Demography and Population Dynamics of Massive Coral Communities in Adjacent High Latitude Regions (United Arab Emirates)

Abstract

Individual massive coral colonies, primarily faviids and poritids, from three distinct assemblages within the southeastern Arabian Gulf and northwestern Gulf of Oman (United Arab Emirates) were studied from 2006-2009. Annual photographic censuses of approximately 2000 colonies were used to describe the demographics (size class frequencies, abundance, area cover) and population dynamics under "normal" environmental conditions. Size class transitions included growth, which occurred in 10-20% of the colonies, followed in decending order by partial mortality (3-16%), colony fission (<5%) and ramet fusion (<3%). Recruitment and whole colony mortality rates were low (<0.7 colonies/m(2)) with minimal interannual variation. Transition matrices indicated that the Arabian Gulf assemblages have declining growth rates (λ<1) whereas the massive coral population is stable (λ = 1) in the Gulf of Oman. Projection models indicated that (i) the Arabian Gulf population and area cover declines would be exacerbated under 10-year and 16-year disturbance scenarios as the vital rates do not allow for recovery to pre-disturbance levels during these timeframes, and (ii) the Gulf of Oman assemblage could return to its pre-disturbance area cover but its overall population size would not fully recover under the same scenarios.

Full text

Demography and Population Dynamics of Massive Coral
Communities in Adjacent High Latitude Regions (United
Arab Emirates)
Kristi A. Foster*, Greg Foster
Nova Southeastern University Oceanographic Center, Dania Beach, Florida, United States of America
Abstract
Individual massive coral colonies, primarily faviids and poritids, from three distinct assemblages within the southeastern
Arabian Gulf and northwestern Gulf of Oman (United Arab Emirates) were studied from 2006–2009. Annual photographic
censuses of approximately 2000 colonies were used to describe the demographics (size class frequencies, abundance, area
cover) and population dynamics under ‘‘normal’’ environmental conditions. Size class transitions included growth, which
occurred in 10–20% of the colonies, followed in decending order by partial mortality (3–16%), colony fission (,5%) and
ramet fusion (,3%). Recruitment and whole colony mortality rates were low (,0.7 colonies/m
2
) with minimal interannual
variation. Transition matrices indicated that the Arabian Gulf assemblages have declining growth rates (l,1) whereas the
massive coral population is stable (l = 1) in the Gulf of Oman. Projection models indicated that (i) the Arabian Gulf
population and area cover declines would be exacerbated under 10-year and 16-year disturbance scenarios as the vital rates
do not allow for recovery to pre-disturbance levels during these timeframes, and (ii) the Gulf of Oman assemblage could
return to its pre-disturbance area cover but its overall population size would not fully recover under the same scenarios.
Citation: Foster KA, Foster G (2013) Demography and Population Dynamics of Massive Coral Communities in Adjacent High Latitude Regions (United Arab
Emirates). PLoS ONE 8(8): e71049. doi:10.1371/journal.pone.0071049
Editor: Maura Geraldine Chapman, University of Sydney, Australia
Received February 5, 2013; Accepted June 27, 2013; Published August 21, 2013
Copyright: ß 2013 Foster, Foster. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: National Coral Reef Institute (NCRI) at Nova Southeastern University’s Oceanographic Center provided funding to the authors for the duration of this
study. Additionally, funding for this study was provided by Dolphin Energy Ltd., in association with the World Wide Fund for Nature - Emirates Wildlife Society
(WWF-EWS), in 2006–2007 as part of a project called ‘‘Coral Reef Investgations in Abu Dhabi and Eastern Qatar’’, for which NCRI was the technical advisor. Field
support including the use of boats, staff, research stations and other equipment was provided by NCRI, the Environmental Agency - Abu Dhabi (EAD), Dibba
Marine Centre of the Ministry of Environment and Water, Dibba-Fujairah Municipality, Fujairah Municipality, and Fujairah Marina C lub. The funders were involved
in the study design as follows: 1) The ‘‘Coral Investigations’’ project began in 2005, a year prior to this study. NCRI and EAD had selected several of the Arabian
Gulf monitoring station locations in 2005. The authors selected additional Arabian Gulf sites in 2006 and 2007 and all Gulf of Oman sites in 2006 with the approval
of NCRI and the respective team members and agencies in Abu Dhabi, Dibba and Fujairah. 2) The configuration of the three belt transects which radiated from a
central point and from which all photo mosaics were created was previously used by Dr. Bernhard Riegl of NCRI. This patte rn was replicated throughout this
study. All monitoring stations were physically installed by the authors and local team members. The funders had no role in data collection and analysis, decision
to publish, or preparation of the manuscript.URLs for funders: http://www.nova.edu/ocean/ncri/index.html; http://www.dolphinenergy.com/; http://uae.panda.
org/ who_we_are .
Competing Interests: The authors wish to make note of indirect funding from a commercial source, Dolphin Energy Limited, for a portion of the research. This
does not alter their adherence to all the PLOS ONE policies on sharing data and materials. Dolphin Energy provided funding to the Emirates Wildlife Society-World
Wide Fund for Nature (EWS-WWF) for which the National Coral Reef Institute at Nova Southeastern University’s Oceanographic Center provided technical
expertise. The authors were employees of Nova Southeastern University through which the research efforts described within the manuscript continued well
beyond the Dolphin Energy project.
* E-mail: kfoster@nova.edu
Introduction
Global climate change is predicted to increase the frequency,
intensity and duration of disturbances that impact coral reefs (e.g.
[1–4]). As coral communities have been shown to require 10–30
years to recover after a major disturbance (e.g. [5–8]), it is possible
that taxa susceptible to environmental disturbances (i.e. branching
and tabular acroporids and pocilloporids) diminish or become
locally extirpated while the resistant taxa (i.e. massive poritids and
faviids) would become the dominant reef builders. Under such
circumstances, it will be the fates of the surviving massive corals
that shape future coral communities in the southeastern Arabian
Gulf, the northwestern Gulf of Oman, and, by extension, other
similarly structured coral reefs (if high latitude communities are
indeed the precursors to tropical coral reefs influenced by climate
change [9–10]).
Coral communities in the territorial waters of the United Arab
Emirates have recently been exposed to a series of natural
disturbances that have had significant impacts on branching and
tabular Acropora and Pocillopora spp. colonies. Elevated temperature
anomalies in 1996, 1998 and 2002 were associated with the mass
mortality of up to 99% of the acroporids in the southeastern
Arabian Gulf (i.e. Abu Dhabi and Dubai) [11–13]. Cyclone Gonu
damaged .50% of the acroporids in the northwestern Gulf of
Oman (e.g. Fujairah) in 2007 [14] and was followed by a
Cochlodinium polykrikoides harmful algal bloom (HAB) in 2008–09
which resulted in mass mortality of Pocillopora damicornis [14–16].
The aforementioned disturbances had lesser effects on massive
coral populations, with greater than 75% survival of poritids and
faviids during each event [11], [13–14]. Coral dominance in both
regions has shifted from highly susceptible branching and tabular
species to more resistant massive species. Whether these shifts are
short-lived or persistent depends on many factors including (i)
PLOS ONE | www.plosone.org 1 August 2013 | Volume 8 | Issue 8 | e71049
ews_wwf/ /

recruitment of new acroporids and pocilloporids from local
surviving colonies and from remote larval sources [5], [9], (ii)
the frequency of disturbance events; and (iii) recruitment, growth
and survival of massive corals.
The objectives of this study are to (i) describe the demographics
and dynamics of the massive coral communities in the southeast-
ern Arabian Gulf and the northwestern Gulf of Oman, (ii) use the
vital rates, based on temporal comparisons of individual colonies,
to develop size class transition probability matrices, and (iii) project
the population sizes and live coral area cover for these
communities over the next 40 years.
Methods
Annual Surveys
Hard coral populations were surveyed annually in the
southeastern Arabian Gulf and northwestern Gulf of Oman
between 2006 and 2009 (Figure 1, Table 1). Permission to conduct
the surveys was granted by the respective regulatory agencies: (i)
Environmental Agency – Abu Dhabi for all Arabian Gulf sites, (ii)
Dibba-Fujairah Municipality and the Dibba Marine Centre of the
Ministry of Environment and Water for the Dibba South site in
the Gulf of Oman, and (iii) Fujairah Municipality for the Mirbah
North site in the Gulf of Oman. Permanent monitoring stations
were installed in order to allow for repetitive photographic surveys
of benthic areas and specific coral colonies. Digital images were
taken along three 10 m61.5 m belt transects at depths ,10 m
within each monitoring station using a rigid photo-framer that
oriented the camera at a fixed distance of 50 cm above the
benthos. The 0.5 m60.75 m base of the framer served as a border
within each image to provide known dimensions for subsequent
image analysis.
Images were joined into a single mosaic for each belt transect. A
number was assigned to each massive coral that appeared as a
whole colony within the photo mosaics. (Branching corals were
also present in certain transects but were excluded from this study
for which the focus was on the slower-growing, disturbance-
resistant massive coral demographics. The status of the branching
corals has been published elsewhere [17]). Each numbered coral
was traced using the Area Analysis function in Coral Point Count
(CPCe) [18], which calculated colony area cover (planar view).
Transect data within each site were pooled to provide percent live
coral cover and coral densities. Data processing for assemblage
classification and ordination included (i) fourth root transforma-
Figure 1. Map of southeastern Arabian Gulf and northwestern Gulf of Oman study areas. Monitoring station locations at Al Hiel (AHL), Bu
Tinah (BTN1&2), Yasat (YST), Saadiyat (SDY), Ras Ghanada (GHN), Dibba South (DS), and Mirbah North (MN).
doi:10.1371/journal.pone.0071049.g001
Massive Coral Demography/Population Dynamics (UAE)
PLOS ONE | www.plosone.org 2 August 2013 | Volume 8 | Issue 8 | e71049

tion for the production of a Bray-Curtis similarity matrix; (ii)
agglomerative, hierarchical cluster analysis using group average
sorting; and (iii) non-metric multi-dimensional scaling (nMDS).
Non-parametric similarity of percentages (SIMPER) tests were
performed to determine which taxa contributed the most to
within-group similarities and among-group dissimilarities. All
multivariate analyses were implemented using PRIMER software
[19].
Size Class Determination
Massive colonies were grouped into five size-dependent
classifications (‘‘SC’’) (Table 2) based on area cover (where areas
were assumed to be based on circular colonies with A = pr
2
). To
determine the most appropriate groupings, size frequency
distributions were compared for areas associated with radius
increments of 1, 2, and 3 cm. The optimal, normally distributed
size-dependent groupings were those based on radius increments
of 2 cm. (Size classes based on radius increments of 1 cm and
3 cm were sub-optimal with frequency distributions skewed to the
left and right, respectively.).
Transition Matrices
Size class transition matrices were developed for faviids, poritids
and all massive corals (i.e. faviids, poritids, siderastreids, and
dendrophylliids) in each of the regional assemblages. The use of
five size classes resulted in 565 matrices in which each element
represents the mean probability of moving from a starting size
class or ‘‘state’’ (column) to ending size class or ‘‘fate’’ (row) [20–
21]. The matrices include growth (G) to the next
`
largest size class,
size class stability (S) by remaining within the same group, or
partial mortality (PM) to a smaller size class. Corals may also
experience fission (the regression of a single colony into multiple
smaller ramets) or fusion (i.e. two or more ramets grow together)
[22–23]. In these cases, area coverages of each ramet set were
pooled and compared to the size class for the respective parent
colony which underwent fission or for the resulting fused colony.
The probabilities of the fission and fusion transitions were added
to the corresponding partial morality, size class stability or growth
elements within the matrices. The resulting probability matrices,
based on 4000+ individual size class transitions, were used to
project the number of corals in each size class during year t+1,
which equals the number in each size class at year t multiplied by
the respective size class transition probabilities plus the mean
number of corals which enter the population through recruitment
(R) (Eq. 1). SC1 colonies that were visible within the belt transect
images during a given year, but had not been visibile the previous
survey, were recorded as recruits. (Image resolution was clear
enough to identify colonies as small as 0.1 cm
2
to genus; however,
some of the smaller Favia and Favites colonies lacked the
morphological characteristics that help to differentiate the species,
so these taxa were pooled into the Favia/Favites group).
SC1
SC2
SC3
SC4
SC5
0
B
B
B
B
B
B
B
B
@
1
C
C
C
C
C
C
C
C
A
(tz1)
~
SPMPMPMPM
G S PM PM PM
0 G S PM PM
00 GSPM
00 0 GS
0
B
B
B
B
B
B
B
B
@
1
C
C
C
C
C
C
C
C
A
|
SC1
SC2
SC3
SC4
SC5
0
B
B
B
B
B
B
B
B
@
1
C
C
C
C
C
C
C
C
A
t
8
>
>
>
>
>
>
>
>
<
>
>
>
>
>
>
>
>
:
9
>
>
>
>
>
>
>
>
=
>
>
>
>
>
>
>
>
;
z
R
0
0
0
0
0
B
B
B
B
B
B
B
B
@
1
C
C
C
C
C
C
C
C
A
ð1Þ
A565 matrix has five eigenvalues, l
i
, or solutions to the matrix.
The dominant eigenvalue (i.e. the largest, positive eigenvalues that
is a real number) is the growth rate of the size class-structure
population [20], [24]: (i) for l .1, the population is growing, (ii)
for l = 1, the population is stable, and (iii) for l ,1, the population
is declining. The ratio of the dominant eigenvalue to the absolute
value of the second largest eigenvalue, known as the damping
Table 1. Descriptions of repetitive monitoring sites in the southeastern Arabian Gulf and northwestern Gulf of Oman.
Station Site Name Depth (m) Location Region Assemblage Year(s)
YST Yasat Ali 3.0–4.7 Island SE Arabian Gulf AG1 2006–09
BTN1 Bu Tinah North 1.8–3.6 Island SE Arabian Gulf AG1 2006–09
BTN2 Bu Tinah West* 2.0–3.5 Island SE Arabian Gulf AG1 2006–08
AHL Al Hiel 2.6–4.2 Island SE Arabian Gulf AG1 2006–09
SDT Saadiyat 5.7–7.2 Coastal SE Arabian Gulf AG2 2007–09
GHN Ras Ghanada 7.6–8.5 Coastal SE Arabian Gulf AG2 2007–09
DS Dibba South 6.7–8.1 Coastal NW Gulf of Oman GO 2007–09
MN Mirbah North 4.5–6.9 Coastal NW Gulf of Oman GO 2007–09
*The Bu Tinah West monitoring station was damaged between 2007 and 2008, presumably as a result of winter storms; therefore, 2007–2008 and 2008–2009 temporal
comparisons for Assemblage AG1 were based on the three remaining sites.
doi:10.1371/journal.pone.0071049.t001
Table 2. Size-dependent classifications for massive coral
colonies.
Size Class Area Cover (cm
2
) Est. Radius (cm)
1 ,12.7 ,2
2 12.7–50.2 2–4
3 50.3–113.0 4–6
4 113.1–201.1 6–8
5 .201.1 .8
Massive coral colonies were grouped into five size classes based on their
measured area cover and estimated radii (assuming circular colonies, A = pr
2
).
doi:10.1371/journal.pone.0071049.t002
ð1Þ
Massive Coral Demography/Population Dynamics (UAE)
PLOS ONE | www.plosone.org 3 August 2013 | Volume 8 | Issue 8 | e71049

ratio, provides the rate of convergence of the population toward a
stable stage distribution (i.e. the larger the damping ratio, the
quicker a population will return to its stable state after a
disturbance) [20–21]. Sensitivities and elasticites, represented as
surface plots, are measures of perturbation analyses that quantify
the relative contribution of each vital rate to the population growth
by adjusting each rate by a specific amount and by a specific
proportion, respectively [20–21]. The dominant eigenvalues (i.e.
population growth rates), stable size class distributions, sensitivities,
and elasticities for the transition matrices were calculated using
PopTools add-in for Excel [25].
Projection Models
Projections were modeled through 2050 as idealized, best-case
scenario forecasts of massive coral populations and live area cover
[20–21]. Such projections assumed that (i) current parameters
remain unchanged over time, (i.e. during normal, disturbance-free
intervals); (ii) coral vital rates (e.g. growth, stability, fission, fusion,
mortality) include the interactions among corals and other benthic
organisms, responses to the surrounding environment and other
factors that affect population structures; and (iii) the mean
recruitment rates between 2006–2009 occur annually throughout
the projection period.
For comparison, alternative disturbance scenarios were calcu-
lated for the assemblages whereby mass mortality events occur
every 16 years (i.e. the midpoint between the historical 15–17 year
disturbance intervals for the southeastern Arabian Gulf region [9])
and every 10 years (i.e. the timeframe between the two most recent
disturbances which occurred in 2002 and 2012). Both disturbance
internals were presumed to result in the death of 25% of the
massive corals [11], [13] while the population dynamics for the
surviving 75% of the corals remain unchanged.
Results and Discussion
Hard Coral Assemblages
Cluster analysis differentiated three hard coral assemblages
(designated AG1, AG2 and GO1), each with .80% between-site
similarity (Figure 2). AG1 and AG2 are subsets of the southeastern
Arabian Gulf sites whose hard coral populations were sparse and
moderate populations, respectively, of Porites harrisoni and other
massives. Assemblage GO1 consisted of two sites along the
northwestern coast of the Gulf of Oman that were moderately
populated by Platygyra daedalea, Favia spp. and other massive corals.
Site selections for this study were made haphazardly to include a
cross-section of known coral community locations (i.e. frequently
visited coastal sites as well as locations near offshore ranger
stations) and independently of population demographics. Howev-
er, it was not surprising that the Gulf of Oman sites comprised an
assemblage separate from the Arabian Gulf sites. Exposure to
salinity and seawater temperature extremes (i.e. $40 ppt and 14–
36uC in the Arabian Gulf [26–27] compared to 36.5 ppt and 22–
31uC [28–29] in the Gulf of Oman) has limited the number of
species in the Arabian Gulf to approximately one-third of those
found in the Gulf of Oman [11], [29–31]. In this study, only 10 of
the 15 scleractinian genera recorded at the Gulf of Oman
monitoring stations were also observed in the Arabian Gulf.
AG1 and AG2 were located in the southwest corner of the study
area and near the Abu Dhabi coast, respectively (Figure 1).
Further studies are needed to determine whether this constitutes a
true west-east geographic gradient or if other factors contribute to
the different community compositions (e.g. proximity to the
prevailing surface current, coastal versus island dynamics). Prior
surveys have characterized the coral communities near Dubai
(approximately 115 km eastward of this study) into five well-
separated assemblages [5], [12]. AG1 and AG2 were composi-
tionally similar to the massive coral understories of two of these
Dubai assemblages [17] which may suggest that these assemblages
(and possible others) are distributed throughout the region and
that the apparent geographic groupings of AG1 and AG2 were
coincidental.
Population Structures
AG1. A sparsely populated assemblage (7% area cover)
dominated by Porites harrisoni, Platygyra daedalea and Cyphastrea
microphthalma (Table 3). Mean coral density was 2.8 live colonies/
m
2
, comprised primarily of size class (SC) 1–2 colonies (i.e. area
cover #50.2 cm
2
) (Figure 3). Subordinate taxa included faviids
(Favia, Favites, Leptastrea spp.), other poritid species (P. solida, P. lutea)
and two Siderastrea savignyana colonies. Live acroporids were not
observed within the vicinities of the monitoring stations; however,
consolidated rubble indicated that acroporids had existed within
these sites at one time.
AG2. A moderately populated assemblage (32% area cover)
dominated by P. harrisoni, P. daedalea and the Favia/Favites group
(Table 3). Mean massive coral density was 12.7 live colonies/m
2
,
consisting primarily of SC5 poritids (i.e. area covers .201.1 cm
2
)
and SC1–2 faviids (Figure 3). Subordinate taxa included other
faviids ( C. microphthalma, Leptastrea transversa), other poritids (P. solida,
P. lutea, P. nodifera) and other massive coral species (S. savignyana,
Coscinaraea columna, Turbinaria reniformis). Acroporids were also
observed within this assemblage, comprising ,2.2% of the total
benthic area cover, but these were excluded from this study as the
focus was on the massive coral demographics. (The acroporids
were subordinate to the massive corals in number and area cover
and are likely to remain subordinate unless this assemblage
experiences an extended disturbance-free period of .15 years, one
or more recruitment pulses of .6 recruits per year, or both [17]).
GO1. A moderately populated assemblage (32% area cover)
dominated by P. daedalea, the Favia/Favites group and mixed Porites
spp. (Table 3). Mean massive coral density was 4.6 live colonies/
m
2
, comprised primarily of SC5 Platygyra spp. and SC1–2 mixed
faviids (Figure 3). Subordinate taxa included other faviids
(Cyphastrea, Leptastrea, Plesiastrea), poritids (Goniopora) and side-
rastreids (Coscinaraea, Psammacora, Pseudosiderastrea, Siderastrea ). Spo-
radic pocilloporids and acroporids were observed within this
assemblage. The maximum branching coral cover (3.8%) was
observed at the Mirbah North monitoring station in 2006 (i.e.
prior to Cyclone Gonu) during a 12–40 year disturbance-free
period [14] within which the massive corals established and
retained dominance over the branching corals. Annual turnover of
post-cyclone pocilloporid and acroporid recruits indicates that
branching corals are likely to may remain subordinate to the
massive corals in the near future [17].
Recruitment. Faviid and poritid recruit sizes ranged between
0.1 and 12.6 cm
2
, with a mean area cover of 4.463.2 cm
2
. First
year recruits near Abu Dhabi and in the Gulf of Oman were
approximately half the size of those recorded as juveniles/recruits
in other regional studies (e.g. #4 cm max diameter in Dubai [5]
and ,5 cm diameter in the Red Sea [32]). Use of the broader
juvenile/recruit grouping would have included SC1 and SC2
colonies herein; however, only 15% of the combined SC1 and
SC2 size classes were first year recruits. A similar analysis of the
SC1 colonies indicated that first year recruits comprised 32% of
SC1 colonies, with the remainder being juveniles or small adults
that exhibited size class stability (43%) or shrinkage from larger
size classes (25%). Such results may aid future regional studies
Massive Coral Demography/Population Dynamics (UAE)
PLOS ONE | www.plosone.org 4 August 2013 | Volume 8 | Issue 8 | e71049

derive recruitment estimates from datasets that cannot differen-
tiate first-year recruits from juveniles or small adults.
Recruit abundance ranged between 0.0 and 0.3 colonies/m
2
,
with an exception of 0.70 faviids/m
2
in 2008–09 within AG2
(Table 4). The mean winter seawater temperature in the
southeastern Arabian Gulf was 1–2uC warmer in 2008–09 than
in 2006–07 and 2007–08 [33], which perhaps contributed to the
more favorable faviid recruit survival in the region. While seawater
temperature has been reported as a significant factor related to
gamete maturation and spawning in the Arabian Gulf [34], further
investigations are needed to determine whether winter seawater
Figure 2. Coral assemblages by cluster analysis and multi-dimensional scaling. (upper) Bray-Curtis similarity cluster analysis depicting
three assemblages; AG1, AG2 and GO1. (lower) MDS graphic representation with ovals around assemblages indentified by dendrogram. AG1 is
comprised of Al Hiel (AHL), Bu Tinah (BTN1&2) and Yasat (YST). AG2 is comprised of Saadiyat (SDY) and Ras Ghanada (GHN). GO1 is comprised of
Dibba South (DS) and Mirbah North (MN).
doi:10.1371/journal.pone.0071049.g002
Massive Coral Demography/Population Dynamics (UAE)
PLOS ONE | www.plosone.org 5 August 2013 | Volume 8 | Issue 8 | e71049

temperatures impact the survival and growth of recently settled
larvae into SC1 colonies the following year.
In general, the numbers of recruits were comparable between
assemblages in a given year, despite the population density in AG2
being approximately four times greater than in AG1 and GO1,
suggesting that factors other than adult densities within the local
populations are influencing recruitment success. For example,
faviid recruit:adult ratios in 2009 were 1:12 in both AG1 and AG2
despite a greater than tenfold difference in adult densities. In
contrast, poritid recruit:adult ratios during 2009 were 1:7 and 1:16
for AG1 and AG2, respectively, although the adult densities were
of the same order of magnitude. Recruit:adult ratios were
considerably lower (1:42) for faviids and higher (3:2) for poritids
in the Gulf of Oman compared to the Arabian Gulf. Further
investigations are required to identify the spatial variation patterns,
if any, and possible contributing factors.
Table 3. Taxa groups responsible for .90% within-group similarities and among-group dissimilarities based on SIMPER analysis.
Arabian Gulf 1 (AG1) Groups: AG1/AG2
Average similarity: 70.89 Cont. (%) Cum. (%) Average dissimilarity: 48.15 Cont. (%) Cum. (%)
Porites 59.4 59.4 Platygyra 23.5 23.5
Platygyra 22.9 82.3 Favia/Favites 22.5 45.9
Cyphastrea 8.3 90.6 Porites 20.0 66.0
Turbinaria 15.9 81.8
Cyphastrea 8.3 90.2
Arabian Gulf 2 (AG2) Groups: AG1/GO1
Average similarity: 84.23 Average dissimilarity: 56.71
Porites 40.5 40.5 Platygyra 35.4 35.4
Platygyra 28.5 69.0 Favia/Favites 21.8 57.2
Favia/Favites 21.1 90.1 Porites 16.9 74.2
Siderastrea 9.6 83.8
Cyphastrea 6.7 90.5
Gulf of Oman (GO1) Groups: AG2/GO1
Average similarity: 81.05 Average dissimilarity: 34.9
Platygyra 52.9 52.9 Porites 41.9 41.9
Favia/Favites 29.0 81.9 Turbinaria 18.0 59.9
Porites 10.3 92.2 Platygyra 14.1 74.0
Siderastrea 11.2 85.2
Coscinaria 5.7 90.9
Cont. (%) is the percentage contributed by the respective taxa group to the (dis)similarity. Cum (%) is the cumulative percentage of (dis)similarity.
doi:10.1371/journal.pone.0071049.t003
Figure 3. Size frequency distributions for massive corals by region. POR = poritids, FAV = faviids; ALL = all massive coral taxa Size Class color
coding: SC1 = red, SC2 = orange, SC3 = yellow, SC4 = green, SC5 = blue.
doi:10.1371/journal.pone.0071049.g003
Massive Coral Demography/Population Dynamics (UAE)
PLOS ONE | www.plosone.org 6 August 2013 | Volume 8 | Issue 8 | e71049

Whole colony mortality. Whole colony mortality ranged
between 0.0 and 0.6 colonies/m
2
, equivalent to #16% of the
population (Table 5), demonstrating that minor levels of mortality
occur as part of ‘‘normal’’ turnover in these populations (i.e. in the
absence of a major disturbance) [22]). Whole colony mortality
occurred most frequently in SC1 and SC2 corals (63% and 21% of
all mortalities, respectively). In most cases, these colonies were no
longer visible in subsequent years, indicating either removal from
the substrate or overgrowth. The probability of mortality
decreased with increasing colony size, a pattern that has been
reported in other studies (e.g. [22–23], [35–38]).
The corals within the GO1 were exposed to a red tide event
during the 2008–2009 sample period [14], yet whole colony death
was similar to that for AG1 faviids and all massive corals which
were not exposed to a similar disturbance. Such results indicate
that additional studies are needed to determine the proportion of
deaths attributable to a disturbance above and beyond normal
population losses.
Size Class Stability, Growth and Partial Mortality
Size class stability was the most likely fate for colonies; 45–65%
of the colonies in AG1, 70–74% in AG2, and 70% in GO1
Table 4. Populations and sexual recruitment of faviids, poritids, and all massive corals.
Faviids Poritids All Massives
06–07* 07–08 08–09 06–07* 07–08 08–09 06–07* 07–08 08–09
AG1
Live colonies 68 68 63 455 296 284 523 365 348
Recruits 2 0 5 29 11 38 31 11 43
Live colonies/m
2
0.4 0.5 0.5 2.5 2.2 2.1 2.9 2.7 2.6
Recruits/m
2
,0.1 N/A ,0.14 0.2 0.1 0.3 0.2 0.1 0.3
AG2
Live colonies 764 769 343 368 1128 1159
Recruits 1 63 7 23 8 88
Live colonies/m
2
8.5 8.5 3.8 4.1 12.5 12.9
Recruits/m
2
,0.1 0.7 ,0.1 0.3 ,0.1 1.0
GO1
Live colonies 294 252 6 2 304 260
Recruits 3 6 0 3 3 9
Live colonies/m
2
3.3 5.6 ,0.1 ,0.1 3.4 5.8
Recruits/m
2
,0.1 ,0.1 N/A ,0.1 ,0.1 0.2
Assemblages: AG1– Arabian Gulf 1; AG2 = Arabian Gulf 2; GO1 = Gulf of Oman.
* = includes Bu Tinah West monitoring station.
doi:10.1371/journal.pone.0071049.t004
Table 5. Whole colony mortalities.
Faviids Poritids All Massives
06–07* 07–08 08–09 06–07* 07–08 08–09 06–07* 07–08 08–09
AG1
Whole colony deaths deaths 4 6 10 65 45 31 69 52 42
Deaths/m
2
,0.1 ,0.1 ,0.1 0.4 0.3 0.2 0.4 0.4 0.3
Percent Mortality (%) 5.6 8.8 15.9 14.3 15.2 10.9 13.2 14.2 12.1
AG2
Whole colony deaths deaths 6 21 4 19 10 40
Deaths/m
2
,0.1 0.2 ,0.1 0.2 0.1 0.4
Percent Mortality (%) 0.8 2.7 1.2 5.2 0.9 3.4
GO1
Whole colony deaths 1 27 0 1 1 29
Deaths/m
2
,0.1 0.6 N/A ,0.1 ,0.1 0.6
Percent Mortality (%) 0.3 10.7 N/A 50.0 0.3 11.2
Assemblages: AG1– Arabian Gulf 1; AG2 = Arabian Gulf 2; GO1 = Gulf of Oman,
* = includes Bu Tinah West; Percent Mortality is the percent of the population that experienced whole colony death.
doi:10.1371/journal.pone.0071049.t005
Massive Coral Demography/Population Dynamics (UAE)
PLOS ONE | www.plosone.org 7 August 2013 | Volume 8 | Issue 8 | e71049

(faviids). The probability of size class stability increased with
increasing colony size (Table 6), with mean annual probabilities
$0.845 for SC5 colonies. A high proportion of size stability is not
unexpected as it may take a colony several years to transition into
a larger size class based on an annual growth rate of 1–2 cm for
massive corals in this region [9].
Growth was the second most likely transition with 17–24% of
the colonies in AG1, 12–15% in AG2, and 9–21% in GO1
(faviids) moving into the next larger size class (Table 7). The mean
annual probability of growth increased with increasing size class
for AG1 and AG2 poritids. No discernible trends were observed
for faviids in any of the assemblages. It is interesting to note that
faviid growth continued to occur with the GO1 population despite
prolonged exposure to the red tide between 2008 and 2009 [14].
A slightly smaller percentage of all colonies, 3–16%, experi-
enced partial mortality (i.e. shrinkage of live tissue area,
unfragmented by bare skeleton) and transitioned into smaller size
classes (Table 8). In the Arabian Gulf, .86% of the colonies that
underwent partial mortality regressed only one size class (SC5 = .
SC4, SC4 = . SC3, SC3 = . SC2, SC2 = . SC1) rather than
multiple size classes, which provides a baseline for negative size
class transitions under ‘‘normal’’ environmental conditions. In the
Gulf of Oman, faviid partial mortality doubled between 2008 and
2009; however, additional studies are needed to determine
whether this difference was due, in part or entirely, to the red
tide or if it was within the range of interannual variability.
Fission and Fusion
Fission and fusion played minor roles in the dynamics of the
AG1, AG2 and GO1 populations, with respective mean annual
probabilities of 0.00–0.06 and 0.0–0.03 (Tables 9–10). Similar
fission probabilities for other massive and foliaceous species were
reported in Jamaica (0.02–0.10; [23]) and in Australia (0.01–0.06;
[22]), indicating that low rates of fission occur among subtropical
and tropical coral communities even in the absence of environ-
mental stresses such as those associated with seawater temperature
extremes, hurricanes and other natural disturbances. Low
probabilities of fusion, in some circumstances, may be attributed
to the rates of tissue reconnection/growth which are currently
understudied. Certainly the extent of tissue loss during fission and
the distance between ramets will impact whether fusion in a
subsequent year is possible. Several years of growth may be
Table 6. Size class stabilities (no size class transitiona).
Faviids Poritids All Massives
06–07* 07–08 08–09 06–07* 07–08 08–09 06–07* 07–08 08–09
AG1
SC1 7 12671310741412
SC2 1163543029653632
SC3 12 181426 201738 3831
SC4 1186221720332526
SC5
3 4 9 62 54 82 65 58 91
Total 44 37 34 231 134 158 275 171 192
% of population 64.7% 54.4% 54.0% 50.8% 45.3% 55.6% 52.6% 46.8% 55.2%
Colonies/m
2
0.2 0.3 0.3 1.3 1.0 1.2 1.5 1.3 1.4
AG2
SC1 112 97 10 10 123 107
SC2 209 227 24 27 240 259
SC3 92 102 25 16 119 120
SC4 3842 1624 5468
SC5
86 78 180 193 268 273
Total 537 546 255 270 804 827
% of population 70.3% 71.0% 74.3% 73.4% 71.3% 71.4%
Colonies/m
2
6.0 6.1 2.8 3.0 8.9 9.2
GO1
SC1 15 6 0 0 15 6
SC2 3936 2 0 4338
SC3 2118 1 1 2319
SC4 9 5 0 0 9 5
SC5
123 97 2 0 125 97
Total 207 162 5 1 215 165
% of population 70.4% 64.3% 83.3% 50.0% 70.7% 63.5%
Colonies/m
2
2.3 3.6 ,0.1 ,0.1 2.4 3.7
Assemblages: AG1– Arabian Gulf 1; AG2 = Arabian Gulf 2; GO1 = Gulf of Oman.
SC = Size class;
* = includes Bu Tinah West.
doi:10.1371/journal.pone.0071049.t006
Massive Coral Demography/Population Dynamics (UAE)
PLOS ONE | www.plosone.org 8 August 2013 | Volume 8 | Issue 8 | e71049

required before ramets are capable of reconnecting during which
barriers (e.g. algal growth on exposed skeleton) may prevent
fusion. Other hindrances to fusion include additional shrinkage
and mortality of the ramets since previously damaged corals have
an increased likelihood of further damage [22].
On average, 2–3 ramets were generated when a parent colony
underwent fission. The majority (79–89%) of the pooled ramet
surface areas were in the same size classes as their respective
parent colonies whereas 11–21% of the fissions resulted in
transitions to smaller size classes. Similarly, 2–3 ramets grew
together to generate a fused colony. The majority (66–76%) of the
fused AG1 and AG2 colonies were in the same size classes as their
respectively pooled ramets whereas 24–33% of the fusions resulted
in transitions to larger size classes. All GO1 faviid fusions were
recorded as size class stability transitions.
Size class transition matrices. Mean transition probability
matrices were developed for faviids, poritids and all massive corals
(Table 11). Little information has previously been published
regarding the life histories of the massive coral species within the
Arabian Gulf [9] and the Gulf of Oman. The vital rates presented
herein may provide actual data for other predictive models that
would otherwise utilize estimations of recruitment, mortality, or
growth.
Although seemingly short, the 2–4 years of repetitive monitor-
ing used to generate the size class transition probability matrices
for AG1 and AG2 is comparable to other vital rate studies for
corals, gorgonians and sponges [22–23], [39–44]. Ideally, annual
data collection would continue in order to determine whether the
coral communities follow predictable cycles or whether irregular
patterns are the norm.
GO1 was exposed to a prolonged red tide event that persisted
between August 2008 and May 2009 [14]. The impacts of this
disturbance on vital rates (e.g. possible increased whole colony
death and partial mortality, decreased growth and size class
stability) were not independently tested. Because the focus of this
study was on the fate of massive corals under ‘‘normal’’
environmental conditions, transition matrices and projections for
GO1 were based on surveys in 2007 and 2008 only. These results
are included herein as a first published report of vital rates for the
massive corals in this region but should be considered as
preliminary.
The stable size class distributions (i.e. the eigenvectors associated
with the dominant eigenvalues), dominant eigenvalues and damping
ratios were determined for each assemblage (Table 12). The
dominant eigenvalues (l) for AG1 and AG2 were ,1, which result
in gradual population decay, whereas the GO1 eigenvalue (l =1)
Table 7. Growth profiles (postive size class transitions).
Faviids Poritids All Massives
06–07* 07–08 08–09 06–07* 07–08 08–09 06–07* 07–08 08–09
AG1
SC1 = . SC2 1 1 2 33 14 6 34 15 8
SC2 = . SC3 9 63191420282023
SC3 = . SC4 2 36162213182519
SC4 = . SC5
2 5 2 11 22 18 13 27 20
Total 14 15 13 79 72 57 93 87 70
% of population 20.6% 22.1% 20.6% 17.4% 24.3% 20.1% 17.8% 23.8% 20.1%
Colonies/m
2
,0.1 0.1 0.1 0.4 0.5 0.4 0.5 0.6 0.5
AG2
SC1 = . SC2 5947 5 3 6550
SC2 = . SC3 2737 1411 4149
SC3 = . SC4 1710 1411 3422
SC4 = . SC5
9 7 15 19 25 26
Total 112 101 48 44 165 147
% of population 14.7% 13.1% 14.0% 12.0% 14.6% 12.7%
Colonies/m
2
1.2 1.1 0.5 0.5 1.8 1.6
GO1
SC1 = . SC2 15 4 0 0 15 4
SC2 = . SC3 15 5 0 0 15 5
SC3 = . SC4 17 9 1 0 18 10
SC4 = . SC5
14 5 0 0 14 5
Total 61 23 1 0 62 24
% of population 20.7% 9.1% 16.7% 0.0% 20.4% 9.2%
Colonies/m
2
0.7 0.5 ,0.1 N/A 0.697 0.5
Values exclude colonies and ramets that underwent fusion.
Assemblages: AG1– Arabian Gulf 1; AG2 = Arabian Gulf 2; GO1 = Gulf of Oman.
SC = Size class;
* = includes Bu Tinah West.
doi:10.1371/journal.pone.0071049.t007
Massive Coral Demography/Population Dynamics (UAE)
PLOS ONE | www.plosone.org 9 August 2013 | Volume 8 | Issue 8 | e71049

Table 8. Partial mortalities (negative size class transitions).
Faviids Poritids All Massives
06–07* 07–08 08–09 06–07* 07–08 08–09 06–07* 07–08 08–09
AG1
SC5 = . SC4 1 106 567 66
SC5 = . SC3 0 004 114 11
SC5 = . SC2 0 000 100 10
SC5 = . SC1 0 000 000 00
SC4 = . SC3 0 1010 5610 66
SC4 = . SC2 0 110 010 12
SC4 = . SC1 0 000 000 00
SC3 = . SC2 1 4 1 6 8 2 7 12 3
SC3 = . SC1 0 013 113 12
SC2 = . SC1
0 0 0 26 6 6 26 6 6
Total 2 7 3 55 27 23 57 34 26
% of population 2.9% 10.3% 4.8% 12.1% 9.1% 8.1% 10.9% 9.3% 7.5%
Colonies/m
2
,0.1 ,0.1 ,0.1 0.3 0.2 0.2 0.3 0.3 0.2
AG2
SC5 = . SC4 1010 2013 3025
SC5 = . SC3 2 3 0 1 2 4
SC5 = . SC2 1 3 2 0 3 4
SC5 = . SC1 0 1 1 1 1 2
SC4 = . SC3 1815 6 7 2622
SC4 = . SC2 0 3 2 4 2 7
SC4 = . SC1 0 1 0 0 0 2
SC3 = . SC2 3821 4 5 4427
SC3 = . SC1 1 2 0 1 1 3
SC2 = . SC1
39 42 1 3 40 49
Total 109 101 36 35 149 145
% of population 14.3% 13.1% 10.5% 9.5% 13.2% 12.5%
Colonies/m
2
1.2 1.1 0.4 0.4 1.7 1.6
GO1
SC5 = . SC4 7 6 0 0 7 6
SC5 = . SC3 0 4 0 0 0 4
SC5 = . SC2 0 4 0 0 0 4
SC5 = . SC1 0 0 0 0 0 0
SC4 = . SC3 5 11 0 0 5 11
SC4 = . SC2 4 1 0 0 4 1
SC4 = . SC1 0 0 0 0 0 0
SC3 = . SC2 6 6 0 0 6 7
SC3 = . SC1 0 2 0 0 0 2
SC2 = . SC1
3 6 0 0 4 6
Total 25 40 0 0 26 41
% of population 8.5% 15.9% 0.0% 0.0% 8.6% 15.8%
Colonies/m
2
0.3 0.9 N/A N/A 0.3 0.9
Values exclude colonies that underwent fission.
Assemblages: AG1– Arabian Gulf 1; AG2 = Arabian Gulf 2; GO1 = Gulf of Oman.
SC = Size class;
* = includes Bu Tinah West.
doi:10.1371/journal.pone.0071049.t008
Massive Coral Demography/Population Dynamics (UAE)
PLOS ONE | www.plosone.org 10 August 2013 | Volume 8 | Issue 8 | e71049

indicates a stable population [20]. The damping ratios were 1.1–1.3,
indicating that faviids and poritids approach asymptotic behavior
(stability) at similar rates among the assemblages (i.e. similar
resilience/recovery following a disturbance) [45].
Sensitivities and Elasticities
Senstivities and elasticities are measures of perturbation analyses
that quantify the relative contribution of each vital rate to the
population growth by adjusting each rate by a specific amount and
by a specific proportion, respectively [20–21]. All sensivity and
elasticity matrices, displayed graphically as surface plots (Figure 4),
indicated that the dominant eigenvalues, l, were most affected by
changes in the upper right corners of the transition matrices which
correspond to the stability of SC5 colonies, partial mortality of
SC5 into SC4, and growth of SC4 into SC5. Sensitivities in AG2
were affected, in decreasing order, by the growth of SC2, SC2
stability, and growth of SC3, due to the large population of smaller
faviids within this assemblage. Sensitivies in GO1 were affected by
recruitment of SC1 colonies into the population.
Population Projections
The size class transition matrices were used to project
populations through 2050 (Figure 5). The following projections
are idealized and are not expected to occur but, rather, are shown
as best case, disturbance-free scenarios:
1. AG1 corals are not projected to reach a stable size class
distribution due to the continual change in the number and
proportion of the SC4–SC5 colonies. The number of colonies
decline by 60% through 2050 due to the mean annual
probabilities of mortality exceeding those for recruitment.
Despite the decrease in colony density, area cover (7.0% in
2009) will temporarily increase to a maximum of 8.0% through
2014–2017, due to the temporary increase in the number of
SC5 colonies, then gradually decrease to 4.8% by 2050.
2. AG2 faviids are projected to reach stable distributions,
dominated by SC2 colonies, around 2015–2020. AG2 poritids
are not projected to reach a stable size class distribution, due
primarily to the changing number and proportion of SC5
colonies. The number of faviids increase by 3% while the
poritids decrease by 45%; the net result is a 23% decrease in
Table 9. Fission – Parent colonies that underwent fission and mean number of ramets generated.
Faviids Poritids All Massives
06–07* 07–08 08–09 06–07* 07–08 08–09 06–07* 07–08 08–09
AG1
SC1 parents 0 0 0 1 0 0 1 0 0
SC2 parents 1 1 0 9 2 2 10 3 2
SC3 parents 2 2 2 1 1 2 3 3 4
SC4 parents 1 0 1 1 4 1 2 4 2
SC5 parents 0 0 0 1 5 2 1 5 2
% of population 5.9% 4.4% 4.8% 2.9% 4.1% 2.5% 3.2% 4.1% 2.9%
Colony fission/m
2
,0.1 ,0.1 ,0.1 ,0.1 ,0.1 ,0.1 ,0.1 0.1 ,0.1
Mean # of ramets 2.0 3.0 2.3 2.1 2.5 2.3 2.2 2.5 2.3
AG2
SC1 parents 0 0 0 0 0 0
SC2 parents 2 0 1 0 3 3
SC3 parents 2 1 3 0 5 6
SC4 parents 2 1 1 1 3 4
SC5 parents 2 1 15 1 17 19
% of population 1.0% 0.4% 5.8% 0.5% 2.5% 1.6%
Colony fission/m
2
,0.1 ,0.1 0.2 ,0.1 0.3 0.2
Mean # of ramets 2.1 2.0 2.3 2.0 2.2 2.2
GO1
SC1 parents 0 0 0 0 0 0
SC2 parents 1 4 0 0 1 4
SC3 parents 1 0 0 0 1 0
SC4 parents 1 0 0 0 1 0
SC5 parents 7 13 0 0 7 13
% of population 3.4% 6.7% 0.0% 0.0% 3.3% 5.0%
Colony fission/m
2
0.1 0.4 N/A N/A 0.1 0.4
Mean # of ramets 2.1 2.7 0.0 0.0 2.1 2.7
Assemblages: AG1– Arabian Gulf 1; AG2 = Arabian Gulf 2; GO1 = Gulf of Oman.
SC = Size class;
* = includes Bu Tinah West.
doi:10.1371/journal.pone.0071049.t009
Massive Coral Demography/Population Dynamics (UAE)
PLOS ONE | www.plosone.org 11 August 2013 | Volume 8 | Issue 8 | e71049

the massive coral population and a gradual decline from 32%
to 22% area cover over the projection period.
3. The distribution of SC1–SC4 faviids in GO1 are projected to
stabilize around 2020; however the number of SC5 colonies
will continue to gradually increase through 2050. (The poritid
community was too small for projections.) The GO1 projected
faviid area cover approaches 60% under idealized conditions;
however, this is likely an overestimation resulting from
transition probabilities that were based on a two-year data set.
The most recent 10-year and the historical 16-year [9]
disturbance intervals for this region were projected through
2050 (Figure 5) with the following results:
1. The 10-year and 16-year intervals are insufficient to allow AG1
and AG2 massives to recover from the population and area
cover losses associated with each disburbance.
2. GO1 populations approach but fall short of the predisturbance
levels within the 10-year and 16-year scenarios; however, borh
invervals are sufficient to return to the respective pre-
disturbance area covers.
The fates of all three assemblages depend heavily on the
continued health of the SC5 colonies. With declining populations
in both Arabian Gulf assemblages (plus the low area cover in AG1)
during normal environmental conditions, these populations are at
risk of collapse should a large proportion of the SC5 colonies
become compromised due to natural or anthropogenic stresses
(e.g. mass mortality, disease outbreaks, coastal development).
Current recruitment levels are insufficient to replace losses
associated with major disturbance events (e.g. up to 25% loss of
massive corals [11], [13]) as demonstrated in the 10- and 16-year
disturbance frequency scenarios (Figure 5).
Conclusions
Little information pertaining to hard coral vital rates within the
Arabian Gulf and the Gulf of Oman has been published to date.
This study documents the population dynamics during ‘‘normal’’
environmental conditions which may be used as baseline
comparisons when conducting coral community health surveys,
when reporting the effects of disturbance events (e.g. temperature
anomalies, cyclonces, red tides, disease outbreaks) or when
developing predictive ecological models for this region. Important
Table 10. Fusion –Fused colonies and mean number of ramets that fuse together.
Faviids Poritids All Massives
06–07* 07–08 08–09 06–07* 07–08 08–09 06–07* 07–08 08–09
AG1
SC1 0 002 002 00
SC2 0 001 101 10
SC3 0 008 118 11
SC4 0 000 020 02
SC5 0 000 350 35
% of population 0 0 0 2.4% 1.7% 2.8% 2.1% 1.4% 2.3%
Colonies/m
2
000,0.1 ,0.1 ,0.1 ,0.1 ,0.1 ,0.1
Mean # of ramets 0 0 0 2.5 2.0 2.4 2.5 2.0 2.4
AG2
SC1 1 0 0 1 1 1
SC2 0 5 1 1 1 6
SC3 3 2 3 1 6 3
SC4 1 3 1 1 2 4
SC5 3 3 3 8 6 11
% of population 1.0% 1.7% 2.3% 3.3% 1.4% 2.2%
Colonies/m
2
,0.1 0.1 ,0.1 0.1 0.2 0.3
Mean # of ramets 2.3 2.2 2.1 2.3 2.2 2,2
GO1
SC1 0 0 0 0 0 0
SC2 1 0 0 0 1 0
SC3 0 2 0 0 0 2
SC4 0 0 0 0 0 0
SC5 0 2 0 0 0 2
% of population 0.3% 1.6% 0.0% 0.0% 0.3% 1.5%
Colonies/m
2
,0.1 ,0.1 N/A N/A ,0.1 ,0.1
Mean # of ramets 2.0 2.3 0.0 0.0 2.0 2.3
Assemblages: AG1– Arabian Gulf 1; AG2 = Arabian Gulf 2; GO1 = Gulf of Oman.
SC = Size class;
* = includes Bu Tinah West.
doi:10.1371/journal.pone.0071049.t010
Massive Coral Demography/Population Dynamics (UAE)
PLOS ONE | www.plosone.org 12 August 2013 | Volume 8 | Issue 8 | e71049

Table 11. Transition probability matrices.
Faviids Poritids All Massive Corals
AG1
SC1 SC2 SC3 SC4 SC5 SC1 SC2 SC3 SC4 SC5 SC1 SC2 SC3 SC4 SC5
SC1 0.412 0.000 0.025 0.000 0.000 SC1 0.330 0.136 0.033 0.000 0.000 SC1 0.333 0.114 0.031 0.000 0.000
SC2 0.230 0.413 0.081 0.056 0.000 SC2 0.223 0.490 0.106 0.014 0.005 SC2 0.224 0.475 0.100 0.022 0.005
SC3 0 0.345 0.699 0.056 0.000 SC3 0 0.244 0.470 0.162 0.026 SC3 0 0.260 0.540 0.137 0.025
SC4 0 0 0.151 0.663 0.150 SC4 0 0 0.337 0.452 0.080 SC4 0 0 0.279 0.499 0.083
SC5 0 0 0 0.225 0.850 SC5 0 0 0 0.359 0.876 SC5 0 0 0 0.331 0.875
s
0.642 0.758 0.956 1.000 1.000
s
0.553 0.870 0.946 0.987 0.987
s
0.557 0.849 0.950 0.989 0.988
d
0.358 0.242 0.044 0.000 0.000
d
0.447 0.130 0.054 0.013 0.013
d
0.443 0.151 0.050 0.011 0.012
AG2
SC1 SC2 SC3 SC4 SC5 SC1 SC2 SC3 SC4 SC5 SC1 SC2 SC3 SC4 SC5
SC1 0.623 0.139 0.010 0.006 0.005 SC1 0.486 0.045 0.015 0.000 0.005 SC1 0.606 0.129 0.011 0.007 0.004
SC2 0.315 0.746 0.205 0.022 0.021 SC2 0.201 0.600 0.118 0.062 0.005 SC2 0.303 0.727 0.188 0.037 0.011
SC3 0 0.109 0.684 0.247 0.026 SC3 0 0.297 0.519 0.141 0.002 SC3 0 0.131 0.642 0.208 0.010
SC4 0 0 0.094 0.597 0.103 SC4 0 0 0.320 0.423 0.080 SC4 0 0 0.148 0.520 0.090
SC5 0 0 0 0.120 0.845 SC5 0 0 0 0.365 0.905 SC5 0 0 0 0.219 0.883
s
0.938 0.994 0.993 0.992 1.000
s
0.687 0.942 0.972 0.991 0.998
s
0.909 0.987 0.989 0.991 0.998
d
0.062 0.006 0.007 0.008 0.000
d
0.313 0.058 0.028 0.008 0.002
d
0.091 0.013 0.011 0.009 0.002
GO1
SC1 SC2 SC3 SC4 SC5 SC1 SC2 SC3 SC4 SC5
SC1 0.484 0.053 0.001 0.000 0.000 SC1 0.484 0.065 0.000 0.000 0.000
SC2 0.484 0.684 0.136 0.125 0.000 SC2 0.484 0.694 0.128 0.125 0.000
SC3 0 0.263 0.477 0.156 0.000 SC3 0 0.242 0.489 0.156 0.000
SC4 0 0 0.386 0.281 0.054 SC4 0 0 0.383 0.281 0.053
SC5 0 0 0 0.438 0.946 SC5 0 0 0 0.438 0.947
s
0.968 1.000 1.000 1.000 1.000
s
0.968 1.000 1.000 1.000 1.000
d
0.032 0.000 0.000 0.000 0.000
d
0.032 0.000 0.000 0.000 0.000
Assemblages: AG1– Arabian Gulf 1; AG2 = Arabian Gulf 2; GO1 = Gulf of Oman. SC = Size Class. Columns depict starting state, rows depict ending fate. s = probability of
survival within the respective size class, equal to the sum of probabilities in each column;d= probability of whole colony death (1-s).
doi:10.1371/journal.pone.0071049.t011
Table 12. Stable size class distributions, dominant eigenvalues and damping ratios.
AG1 AG2 GO1
FAV POR ALL FAV POR ALL FAV ALL
Stable SC Distributions Stable SC Distributions Stable SC Distributions
SC1 0.005 0.015 0.014 SC1 0.183 0.016 0.120 SC1 0.008 0.010
SC2 0.049 0.043 0.045 SC2 0.453 0.069 0.318 SC2 0.078 0.081
SC3 0.120 0.111 0.122 SC3 0.229 0.092 0.199 SC3 0.066 0.066
SC4 0.304 0.173 0.184 SC4 0.072 0.149 0.113 SC4 0.093 0.091
SC5 0.523 0.658 0.636 SC5 0.062 0.674 0.250 SC5 0.755 0.753
Dominant Eigenvalue Dominant Eigenvalue Dominant Eigenvalue
l 0.981 0.970 0.971 l 0.984 0.986 0.981 l 1.000 1.000
Real or complex Real Real Real Real or complex Real Real Real Real or complex Real Real
Damping Ratio Damping Ratio Damping Ratio
l
1
/|l
2
| 1.2 1.3 1.3 l
1
/|l
2
| 1.1 1.2 1.1 l
1
/|l
2
| 1.1 1.1
Assemblages: AG1– Arabian Gulf 1; AG2 = Arabian Gulf 2; GO1 = Gulf of Oman. Coral taxa: FAV = faviids; POR = poritids; ALL – all massive coral taxa. SC = Size Class.
doi:10.1371/journal.pone.0071049.t012
Massive Coral Demography/Population Dynamics (UAE)
PLOS ONE | www.plosone.org 13 August 2013 | Volume 8 | Issue 8 | e71049

Figure 4. Sensitivity and elasticity surface plots for all massive corals by region. Fate and state axes represent transitions between size
classes 1–5. Vertical axes represent the sensitivity and elasticity of the respective population growth rates, l, to perturbation analyses. Sensitivity/
Elasticity color coding: 0.0–0.2 = green, 0.2–0.4 = yellow, 0.4–0.6 = red, 0.6–0.8 = blue.
doi:10.1371/journal.pone.0071049.g004
Massive Coral Demography/Population Dynamics (UAE)
PLOS ONE | www.plosone.org 14 August 2013 | Volume 8 | Issue 8 | e71049

findings related to the massive corals in the UAE are summarized
as follows:
1. First year recruits have maximum radii #2 cm; however, only
32% of the colonies within this size range are recruits whereas
the remainder is comprised of juveniles and small adults. Mean
annual recruit abundance is typically low (#0.7 recruits/m
2
),
exclusive of possible recruitment pulses which were not
recorded during this study. Recruitment success does not
appear to be heavily influenced by adult densities within the
local population.
2. Whole colony mortality and partial mortality (i.e. shrinkage
into a smaller size class) may each effect up to 16% of the
population in a given year as part of ‘‘normal’’ turnover.
3. Colonies may take several years to transition into a larger size
class due to the slow growth rate for massive corals; only 9–
24% of the population experiences growth whereas 45–74%
maintains size class stability in a given year.
4. Fission and fusion play minor roles in the population dynamics,
effecting 0–6% and 0–3% of the colonies, respectively.
The size class transition probability matrices developed in this
study indicate that the Arabian Gulf massive coral assemblages
have negative population growth rates (l ,1) under ‘‘normal’’
environmental conditions. Projection models show that 10-year
and 16-year disturbance intervals further exacerbate the popula-
tion declines. It is, therefore, critical that these assemblages be
protected, to whatever extent possible, from disturbances that are
detrimental to their demographics or population dynamics (e.g.
Figure 5. Population and area cover projections through 2050 for all massive corals. Red MM(0) lines represesent optimal, disturbance-
free projections (zero mass mortality events). Black MM(16) and blue MM(10) lines represent mass mortality every 16 years and 10 years, respectively,
with 25% loss of massive corals during each disturbance.
doi:10.1371/journal.pone.0071049.g005
Massive Coral Demography/Population Dynamics (UAE)
PLOS ONE | www.plosone.org 15 August 2013 | Volume 8 | Issue 8 | e71049

disturbances resulting in decreased recruitment, the loss of SC5
colonies, or increased whole colony or partial mortality). This is
especially true in locations where poritids and faviids take the place
of acroporids and pocilloporids as the dominant reef builders (i.e.
following temperature anomalies, cyclones, and red tides to which
the branching and tabular colonies are more susceptible) because
it will be the massive taxa that sustain the coral communities and
their associated biota during such recovery periods.
Acknowledgments
This study was an extension of ongoing work in the southeastern Arabian
Gulf by Drs. Bernhard Riegl and Sam Purkis (NCRI), who provided
guidance, support and encouragement. Project support, including the use
of boats, field stations, equipment and personnel, was provided by National
Coral Reef Institute, Environmental Agency – Abu Dhabi, Dibba Marine
Centre of the Ministry of Environment and Water, Dibba-Fujairah
Municipality, Fujairah Municipality, and the Fujairah Marina Club.
Invaluable assistance was provided by Suaad Al Harthi and Ibrahim Bugla
(EAD); Dr. Christophe Tourenq (EWS-WWF); Maral Shuriq (Fujairah
Municipaliy); Debra Rein; and the boat drivers, rangers, agency staff and
management in the United Arab Emirates.
Author Contributions
Conceived and designed the experiments: KAF GF. Performed the
experiments: KAF GF. Analyzed the data: KAF. Contributed reagents/
materials/analysis tools: KAF GF. Wrote the paper: KAF GF.
References
1. Hoegh-Guldberg O (1999) Climate change, coral bleaching and the future of the
world’s coral reefs. Marine and Freshwater Research 50: 839–866.
2. Sheppard CRC (2003) Predicted recurr ences of mass coral mortality in the
Indian Ocean. Nature 425: 294–297.
3. Emanual K (2005) Increasing destructiveness of tropical cyclones over the past
30 years. Nature 436: 686–688.
4. Webster P, Holland G, Curry JA, Chang H-R (2005) Changes in tropical
cyclone number, duration, and intensity in a warming environment. Science
309: 1844–1847.
5. Burt J, Bartholomew A, Usseglio P (2008). ‘‘Recovery of corals a decade after a
bleaching event in Dubai, United Arab Emirates.’’ Marine Biology 154: 27–36.
6. Connell JH (1997) Disturbance and recovery of coral assemblages. Coral Reefs
16(Supplement): S101–S113.
7. Nin˜io R, Meekan M, Done T, Sweatman H (2000) Temporal patterns in coral
assemblages on the Great Barrier Reef from local to large spatial scales. Marine
Ecology Progress Series 194: 65–74.
8. Bruno JF, Selig ER (2007) Regional decline of coral cover in the Indo-Pacific:
timing, extent and subregional comparisons. PLoS One e711(8).
9. Riegl BM, Purkis SJ (2009) Model of cor al population response to accelerated
bleaching and mass mortality in a changing climate. Ecological Modelling 220:
192–208.
10. Perry CT, Larcombe P (2003) Marginal and non-reef buildi ng co ral
environments. Coral Reefs 22: 427–432.
11. Riegl B (1999) Coral in a non-reef setting in the southern Arabian Gulf (Dubai,
UAE): fauna and community structure in response to recurring mass mortality.
Coral Reefs 18: 63–73.
12. Riegl B (2002) Effects of the 1996 and 1998 positive sea-surface temperature
anomalies on corals, coral diseases and fish in the Arabian Gulf (Dubai, UAE).
Marine Biology 140: 29–40.
13. Riegl B (2003) Climate change and coral reefs: different effects in two high-
latitude areas (Arabian Gulf, South Africa). Coral Reefs 22: 433–446.
14. Foster KA, Foster G, Tourenq C, Shuriqi MK (2011) Shifts in coral community
structures following cyclone and red tide disturbances within the Gulf of Oman
(United Arab Emirates). Marine Biology 158: 955–968.
15. Marquis E, Trick C (2009) Harmful algal bloom in the coastal waters of Dibba
Fujairah - November 2008: Preliminary Report, United Nations University -
INWEH.
16. Richlen M, Morton S, Jamali EA, Rajan A, Anderson DM (2010) The
catastrophic 2008–2009 red tide in the Arabian Gulf region, with observations
on the identification and phylogeny of the fish-killing dinoflagellate Cochlodinium
polykrikoides. Harmful Algae 9: 163–172.
17. Foster KA, Foster G, Al-Harthi S (2013). Coral community structures in the
southeastern Arabian Gulf (Qatar and Abu Dhabi, UAE): various stages of
Acropora recovery a decade after recurrent elevated temperature anomalies. Open
Journal of Marine Science: In press.
18. Kohler KE, Gill SM (2006) Coral Point Count with Excel extensions (CPCe): a
visual basic program for the determination of coral and substrate coverage using
random point count methodology. Computers and Geosciences 32(9): 1259–
1269.
19. Clarke KR, Gorley RN (2006) PRIMER v6: User Manual/Tutorial. Plymouth,
England, PRIMER-E, Inc.
20. Caswell H (2001) Matrix population models: construction, analysis, and
interpretation. Sunderland, MA, Sinauer Associates, Inc.
21. Owen-Smith N (2007) Introduction to modeling in wildlife and resource
conservation. Malden, MA, Blackwell Publishing.
22. Babcock RC (1991) Comparative demography of three species of scleractinian
corals using age- and size-dependent classifications. Ecological Monographs
61(3): 225–244.
23. Hughes TP, Jackson J (1985) Population dynamics and life histories of foliaceous
corals. Ecological Monographs 55(2): 141–166.
24. Hughes TP (1984) Population dynamics based on individual size rather than age:
a general model with a reef coral example. The American Naturalist 123(6):
778–795.
25. Hood GM (2010) PopTools version 3.2.5. Available: http://www.poptools.org.
26. Kinzie III RA (1973) The zonation of West Indian gorgonians. Bulletin of
Marine Science 23(1): 93–155.
27. Shinn EA (1976) Coral reef recovery in Florida and the Persian Gulf.
Environmental Geology 1(4): 241–254.
28. Shriadah MA (2001) Hydrochemical characteristics of the United Arab Emirates
waters along the Arabian Gulf and the Gulf of Oman. Acta Adriatica 42(2): 93–
102.
29. Rezai H, Wilson SC, Claereboudt M, Riegl B (2004). Coral reef status in the
ROPME Sea area: Arabian/Persian Gulf, Gulf of Oman and Arabian Sea.
Status of
Coral Reefs of the World: 2004 1: 155–170.
30. Coles SL (2003) Coral species diversity and environmental factors in the Arabian
Gulf and the Gulf of Oman: a comparison to the Indo-Pacific region. Atoll
Research Bulletin 507: 497–508.
31. Claereboudt MR (2006). Reef corals and coral reefs of the Gulf of Oman.
Sultanate of Oman, The Historical Association of Oman.
32. Riegl B, Berumen M, Bruckner A (2013) Coral population trajectories, increased
disturbance and management intervention: a sensitivity analysis. Ecology and
Evolution 3(4): 1050–1064.
33. Foster K, Foster G, Al-Cibahy A, Al-Harthi S, Purkis SJ, et al. (2012)
Environmental setting and temporal trends in southeastern Gulf coral
communities. In: Riegl BM & Purkis SJ (eds), Coral Reefs of the Gulf:
Adaptations to Climatic Extremes, Coral Reefs of the World, Vol 3, 51–70.
34. Al-Hashmi A, Hamza W, Ramadan G (2012) Coral reef reproduction and its
features in the Arabian Gulf (Jebel Ali, UAE). International Journal of
Environment and Sustainability 1(3): 12–21.
35. Bak RPM, Engel MS (1979) Distribution, abundance and survival of juvenile
hermatypic corals (Scleractinia) and the importance of life history strategies in
parent coral communities. Marine Biology 54: 341–352.
36. Hughes TP, Jackson J (1980) Do corals lie about their age? Some demographic
consequences of partial mortality, fission, and fusion. Science 209: 713–715.
37. Rylaarsdam KW (1983) Life histories and abundance patterns of colonial corals
on Jamaican reefs. Marine Ecology Progress Series 13: 249–260.
38. Sammarco PW (1980) Diadema and its relationship to coral spat mortality:
grazing, competition, and biological disturbances. Journal of Ex perimental
Marine Biology and Ecology, 45, 245–272.
39. Mercado-Molina AE, Sabat AM, Yoshioka PM (2011) Demogra phy of the
demosponge Amphimedon compressa: evaluation of the importance of sexual versus
asexual recruitment to its population dynamics. Journal of Experimental Marine
Biology and Ecology 407: 355–362.
40. Linares C, Doak DF, Coma R, Diaz D, Zabala M (2007) Life history and
viability of long-lived marine invertebrate: the octocoral Paramuricea clavata.
Ecology 88: 918–928.
41. Lirman D (2003) A simulation model of the population dynamics of the
branching corals Acropora palmata effects of storm intensity and frequency.
Ecological Modelling 161: 169–182.
42. Gotelli NJ (1991) Demographic models for Leptogorgia virgulata, a shallow-water
gorgonian. Ecology 72: 457–467.
43. McFadden CS (1991) A comparative demographic analysis of clonal
reproduction in a temperate soft coral. Ecology 72: 1849–1866.
44. Hughes TP (1984) Population dynamics based on individual size rather than age:
a general model with a reef coral example. American Naturalist 123: 778–795.
45. Westerberg L, Wennergren U (2007) Matrix Models: A Tool for Landscape
Management? In: Frontiers in Ecology Research, Antonello SD (ed), Nova
Science Publishers, New York. 141–163.
Massive Coral Demography/Population Dynamics (UAE)
PLOS ONE | www.plosone.org 16 August 2013 | Volume 8 | Issue 8 | e71049