1 23

Marine BiologyInternational Journal on Life in Oceansand Coastal Waters ISSN 0025-3162 Mar BiolDOI 10.1007/s00227-012-2066-7

Temporal and spatial variability in coralrecruitment on two Indonesian coralreefs: consistently lower recruitment to adegraded reef

P. Salinas-de-León, C. Dryden,D. J. Smith & J. J. Bell

1 23

Your article is protected by copyright and

all rights are held exclusively by Springer-

Verlag. This e-offprint is for personal use only

and shall not be self-archived in electronic

repositories. If you wish to self-archive your

work, please use the accepted author’s

version for posting to your own website or

your institution’s repository. You may further

deposit the accepted author’s version on a

funder’s repository at a funder’s request,

provided it is not made publicly available until

12 months after publication.

ORIGINAL PAPER

Temporal and spatial variability in coral recruitment on twoIndonesian coral reefs: consistently lower recruitmentto a degraded reef

P. Salinas-de-Leon • C. Dryden • D. J. Smith •

J. J. Bell

Received: 25 January 2012 / Accepted: 30 August 2012

� Springer-Verlag 2012

Abstract Corals are the primary reef-building organisms,

therefore it is key to understand their recruitment patterns

for effective reef management. Coral recruitment rates and

juvenile coral abundance were recorded in the Wakatobi

National Marine Park, Indonesia, on two reefs (Sampela

and Hoga) with different levels of environmental degra-

dation (12.5 vs. 44 % coral cover with high and low sed-

imentation rates, respectively) to examine consistencies in

recruitment patterns between years and seasons. Recruit-

ment was measured on multiple panels at two sites on each

reef (6–7 m depth) and cleared areas of natural reef.

Although coral recruitment was twofold higher in

2008–2009 than in 2007–2008, and seasonal differences

were identified, consistent significant differences in

recruitment rates were found between the two reefs even

though they are separated by only *1.5 km. Recruitment

rates and juvenile abundance were lower on the more

degraded reef. These patterns are likely a consequence of

differential pre- and post-settlement mortality as a result of

the high sedimentation rates and degraded conditions and

possibly reduced larval supply.

Introduction

Understanding temporal and spatial variation in recruit-

ment patterns is critical for measuring population dynamics

and effectively managing marine populations, and for

understanding the resilience of coral reef communities to

disturbance (Hughes et al. 2010). Explaining such

dynamics are particularly important for coral reef systems

as they are in decline worldwide as a consequence of

natural and anthropogenic disturbance (e.g. Hughes 1994;

Bellwood et al. 2004; Mora 2008). For example, Indo-

Pacific coral reefs are seriously threatened, and recent

estimates show a 32 % region-wide decline in coral cover

since the 1970s (Bruno and Selig 2007).

Recruitment has long been recognised as one of the most

important factors driving the ecology of marine inverte-

brates and is critical for the maintenance of viable reef

populations and for promoting the recovery of coral reefs

after disturbance (Babcock and Mundy 1996; Connell et al.

1997; Hughes et al. 2000). Here, we define recruitment as

the number of new individuals that settle and survive until

the time of observation, which reflects patterns of larva

availability, larva substrate selection and post-settlement

mortality rates (Harrison and Wallace 1990; Connell et al.

1997).

Despite Burke et al. (2002) suggesting that [85 % of

Indonesia’s reefs are threatened by anthropogenic impacts

and Edinger et al. (1998) showing a drastic loss of coral

biodiversity in Indonesian reefs due to land-based

Communicated by J. P. Grassle.

Electronic supplementary material The online version of thisarticle (doi:10.1007/s00227-012-2066-7) contains supplementarymaterial, which is available to authorized users.

P. Salinas-de-Leon (&) � J. J. Bell

Centre for Marine Environmental and Economic Research,

School of Biological Sciences, Victoria University of

Wellington, PO Box 600, Wellington, New Zealand

e-mail: pelayosalinas@gmail.com

C. Dryden

School of Marine Science and Technology, Newcastle

University, Armstrong Building, Newcastle Upon

Tyne NE1 7RU, UK

D. J. Smith

Coral Reef Research Unit, Department of Biological Sciences,

University of Essex, Wivenhoe Park, Colchester CO4 3SQ, UK

123

Mar Biol

DOI 10.1007/s00227-012-2066-7

Author's personal copy

pollution, only a few coral recruitment studies have been

conducted in the Coral Triangle region (Tomascik et al.

1996; Reyes and Yap 2001; Fox et al. 2003; Fox 2004;

Schmidt-Roach et al. 2008). Of these studies, only Fox

et al. (2003) and Fox (2004) were conducted in more than

one locality and were temporally replicated. Therefore,

more information is needed on hard coral population

dynamics in this biodiverse region to assist in the devel-

opment of effective management plans.

Scleractinian coral recruitment is influenced by a num-

ber of factors including the amount of live coral cover in

source populations (Hughes et al. 2000), abundance and

diversity of coral larvae (Potts et al. 1985), hydrodynamic

variability (Amar et al. 2007), sedimentation (Babcock and

Davies 1991), temperature (Nozawa and Harrison 2007),

recruitment cues and inhibition from other benthic taxa

(Harrington et al. 2004), eutrophication levels (Tomascik

1991), grazing pressure (Sammarco 1980), light levels

(Mundy and Babcock 1998) and connectivity between reefs

(Roberts 1997). Some of these, such as the presence of

crustose coralline algae (CCA), positively affect coral

recruitment, while others, such as increased sedimentation

or competition with macroalgae and/or soft corals, nega-

tively affect recruitment (Benayahu and Loya 1985; Bab-

cock and Davies 1991; Maida et al. 1995).

In a preliminary 1-year study in 2007–2008 (Salinas-de-

Leon et al. 2011) on the same reefs used in the present

study (Hoga and Sampela reefs in Indonesia), we charac-

terised the benthic habitats with respect to coral taxonomic

composition and coverage by different organisms. We also

examined coral recruitment on cleared areas of vertical reef

wall and on concrete and terracotta settlement panels. From

this study, we concluded that there were significant dif-

ferences between the Hoga and Sampela sites with respect

to coral cover (*44 vs. 12.5 %, respectively) and taxo-

nomic composition, with Acroporidae and Poritidae being

most abundant at Hoga, and Faviidae and Poritidae being

most abundant at Sampela. Coral recruitment was similar

on the different substrates but differed significantly

between Hoga and Sampela reefs. We concluded that ter-

racotta panels secured to vertical reef faces provide a

suitable substrate for assessing relative recruitment rates in

this environment. Here, we extended this earlier study into

a second year (2008–2009) and also conducted a seasonal

study to determine whether the differences between the

Hoga and Sampela reefs were consistent. The objectives of

this study were to (1) compare coral recruitment rates to

reefs with different levels of environmental degradation in

the Wakatobi National Marine Park (WNMP), SE Sulaw-

esi, Indonesia (using the same sites as Salinas-de-Leon

et al. 2011); and (2) examine the consistency of any pat-

terns between 2 years and among three seasons. We used

three approaches to these objectives including measuring

coral recruitment rates to natural reef substrate (annual

patterns only) and artificial settlement panels and juvenile

coral abundance. The seasonal study in 2008–2009 exam-

ined three consecutive 4-month periods. Indonesia’s cli-

mate is dominated by two monsoon seasons each year,

driven by the Inter-tropical Convergence Zone, with the

north-west monsoon (wet season) occurring between

November and March and the south-east monsoon (dry

season) between May and September (Tomascik et al.

1997).

Materials and methods

Study sites

The WNMP is in the Tukang Besi Archipelago, an island

group off south-east Sulawesi in Indonesia (Fig. 1). The

park represents Indonesia’s third largest Marine Protected

Area, which was designated in 1992 and covers 1.39 mil-

lion hectares. The park lies at the heart of the ‘Coral Tri-

angle’ and supports extremely high coral diversity with

more than 390 hermatypic scleractinian coral species

belonging to 68 genera and 13 families (Turak 2003).

Sampling

Our study was conducted at four sites on two reefs (con-

sidered two replicate sites on each reef). Two sites (B3, B4;

terminology for sites is based on earlier studies in this

region) were on the Hoga reef (5�2803000S, 123�4501800E)

and two sites (S1, S2) on the Sampela reef (5�2805100S,

123�44’3900E); (Fig. 1). The reefs are *1.5 km apart and

Fig. 1 Sampling sites within Wakatobi National Marine Park in SE

Sulawesi, Indonesia. Sites B3 and B4 were on Hoga reef, and sites S1

and S2 on Sampela reef, adjacent to a Bajau village (represented on

map by three white houses). Black areas on map represent land and

grey areas show coral reefs. Map modified from Salinas-de-Leon

et al. (2011)

Mar Biol

123

Author's personal copy

separated by a deep channel ([60 m). Sites within reefs were

at least 250 m apart and were selected based on earlier ben-

thic surveys that revealed higher live coral coverage at Hoga

B3 and B4 (*44 %) compared to Sampela S1 and S2

(*12.5 %); (Salinas-de-Leon et al. 2011). Sites S1 and S2

are near the Bajau (Sea Gypsy) village of Sampela and are

subjected to severe anthropogenic disturbance, including

overfishing, blast and cyanide fishing, coral mining and

sedimentation due to untreated sewage discharge and man-

grove removal (Crabbe and Smith 2005).

Site characterisation

Benthic habitat composition was characterised in July 2007

at each site with respect to live coral cover, taxonomic

composition of the coral community and abundance of

other benthic taxa. This information was collected because

in some taxa, the relative abundance of spat populations is

similar to that of the adult populations (although not for all

species, reviewed by Pineda 2000), and the presence of

other benthic groups also influences coral recruitment

(Smith 1992; Harrington et al. 2004). Benthic habitat

composition was determined using the continual line

intercept method (English et al. 1997). Ten haphazardly

placed 10-m transects were sampled at 6 m depth using the

line intercept method at each of the four study sites. For

each transect, all coral colonies lying underneath the

transect tape were identified to genus level and recorded.

Other benthic categories included bare substrate, sand,

coral rubble, dead coral, soft coral, algae, crustose coralline

algae (CCA), sponges and ascidians.

Temperature was recorded throughout the study period

using HOBO pendant temperature loggers (Onset Corpora-

tion, USA) placed at 6 m depth at two sites (B3, S1), one on

each reef system. Temperature data were smoothed based on

mean data averaged over 10-min intervals, and averaged

across the four sites, by the logger software (Onset). Current

strength and direction were measured using a Valeport model

106 flow meter (Valeport Oceanographic instruments, UK)

positioned at 7–8 m depth. Only 3 9 24 h deployments of the

flow meter, during both neap and spring tides, at each of the

four study sites were conducted due to logistical limitations.

Sedimentation rates were estimated by placing three replicate

sediment traps (plastic cylinders 7 cm diameter and 20 cm tall

(English et al. 1997) at 6 m depth for 7 days at each site during

November 2008 and July 2009. Sediment was washed out of

the traps and filtered, and then oven dried for 24 h and

weighed to the nearest 0.01 g.

Recruitment

The panels deployed to monitor coral recruitment were

made of locally available unglazed terracotta

(20 9 10 9 0.7 cm). Panels were directly attached to the

reef substrate using a modification of the method of Mundy

(2000). Terracotta panels were cable tied to thin (0.5 cm)

hardwood planks. The wood was then drilled, and

recruitment panels were attached to vertical surfaces on the

reef using two galvanised nails, a material widely used in

previous recruitment studies (e.g. Harriott and Fisk 1987;

Mangubhai et al. 2007). A small piece of hardwood

(2 9 2 9 2 cm) was placed between each panel and the

reef to leave a 2-cm gap, as this is known to facilitate coral

recruitment (Harriott and Fisk 1987); (See ESM 1 for a

diagrammatic representation of the panel set up). Eight

panels were haphazardly placed between 6 and 7 m depth

at each of the four study sites, and at least 5 m separated

the panels from one another; panels were attached to ver-

tical reef surfaces based on Salinas-de-Leon et al. (2011).

Only corals recruiting to the back surface of the panels

(closest to the reef) were included in the analysis as 99 %

of all corals recruited to this side of the tiles.

Panels were submerged for two periods of time to record

yearly and seasonal coral recruitment rates and therefore

examine the consistency of the patterns on the two different

reefs over time. (A) Panels were deployed for 12 mo

between July 2007 and July 2008; these panels were a

subset of those used by Salinas-de-Leon et al. (2011). Upon

retrieval, new panels were placed during July 2008 and

retrieved 12 mo later in July 2009. (B) Panels were also

deployed in July 2008 to assess seasonal recruitment, and

were retrieved and replaced by 8 more panels every 4 mo,

providing three 9 4 mo recruitment data sets: July 2008 to

November 2008, November 2008 to March 2009 and

March 2009 to July 2009. These periods of time coincided

with major climatic changes within the study region.

When panels were collected, they were labelled and

transported back to the laboratory in sea water. They then

were photographed and panels were bleached in a chlorine

solution for 24 h, then rinsed in fresh water to remove

excess bleach before being air dried. Each panel was

searched twice by two different observers at two different

microscope magnifications (109 and 209), and all coral

recruits were recorded. The panel edges were not analysed

given their small area (i.e. panels were 0.7 cm thick).

Recruits were identified as belonging to the families

Acroporidae, Pocilloporidae and Poritidae based on pho-

tomicrographs in English et al. (1997) and Babcock et al.

(2003), with the remaining spat, including those too dam-

aged to identify, being designated as ‘others’.

Recruitment to natural reef cleared areas

Four reef areas (25 9 25 cm) were cleared of all living

organisms at each of the four sites, to determine coral

recruitment rates to natural substrata. Permanent cleared

Mar Biol

123

Author's personal copy

areas were created in July 2007 and July 2008, and were

surveyed after 12 mo in July 2008 and July 2009, respec-

tively. These areas were haphazardly located on vertical

walls near the recruitment panels and were permanently

marked with four galvanised nails. Cleared areas were at

least 5 m apart.

Cleared areas were surveyed by eye during the day and

at night. Initial surveys were conducted during the day, and

all coral recruits present in the quadrats were recorded and

identified to the lowest taxonomic level possible. Observed

recruits were marked using a small nail. A second survey

was conducted at night using the Underwater Fluorescence

Technique of Piniak et al. (2005). Permanent cleared areas

were resurveyed with a blue light (Nightsea Inc.) and with

a yellow barrier filter placed over the diver’s mask

(Nightsea Inc.) to facilitate the detection of coral recruits.

False positives can be caused by a number of other

organisms that also fluoresce, but we minimised this effect

by using a strong white light as described by Baird et al.

(2006). Recruits missed under daylight conditions were

recorded, measured and identified to the lowest taxonomic

level possible; results were presented as the total number of

recruits recorded. The data collected from areas cleared in

2007 and surveyed in 2008 were the same as those used by

Salinas-de-Leon et al. (2011) to compare with recruitment

rates on panels made of different materials.

Juvenile coral abundance and diversity

Juvenile coral abundance and diversity were assessed in

July 2008 using 20 0.5 m 9 0.5 m (0.25 m2) quadrats

positioned haphazardly along 50-m transects at 6 m depth

at each of the four sampling sites. Prior to surveying the

quadrats, sediment was removed from the quadrat by

gently wafting the quadrat area; this facilitated counts of

the small juvenile corals. Juvenile corals were defined as

colonies \40 mm in diameter that were attached to the

substratum and did not have the fractured surface charac-

teristic of asexual recruits (Edmunds 2000). Corals present

in the quadrats were identified to genus level.

Statistical analysis

Data were analysed by the statistical packages MINITAB

v16, PRIMER v6 and PERMANOVA v1.0.2. Differences

in benthic cover composition among sites were analysed

using a Kruskal–Wallis test for each of the benthic cate-

gories (as data were not normally distributed). The

recruitment and juvenile coral abundance data were not

normally distributed even after transformation and vari-

ances were unequal, so the data were analysed with a

permutational ANOVA. PERMANOVA is a permutation-

based version of analysis of variance (Anderson 2001). It

uses the distances between samples to partition variance

and randomisations or permutations of the data to produce

the p value for the hypothesis test. It is non-parametric and

is therefore robust to the assumption of multivariate nor-

mality, making it less prone to Type I errors. A preliminary

PERMANOVA analysis was used to investigate the dif-

ferences in recruitment rates to panels, cleared reef areas

and abundance of juvenile corals including the factors year

(season), reef (Hoga, Sampela) and site (B3, B4, S1, S2).

With this preliminary analysis, we found no significant

differences at the site level, so the factor site was removed

from the final analysis (data not shown).

Differences in the yearly recruitment rates to panels and

to cleared reef areas were analysed using a PERMANOVA

model with two fixed factors: year (two levels: 2008, 2009)

and reef (two levels: Hoga and Sampela). The PERMA-

NOVA seasonal recruitment data statistical design had two

fixed factors: season (three levels: November, March and

July) and reef (two levels: Hoga and Sampela). The PER-

MANOVA model created for the juvenile abundance data

included just the factor reef (two levels: Hoga and Sam-

pela), given that surveys were only conducted in 2008.

Results

Biological and physical characteristics of the study sites

The Hoga and Sampela reefs had significantly different

benthic substrata and community composition (Fig. 2a, b).

The Hoga reef was dominated by hard coral, while the

Sampela reef was dominated by bare substrate and coral

rubble (Table 1). There were also significant differences

(p \ 0.05, Kruskal–Wallis) in the percentage cover of

CCA and sand between the two reefs (Fig. 2a), with higher

CCA abundance at Hoga and higher sand cover at Sampela.

The Hoga reef coral assemblage (at 6-m depth) was dom-

inated by Acroporidae and Poritidae, while the Sampela

reef coral assemblage was dominated by members of the

family Faviidae and Poritidae (Fig. 2b).

The Hoga and Sampela reefs had similar water flow

rates and water temperatures (2 years range: 26.8–29.9 �C)

over the study period (Table 1). Sedimentation rates were

consistently higher on the Sampela reef, compared to the

Hoga reef during both the wet and dry seasons (Table 1).

Annual recruitment

A total of 114 and 268 coral spat were recorded on the 32

panels across all sites in 2008 (0.89 ± 0.28 SE spat per

cm2) and 2009 (2.09 ± 0.57 SE spat per 100 cm2),

respectively. Inter-annual variation in coral recruitment

was highly significant (Fig. 3a, Table 2), and the number

Mar Biol

123

Author's personal copy

of coral spat on panels also varied consistently between

reefs, with higher abundance at Hoga than at Sampela in

both years (Fig. 3a, Table 2). There was also a significant

interaction between year and reef, with the difference being

much greater in the second year than the first. For the 2008

panels, 25 % of all coral spat (N = 114) belonged to the

family Acroporidae, 21 % to Pocilloporidae, 13 % to the

Poritidae and 41 % were classified as ‘others’. In 2009,

12 % of all coral spat (N = 268) belonged to the Acro-

poridae, 10 % to the Pocilloporidae, 16 % to the Poritidae

and 62 % were classified as ‘others’.

Recruitment to cleared reef areas

A total of 91 and 167 coral recruits were recorded on the

cleared reef areas across all sites in 2008 (1.07 ± 0.34 SE

spat per 100 cm2) and 2009 (2.04 ± 0.37 SE spat per

100 cm2), respectively. Consistent with the panel data, a

significantly higher number of recruits were recorded in

both years at Hoga compared to Sampela (Fig. 3b,

Table 3). Significant differences (Fig. 3b, Table 3) were

also recorded between years, with higher recruitment rates

recorded in 2008–2009 than in 2007–2008.

Juvenile coral abundance and diversity

Juvenile coral abundance was 0–19 per 0.25 m2 in the 80

quadrats surveyed (20 m2), with 443 juvenile corals

recorded in total. Significantly (p \ 0.05, PERMANOVA)

higher numbers of juveniles were recorded at Hoga com-

pared to Sampela (Fig. 4). Faviidae juveniles were the

most abundant across all sites (28.3 %), followed by Por-

itidae (24.5 %), Agariciidae (18.9 %), Acroporidae

(10.7 %) and Pocilloporidae (4.22 %).

Seasonal recruitment

Similar to the patterns in yearly recruitment, significantly

higher numbers of coral spat were recorded at Hoga

compared to Sampela across all seasons (Fig. 5, Table 4).

There was also a significant interaction between reef and

season with larger differences between the reefs during

some seasons (Table 4). We found significantly higher

coral recruitment rates in November 2008–March 2009

(2.09 ± 0.34 SE spat per 100 cm2) compared to July–

November 2008 (0.67 ± 0.21 SE spat per 100 cm2) and

March–July 2009 (0.35 ± 0.09 SE spat per 100 cm2);

Fig. 2 Mean per cent cover (±SE) of major benthic groups (a) and

coral families (b) at four sites (B3 and B4 at Hoga reef; S1 and S2 at

Sampela reef) within Wakatobi National Marine Park, SE Sulawesi,

Indonesia. (*) indicates significant differences between reefs

(Kruskal–Wallis, p \ 0.05). HC hard coral, BA bare, RU rubble,

CCA coralline algae, SA sand, SPO sponge, SC soft coral, ASCascidian, ALG algae, OT other. Benthic composition data are the same

as those used by Salinas-de-Leon et al. (2011)

Table 1 Site characterisation table showing mean percentage coral

cover (±Standard Error); water flow rate range (cm s-1); and mean

sedimentation rates (g day-1); (±Standard Error) at four sites (B3 and

B4 at Hoga reef; S1 and S2 at Sampela reef) in the Wakatobi National

Marine Park, SE Sulawesi, Indonesia

Reef Site Percentage coral cover Water flow Sedimentation

Dry Wet

Hoga B3 43.7 ± 0.97

10–20

0.14 ± 0.04 0.12 ± 0.02

B4 44 ± 2.58 0.16 ± 0.04 0.18 ± 0.05

Sampela S1 14 ± 1.71 0.23 ± 0.04 0.33 ± 0.05

S2 11 ± 1.59 0.29 ± 0.09 0.32 ± 0.11

‘Dry’ and ‘Wet’ refer to data collected in the dry season (July 2009) and wet season (November 2008), respectively

Mar Biol

123

Author's personal copy

(Fig. 5, Table 4). The higher recruitment rates from November

to March correlated with a higher sea surface water tempera-

ture during the wet north-west monsoon season (Fig. 5).

Recruitment rates of acroporids were also higher in Novem-

ber–March (0.58 ± 0.08 SE spat per 100 cm2) compared with

July–November (0.14 ± 0.04 SE spat per 100 cm2) and

March–July (0.04 ± 0.01 SE spat per 100 cm2).

Discussion

We found significantly lower recruitment rates of hard

corals to a reef with higher levels of environmental

degradation, and these differences were consistent over

time and for different estimation methods. We also found

significant inter-annual and seasonal variation in coral

a bFig. 3 Mean (±SE) number of

coral recruits (per 100 cm2) on

recruitment panels (a) and

cleared reef areas (b) deployed

in 2008 and 2009 on Hoga and

Sampela reefs within Wakatobi

National Marine Park, SE

Sulawesi, Indonesia. Cleared

reef area data standardised to

100 cm2 to make them

comparable to recruitment

panels

Table 2 Results of the PERMANOVA comparing the factors: year

(2008, 2009) and reef (Hoga, Sampela) and their interaction for hard

coral recruitment to recruitment panels in the Wakatobi National

Marine Park, Indonesia

Source of

variation

df SS MS Pseudo-F p (perm)

Year 1 6,290.2 6,290.2 10.166 0.001

Reef 1 15,497 15,497 25.045 0.001

Year 9 reef 1 3,093.4 3,093.4 4.99 0.022

Res 124 76,726 618.76

Total 127 1.0161E5

Bold indicates significance at the 0.05 level

Table 3 Results of the PERMANOVA comparing the factors: year

(2008, 2009) and reef (Hoga, Sampela) and their interaction for hard

coral recruitment to cleared reef areas in the Wakatobi National

Marine Park, Indonesia

Source of variation df SS MS Pseudo-F p (perm)

Year 1 4,504.4 4,504.4 6.7392 0.007

Reef 1 10,232 10,232 15.308 0.001

Year 9 site 1 5,293 1,764.3 2.0592 0.122

Res 28 18,715 668.39

Total 31 34,828

Bold indicates significance at the 0.05 level

Fig. 4 Mean (±SE) number of coral juveniles (per 0.25 m2) for each

taxonomic group on Hoga and Sampela reefs within Wakatobi

National Marine Park, SE Sulawesi, Indonesia

Fig. 5 Left Axis: mean (±SE) number of coral recruits (per 100 cm2)

on recruitment panels in three submersion periods at two reefs (Hoga

and Sampela) within Wakatobi National Marine Park, SE Sulawesi,

Indonesia. Right axis: mean monthly sea temperature (�C); (averaged

among four sites)

Mar Biol

123

Author's personal copy

recruitment rates in the WNMP. Understanding recruitment

dynamics and identifying the factors responsible for tem-

poral and spatial variation in such patterns are important

for effectively managing marine populations. The first

stage in understanding such dynamics is to describe vari-

ation patterns and propose hypotheses to explain the pat-

terns observed.

We propose a number of hypotheses to explain the

consistent differences in the number of coral recruits

between the Hoga and Sampela reefs. A likely explanation

for the differences we report is the effect of settlement-

related processes (such as habitat difference or larval

selection) and post-settlement mortality. Sedimentation

levels were much higher at Sampela reef during both the

dry (July) and wet (November) seasons. Sedimentation is

known to affect the growth and calcification (Tomascik and

Sander 1985), reproduction (Tomascik and Sander 1987b),

respiration, feeding and photosynthesis (Fabricius 2005),

fecundity (Tomascik and Sander 1987b), community

structure (Tomascik and Sander 1987a) and more impor-

tantly, the larval settlement of scleratinian corals (Hunte

and Wittenberg 1992). The effects of sedimentation vary

considerably among coral species, but also between sedi-

ment types and environmental conditions (Fabricius 2005).

The lower recruitment rates reported at Sampela might in

part be influenced by the higher sedimentation rates

recorded there. Sedimentation and associated turbidity are

known to affect planula settlement behaviour through

direct physical action in the water column (Tomascik 1991)

and by inhibiting actual planula settlement (Babcock and

Davies 1991). We found that the majority of the recruits

settled on the back (cryptic) side of the panels (Harriott and

Fisk 1987), and this habitat has low sediment accumulation

rates, but the effects of suspended sediment are still pres-

ent, particularly the light reduction, which is known to be

important for corals (Mundy and Babcock 1998).

Habitat differences might also explain the higher post-

recruitment mortality rates at Sampela compared to Hoga.

Our benthic surveys revealed higher levels of coral rubble

at Sampela, and rubble fields created by chronic blast

fishing are unsuitable for coral recruit survival. This is

because spat settling onto coral rubble are killed by the

periodic shifting of the rubble with strong currents (Fox

et al. 2003). Finally, differences in predation rates could

explain the recruitment rate differences between reefs, as

fish abundance (see Bell and Smith 2004) and assemblage

composition, along with other potential predators are likely

to be different between the sites, and predation influences

coral recruitment (e.g. Sammarco 1980). However, since

most coral recruits were on the backs of the panels, we

were able to compare recruitment between the two reefs. A

future experimental approach will investigate these factors

further to determine which factor or combination of factors

explains the patterns we have reported.

These recruitment differences could also be affected by

differences in the live coral coverage between the reefs,

being significantly lower at Sampela than at Hoga. How-

ever, while there is some evidence that a high proportion of

the coral larvae produced locally might be retained close to

the parental population (Sammarco and Andrews 1988;

Underwood et al. 2007; Underwood et al. 2009), the rela-

tively close proximity of the two reefs (1.5 km) also sug-

gests they may share a common pool of larvae, particularly

for species with long-lived pelagic larvae. In contrast to our

results, other authors have not found recruitment differ-

ences between reefs with different levels of coral cover or

taxonomic composition. For example, Fox (2004) found no

differences in coral recruitment rates to panels at blasted

and unblasted reefs, despite the reefs having greater spatial

separation than our sites; this suggests that low-coral-cover

sites are not necessarily limited by larval supply.

The number of coral spat per panel recorded at sites

within the Wakatobi is slightly lower than those recorded

in other recruitment studies within Indonesia (Manado and

Komodo), and much lower than those at many locations

along the central Great Barrier Reef (Baird and Hughes

2000; Fox 2004; Schmidt-Roach et al. 2008). However, it

is difficult to interpret differences in recruitment rates

because of methodological and site-specific differences

between studies, which are known to influence those rates

(Glassom et al. 2004). Importantly, the patterns we repor-

ted and the recruitment rates measured were consistent

between natural areas of reef and panels on the same reef

suggesting that the panels provided a suitable surrogate for

recruitment rates to these reefs (for further discussion, see

Salinas-de-Leon et al. 2011).

Recruitment was twofold higher during the second year

of our study compared to the first, and the differences

between reefs were also greater in year 2. Such inter-

annual variation has been recorded in other studies

including in the Komodo National Park in Indonesia and

also in other biogeographical regions, such as the Red Sea

Table 4 Results of the PERMANOVA comparing the factors: season

(November, March and July) and reef (Hoga, Sampela) and their

interactions for hard coral recruitment to terracotta recruitment panels

in the Wakatobi National Marine Park, Indonesia

Source of

variation

df SS MS Pseudo-F p (perm)

Season 2 55,222 27,611 46.959 0.001

Reef 1 29,616 29,616 50.369 0.001

Season 9 reef 2 4,204.9 2,102.4 3.5757 0.014

Res 186 1.09E ? 05 587.98

Total 191 1.98E ? 05

Bold indicates significance at the 0.05 level

Mar Biol

123

Author's personal copy

(Glassom et al. 2004) and the GBR, Australia (Harriott and

Banks 1995; Dunstan and Johnson 1998). This annual

variability might be caused by a number of factors, for

example, some corals show considerable variation in their

reproductive output between years (Hughes et al. 2000), or

by differences in the physical processes that influence

larval transport and supply (Pineda 2000).

Many of the seasonal differences reported in our study

are likely to be driven by differences between coral fami-

lies in their reproductive modes (i.e. brooding vs. broad-

casting) and because different species are likely to spawn at

different times of the year. Broadcasting corals often

exhibit spawning synchrony within populations (Harrison

et al. 1984). Richmond and Hunter (1990) predicted that

for equatorial regions, where sea surface temperature

variations and tidal amplitude are small, reproductive

seasonality and synchrony among species would be

reduced. In contrast to this prediction, recent work on coral

reefs close to equatorial regions, including Singapore

(Guest et al. 2005) and several locations within Eastern

Indonesia (Baird et al. 2009), has revealed that multiple

species show synchronous spawning or ‘mass’ coral

spawning. From these studies, it is now believed that

acroporid spawning occurs in Indonesia during the days

following the full moon between either March and April, or

October and November, depending on the species (Baird

et al. 2009). Although we cannot specify exactly when

acroporid spawning occurs in the WNMP (given that

panels were replaced every 4 months), the recruitment

results support a spawning season somewhere between

November and March. This seasonal peak coincided with

the period of warmest sea surface temperature during the

north-west monsoon season, and our results are similar to

those of Fox (2004) in Komodo (Central Indonesia).

In conclusion, we found consistent differences in coral

recruitment rates between two reef systems with different

levels of environmental degradation, which is likely to be

explained by different post-settlement mortality effects,

although we cannot rule out some differences in larval

supply as a partial explanation. Our results demonstrate

how major environmental remediation would be needed for

degraded reefs like Sampela to recover.

Acknowledgments P. Salinas-de-Leon was supported by a Victoria

University PhD Scholarship and wishes to thank the Salinas de Leon

foundation, the Centre for Marine Environmental and Economic

Research, the Centre for Biodiversity and Restoration, the New

Zealand Postgraduate Study Abroad scheme and Operation Wallacea

through collaboration with the Wakatobi Taman National for funding

this study. P. Salinas-de-Leon would also like to thank S. Rowley,

A. Costales-Carrera, Jufri, Magliani, Ludi, Ilu, Arif and other

volunteers at the Hoga Island research centre. Also, thanks to the

Wallacea Foundation and the Indonesian Institute of Sciences (LIPI

research permit granted to DJ Smith). J.J. Bell is also grateful to the

PADI Foundation for providing funding for travel.

References

Amar KO, Chadwick NE, Rinkevich B (2007) Coral planulae as

dispersion vehicles: biological properties of larvae released early

and late in the season. Mar Ecol Prog Ser 350:71–78

Anderson MJ (2001) Permutation tests for univariate or multivariate

analysis of variance and regression. Can J Fish Aquat Sci 58:626–639

Babcock R, Davies P (1991) Effects of sedimentation on settlement of

Acropora millepora. Coral Reefs 9:205–208

Babcock R, Mundy C (1996) Coral recruitment: consequences of

settlement choice for early growth and survivorship in two

scleractinians. J Exp Mar Biol Ecol 206:179–201

Babcock RC, Baird AH, Piromvaragorn S, Thomson DP, Willis BL

(2003) Identification of scleractinian coral recruits from Indo-

Pacific reefs. Zool Stud 42:211–226

Baird AH, Hughes TP (2000) Competitive dominance by tabular

corals: an experimental analysis of recruitment and survival of

understorey assemblages. J Exp Mar Biol Ecol 251(1):117–132

Baird AH, Salih A, Trevor–Jones A (2006) Fluorescence census

techniques for the early detection of coral recruits. Coral Reefs

25:73–76

Baird AH, Guest JR, Willis BL (2009) Systematic and biogeograph-

ical patterns in the reproductive biology of scleractinian corals.

Annu Rev Ecol Evol Syst 40:551–571

Bell JJ, Smith DJ (2004) Ecology of sponge assemblages (Porifera) in

the Wakatobi region, south-east Sulawesi, Indonesia: richness

and abundance. J Mar Biol Assoc UK 84:581–591

Bellwood DR, Hughes TP, Folke C, Nystrom M (2004) Confronting

the coral reef crisis. Nature 429:827–833

Benayahu Y, Loya Y (1985) Settlement and recruitment of a soft

coral: why is Xenia macrospiculata a successful colonizer? Bull

Mar Sci 36:177–188

Bruno JF, Selig ER (2007) Regional decline of coral cover in the

Indo-Pacific: timing, extent, and subregional comparisons. PLoS

ONE 2(8):e711. doi:10.1371/journal.pone.0000711

Burke L, Selig E, Spalding M (2002) Reefs at risk in south-east Asia.

World Resources Institute, Washington, DC

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

abundance, recruitment, and disturbance at several scales in

space and time. Ecol Monogr 67:461–488

Crabbe MJC, Smith DJ (2005) Sediment impacts on growth rates of

Acropora and Porites corals from fringing reefs of Sulawesi,

Indonesia. Coral Reefs 24(3):337–441

Dunstan PK, Johnson CR (1998) Spatio-temporal variation in coral

recruitment at different scales on Heron Reef, southern Great

Barrier Reef. Coral Reefs 17:71–81

Edinger EN, Jompa J, Limmon GV, Widjatmoko W, Risk MJ (1998)

Reef degradation and coral biodiversity in Indonesia: effects of

land-based pollution, destructive fishing practices and changes

over time. Mar Pollut Bull 36:617–630

Edmunds PJ (2000) Patterns in the distribution of juvenile corals and

coral reef community structure in St. John, US Virgin Islands.

Mar Ecol Prog Ser 202:113–124

English S, Wilkinson C, Baker V (1997) Survey manual for tropical

marine resources. Australian Institute of Marine Science,

Townsville

Fabricius KE (2005) Effects of terrestrial runoff on the ecology of

corals and coral reefs: review and synthesis. Mar Pollut Bull

50:125–146

Fox HE (2004) Coral recruitment in blasted and unblasted sites in

Indonesia: assessing rehabilitation potential. Mar Ecol Prog Ser

269:131–139

Fox HE, Pet JS, Dahuri R, Caldwell RL (2003) Recovery in rubble

fields: long-term impacts of blast fishing. Mar Pollut Bull

46:1024–1031

Mar Biol

123

Author's personal copy

Glassom D, Zakai D, Chadwick-Furman NE (2004) Coral recruit-

ment: a spatio-temporal analysis along the coastline of Eilat,

northern Red Sea. Mar Biol 144:641–651

Guest JR, Baird AH, Goh BPL, Chou LM (2005) Reproductive

seasonality in an equatorial assemblage of scleractinian corals.

Coral Reefs 24:112–116

Harrington L, Fabricius K, De’Ath G, Negri A (2004) Recognition

and selection of settlement substrata determine post-settlement

survival in corals. Ecology 85:3428–3437

Harriott VJ, Banks SA (1995) Recruitment of scleractinian corals in

the Solitary Islands Marine Reserve, a high latitude coral-

dominated community in Eastern Australia. Mar Ecol Prog Ser

123:155–161

Harriott VJ, Fisk DA (1987) A comparison of settlement plate types

for experiments on the recruitment of scleractinian corals. Mar

Ecol Prog Ser 37:201–208

Harrison P, Wallace C (1990) Reproduction, dispersal, and recruit-

ment of scleractinian corals. In: Dubinsky Z (ed) Ecosystems of

the world: coral reefs. Elsevier, Amsterdam, pp 133–207

Harrison PL, Babcock RC, Bull GD, Oliver JK, Wallace CC, Willis

BL (1984) Mass spawning in tropical reef corals. Science

223:1186–1189

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

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

Hughes TP, Baird AH, Dinsdale EA, Moltschaniwskyj NA, Pratchett

MS, Tanner JE, Willis BL (2000) Supply-side ecology works

both ways: the link between benthic adults, fecundity, and larval

recruits. Ecology 81:2241–2249

Hughes TP, Graham T, Jackson JBC, Mumby PJ, Steneck RS (2010)

Rising to the challenge of sustaining coral reef resilience. Trends

Ecol Evol 25:633–642

Hunte W, Wittenberg M (1992) Effects of eutrophication and

sedimentation on juvenile corals. II Settlement. Mar Biol

114:625–631

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

scleractinian coral recruitment.1. Directional allelopathy and

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

Mangubhai S, Harrison PL, Obura DO (2007) Patterns of coral larval

settlement on lagoon reefs in the Mombasa Marine National Park

and Reserve, Kenya. Mar Ecol Prog Ser 348:149–159

Mora C (2008) A clear human footprint in the coral reefs of the

Caribbean. Proc R Soc Biol Sci Ser B 275:767–773

Mundy CN (2000) An appraisal of methods used in coral recruitment

studies. Coral Reefs 19:124–131

Mundy CN, Babcock RC (1998) Role of light intensity and spectral

quality in coral settlement: implications for depth-dependent

settlement? J Exp Mar Biol Ecol 223:235–255

Nozawa Y, Harrison PL (2007) Effects of elevated temperature on

larval settlement and post-settlement survival in scleractinian

corals, Acropora solitaryensis and Favites chinensis. Mar Biol

152:1181–1185

Pineda J (2000) Linking larval settlement to larval transport:

assumptions, potentials and pitfalls. Oceanogr East Pac I:84–105

Piniak GA, Fogarty ND, Addison CM, Kenworthy WJ (2005)

Fluorescence census techniques for coral recruits. Coral Reefs

24:496–500

Potts DC, Done TJ, Isdale PJ, Fisk DA (1985) Dominance of a coral

community by the genus Porites (Scleractinia). Mar Ecol Prog

Ser 23:79–84

Reyes MZ, Yap HT (2001) Effect of artificial substratum material and

resident adults on coral settlement patterns at Danjugan Island,

Philippines. Bull Mar Sci 69:559–566

Richmond RH, Hunter CL (1990) Reproduction and recruitment of

corals: comparisons among the Caribbean, the Tropical Pacific,

and the Red Sea. Mar Ecol Prog Ser 60:185–203

Roberts CM (1997) Connectivity and management of Caribbean coral

reefs. Science 278:1454–1457

Salinas-de-Leon P, Costales-Carrera A, Zeljkovic S, Smith DJ, Bell JJ

(2011) Scleractinian settlement patterns to natural cleared reef

substrata and artificial settlement panels on an Indonesian coral

reef. Estuar Coast Shelf Sci 93:80–85

Sammarco PW (1980) Diadema and its relationship to coral spat

mortality: grazing, competition, and biological disturbance.

J Exp Mar Biol Ecol 45:245–272

Sammarco PW, Andrews JC (1988) Localized dispersal and recruit-

ment in Great Barrier Reef corals: the Helix Experiment. Science

239:1422–1424

Schmidt-Roach S, Kunzmann A, Arbizu PM (2008) In situ observa-

tion of coral recruitment using fluorescence census techniques.

J Exp Mar Biol Ecol 367:37–40

Smith SR (1992) Patterns of coral recruitment and postsettlement

mortality on Bermudas reefs—comparisons to Caribbean and

pacific reefs. Am Zool 32:663–673

Tomascik T (1991) Settlement patterns of Caribbean scleractinian

corals on artificial substrata along a eutrophication gradient,

Barbados, West Indies. Mar Ecol Prog Ser 77:261–269

Tomascik T, Sander F (1985) Effects of eutrophication on reef

building corals. I. Growth rate of the reef building coral

Montastrea annularis. Mar Biol 87:143–155

Tomascik T, Sander F (1987a) Effects of eutrophication on reef

building corals. II. Structure of scleractinian coral communities

on fringing reefs, Barbados, West Indies. Mar Biol 94:53–75

Tomascik T, Sander F (1987b) Effects of eutrophication on reef

building corals. III. Reproduction of the reef building coral

Porites porites. Mar Biol 94:77–94

Tomascik T, vanWoesik R, Mah AJ (1996) Rapid coral colonization

of a recent lava flow following a volcanic eruption, Banda

Islands, Indonesia. Coral Reefs 15:169–175

Tomascik T, Mah AM, Nontji A, Moosa MK (1997) The ecology of

Indonesian seas. Oxford University Press, Singapore

Turak E (2003) Coral diversity and distribution in the Wakatobi

National Marine Park. In: Pet-Soede L, Erdmann MV (eds)

Rapid ecological assessment Wakatobi National Park, WWF and

TNC Joint Publication, Indonesia

Underwood JN, Smith LD, Van Oppen MJH, Gilmour JP (2007)

Multiple scales of genetic connectivity in a brooding coral on

isolated reefs following catastrophic bleaching. Mol Ecol

16:771–784

Underwood JN, Smith LD, van Oppen MJH, Gilmour JP (2009)

Ecologically relevant dispersal of corals on isolated reefs:

implications for managing resilience. Ecol Appl 19:18–29

Mar Biol

123

Author's personal copy