luna, which happens
to be a permanent stationfor the PuertoRico Coral Reef Monitoring
Program (PRCMP). Since the reintroduction, D. antillarum densi-
ties havecontinuallyincreased fromzero in 2015, to 8.0ind 30 m
in 2017 and 13.2 ind 30 m
in 2019 (García-Sais et al. 2017,
2019), resulting in the highest densities ever recorded for any per-
manent stations in the PRCRMP (García-Sais et al. 2017). There
is evidence for positive density-dependent effects in recovery of
D. antillarum populations (Hunte & Younglao 1988; Miller
et al. 2007). I have observed many times small D. antillarum
recruitsinside the corralswith the restockedindividuals.Restocked
D. antillarum at Media Luna may be attracting the natural recruit-
ment and local recovery of this species or the recovery may be
attributedto density-dependent habitat selection (Rogers& Loren-
zen 2016). Even though this study shows success, there are some
limitations, such as the short duration of the monitoring. Monitor-
ing for a longer period needs to occur to identify the long-term
impactsof restocking D. antillarum. Additional limitations include
the retentionof individuals insidecorrals for the study duration and
D. antillarum behavior. Another limitation of this study was the
lack ofa procedural control(corral without sea urchins). The caging
treatment may have affected the algal abundance and composition
through the study. However, I do not believe that was the case
because the Olmeda-Saldaña et al. (2021) study was conducted
during the same time and found no caging effects on algal compo-
sition and abundance. Also,there was at least one corral at each site
(except at El Coral) where all the D. antillarum escaped during the
rst week of restocking. The algal composition and abundance did
not change through time inside these corrals without sea urchins.
Lastly, due to the natural dynamics, algae increase in abundance
during the summer with higher water temperatures and light
(Ferrariet al. 2012). This study occurred during late summer,when
algal covershould be at its highest. Algae increased in abundance
only in the control plots, not inside the corrals with the sea
urchins. Given the evidence of this study and others
(Olmeda-Saldaña et al. 2021; Williams & Olmeda-Sal-
daña 2021), the restoration of native sea urchins is a non-
invasive and useful approach to aid the mitigation of algae on
coral reefs. However, precaution should be taken when rest or-
ing urchin densities, as pre-mortality densities negatively
affected coral spat (Sammarco 1980).
The restoration of coral reefs has taken a mostly monospecic
approach, by only outplanting scleractinian corals. Many of the
reefs receiving the coral outplants are overwhelmed by benthic
algae, which are manually removed by divers. Protection of herbiv-
orous shes, such as parrotshes, has been the focus to alleviate the
algal problems on Caribbean coral reefs (Bellwood et al. 2004;
Mumby 2006). Not all herbivores are the same in reducing algae,
as they have different behavioral and morphological traits which
affect what they consume and the efciency of their grazing (Car-
penter 1986; Burkepile & Hay 2011; Wilson et al. 2021). Studies
have shown that herbivorous shes are not effective in reducing
macroalgae, as they tend to target more palatable algae, like la-
mentous turf algae (Burkepile & Hay 2010; Briggs et al. 2018).
As seen in this study, D. antillarum effectively reduces Dictyota
and other eshy algae, but will not consume Halimeda,another
prevalent alga on coral reefs. The Caribbean king crab, Maguimi-
thrax spinosissimus,mayll this niche, as they effectively consume
Halimeda spp. and eshy macroalgae (Spadaro & Butler 2021). If
restocking D. antillarum is not an option, other sea urchins, specif-
ically Echinometra viridis and Tripneustes ventricosus (Francis
et al. 2019), may also be used to control Ramicrusta and other algae
(respectively). E. viridis have been proven to be as efcient at graz-
ing algae as D. antillarum when at higher densities (Sangil & Guz-
man 2016; Kuempel & Altieri 2017). In the eld, I have observed
lower abundance of Ramicrusta at sites where E. viridis were prev-
alent (>5 ind m
). Therefore, to effectively reduce benthic algae,
protection and/or restoration should take a more diverse approach
by enhancing a mixed herbivore assemblage, such as crabs, sh,
and sea urchins (Lubchenco & Gaines 1981; Burkepile &
Hay 2008).
This project was possible with the funding from the Cooperative
Agreement between the Department of Natural and
Environmental Resources and the National Oceanica and Atmo-
spheric Administration Coral Reef Conservation Program
Restoration Ecology 9of11
Grazing effectiveness by Diadema antillarum
(NA17NOS4820037). A big thanks to Manuel Olmeda, who
helped with the collection and care of the sea urchins in the lab-
oratory. Also, thank you to Katie Flynn, Orlando Espinosa, Mil-
ton Carlo, Luis Rodriguez, Liajay Rivera, Catalina Morales,
Francisco Gonz
alez, Jaaziel Garcia, and Nicolle Lebr
on for their
help in the eld and the laboratory. In addition, thank you to the
Department of Marine Science, the University of Puerto Rico for
the facilities, and all the employees who helped with the project.
Lastly, I would like to thank two anonymous reviewers for their
helpful comments, which made the manuscript stronger.
AGRRA (2020) Coral reef monitoring: benthic indicators. https://www.agrra.
Anderson MJ (2001) A new method for non-parametric multivariate analysis of
variance. Austral Ecology 26:3246
Anderson MJ, Gorley RN, Clarke KR (2008) PERMANOVA+for PRIMER:
guide to software and statistical methods. PRIMER-E Ltd., Plymouth, UK
Bak RPM, Carpay MJE, de Ruyter van Steveninck E (1984) Densities of the sea
urchin Diadema antillarum before and after mass mortalities on the coral
reefs of Curacao. Marine Ecology Progress Series 17:105108
Ballantine DL, Appeldoorn RS, Yoshioka P, Weil E, Armstrong R, et al. (2008)
Biology and ecology of Puerto Rico coral reefs. In: Riegl BM, Dodge RE
(eds) Coral reefs of the USA. Springer, New York
Ballantine DL, Athanasiadis A, Ruiz H (2011) Notes on the benthic marine algae
of Puerto Rico X. Additions to the ora. Botanica Marina 54:292302
Bauer JC (1980) Observations on geographical variations in population density of
the echinoid Diadema antillarum within the western North Atlantic. Bulle-
tin of Marine Science 30:509515
Bellwood DR, Hughes TP, Folke C, Nyström M (2004) Confronting the coral reef
crisis. Nature 429:827833
Birrell CL, McCook LJ, Willis BL (2005) Effects of algal turfs and sediment on
coral settlement. Marine Pollution Bulletin 51:408414
Birrell CL, McCook LJ, Willis BL, Diaz-Pulido GA (2008) Effects of benthic algae
on the replenishment of corals and the implications for the resilience of coral
reefs. Oceanography and Marine Biology: An Annual Review 43:2563
Box SJ, Mumby P (2007) Effect of macroalgal competition on growth and survival of
juvenile Caribbean corals. Marine Ecology Progress Series 342:139149
Briggs CJ, Adam TC, Holbrook SJ, Schmitt RJ (2018) Macroalgae size refuge
from herbivory promotes alternative stable states on coral reefs. PLoS
ONE 13(9):e0202273
Burkepile DE, Hay ME (2008) Herbivore species richness and feeding comple-
mentarity affect community structure and function on a coral reef. Proceed-
ings of the National Academy of Sciences 105:10516206
Burkepile DE, Hay ME (2010) Impact of herbivore identify on algal succession
and coral growth on a Caribbean reef. PLoS One 5:e8963
Burkepile DE, Hay ME (2011) Feeding complementarity versus redundancy
among herbivorous shes on a Caribbean reef. Coral Reefs 30:351362
Butler MJ IV, Kintzing MD (2016) An exception to the rule: top-down control of
a coral reef macroinvertebrate community by a tropical spiny lobster. Bul-
letin of Marine Science 92:137152
Butler MJ, Mojica AM (2012) Herbivory by the Caribbean king crab on coral
patch reefs. Marine Biology 159:26972706
Bythell JC, Sheppard CR (1993) Mass mortality of Caribbean shallow corals.
Marine Pollution Bulletin 26:296297
Carpenter RC (1981) Grazing by Diadema antillarum (Philippi) and its effects on
the benthic algal community. Journal of Marine Research 39:749765
Carpenter RC (1986) Partitioning herbivory and its effects on coral-reef algal
communities. Ecological Monographs 56:345363
Clarke KR, Gorley RN (2015) PRIMER v7: user manual/tutorial. Plymouth,
England: PRIMER-E Plymouth.
Eckrich CE, Engel S (2013) Coral overgrowth by an encrusting red alga (Rami-
crusta sp.): a threat to Caribbean reefs? Coral Reefs 32:8184
Edmunds PJ, Carpenter RC (2001) Recovery of Diadema antillarum reduces
macroalgal cover and increases abundance of juvenile corals on a Carib-
bean reef. Proceedings of the National Academy of Science 98:48224824
Edmunds PJ, Zimermann SA, Bramanti L (2019) A spatially aggressive peysson-
nelid algal crust (PAC) threatens shallow coral reefs in St. John, US Virgin
Islands. Coral Reefs 38:13291341
Ferrari R, Gonzalez-Rivero M, Carlos Ortiz J, Mumby P (2012) Interaction of
herbivory and seasonality on the algal dynamics of Caribbean macroalgae.
Coral Reefs 31:683692
Francis FT, Filbee-Dexter K, Yan HF, Côté IM (2019) Invertebrate herbivores:
overlooked allies in the recovery of degraded coral reefs? Global Ecology
and Conservation 17:e00593
García-Sais JR, Williams SM, Sabater-Clavell J, Carlo M (2017) Puerto Rico
coral reef monitoring program: 2017 survey. Final report for the Depart-
ment of Natural and Environmental Resources (DNER). DNERs Coral
Reef Conservation and Management Program
García-Sais JR, Williams SM, Sabater-Clavell J, Carlo M (2019). Puerto Rico
coral reef monitoring program: 2019 survey. Final report for the Depart-
ment of Natural and Environmental Resources (DNER). DNERs Coral
Reef Conservation and Management Program.
Hughes TP (1994) Catastrophes, phase shifts, and large-scale degradation of a
Caribbean coral reef. Science 265:15471551
Hughes TP, Baird AH, Bellwood DR, Card M, Connolly SR, Folke C, et al.
(2003) Climate change, human impacts, and the resilience of coral reefs.
Science 301:929933
Hughes TP, Reed DC, Boyle MJ (1987) Herbivory on coral reefs: community
structure following mass mortalities of sea urchins. Journal of Experimental
Marine Biology and Ecology 11:3959
Hughes TP, Rodrigues MJ, Bellwood DR, Ceccarelli D, Hoegh-Guldberg O,
McCook L, Moltschaniwskyj N, Pratchett MS, Steneck RS, Willis B
(2007) Phase shifts, herbivory, and the resilience of coral reefs to climate
change. Current Biology 17:360365
Hunte W, Younglao D (1988) Recruitment and population recovery of Diadema
antillarum (Echinodermata; Echinoidea) in Barbados. Marine Ecology
Progress Series 45:109119
Jackson JBC, Donovan M, Cramer K, Lam V (2014) Status and trends of Carib-
bean coral reefs: 19702012. Global Coral Reef Monitoring Network,
IUCN, Gland, Switzerland
Kapetanaki P (2008) The effects of intruder size on the territoriality of the three-
spot damselsh (Stegastes planifrons). CIEE Bonaire Report in Physis:
Journal of Marine Science III 18:712
Karlson RH, Levin DR (1990) Recruitment-limitation in open populations of
Diadema antillarum: an evaluation. Oecologia 82:4044
Kuempel CD, Altieri AH (2017) The emergent role of small-bodied herbivores in
pre-empting phase shifts on degraded coral reefs. Scientic Reports 7:7
Lessios HA (1995) Diadema antillarum 10 years after mass mortality: still rare,
despite help from a competitor. Proceedings of the Royal Society B: Bio-
logical Science 259:331337
Lessios HA (2016) The great Diadema antillarum die-off: 30 years later. Annual
Review of Marine Science 8:267283
Lidz BH, Hallock P (2000) Sedimentary petrology of a declining reef ecosystem,
Florida reef tract (U.S.A.). Journal of Coastal Research 16:675
Littler MM, Littler DS, Lapointe BE (1993) Modication of tropical reef commu-
nity structure due to cultural eutrophication: the southwest coast of Marti-
nique. Proceedings 7th International Coral Reef Symposium 1:335343
Lubchenco J, Gaines SD (1981) A unied approach to marine plant-herbivore
interactions. I. Populations and communities. Annual Review of Ecology
and Systematics 12:12437
Macia S, Robinson MP, Nalevanki A (2007) Experimental dispersal of recover-
ing Diadema antillarum increases grazing intensity and reduces macroalgal
abundance on a coral reef. Marine Ecology Progress Series 348:173182
Mercado-Molina AE, Montañez-Acuña A, Rodríguez-Barreras R, Col
Miranda R, Díaz-Ortega G, Martínez-Gonz
alez N, Schleier-Hern
andez S,
Restoration Ecology10 of 11
Grazing effectiveness by Diadema antillarum
Sabat AM (2015) Revisiting the population status of the sea urchin Dia-
dema antillarum in northern Puerto Rico. Journal of the Marine Biological
Association of the United Kingdom 95:10171024
Miller MW, Szmant AM, Precht WF (2006) 12 lessons learned from experimental
key-species restoration. Pages 219234. In: Coral Reef Restora tionH andbook.
CRC Press, Boca Raton, Florida
Miller RJ, Adams AJ, Ebersole JP, Ruiz E (2007) Evidence for positive density-
dependent effects in recovering Diadema antillarum populations. Journal
of Experimental Marine Biology and Ecology 349:215222
Morrow KM, Bromhall K, Motti CA, Munn CB, Bourne DG (201 7) Allelochemicals
produced by brown macroalgae of the Lobophora genus are active against
coral larvae and associated bacteria, supporting pathogenic shifts to Vibrio
dominance. Applied and Environmental Microbiology 83:e02391e02316
Mumby PJ (2006) The impact of exploiting grazers (Scaridae) on the dynamics of
Caribbean coral reefs. Ecological Applications: a publication of the Eco-
logical Society of America 16:747769
Nedimyer K, Moe MA (2003) Technique development for the reestablishment of
the long-spined sea urchin, Diadema antillarum, on two small patch reefs
in the upper Florida Keys. Science Report: An Ecosystem Report Card
After Five Years of Marine Zoning, p 268.
OBrien J, Scheibling RE (2018) Turf wars: competition between foundation and
turf-forming species on temperate and tropical reefs and its role in regime
shifts. Marine Ecology Progress Series 590:117
Odgen JC, Brown RA, Salesky N (1973) Grazing by the echinoid Diadema
antillarum Philippi: formation of halos around West Indian patch reefs. Sci-
ence 182:715717
Olmeda-Saldaña M, Williams SM, Weil E, Cruz-Motta JJ (2021) Experimental evalu-
ation of Diadema antillarum herbivory effects on benthic community assem-
blages. Journal of Experimental Marine Biology and Ecology 541:151566
Phinney JT, Muller-Karger F, Dustan P, Sobel J (2001) Using remote sensing to
reassess the mass mortality of Diadema antillarum 1983-1984. Conserva-
tion Biology 15:885881
Pueschel CM, Saunders GW (2009) Ramicrusta textilis nov. (Peyssonneliaceae,
Rhodophyta), an anatomically complex Caribbean alga that overgrows
corals. Phycologia 48:480491
Rogers A, Lorenzen K (2016) Does slow and variable recovery of Diadema antil-
larum on Caribbean fore-reefs reect density-dependent habitat selection?
Frontiers in Marine Science 3:63.
Rogers CS, Miller J (2006) Permanent phase shiftsor reversible declines in
coral cover? Lack of recovery of two coral reefs in St. John, US Virgin
Islands. Marine Ecology Progress Series 306:103114
Ruiz H (2015) The ecology of the red algae Ramicrusta textilis, its dynamic with
corals and the evaluation of possible management strategies to minimize its
threat to coral reefs around Puerto Rio. Final report to Puerto Rico Depart-
ment of Natural and Environmental Resources.
Sammarco PW (1980) Diadema and its relationship to coral spat mortality: graz-
ing, competition, and biological disturbance. Journal of Experimental
Marine Biology and Ecology 45:245272
Sammarco PW (1982) Effects of grazing by Diadema antillarum Philippi
(Echinodermata: Echinoidea) on algal diversity and community structure.
Journal of Experimental Marine Biology and Ecology 65:83105
Sammarco PW, Williams AH (1982) Damselsh territoriality: inuence on Dia-
dema distribution and implications for coral community structure. Marine
Ecology Progress Series 8:5359 Available at:
Sangil C, Guzman HM (2016) Assessing the herbivore role of the sea-urchin
Echinometra viridis: keys to determine the structure of communities in dis-
turbed coral reefs. Marine Environmental Research 120:202213
Scofn TP, Stearn CW, Boucher D, Frydl P, Hawkings CM, Hunter IG,
MacGeachy JK (1980) Calcium carbonate budget of a fringing reef on
the west coast of Barbados. II. Erosion, sediments and internal structure.
Bulletin of Marine Science 30:475508
Spadaro AJ, Butler MJ (2021) Herbivorous crabs reverse the seaweed dilemma
on coral reefs. Current Biology 31:31859.e3
Speare KE, Duran A, Miller MW, Burkepile DE (2019) Sediment associated with
algal turfs inhibits the settlement of two endangered coral species. Marine
Pollution Bulletin 144:189195
Tuohy E, Wade C, Weil E (2020) The lack of recovery of the long-spined sea
urchin Diadema antillarum Philippi in Puerto Rico 30 years after the Carib-
bean-wide mass mortality. PeerJ 8:e8428
Vicente VP, Goenaga C (1984) Mass mortalities of the sea urchin Diadema antillarum
(Philippi) in Puerto Rico. Center for Energy and Environmental Research, Uni-
versity of Puerto Rico, US Department of Energy, CREER-M p 30.
Weil E, Cr
oquer A, Urreiztieta I (2009) Temporal variability and impact of coral
diseases and bleaching in La Parguera, Puerto Rico from 2003-2007. Carib-
bean Journal of Science 45:221246
Wilkinson C (2008) Status of coral reefs of the world: 2008. Global Coral Reef
Monitoring Network and Reef and Rainforest Research Centre,
Townsville, Australia
Williams AH (1979) Interference behavior and ecology of threespot damselsh
(Eupimacentrus planifrons). Oecologia 38:223230
Williams SL, Carpenter RC (1990) Photosynthesis/photon ux density relation-
ships among components of coral reef algal turfs. Journal of Phycology
Williams SM (2016) A novel approach to the restoration of Diadema antillarum
on coral reefs in the Caribbean. Reef Encounter 31:4850
Williams SM (2018) The control of algal abundance on coral reefs through the
reintroduction of Diadema antillarum. Final Report for Puerto Rico Depart-
ment of Natural and Environmental Resources. San Juan, Puerto Rico
Williams SM, García-Sais J, Yoshioka PM (2011) Spatial variation of Diadema
antillarum settlement in La Parguera, Puerto Rico. Bulletin of Marine Sci-
ence 87:531540
Williams SM, García-Sais R (2020)A potentialnew threaton the coral reefsof Puerto
Rico: the recent emergence of Ramicrusta spp. Marine Ecology 41:e12592
Williams SM, Olmeda-Saldaña M (2021) The restocking of the long spined sea
urchin, Diadema antillarum, to control macroalgae on two coral reefs in
two natural reserves in Cabo Rojo, Puerto Rico. San Juan, Puerto Rico: Final
Report submitted to the Puerto Rico Department of Natural and Environmental
Williams SM, Yoshioka PM, García-Sais J (2010) Recruitment pattern of Dia-
dema antillarum in La Parguera, Puerto Rico. Coral Reefs 29:809812
Wilson MW, Gaines SD, Stier AC, Halpern BS (2021) Variation in herbivore
grazing behavior across Caribbean reef sites. Marine Biology 168:53
Supporting Information
The following information may be found in the online version of this article:
Table S1. The results of the three-way Permutational Analysis of Variance (PERMA-
NOVA) test to examine the benthic composition change between sites (Si), plots (Co),
and sampling time (Ti) at Fajardo, Puerto Rico.
Table S2. The results of the three-way Permutational Analysis of Variance (PERMA-
NOVA) test to examine the change in Ramicrusta cover between sites (Si), corrals
(Co), and sampling time (Ti) at La Parguera, Puerto Rico.
Table S3. The results of the three-way Permutational Analysis of Variance
(PERMANOVA) test to examine the benthic composition change between sites (Si),
plots (Co), and sampling time (Ti) at La Parguera, Puerto Rico.
Coordinating Editor: Phanor Montoya-Maya Received: 11 March, 2021; First decision: 18 May, 2021; Revised: 30 May,
2021; Accepted: 5 June, 2021
Restoration Ecology 11 of 11
Grazing effectiveness by Diadema antillarum
... Studying settlement rates is also meaningful for other purposes, as the collected settlers can be used for D. antillarum restoration (Williams, 2017;Williams, 2021). While the effectiveness of settlement collectors has been studied for other sea urchins (Balsalobre et al., 2016), it is unknown which substrate is most effective for D. antillarum settlement. ...
... The collection of D. antillarum settlers is the key activity of a relatively new reef restoration method in which the settlers are collected in the field, raised in a land-based nursery and then returned to the reefs once they reach a young adult size (2-4 cm) (Williams, 2017;Williams, 2021). To make this method economically feasible, it is important to maximize the number of settlers that can be collected. ...
Full-text available
The massive die-off of the herbivorous sea urchin Diadema antillarum in 1983 and 1984 resulted in phase shifts on Caribbean coral reefs, where macroalgae replaced coral as the most dominant benthic group. Since then, D. antillarum recovery has been slow to non-existent on most reefs. Studying settlement rates can provide insight into the mechanisms constraining the recovery of D. antillarum, while efficient settlement collectors can be used to identify locations with high settlement rates and to collect settlers for restoration practices. The aim of this study was to compare pre and post die-off settlement rates and to determine possible settlement peaks in the Eastern Caribbean island of St. Eustatius. Additionally, we aimed to determine the effectiveness and reproducibility of five different settlement collectors for D. antillarum. D. antillarum settlement around St. Eustatius was highest in May, June and August and low during the rest of the study. Before the die-off, settlement recorded for Curaçao was high throughout the year and was characterized by multiple settlement peaks. Even though peak settlement rates in this study were in the same order of magnitude as in Curaçao before the die-off, overall yearly settlement rates around St. Eustatius were still lower. As no juvenile or adult D. antillarum were observed on the reefs around the settlement collectors, it is likely that other factors are hindering the recovery of the island's D. antillarum populations. Of all five materials tested, bio ball collectors were the most effective and reproducible method to monitor D. antillarum settlement. Panels yielded the least numbers of settlers, which can partly be explained by their position close to the seabed. Settler collection was higher in mid-water layers compared to close to the bottom and maximized when strings of bio balls were used instead of clumps. We recommend research into the feasibility of aiding D. antillarum recovery by providing suitable settlement substrate during the peak of the settlement season and adequate shelter to increase post-settlement survival of settlers. The bio ball collectors could serve as a suitable settlement substrate for this new approach of assisted natural recovery.
Full-text available
The massive die-off of the sea urchin Diadema antillarum in 1983–1984 is one the main reasons for low coral recruitment and little coral recovery in the Caribbean. As the natural recovery of D. antillarum is slow to non-existent, multiple restoration studies have been attempted. There are currently three different approaches to obtain individuals for restocking: the translocation of wild-collected juveniles or adults, lab-reared juveniles cultured from wild-collected settlers, or lab-reared juveniles cultured from gametes. All three methods are costly and can only be applied on a relatively small scale. We here propose a fourth, new, approach, which we term assisted natural recovery (ANR) of D. antillarum populations. ANR, a concept already applied in terrestrial restoration to restore forests and grasslands, can accelerate succession by removing barriers to natural recovery. In this study, performed on the Dutch Caribbean island of Saba, suitable settlement substrate was provided in the form of bio ball streamers that were attached to the reef shortly before the settlement season. At the end of the experiment, reefs with streamers had significantly higher D. antillarum recruit densities than control reefs without additional settlement substrate, indicating that the lack of settlement substrate is an important factor constraining natural recovery. However, D. antillarum recruit abundance was low compared to the measured settlement rates, possibly due to low post-settlement survival. The size distribution of recruits showed that recruits almost never became larger than 20 mm, which is likely due to predation. We conclude that, next to low settlement availability, low post-settlement survival and high predation on recruits also constrain the natural recovery of D. antillarum populations on Saba. To improve the survival of settlers till adults, we propose to 1) reduce predation on settlers by using bio balls or other substrates that can provide shelter to larger individuals and 2) optimize the reef habitat by removing macroalgae, either manually or by facilitating other herbivores. To improve the survival of recruits, we suggest to 1) choose sites with a known lower predation density or 2) protect recruits with a corral around the reef underneath the streamers. The combination of these measures could improve prospects for ANR, and we expect this new approach can contribute to the recovery of D. antillarum populations in the future.
Full-text available
Recent warm temperatures driven by climate change have caused mass coral bleaching and mortality across the world, prompting managers, policymakers, and conservation practitioners to embrace restoration as a strategy to sustain coral reefs. Despite a proliferation of new coral reef restoration efforts globally and increasing scientific recognition and research on interventions aimed at supporting reef resilience to climate impacts, few restoration programs are currently incorporating climate change and resilience in project design. As climate change will continue to degrade coral reefs for decades to come, guidance is needed to support managers and restoration practitioners to conduct restoration that promotes resilience through enhanced coral reef recovery, resistance, and adaptation. Here, we address this critical implementation gap by providing recommendations that integrate resilience principles into restoration design and practice, including for project planning and design, coral selection, site selection, and broader ecosystem context. We also discuss future opportunities to improve restoration methods to support enhanced outcomes for coral reefs in response to climate change. As coral reefs are one of the most vulnerable ecosystems to climate change, interventions that enhance reef resilience will help to ensure restoration efforts have a greater chance of success in a warming world. They are also more likely to provide essential contributions to global targets to protect natural biodiversity and the human communities that rely on reefs.
Full-text available
Background Marine protected areas (MPAs) usually have both positive effects of protection for the fisheries’ target species and indirect negative effects for sea urchins. Moreover, often in MPAs sea urchin human harvest is restricted, but allowed. This study is aimed at estimating the effect of human harvest of the sea urchin Paracentrotus lividus within MPAs, where fish exploitation is restricted and its density is already controlled by a higher natural predation risk. The prediction we formulated was that the lowest densities of commercial sea urchins would be found where human harvest is allowed and where the harvest is restricted, compared to where the harvest is forbidden. Methods At this aim, a collaborative database gained across five MPAs in Sardinia (Western Mediterranean, Italy) and areas outside was gathered collecting sea urchin abundance and size data in a total of 106 sites at different degrees of sea urchin exploitation: no, restricted and unrestricted harvest sites (NH, RH and UH, respectively). Furthermore, as estimates made in past monitoring efforts (since 2005) were available for 75 of the sampled sites, for each of the different levels of exploitation, the rate of variation in the total sea urchin density was also estimated. Results Results have highlighted that the lowest sea urchin total and commercial density was found in RH sites, likely for the cumulative effects of human harvest and natural predation. The overall rate of change in sea urchin density over time indicates that only NH conditions promoted the increase of sea urchin abundance and that current local management of the MPAs has driven towards an important regression of populations, by allowing the harvest. Overall, results suggest that complex mechanisms, including synergistic effects between natural biotic interactions and human pressures, may occur on sea urchin populations and the assessment of MPA effects on P. lividus populations would be crucial to guide management decisions on regulating harvest permits. Overall, the need to ban sea urchin harvest in the MPAs to avoid extreme reductions is encouraged, as inside the MPAs sea urchin populations are likely under natural predation pressures for the trophic upgrading.
Full-text available
Herbivorous fish can increase coral growth and survival by grazing down algal competitors. With coral reefs in global decline, maintaining adequate herbivory has become a primary goal for many managers. However, herbivore biomass targets assume grazing behavior is consistent across different reef systems, even though relatively few have been studied. We document grazing behavior of two scarid species in Antigua, Barbuda, and Bonaire. Our analyses show significant differences in intraspecific feeding rates, time spent grazing, and intensity of grazing across sites, which may alter the ecological impact of a given scarid population. We suggest several hypothesized mechanisms for these behavioral variations that would benefit from explicit testing in future research. As managers set targets to enhance herbivory on reefs, it is critical that we understand potential differences in scarid grazing impact. Our findings demonstrate the variability of grazing behavior across different reef sites and call for further investigation of the drivers and ecological implications of these inconsistencies.
Full-text available
The rapid appearance of Ramicrusta spp. is described and analyzed from 40 permanent monitoring coral reef stations in Puerto Rico. Before 2016, Ramicrusta had not been observed from any of the reef monitoring stations. By 2018, it was present at 76% of all the monitoring stations. Ramicrusta was the dominant substrata type at all of the shallow reef sites sampled on the east coast (e.g., Fajardo, Culebra, and Vieques), reaching a cover (±SE) as high as 63.0 ± 5.8%. The spread of Ramicrusta occurred at the expense of historically resilient living benthic elements, such as turf algae. Since its detection in 2016, colonization of hard substrata by Ramicrusta remained constant, with the exception of two shallow reefs in Fajardo and Culebra, where the cover was significantly reduced by the scouring and or abrasive effects of two major hurricanes. The ecological implications of Ramicrusta prevalence on Puerto Rican reefs remain unclear; however, increasing herbivory might be a useful mitigation tool in the reduction of Ramicrusta abundance on coral reefs.
Full-text available
Caribbean populations of the long-spined black sea urchin Diadema antillarum Philippi were decimated by a disease-induced mass mortality in the early 1980’s. The present study provides an updated status of the D. antillarum recovery and population characteristics in La Parguera Natural Reserve, Puerto Rico. The last detailed study to assess population recovery in 2001, indicated a slow, and modest recovery, albeit densities remained far below pre-mass mortality levels. Population densities were assessed along three depth intervals in six reef localities and one depth in three lagoonal sea-grass mounds using ten 20 m ² (10 × 2 m) belt-transects at each depth interval. Most of these were previously surveyed in 2001. All individuals encountered along the belt transects were sized in situ with calipers and rulers to characterize the size (age) structure of each population and get insight into the urchin’s population dynamics and differences across localities in the area. Habitat complexity (rugosity) was assessed in all depth intervals. No significant differences in population densities between reef zones (inner shelf and mid-shelf) were observed, but significantly higher densities were found on shallow habitats (<5 m depth; 2.56 ± 1.6 ind/m ² ) compared to intermediate (7–12 m; 0.47 ± 0.8 ind/m ² ) and deep (>12 m; 0.04 ± 0.08 ind/m ² ) reef habitats in almost all sites surveyed. Habitat complexity had the greatest effect on population densities at all levels (site, zone and depth) with more rugose environments containing significantly higher densities and wider size structures. Comparison between survey years revealed that D. antillarum populations have not increased since 2001, and urchins seem to prefer shallower, more complex and productive areas of the reef. Populations were dominated by medium to large (5–9 cm in test diameter) individuals and size-frequency distributions indicated that smaller juveniles were virtually absent compared to 2001, which could reflect a recruitment-limited population and explain in part, the lack of increase in population densities. The limited temporal scale of this study, high horizontal movement of individuals along the complex, shallower reef and inshore habitats could also explain the general decline in mean densities. Other extrinsic factors affecting reproductive output and/or succesful recruitment and survival of juveniles likely contribute to the high variablility in population densities observed over time.
Full-text available
In recent decades, many Caribbean reefs have experienced large declines in abundance of scleractinian corals, and blooms of fleshy macroalgae have often accompanied these trends. In 2010 a new macroalgal threat emerged in Lac Bay, Bonaire, where peyssonnelid algal crusts (PAC) rapidly spread in shallow water and overgrew corals and sponges. Similar growths have been reported in Puerto Rico and the Virgin Islands, and here, we describe the spread of PAC on the shallow reefs of St. John, US Virgin Islands, ~ 760 km northeast of Bonaire. In 2015, PAC covered 0.2–20.6% of hard benthic surfaces at ≤ 7 m depth. By August 2017, at the same sites, PAC had increased to 3.2–61.0% cover, at 5 m depth at 10 other sites along 5 km of shore, it covered 1.4–61.6% of reef surfaces, and at 9 m depth at five sites, it covered 0.8–41.0% of reef surfaces. At 5 m depth in August 2017, scleractinians and octocorals were frequently contacting PAC (42–47% of colonies), and more scleractinians (74%) than octocorals (39%) were overgrown by PAC. Following surveys in August 2017, St. John was hit by two Category 5 hurricanes, yet the shore-wide mean cover of PAC at 5 m depth was only reduced from 26 to 23%. Our results suggest PAC is poised to cause significant ecological change on the reefs of St. John and potentially will promote a community shift favoring octocorals over scleractinians. PAC constitutes an emerging regional threat on shallow Caribbean reefs to which researchers and resource managers will need to quickly respond.
Full-text available
Coral reef systems can undergo rapid transitions from coral-dominated to macroalgae-dominated states following disturbances, and models indicate that these may sometimes represent shifts between alternative stable states. While several mechanisms may lead to alternate stable states on coral reefs, only a few have been investigated theoretically. We explore a model that illustrates that reduced vulnerability of macroalgae to herbivory as macroalgae grow and mature could be an important mechanism: when macroalgae are palatable to herbivores as juveniles, but resistant as adults, coral-dominated and algae-dominated states are bistable across a wide range of parameter space. We compare two approaches to global sensitivity analysis to rank the relative importance of the model parameters in determining the presence and magnitude of alternative stable states, and find that the two most influential parameters are the death rate of coral and the rate of maturation of algae out of the vulnerable stage. The Random Forest approach for global sensitivity analysis, recently adopted by ecologists, provides a more efficient method for ranking the relative importance of parameters than a variance-based approach that has been used frequently by computer scientists and engineers. Our results suggest that managing reefs to reduce chronic stressors that cause coral mortality and/or enhance the growth rates of algae can help prevent reefs from becoming locked in a macroalgae-dominated state.
Diadema antillarum is a keystone herbivore that controls algal cover on coral reefs, as reductions in this species has been associated with significant increases in algal cover. However, most of the evidence that supports this model comes from observational studies where processes have been inferred from descriptions of temporal and spatial relationships. Within this context, a manipulative experiment was conducted to test the effects of D. antillarum densities and rugosity on benthic assemblages. The experiment consisted of fencing 1m 2 experimental plots (metal fence corrals) of high and low rugosity where different densities (0, 1, 5 and 10 individuals/ m 2) of D. antillarum were enclosed per plot. Inside these plots, three permanent and random 100cm 2 photo-quadrats were taken during eight sampling times (six months) to estimate temporal change in benthic cover. Multivariate and univariate analyses indicated that: 1) presence of D. antillarum significantly decreased turf algae (from 90% to 4%) and Dictyota (from 25% to 0%) cover and increased clean substrate (from 0% to 72%) at both sites; 2) herbivory rates for low rugosity were different from high rugosity herbivory rates; 3) densities of 1 D. antillarum individual/m 2 did not change algal cover; 4) a minimum of 5 D. antillarum individuals/m 2 were required to maintain low algal cover on low and high rugosity experimental units. Based on these results we suggest that D. antillarum can effectively enhance coral reef health by decreasing algal cover and creating clean substrate. For restoration efforts, optimal densities of 5 D. antillarum individuals/m 2 should be considered when restocking individuals to coral reefs.
Coral reefs are on a steep trajectory of decline, with natural recovery in many areas unlikely. Eutrophication, overfishing, climate change, and disease have fueled the supremacy of seaweeds on reefs, particularly in the Caribbean, where many reefs have undergone an ecological phase shift so that seaweeds now dominate previously coral-rich reefs. Discovery of the powerful grazing capability of the Caribbean’s largest herbivorous crab (Maguimithrax spinosissimus) led us to test the effectiveness of their grazing on seaweed removal and coral reef recovery in two experiments conducted sequentially at separate locations 15 km apart in the Florida Keys (USA). In those experiments, we transplanted crabs onto several patch reefs, leaving others as controls (n = 24 reefs total; each 10-20 m2 in area) and then monitored benthic cover, coral recruitment, and fish community structure on each patch reef for a year. We also compared the effectiveness of crab herbivory to scrubbing reefs by hand to remove algae. Crabs reduced the cover of seaweeds by 50%-80%, resulting in a commensurate 3-5-fold increase in coral recruitment and reef fish community abundance and diversity. Although laborious hand scrubbing of reefs also reduced algal cover, that effect was transitory unless maintained by the addition of herbivorous crabs. With the persistence of Caribbean coral reefs in the balance, our findings demonstrate that large-scale restoration that includes enhancement of invertebrate herbivores can reverse the ecological phase shift on coral reefs away from seaweed dominance.
A key question for coral reef conservation is whether reefs dominated by macroalgae can recover. Since the near-disappearance of the herbivorous urchin Diadema antillarum in the Caribbean, a prevalent management paradigm has focused on protecting herbivorous fishes to trigger shifts back to a coral-rich state. However, in the absence of D. antillarum, the contribution of other large macroinvertebrates to herbivory intensity has been largely overlooked. We used day and night field surveys and behavioural observations at 16 degraded reef patches in the Bahamas to measure the abundance of large herbivorous macroinvertebrates and their consumption of fleshy macroalgae. Tripneustes sea urchins and Maguimithrax crabs were the main herbivorous macroinvertebrates on our sites and were active mainly at night, with 97% of urchins and 45% of crabs observed consuming fleshy macroalgae. By comparison, < 5% of herbivorous fishes observed ate macroalgae. In the laboratory, Tripneustes sea urchins and Maguimithrax crabs readily consumed macroalgae (at rates of 0.19 g h ⁻¹ and 0.38 g h ⁻¹ , respectively), but their low abundance on patch reefs (4 crabs and 2.3 urchins per reef, on average) translated into low overall rates of macroalgal removal. Perhaps for this reason, there was no relationship between the density of these large macroinvertebrates or their grazing rate and macroalgal cover on patch reefs. Nevertheless, we calculated that macroalgal consumption by Maguimithrax crabs alone could exceed macroalgae production with a doubling of their current low abundance; a 2.6-fold increase in Tripneustes urchin abundance would achieve the same result. Our results suggest that large herbivorous macroinvertebrates, some of which are currently the target of artisanal fishing in many Caribbean countries, could contribute greatly to the recovery of coral reefs with established macroalgal communities, at least in patch reef habitats.
Shifts in competitive balance between key functional groups may drive regime shifts in tropical and temperate marine ecosystems. On shallow reefs, regime shifts increasingly involve changes from spatial dominance by foundation species (e.g. reefbuilding corals, canopy-forming algae) to dominance by turf-forming algae differing in structural complexity. To disentangle competitive inter actions fromother processes that may contribute to these shifts, we conducted a global meta-analysis of manipulative competition experiments between foundation and turf-forming species. Canopy- forming algae had consistently negative effects on abundance of turfforming algae, particularly on subtidal reefs, but with a tendency towards larger effects on delicate filamentous forms compared to articulated coralline and corticated/coarsely branching turf. Competitive effects of turf-forming algae on canopy species were limited to early life-history stages, and similarly varied between turf functional groups and between subtidal and intertidal reefs. Conversely, shorter filamentous turf assemblages typical of tropical reefs had no significant effect on settlement and survival of coral larvae. Interactions between turf-forming algae and established coral colonies were negative overall, but variable in magnitude. Mean effect sizes indicated that corals suppress turf abundance, but not vice versa. However, turf-forming algae significantly im - pacted coral growth and tissue mortality. We suggest reefs with extensive cover of foundation species are resistant to proliferation of turf algae, but competition will inhibit recovery of reefs following disturbances that enable turf algae to establish. Therefore, competitive effects of foundation and turf-forming species must be accounted for to effectively evaluate the stability of these undesirable regime shifts and recovery potential under alternative climate and management scenarios.