37

Marine Pollution Bulletin 110 (2016) 701–717
Contents lists available at ScienceDirect
Marine Pollution Bulletin
journal homepage: www.elsevier.com/locate/marpolbul
Variation in the composition of corals, fishes, sponges, echinoderms,

ascidians, molluscs, foraminifera and macroalgae across a pronounced
in-to-offshore environmental gradient in the Jakarta Bay–Thousand
Islands coral reef complex
D.F.R. Cleary a, A.R.M. Polónia a, W. Renema b, B.W. Hoeksema b,c, P.G. Rachello-Dolmen d,e, R.G. Moolenbeek b,
A. Budiyanto f, Yahmantoro f, Y. Tuti f, Giyanto f, S.G.A. Draisma g, W.F. Prud'homme van Reine b, R. Hariyanto f,
A. Gittenberger b,c,h, M.S. Rikoh f, N.J. de Voogd b,⁎
a
Department of Biology, CESAM, Universidade de Aveiro, Campus Universitário de Santiago, 3810-193 Aveiro, Portugal
b
Naturalis Biodiversity Center, Darwinweg 2, 2333 CR Leiden, The Netherlands
c
Institute of Biology (IBL), Leiden University, P.O. Box 9516, 2300 RA Leiden, The Netherlands
d
Smithsonian Tropical Research Institute, Naos Island Marine Laboratory, Panama City 20521-9100, Panama
e
Dept. of Geology & Geophysics, Texas A&M University, College Station, TX 77843, United States
f
Research Centre for Oceanography, Indonesian Institute of Sciences, Pasir Putih 1, Ancol Timur, Jakarta Utara 11048, Indonesia
g
Center of Excellence for Biodiversity of Peninsular Thailand, Faculty of Science, Prince of Songkla University, Hat Yai, Songkhla 90112, Thailand
h
GiMaRIS, J.H. Oortweg 21, 2333 CH Leiden, The Netherlands
a r t i c l e i n f o a b s t r a c t
Article history: Substrate cover, water quality parameters and assemblages of corals, fishes, sponges, echinoderms, ascidians, mol-
Received 30 September 2015 luscs, benthic foraminifera and macroalgae were sampled across a pronounced environmental gradient in the Ja-
Received in revised form 12 April 2016 karta Bay–Thousand Islands reef complex. Inshore sites mainly consisted of sand, rubble and turf algae with
Accepted 17 April 2016
elevated temperature, dissolved oxygen, pH and chlorophyll concentrations and depauperate assemblages of all
Available online 11 May 2016
taxa. Live coral cover was very low inshore and mainly consisted of sparse massive coral heads and a few encrusting
Keywords:
species. Faunal assemblages were more speciose and compositionally distinct mid- and offshore compared to in-
Composition shore. There were, however, small-scale differences among taxa. Certain midshore sites, for example, housed as-
Redundancy analysis semblages resembling those typical of the inshore environment but this differed depending on the taxon.
Multi-taxon Substrate, water quality and spatial variables together explained from 31% (molluscs) to 72% (foraminifera) of
Urbanisation the variation in composition. In general, satellite-derived parameters outperformed locally measured parameters.
© 2016 Elsevier Ltd. All rights reserved.
1. Introduction herbivory as top-down mechanism and an increasing supply of nu-

trients as bottom-up mechanism (Burkepile et al., 2013; Bruno
A central aim in ecology is to explain the spatial and temporal varia- et al., 2014). In addition to this, corals may also undergo mortality
tion in biodiversity (Mora et al., 2003). Understanding the causes of this as a result of mass bleaching events, which impacts some species
variation is an important issue in ecosystems such as coral reefs, partic- more than others (Carpenter et al., 2008) and therefore may also
ularly those that have been severely affected by over-harvesting, pollu- have a selective effect on coral communities.
tion, disease and coral bleaching (Roberts et al., 2002; Walther et al., The community composition of coral reef taxa is structured by often
2002; Mora et al., 2003; Bellwood et al., 2004). Evidence over the last pronounced in-to-offshore gradients in parameters including depth, sa-
40 years has revealed a widespread loss of coral cover in tropical linity, nutrient concentrations and sedimentation (Cleary et al., 2005,
coastal waters of the Indo-West Pacific and the western Atlantic. In 2008, 2014; Fox and Bellwood, 2007). Coral reef communities have
some areas, entire coral reefs have been overgrown and killed by also been structured by infrequent past events (e.g., hurricanes) and
fast-growing species of macroalgae, leading to ‘coral–algal shifts’ chronic stress related to human activities such as fishing and the release
(Raffaelli et al., 1998; Bellwood et al., 2004; Schaffelke et al., 2005). of heavy metals and other contaminants. Many of these pressures also
This kind of reef degradation has been attributed to a decline in have a pronounced spatial component (Goatley and Bellwood, 2013).
Taxa also interact and can have important effects on the community
⁎ Corresponding author. composition of other taxa (Bellwood et al., 2004; Berumen and
E-mail address: nicole.devoogd@naturalis.nl (N.J. de Voogd). Pratchett, 2006; Hughes et al., 2010). Herbivory is considered to be a
http://dx.doi.org/10.1016/j.marpolbul.2016.04.042
0025-326X/© 2016 Elsevier Ltd. All rights reserved.
702 D.F.R. Cleary et al. / Marine Pollution Bulletin 110 (2016) 701–717
significant selective factor structuring macroalgal assemblages (Littler sides were sampled. This entailed estimating the cover of coral colo-
and Littler, 1984) and can increase species diversity by removing dom- nies and other benthic cover categories (algae, dead coral, coral rub-
inant spatial competitors (Menge and Farrell, 1989), by clearing sub- ble and sand) using the line intercept transect survey method
strate for new individuals (Menge and Lubchenco, 1981), and by (English et al., 1997). In each site, three 30-m transects were laid
maintaining the equilibrium between competing species (Gleeson and out at two depths (3 and 5 m) for a total of six transects per site. A
Wilson, 1986). Herbivory can also reduce diversity by selectively re- total of 43 sites were sampled that included a core set of 28 sites
moving preferred food items (Lubchenco and Gaines, 1981) and alter- (Table 1). For certain taxa, some core sites were not sampled due to
ing rates of succession (McClanahan, 1997). Major grazers include logistical constraints and additional sites were sampled. Along each
fishes, molluscs and sea urchins (Hay, 1981a, 1981b, 1981c). transect, we measured the intercept distance of individual live
The crown-of-thorns starfish, Acanthaster planci, a ferocious coral coral colonies and other substrate cover categories.
predator, has also caused widespread damage to Indo-Pacific coral Consistent with previous studies of the area (DeVantier et al., 1998;
reefs (Baird et al., 2013; Pratchett et al., 2014; Plass-Johnson et al., Cleary et al., 2006, 2008), all sample sites fell within specific zones along
2015). Other taxa such as sponges, ascidians and benthic algae in con- an in-to-offshore gradient. These included an inshore zone (zone 1) to
trast may not prey on corals, but compete with corals for habitat the south of − 5.97° latitude, a midshore zone (zone 2) between
space (Jackson and Buss, 1975; Lirman, 2014), which may be more − 5.77° and − 5.97° latitude, and an offshore zone (zone 3) to the
problematic for some corals than others (van der Ent et al., 2015). In ad- north of −5.77° latitude (Fig. 1). The offshore zone is largely contained
dition to studying communities of coral, it is important to understand within the Pulau Seribu National Marine Park, the first marine park
how the various coral reef communities vary spatially and in relation established in Indonesia (Farhan and Lim, 2012).
to environmental conditions in order to assess management strategies
and the exploitation of natural products (Tuomisto et al., 2003). 2.3. Corals
The main objective of the present study was to assess the composi-
tion of several abundant coral reef taxa across a pronounced in-to- Scleractinian coral species were visually identified during line-
offshore gradient in the Jakarta Bay–Thousand Islands reef complex. To intercept transect surveys (English et al., 1997). Photos of unrecognized
achieve this objective, we sampled data on reef-dwelling corals, fishes, specimens were taken for closer examination and identification using
sponges, echinoderms, ascidians, molluscs, large benthic foraminifera Veron (2000). The line-intercept transect data was analysed in order
(LBF) and macroalgae. We also measured substrate variables including to calculate the percent cover of each coral species and life form sur-
live coral and algal cover, dead coral, rubble and sand cover and local veyed (see below). In each site, we surveyed six transects (30 m long)
water quality variables including water transparency, pH and dissolved at each of the two studied depths (3 and 5 m). In total, 28 sites were sur-
oxygen content. The water quality variables were supplemented with veyed for corals.
satellite-derived parameters including chlorophyll-a concentration, re-
mote sensing reflectance at 645 nm (Rrs_645), sea surface temperature 2.4. Fishes
(SST) and coloured dissolved organic matter index (CDOM). The
satellite-derived parameters were chosen as proxies of potential threats Fishes were visually assessed along six transects (30 m long) at each
to coral reefs, namely eutrophication (chlorophyll-a concentrations), of the two studied depths (3 and 5 m). Individuals observed within 5 m
bleaching (sea surface temperature), sedimentation (remote sensing re- on either side of the transect were identified to species, if possible, and
flectance at 645 nm) and runoff (colored dissolved organic matter recorded. Individuals that left the transect area and re-entered were not
index) (Miller and Mckee, 2004; Chen et al., 2007). In addition to counted again (Hoey and Bellwood, 2008; Dickens et al., 2011). In total,
assessing compositional change across the in-to-offshore gradient, we 28 sites were surveyed for fishes.
also assessed to what extent spatial, substrate and water quality vari-
ables are able to explain variation in the composition of coral reef taxa. 2.5. Sponges
2. Material and methods Thirty patch reefs were visually surveyed during a one-hour dive
from deep to shallow water (3–25 m) using a presence/absence classi-
2.1. Study site fication. Smaller (cryptic, boring, and thinly encrusting b4 cm) speci-
mens were excluded from this study. Species were visually identified
The Jakarta Bay–Thousand Islands reef complex (hereafter re- in the field, and fragments of all species were collected for closer exam-
ferred to as JBTI) consists of cay-crowned platform reefs that extend ination. Voucher specimens were preserved in 70% ethanol and depos-
from Jakarta Bay to more than 80 km to the north in the Java Sea ited in the sponge collection at Naturalis Biodiversity Center, Leiden,
(Fig. 1). Several rivers transport sewage and storm water over a the Netherlands.
2000 km2 catchment area to the central sector of the bay (Williams
et al., 2000; Rinawati et al., 2012), defined by two flanking delta sys- 2.6. Echinoderms
tems, both of which have a large sediment input in a catchment area
of 6000 km2 (Rees et al., 1999; Williams et al., 2000). The river dis- In close proximity of the transects, a visual census was made of as-
charge in JBTI also contains pollutants, such as industrial waste and teroids and echinoids (starfish and sea urchins; referred to as echino-
household litter (Willoughby, 1986; Uneputty and Evans, 1997; derms in the present study for simplicity) using a presence/absence
Willoughby et al., 1997), heavy metals (Hosono et al., 2011), and classification. No attempt was made to locate cryptic species. In total,
nutrients (Baum et al., 2015). Annual precipitation averages 31 sites were surveyed for echinoderms.
1700 mm yr− 1 with a ‘wet’ season during the northwest monsoon
(November–March) and a ‘dry’ season during the southeast mon- 2.7. Ascidians
soon (May–September) (Rees et al., 1999).
In the neighbourhood of the transects, a visual census was made of
2.2. Data collection ascidians using a presence/absence classification. In total, 25 sites
were surveyed for ascidians. It was not possible to identify all ascidians
Sampling took place from September 7th to 23rd 2005, predomi- to species level, as many species have not been described and also
nantly at the northwestern side of the surveyed reefs with the excep- because immature specimens of some species lacked diagnostic charac-
tion of a single reef (Tikus) where the southern and northwestern ters. Following a similar study off Halmahera, Moluccas, Indonesia,
D.F.R. Cleary et al. / Marine Pollution Bulletin 110 (2016) 701–717 703
Fig. 1. Map of the study area (Jakarta Bay and Thousand islands, Indonesia). The names of selected islands adjacent to our sampling sites are shown.
specimens that could not be identified were assigned to morphotypes do not represent total molluscan diversity. In total, 37 sites were sur-
(Gittenberger et al., 2014). Whenever possible, specimens were veyed for molluscs.
photographed underwater revealing diagnostic characters. Sampled
specimens were photographed in the laboratory, preserved in either 2.9. Large benthic foraminifera
96% ethanol and/or formalin 4% as reference material and deposited
in the ascidian collection at Naturalis Biodiversity Center, Leiden, Large benthic foraminifera (hereafter referred to as LBF) were col-
the Netherlands. lected every 3 m of depth along a depth transect starting at the reef
base to the reef crest using methods described in Renema (2008). How-
2.8. Molluscs ever, in this study, only samples collected between 1 and 10 m depth
were included. On the reef slope a sample of the reef substratum was
Sediment samples were collected along each transect. For the pur- collected down to a depth in the sediment/rubble where no LBF were
poses of this study, and due to the available taxonomic expertise, we fo- observed anymore. Afterwards the samples were washed to remove
cused on species belonging to the bivalve families Cardiidae and the LBF from the larger and heavier parts of coral rubble, dried, and fur-
Pectinidae and the gastropod families Conidae, Cypraeidae and ther processed in the laboratory. LBF larger than 500 μm were subjected
Nassariidae. For simplicity, we refer to these as molluscs although they to further study. Following previous studies (Renema, 2008), LBF in all
Table 1
Number of observed species of corals (Cor.sp), fishes (Fis.sp), sponges (Spo.sp), echinoderms (Ech.sp), ascidians (Asc.sp), molluscs (Mol.sp), LBF (LBF.sp) and algae (Alg.sp) and number of
observed individuals for corals (Cor.ab), fishes (Fis.ab) and LBF (LBF.ab). The minimum, maximum, mean and standard deviation (std) are given for all sites (total) and for zones 1, 2 and 3
separately.
Code Location Zone Long Lat Cor.sp Cor.ab Fis.sp Fis.ab Spo.sp Ech.sp Asc.sp Mol.sp LBF.ab LBF.sp Alg.sp
AyB Ayer Besar Island 1 106.775 −6.001 33 90 – – 8 2 7 7 6 4 8

Bid Bidadari Island 1 106.741 −6.026 2 4 16 46 18 8 1 7 520 9 3
DaK Damar Kecil Island 1 106.841 −5.983 3 6 34 376 6 2 3 8 – – 12
Kel Kelor Island 1 106.740 −6.022 2 5 7 16 12 2 3 8 12 3 4
NyB Nyamuk Island 1 106.851 −6.024 9 46 35 783 – 4 – – – – –
Onr Ounrust Island 1 106.726 −6.034 0 0 5 7 14 5 – – – – 7
UbB Ubi Island 1 106.737 −5.993 9 19 45 277 8 3 2 2 – 0 –
UnJ Untung Jawa Island 1 106.702 −5.971 47 122 73 860 16 2 6 1 1210 9 18
Bok Bokor Island 2 106.623 −5.939 27 59 57 1189 24 5 7 14 762 7 18
Bur Burung Island N-side 2 106.601 −5.859 – – – – – 2 – – – – –
DaB Damar Besar Island 2 106.838 −5.953 58 204 70 1456 12 3 8 2 456 7 11
Dap Dapur Island 2 106.721 −5.941 40 212 101 3051 15 3 5 14 1120 13 7
JoE Karang Jong reef E-side 2 106.643 −5.851 – – – – – – 5 22 – – –
JoN Karang Jong reef NW-side 2 106.641 −5.851 – – – – – – 5 2 – – –
Kon Kongsi Island 2 106.593 −5.853 – – – – – – – 3 – – –
KoS Kongsi Island S-side 2 106.607 −5.869 – – – – – – – 7 – – –
LaB Lancang Besar Island 2 106.588 −5.923 40 119 77 1431 23 8 10 3 777 10 26
PaB Payung Besar Island E-side 2 106.558 −5.820 – – – – – – –– 10 – – –
PaE Pari Island E-side 2 106.637 −5.853 – – – – – – 9 13 1469 14 –
PaK Payung kecil Isl 2 106.543 −5.808 56 306 82 2031 29 5 – 16 – – 30
PaO Pari Island S-side offshore 2 106.635 −5.875 – – – – – – – 19 – – –
PaS Pari Island S-side 2 106.607 −5.869 – – – – – 4 – – – – –
PaW Pari Island W-side South of jetty 2 106.607 −5.858 – – – – – – – 7 – – –
PeE Peniki Island E-side 2 106.717 −5.689 – – – – 35 – – 20 – – –
PeW Peniki Island NW-side 2 106.709 −5.691 – – – – 35 – – 21 – – –
TdK Tidung Kc Island 2 106.517 −5.793 55 411 118 4141 39 7 6 16 451 15 36
Tid Tidung Besar Isl 2 106.475 −5.788 121 518 114 1781 37 7 17 20 970 12 35
TkE Tikus Island NE-side 2 106.584 −5.850 – – – – – – – 12 – – –
TkS Tikus Selatan Island 2 106.576 −5.859 41 124 86 2026 27 6 8 12 321 13 15
TkU Tikus Utara Island 2 106.574 −5.852 95 364 118 4329 32 1 – 17 665 15 27
TkW Tikus Island W-side 2 106.569 −5.855 – – – – 28 8 12 15 – – –
Air Air Island 3 106.741 −5.756 74 272 100 2301 33 9 6 17 508 17 37
Bel Belanda Island 3 106.602 −5.602 84 386 127 3047 34 6 7 11 534 11 27
HaB Hantu Besar Island 3 106.537 −5.525 96 350 147 3361 41 8 10 – 656 13 11
HaK Hantu Kecil Island 3 106.526 −5.535 70 242 136 2995 – 6 – – 412 15 16
Juk Jukung Island 3 106.523 −5.567 41 113 119 3933 38 – 9 14 – – 14
KoB Kotok Besar Island 3 106.537 −5.693 54 264 103 3856 31 3 – 17 574 14 33
KoK Kotok Kecil Island 3 106.534 −5.687 67 410 114 3847 37 9 12 20 211 11 29
Kpa Kelapa Island 3 106.555 −5.653 110 305 86 3631 31 3 – 13 674 13 25
Pan Panjang Island 3 106.556 −5.639 69 374 114 3320 42 7 15 9 269 15 29
Put Putri Island 3 106.567 −5.588 71 344 111 9524 37 9 8 1 538 14 22
SeB Sepa Island 3 106.575 −5.572 77 303 97 5117 25 5 – 11 421 14 23
SeD Semak Daun Island 3 106.571 −5.726 91 349 122 3732 37 4 6 6 1488 16 37
Total sites sampled 28 28 28 28 30 31 25 37 24 24 27
Min total 2 4 5 7 6 1 1 1 6 3 3
Max total 121 518 147 9524 42 9 17 22 1488 17 37
Mean total 55.1 225.8 86.2 2588.0 26.8 5.0 7.5 11.3 626.0 11.8 20.7
std total 32.9 149.2 39.5 2029.6 11.0 2.5 3.8 6.3 392.8 3.7 10.7
Min zone 1 (inshore) 2 4 5 7 6 2 1 1 6 3 3
Max zone 1 (inshore) 47 122 73 860 18 8 7 8 1210 9 18
Mean zone 1 (inshore) 15.0 41.7 30.7 337.9 11.7 3.5 3.7 5.5 437.0 6.3 8.7
Std zone 1 (inshore) 17.8 47.2 24.0 359.2 4.5 2.1 2.3 3.1 568.9 3.2 5.6
Min zone 2 (midshore) 27 59 57 1189 12 1 5 2 321 7 7
Max zone 2 (midshore) 121 518 118 4329 39 8 17 22 1469 15 36
Mean zone 2 (midshore) 59.2 257.4 91.4 2381.7 28.0 4.9 8.4 12.6 776.8 11.8 22.8
Std zone 2 (midshore) 30.1 152.7 22.3 1179.4 8.4 2.4 3.6 6.5 364.7 3.1 10.5
Min zone 3 (offshore) 41 113 86 2301 25 3 6 1 211 11 11
Max zone 3 (offshore) 110 410 147 9524 42 9 15 20 1488 17 37
Mean zone 3 (offshore) 75.3 309.3 114.7 4055.3 35.1 6.3 9.1 11.9 571.4 13.9 25.3
Std zone 3 (offshore) 18.5 80.5 17.1 1850.3 4.9 2.3 3.1 5.6 336.7 1.9 8.5
samples were identified to species level using a stereomicroscope but and many (but not all) turf algae were excluded. Algae were identified
only samples containing more than 200 individuals have been included to species level or if unreliable to genus level. Reference vouchers are
in this study. In total, 24 sites were surveyed for LBF. kept at the herbarium (L) of the Naturalis Biodiversity Center, Leiden,
the Netherlands.
2.10. Macroalgae
2.11. Habitat structure
One snorkeler and one SCUBA diver made a presence–absence visual
census of red, brown, and green macroalgae (Rhodophyta, In the present study, habitat variables consisted of 12 distinct sub-
Phaeophyceae and Chlorophyta) during 1 h on the reef slope, on the strate forms. Since the high habitat heterogeneity of reef systems is es-
reef crest, and on the reef flat (lagoon) at 27 sites. Encrusting algae sentially created by the complex and diverse colonial structures of
scleractinian corals, seven of the 12 substrate forms used to characterise argument set to binomial. However, the ratio of residual deviance to re-
the habitat heterogeneity of the JBTI reef system were based on live sidual d.f. in the models substantially exceeded one so we set family to
coral cover. Coral-related variables were not used to predict variation ‘quasibinomial’. In the ‘quasibinomial’ family, the dispersion parameter
in coral composition, nor were algal variables used to predict variation is not fixed at one so that it can model over-dispersion. Using the GLM
in algal composition. The 12 distinct forms of substrate assessed includ- model, we tested for significant variation among habitats using the
ed the cover of Acropora corals, branching corals, encrusting corals, foli- anova() function in R with the F test, which is most appropriate when
ose corals, massive corals, submassive corals, free-living mushroom dispersion is estimated by moments as is the case with quasibinomial fits.
corals, turf algae, macroalgae, dead coral, rubble and sand. Redundancy analysis (RDA) and variance partitioning (Borcard and
Legendre, 2002) were used to assess to what extent spatial, habitat
2.12. Water quality and water quality variables were able to predict variation in taxon com-
position. Spatial variation in the study area was modelled using princi-
We used an aqua Troll water quality instrument (In situ Inc., Fort pal coordinates of neighbour matrices (PCNM). PCNM is a method for
Collins, CO, USA) to measure water temperature (Tmp), pH, dissolved quantifying spatial trends across a range of scales and is based on eigen-
oxygen (DO; Clark DO μg s l−1) and salinity (Sal: parts per thousand value decomposition of a truncated matrix of geographic distances
(ppt) or ‰). Water transparency was also measured once in each site among sampling sites (Borcard and Legendre, 2002). For a detailed de-
during each sampling event using a Secchi disk. In addition to the scription of PCNM see Borcard and Legendre (2002). Significant PCNM
above, we also used satellite imagery to measure four water quality var- eigenvectors were selected using the pcnm() function in the Vegan
iables namely, coloured dissolved organic matter index (CDOM), near- package in R (Oksanen et al., 2009) with 999 permutations. Quantitative
surface chlorophyll-a concentration (Chla), sea surface temperature taxon data matrices (corals, fishes and LBF) were loge (x + 1) trans-
(SST) and remote sensing reflectance at 645 nm (Rrs_645) following formed and further ‘transformed’ using the decostand() function in
previously described methods (Polónia et al., 2015). CDOM was used Vegan. Qualitative (presence/absence) data matrices were only trans-
as a tracer of riverine inputs and Rrs_645 as a proxy for total suspended formed using the decostand() function in Vegan. With the decostand
sediments derived from land-based erosion (Miller and Mckee, 2004; transformation, the species data were adjusted so that subsequent ordi-
Chen et al., 2007). CDOM is largely composed of humic and fulvic sub- nation analyses preserved the chosen distance among sample sites. In
stances resulting either from decaying plant material brought by land the present case, the Hellinger distance was used, as recommended by
run-off in areas with high vegetation productivity or originating from Legendre and Gallagher (2001). For each taxon/group, we then set
mangroves and seagrasses (Carder et al., 1999; Martin, 2004; up two models using redundancy analysis with the Hellinger-
Richardson and LeDrew, 2006). Since the accuracy of satellite data transformed matrix as response variable and spatial, substrate and
(compared to in situ data) tends to be higher with long-term averaging water quality variables as explanatory variables respectively. RDA ar-
(Patt et al., 2003), time series of monthly mean data were generated. ranges the data points in a multidimensional space where the axes rep-
Long-term mean values were generated for the previously mentioned resent gradients in species abundances, constrained by the explanatory
satellite-derived parameters for the years 2002–2005. variables (spatial, substrate and water quality variables) (Makarenkov
and Legendre, 2002; Cleary et al., 2005). Here, the amount of variation
2.13. Analyses in composition explained by the explanatory variables is the sum of all
constrained eigenvalues divided by the total variation in the species
Taxon (species-by-site), water quality, habitat and spatial (latitude data (Cleary et al., 2005). We used the ordistep() function in vegan to se-
and longitude of each sample site) matrices (see Appendices 1–14 in lect significant spatial, substrate and water quality predictors of variation
the online version at http://dx.doi.org/10.1016/j.marpolbul.2016.04. in composition using backward selection, maximum permutations set to
042.) were imported into R (R Core Team, 2013). Species abundance 1000 and a selection criterion of P = 0.10. Variance partitioning (with
matrices were loge (x + 1) transformed (in order to normalise the dis- the varpart function in vegan; Borcard and Legendre, 2002; Cleary
tribution of the data) and a distance matrix was constructed using the et al., 2004) was then used to partition the variance explained (1) purely
Bray–Curtis index for species matrices or the Euclidean index for by habitat variables, (2) purely by water quality variables, (3) purely by
water quality and substrate form matrices with the vegdist() function spatial variables and (4) by combinations of habitat, water quality and
in the VEGAN package (Oksanen et al., 2009) in R. The Bray–Curtis spatial variables. The significance of the RDA ordination axes was tested
index is one of the most frequently applied (dis)similarity indices with the anova() function in Vegan with the ‘by’ argument set to ‘axis’.
used in ecology (Legendre and Gallagher, 2001). Variation in species This function tests the joint effect of constraints in RDA using an
composition, water quality and substrate composition among zones ANOVA like permutation test. All figures were made using R.
was assessed with Principal Coordinates Analysis (PCO) using the
cmdscale() function in R with Bray–Curtis (species composition) or 3. Results
Euclidean (water quality and substrate composition) distance matrices
as input. Variation among zones was tested for significance using the There were highly significant differences in substrate cover among
adonis() function in VEGAN. In the adonis analysis, the Bray–Curtis or zones (Fig. 2). Acropora, branching, foliose, submassive and dead coral
Euclidean distance matrices of species composition, water quality or cover was higher in mid- and offshore sites than inshore sites whereas
substrate composition were the response variables with the zone as in- sand cover was higher inshore than mid- and offshore. There was no
dependent variable. The number of permutations was set at 999; all significant difference in the cover of encrusting and massive corals,
other arguments used default values, which were set in the function. macro- and turf algae or rubble among zones. There were also signifi-
We also used the adonis() function to test for significant differences cant differences in selected locally measured environmental parameters
among zones for selected individual water quality variables (transpar- among zones with salinity and water transparency higher mid- and off-
ency, Tmp, pH, Sal and DO). Weighted averages scores were computed shore than inshore and temperature, pH and dissolved oxygen concen-
for species, water quality or substrate variables on the first two PCO tration higher inshore than mid- and offshore (Fig. 3).
axes using the wascores() function in the vegan package. We tested Substrate composition differed significantly among zones (F2,26 =
for significant differences in substrate cover (Acropora coral, branching 11.85, P b 0.001, R2 = 0.477). The ordination based on substrate form
coral, encrusting coral, foliose coral, massive coral, submassive coral, showed clear separation between inshore and mid- and offshore sites.
mushroom coral, macroalgae, turf algae, dead coral, rubble and sand) The substrate composition of the midshore site Bok, however, resem-
among zones with an analysis of deviance using the glm() function in bled that of inshore sites whereas that of the midshore site Dap resem-
R. As data were proportional, we first applied a GLM with the family bled that of offshore sites. Inshore sites were characterised by high sand
Fig. 2. Mean percentage cover (error bars represent a single standard deviation) of selected substrate forms in inshore (In), midshore (Md) and offshore (Of) zones. Results of the GLM
analysis are shown in the top-right corner of each sub-figure.
cover whereas the more distant offshore sites (and Dap) were (Kel, inshore) to 17 (Air, offshore) and macroalga species from 0 (UbB,
characterised by high Acropora, branching and foliose coral cover inshore) to 37 (Air and SeD, both offshore). The mean number (and
(Fig. 4a). The other midshore sites closest to shore (LaB and DaB) had standard deviation) of coral species encountered per site varied from
values close to zero on the first PCO axis indicating substrate structure 13.1 (±17.3) inshore to 59.2 (±30.07) midshore and 78.5 (±15.7) off-
intermediate between inshore and the more distant mid-shore and offshore. The mean number of fish species per site varied from 30.7
shore sites (Fig. 4b). As was the case with substrate form, water quality (±24.0) inshore to 91.4 (±22.3) midshore and 114.3 (±17.9) offshore.
also differed significantly among zones (F 2,27 = 13.63, P b 0.001, All taxa, with the exception of molluscs, showed a similar pattern of in-
R2 = 0.522). Inshore sites (and the midshore site Bok) were associ- creasing number of species encountered (species density) per site from
ated with high Chla values whereas mid- and offshore sites were as- inshore to mid- and offshore.
sociated with higher CDOM values and greater water transparency The composition of all taxa differed significantly among zones
(Fig. 4c, d). (Figs. 6 and 7) and was primarily structured along the pronounced in-
The number of species encountered per site varied widely across to-offshore gradient in JBTI with the most pronounced differences be-
zones and taxa (Table 1 and Fig. 5). The number of coral species varied tween inshore versus mid- and offshore sites (Figs. 8 and 9). The
from 0 (Onr, inshore) to 121 (Tidung, midshore), fish species from 5 midshore sites closest to shore (LaB, Bok, Dap and DaB), however, ap-
(Onr, inshore) to 147 (HaB, offshore), sponge species from 6 (DaK, in- pear to represent a transition zone. Coral and fish assemblages in Bok,
shore) to 42 (Pan, offshore), echinoderm species from 1 (TkU, for example, were more similar to inshore sites, whereas assemblages
midshore) to 9 (Air, KoK and Put, all offshore), ascidian species from 1 in Dap were more similar to offshore sites and assemblages in LaB and
(Bid, inshore) to 17 (Tid, midshore), mollusc species from 1 (UnJ and DaB were intermediate. For sponges and echinoderms, in contrast, as-
Put, inshore and offshore) to 22 (JoE, midshore), LBF species from 3 semblages in LaB and Bok were more similar to offshore sites whereas
Fig. 3. Mean (error bars represent a single standard deviation) a) water transparency, b) temperature, c) pH, d) salinity and e) dissolved oxygen in inshore (In), midshore (Md) and
offshore (Of) zones. Results of the adonis analysis are shown in the top-right corner of each sub-figure.
assemblages in Dap and DaB were more similar to inshore sites (Fig. 8). (Ca-co), C. valentini (Ca-va), Scolopsis monogramma (Sp-mo), Diploprion
For ascidians assemblages in all four of these midshore sites were more bifasciatus (Di-bi) and Monacanthus sp. (Mo-sp). Most fish species, includ-
similar to inshore sites (Fig. 9). In contrast to the other taxa, assem- ing all of the most abundant (N1000 individuals) species, were, however,
blages of echinoderms and molluscs in certain offshore sites (e.g., PaS mainly found in mid- and offshore sites, e.g., Pomacentrus amboinensis
and SeD) were more similar to inshore than offshore sites. The midshore (Po-am), Pomacentrus alexanderae (Po-al), Cirrhilabrus cyanopleura (Ci-
site Dap represents an interesting site with very high coral cover and the cy), Amblyglyphidodon leucogaster (Am-le), Amblyglyphidodon leucogaster
highest foliose coral cover overall at 48.4%. At this site assemblages of (Am-le), Caesio teres (Ca-te), C. cyanopleura (Ci-cy) and Chromis
corals, fishes, molluscs and LBF were similar to offshore sites whereas ternatensis (Cr-te). Sponge species associated with inshore sites included
assemblages of sponges, echinoderms, ascidians and algae were similar Chondrilla sp. (Cn-sp), Haliclona cf. cymaeformis and Spongia ceylonensis
to inshore sites. This suggests that although taxa may globally respond (Sp-ce). The species Callyspongia joubini (Ca-jo) and Ptilocaulis sp. (Pt-
to environmental conditions in a similar fashion, there may still be sp) were mainly found inshore and midshore whereas the species
smaller-scale differences in response. Amphimedon cf. paraviridis (Am-pa), Spheciospongia inconstans (Sp-in),
Coral species mainly found inshore included the coral species Xestospongia testudinaria (Xe-te), Petrosia nigricans (Pe-ni), Aaptos
Oulastrea crispata (Ou-cr) and Polyphyllia talpina (Po-ta) (Fig. 6). Porites suberitioides (Aa-su), Clathria reinwardtii (Cl-re), Cinachyrella australiensis
lobata (Po-lo), P. lutea (Po-lu) and Galaxea fascicularis (Ga-fa), in contrast, (Ci-au) and Dysidea arenaria (Dy-ar) were found across a wide range of
were found across a wide range of sites. Most species, however, were as- sites from inshore to offshore suggesting tolerance to a wide range of eco-
sociated with mid- and offshore reefs, e.g., Seriatopora hystrix (Se-hy), logical conditions. Other sponge species were restricted to mid- and off-
Montipora foliosa (Mo-fo), Porites rus (Po-ru) and Acropora millepora shore sites including Hyrtios erectus (Hy-er), Stylissa carteri (St-ca),
(Ac-mp). Fish species observed inshore included Canthigaster compresus Callyspongia aerizusa (Ca-ae), Dysidea frondosa (Dy-fr), Axinyssa cf.
Fig. 4. Ordination showing the first two axes of the PCO analysis for a) substrate variables and c) water quality variables. Symbols represent sites from inshore, midshore and offshore
zones. Variation in site scores of the first PCO axis on a map of the study area for b) substrate variables and d) water quality variables. Open symbols indicate positive values along the
first PCO axis and shaded symbols negative values. The size of the symbol is proportional to the score. Codes for the midshore sites closest to shore are given: LaB, Bok, Dap and DaB.
Codes for a) transparency (Tra), salinity (Sal), temperature (Tmp), dissolved oxygen (DO), sea surface temperature (SST) and Rrs (remote sensing reflectance and 645 nm). Codes for
b) branching coral (Cor.bra), encrusting coral (Cor.enc), foliose coral (Cor.fol), massive coral (Cor.mas), mushroom coral (Cor.mus), submassive coral (Cor.sub), dead coral (Dead.cor),
macroalgae (Alg.mac) and turf algae (Alg.tur).
aculeata (Ax-ac) and Clathria cervicornis (Cl-ce). Echinoderm species javensis (Rh-ja), Lobophora papenfussii (Lo-pa) and Bryopsis pennata (Br-
found in inshore sites included the sand dollar Laganum depressum (La- pe).
de) and the sea urchins Brissus agassizii (Br-ag), Echinothrix calamaris Significant predictors of variation in composition included satellite-
(Et-ca) and Diadema setosum (Di-se) whereas the starfish species Linckia derived CDOM concentrations (all taxa except ascidians and LBF),
laevigata (Li-la), A. planci (Ac-pl), and Culcita novaeguinea (Cu-no) were water transparency (corals, fishes, sponges, molluscs and algae), salinity
mainly found in offshore sites. (ascidians), branching coral cover (echinoderms), free-living mush-
The ascidian species Aplidium sp. (Ap-sp) and Botrylloides sp. (Bo-sp) room coral cover (echinoderms and algae) and foliose coral cover (as-
were only found in inshore and the midshore sites closest to the coast cidians) (Table 2). Significant predictors of variation in composition
(Fig. 7). The species Pseudodistoma fragile (Ps-fr), Didemnum molle (Di- associated with inshore sites included satellite-derived chlorophyll-a
mo), D. rubeum (Di-ru), Archidistoma richerie (Ar-ri) and Leptoclinides (corals, fishes, ascidians, LBF and algae), dissolved oxygen (fishes and
reticulates (Le-re) were, in contrast, found in mid- and offshore sites. forams), temperature (corals, fishes, ascidians and LBF), SST (fishes,
The mollusc species Nassarius pauper (Na-pa) and Acrosterigma flava sponges, echinoderms, LBF and algae), sand (all taxa) and rubble
(Ac-fl) were mainly restricted to inshore and midshore sites with high (corals, fishes and molluscs). Ordinations based on redundancy analysis
sand cover. Other mollusc species found in inshore sites included that show the relationship between species and environmental vari-
Conus geographus (Co-ge), Conus glans (Co-gl) and Fulvia papyracea ables can be seen in the Supplementary material (Supplementary
(Fu-pa). Mollusc species found in mid- and offshore sites included Figs. 1 and 2 in the online version at http://dx.doi.org/10.1016/j.
Laevichlamys cuneata (La-cu), Nassarius silvardi (Na-si), Semipallium marpolbul.2016.04.042.).
fulvicostatum (Se-fu), Acrosterigma rubicundum (Ac-ru), Excellichlamys The amount of variation in species composition only explained by
histrionica (Ex-hi), Afrocardium richardi (Af-ri) and Decatopecten radula substrate variables, thus after factoring out variation explained by
(De-ra). The LBF species Calcarina sp. (Ca-s2) was found in inshore water quality and spatial variables (the pure substrate component) var-
sites whereas most other species were strongly associated with or re- ied from 6% (fishes) to 16% (ascidians). The amount of variation ex-
stricted to mid- and offshore sites, e.g., Amphistegina lobifera (Am-lo), plained by the pure water quality component (thus after factoring out
Amphistegina lessonii (Am-le), Sorites orbiculus (So-or). Most macroalga the variation explained by substrate and spatial variables) varied from
species were associated with mid- and offshore sites, for example, 8% (echinoderms and molluscs) to 23% (LBF). The pure spatial compo-
Actinotrichia fragilis (Ac-fr), Caulerpa cupressoides (Ca-cu), Bornetella nent explained from 5% (echinoderms) to 17% (LBF) of variation in
nitida (Bo-ni) and Valonia ventricosa (Ve-ve). Species mainly found in- taxon composition (Supplementary material: Supplementary Figs. 3
shore and in perturbed midshore sites (e.g., Bok) included Rhipidosiphon and 4 in the online version at http://dx.doi.org/10.1016/j.marpolbul.
Fig. 5. Mean species density (error bars represent a single standard deviation) of focal taxa in inshore (In), midshore (Md) and offshore (Of) zones.
2016.04.042.). The combined effects of all three of these parameters ex- salinity and introducing a heavy nutrient load. The coast adjacent to Ja-
plained from 3% (molluscs) to 17% (corals) of the variation in composi- karta has long been subjected to occasional severe algal blooms (Rees
tion. The combined effects of habitat and water quality, habitat and et al., 1999). The introduction of nutrients may lead to some initial acid-
space and water quality and space explained from 0% (corals, sponges ification. Phytoplankton, however, quickly utilise the available nutrients
and molluscs) to 7% (sponges). The percentage of unexplained variation and the byproducts of their photosynthesis increase seawater pH. Like-
varied from 28% (LBF) to 69% (molluscs). Substrate variables explained wise, algal blooms can cause super saturation of oxygen and conse-
more of the compositional variation of echinoderms (14%) and molluscs quently high dissolved oxygen levels. This can cause substantial
(9%) whereas water quality variables explained more of the variation of physiological stress for resident fauna (Ross et al., 2001). Die-off, how-
corals (16%), fishes (21%), sponges (11%), ascidians (19%), LBF (23%) ever, following a bloom can deplete oxygen leading to hypoxic condi-
and algae (18%). tions detrimental to the resident fauna. Eutrophied waters may also
experience a daily cycle of high oxygen concentrations during the day
4. Discussion due to photosynthesising phytoplankton and macroalgae and occasion-
al severe oxygen depletion at night when both of these groups respire.
A number of previous studies have reported on the strong in-to- This seasonal and daily variation in oxygen concentration and other var-
offshore gradient in JBTI (Moll and Suharsono, 1986; DeVantier et al., iables such as salinity make eutrophied inshore waters an inhospitable
1998; Cleary et al., 2008, 2014; de Voogd and Cleary, 2008; Renema, environment for numerous taxa.
2008; Maduppa et al., 2013). In the present study, inshore sites The number of species encountered inshore was a fraction of that
were characterised by poor water transparency and higher mean tem- encountered mid- and offshore. This indicates that many of these taxa
perature, pH, dissolved oxygen concentrations, satellite-derived are sensitive to the poor water quality (high chl-a concentrations and
chlorophyll-a concentrations and lower mean salinity. The substrate in- poor water transparency) and degraded habitat (high sand and low
shore mainly consisted of sand, rubble and turf algae with very low live coral cover) of inshore reefs. Although all taxa appeared to be adversely
coral cover. The inshore waters are subject to riverine runoff lowering affected, these conditions seemed to be particularly detrimental to
Fig. 6. Ordination showing the first two axes of the PCO analysis for a) corals, b) fishes, c) sponges and d) echinoderms. Symbols represent samples from the inshore, midshore and offshore
zones. For corals and fishes, large grey symbols represent species ≥50 individuals. Codes for a) Acropora millepora (Ac-mp), Favia halicora (Fa-ha), Galaxea fascicularis (Ga-fa), Leptastrea
purpurea (Le-pu), L. transversa (Le-tr), Montastrea sp. (Mn-sp), Montipora digitata (Mo-di), M. foliosa (Mo-fo), Oulastrea crispata (Ou-cr), Porites lobata (Po-lo), P. lutea (Po-lu), P. rus (Po-
ru), Polyphyllia talpina (Po-ta), Seriatopora hystrix (Se-hy) and Turbinaria sp. (Tu-sp). Codes for b) Abudefduf sexfasciatus (Ab-sx), Aeoliscus strigosus (Ae-st), Amblyglyphidodon curacao (Am-
cu), A. leucogaster (Am-le), Apogon compressus (Ap-co), Canthigaster compresus (Ca-co), C. valentini (Ca-va), Caesio teres (Ca-te), Cirrhilabrus cyanopleura (Ci-cy), Chromis ternatensis (Cr-te),
Diploprion bifasciatus (Di-bi), Halichoeres scapularis (Ha-sc), Monacanthus sp. (Mo-sp), Pomacentrus alexanderae (Po-al), P. amboinensis (Po-am), Siganus javus (Si-ja), Scolopsis ciliatus (Sp-
ci) and S. monogramma (Sp-mo). Codes for c) Aaptos suberitioides (Aa-su), Acanthella cavernosa (Ac-ca), Amphimedon cf. paraviridis (Am-pa), Axinyssa cf. aculeata (Ax-ac), Callyspongia
aerizusa (Ca-ae), Callyspongia sp ‘bug’ (Ca-bu), Callyspongia joubini (Ca-jo), Cinachyrella australiensis (Ci-au), Clathria cervicornis (Cl-ce), C. reinwardtii (Cl-re), Chondrilla sp. (Cn-sp),
Dasychalina fragilis (Da-fr), Dysidea arenaria (Dy-ar), D. frondosa (Dy-fr), Haliclona sp. ‘black-tubes’ (Ha-bt), Haliclona cf. cymaeformis (Ha-cm), Haliclona ‘pink’ (Ha-pi), Hyrtios erectus
(Hy-er), Lamellodysidea herbacae (La-he), Liosina paradoxa (Li-pa), Petrosia nigricans (Pe-ni), Pseudoceratina purpurea (Ps-pu), Ptilocaulis sp. (Pt-sp), Spongia ceylonensis (Sp-ce),
Spheciospongia inconstans (Sp-in), Stylissa carteri (St-ca) and Xestospongia testudinaria (Xe-te). Codes for d) Acanthaster planci (Ac-pl), Asthenosoma varium (As-va), Brissus agassizii (Br-
ag), Culcita novaeguinea (Cu-no), Diadema setosum (Di-se), Echinothrix calamaris (Et-ca), E. diadema (Et-di), Laganum depressum (La-de), Linckia laevigata (Li-la). Results of the adonis
analysis comparing composition among zones: corals: F2,27 = 6.40, P b 0.001, R2 = 0.339; fishes: F2,25 = 7.70, P b 0.001, R2 = 0.381; sponges: F2,27 = 5.98, P b 0.001, R2 = 0.307; and
echinoderms: F230, = 3.72, P b 0.001, R2 = 0.210).
photosymbiont-hosting organisms such as corals and LBF (although not Water quality variables were more important predictors of varia-
all of the LBF included in the present study are symbiont-bearing) and tion in composition for corals, fishes, sponges, ascidians, LBF and
taxa that depend on intact coral reefs for food and cover. Corals were algae whereas substrate variables were more important predictors
completely absent from one inshore site and sparse in most other in- for echinoderms and molluscs. In line with this, the biotic composi-
shore sites. Chronic exposure to land-based perturbations have also tion of selected taxa (ascidians, macroalgae, and corals) on the
led to the severe degradation of the habitat matrix whereby the com- west side of Halmahera (Moluccas, Indonesia) appeared also to de-
plex three dimensional structure of a healthy coral reef has been re- pend on water quality parameters, such as salinity and turbidity
placed by a much more simple habitat dominated by sand, turf algae (Gittenberger et al., 2014). This contrasts somewhat with Polónia
and coral rubble with a few remaining massive coral heads. Importantly, et al. (2015) who studied assemblages of corals, sponges, LBF and
the inshore reefs of Jakarta Bay formerly housed diverse and thriving sediment archaea and bacteria in the Spermonde Archipelago. In
coral reefs. Umbgrove (1939) noted that reef faunas such as those sur- the Spermonde, habitat variables were more important explanatory
rounding the inshore island of Nyamuk Besar consisted of acroporid factors than water quality variables with the exception of sediment bac-
corals such as Montipora digitata, M. foliosa and Acropora aspera. Fur- teria. The present work, however, confirms previous work in JBTI
thermore, corals of JBTI have suffered from severe coral bleaching in (Cleary et al., 2006) where proxies of water quality variables
1983, which affected some species more than others (Brown and (e.g., distance to shore) were the primary explanatory factors of coral
Suharsono, 1990; Hoeksema, 1991). Coral mortality has also been re- composition. This discrepancy may be related to the much poorer
ported as a result of predation by A. planci (e.g., Baird et al., 2013) and water quality of inshore reefs in JBTI compared to the Spermonde archi-
by overgrowth of an aggressive sponge species (de Voogd et al., pelago. The present study, in which water quality variables were more
2013). Since coral species vary in their susceptibility to these factors, important predictors of the composition of ascidians than habitat vari-
this will affect some coral species more than others, adding to the variables, is in line with Gittenberger et al. (2014) where various ascidian
ation in coral degradation. species were found to be very good indicators for bioregions, varying
Fig. 7. Ordination showing the first two axes of the PCO analysis for a) ascidians, b) molluscs, c) forams and d) algae. Symbols represent samples from inshore, midshore and offshore zones.
For forams, large grey symbols represent species ≥50 individuals. Codes for a) Aplidium breviventer (Ap-br), Archidistoma richerie (Ar-ri), Ascidia ornata (As-or), Botrylloides sp. (Bo-sp),
Clavelina cyclus (Cl-cy), C. moluccensis (Cl-mo), Didemnum molle (Di-mo), D. rubeum (Di-ru), Leptoclinides reticulates (Le-re), Pseudodistoma fragile (Ps-fr). Codes for b) Acrosterigma
angulata (Ac-an), A. flava (Ac-fl), A. rubicundum (Ac-ru), Afrocardium richardi (Af-ri), Conus geographus (Co-ge), C. glans (Co-gl), Excellichlamys histrionica (Ex-hi), Frigidocardium torresi
(Fr-to), Fulvia papyracea (Fu-pa), Laevichlamys cuneata (La-cu), Nassarius canaliculatus (Na-ca), N. pauper (Na-pa), Semipallium fulvicostatum (Se-fu). Codes for c) Amphistegina lessonii
(Am-le), Amphistegina lobifera (Am-lo), Calcarina defrancei (Ca-de), C. sp2 (Ca-s2), Heterostegina depressa (He-de), Laevipeneroplis malayensis (La-ma), Operculina discoidalis (Op-di),
Sorites orbiculus (So-or). Codes for d) Actinotrichia fragilis (Ac-fr), Amphiroa spp. (Am-sp), Bornetella nitida (Bo-ni), Bryopsis pennata (Br-pe), Caulerpa brachypus (Ca-br), C. cupressoides
(Ca-cu), C. racemosa species complex (Ca-ra), Dictyota spp. (Dy-sp), Halimeda spp. (Ha-sp), Laurencia spp. (La-sp), Lobophora papenfussii (Lo-pa), L. variegata (Lo-va), Padina spp. (Pa-
sp), Portieria hornemannii (Po-ho), Rhipidosiphon javensis (Rh-ja), Sargassum spp. (Sa-sp), Spongocladia (Cladophoropsis) vaucheriaeformis (Sp-va), Turbinaria spp. (Tu-sp), Ventricaria
(Valonia) ventricosa (Ve-ve). Results of the adonis analysis comparing composition among zones: ascidians: F2,22 = 3.56, P b 0.001, R2 = 0.244; molluscs: F2,34 = 1.75, P = 0.006,
R2 = 0.094; LBF: F2,21 = 4.07, P b 0.001, R2 = 0.279; and algae: F2,27 = 5.82, P b 0.001, R2 = 0.327).
in their distance offshore, while no correlations were found between as- The pure water quality component was a more important predictor
cidian communities and habitats. of variation in fish (21% versus 6%) and LBF (23% versus 9%) composition
The results of the 2005 survey showed that the inshore reefs of Ja- than the pure substrate component. This contrasts somewhat with pre-
karta Bay were clearly the poorest with regard to coral species numbers vious studies of coral reef fishes. Substrate structure, for example, as op-
(Table 1); this was also observed during similar surveys in 1985, 1995 posed to just the amount of live coral cover has previously been shown
and 2011 (Cleary et al., 2006, 2008, 2014). This low species density is to strongly affect the composition of fish assemblages (Lewis, 1997;
most probably related to the river discharge along the shore line of Ja- Syms and Jones, 2000; Emslie et al., 2008; Baum et al., 2015). Likewise,
karta Bay, causing an influx of sediment, nutrients and other pollutants in Bocas del Toro, Panama, fish with a given set of traits were associated
(Verstappen, 1988; Farhan and Lim, 2012; Johan et al., 2015). It is im- with specific habitats. For example, highly mobile species and oceanic
portant to note, however, that a study of historical museum collections planktivores were more abundant in exposed reef environments while
has indicated that coral species richness in Jakarta Bay was much higher demersal zooplankton feeders, territorial herbivores and sedentary spe-
in the 1920s (van der Meij et al., 2009). A similar decline in species rich- cies were found in sheltered environments such as massive and folia-
ness was observed for molluscs between the 1930s and 2005 (van der ceous coral zones. Omnivores were more abundant in areas of sand
Meij et al., 2010). The zebra coral, O. crispata, is an exception by being and rubble (Dominici-Arosemena and Wolff, 2005). In the Ryukyu
the most common scleractinian coral species in most of the inshore Islands Japan, however, fish assemblages varied with reef zone, water
reefs, whereas it was absent in mid- and offshore reefs, as also observed depth and offshore distance but not with coral abundance or density
in 1985 (Cleary et al., 2006). This coral species is known to occur pre- (Lecchini et al., 2003).
dominantly in shallow, turbid environments where other corals are Symbiont bearing larger foraminifera are among the most important
rare (Yamashiro, 2000; Chen et al., 2011). In the Spermonde archipela- calcifiers in healthy reef systems (Scoffin and Tudhope, 1985; Renema,
go, coral species density was also relative low inshore; this has been at- 2006). They are sensitive to environmental parameters and widely
tributed to the close proximity of inshore reefs to river outlets and the used as indicators of water quality (Hallock et al., 2003; Uthicke and
shallower depth of the surrounding sea floor (Cleary et al., 2005; Nobes, 2008). Poor water quality results in fewer obligate symbiont-
Becking et al., 2006; Hoeksema, 2012a, 2012b; Polónia et al., 2015). bearing taxa relative to non-symbiont-bearing taxa (Hallock et al.,
Fig. 8. Variation in site scores of the first PCO axis on a map of the study area for a) corals, b) fishes, c) sponges and d) echinoderms. Open symbols indicate positive values along the first axis
and shaded symbols negative values. The size of the symbol is proportional to the score. Codes for the midshore sites closest to shore are given: LaB, Bok, Dap and DaB. For echinoderms
additional extra codes are shown for the mid- and offshores sites: PaS and SeD.
2003), as observed in the inshore reefs. Depth is an important compo- seagrasses to frondose macroalgae to turf algae to faster growing phyto-
nent determining the assemblage composition of LBF within reef sys- plankton (Duarte, 1995). Although macroalgae may temporarily benefit
tems, and most variation in assemblage composition is observed in from high nutrient levels, increasing nutrification eventually leads to
either shallow (reef flat) assemblages (Renema, 2010) or deeper on them being out-competed by turf algae and phytoplankton. The latter
the reef slope (Hohenegger, 1994). In Caribbean reef systems, members prevent sufficient light reaching the benthic seaweeds eventually lead-
of the genus Amphistegina are an important component of LBF assem- ing to their demise. Moreover, most macroalgae need a hard substrate
blages. Variation in the proportion of symbiont-bearing LBF in the to attach to and the dominant substrate inshore mainly consisted of
total benthic LBF assemblage is for a large part determined by the abun- sand, silt and shifting rubble (Rachello-Dolmen and Cleary, 2007). Fur-
dance of Amphistegina species. In the Indo-West-Pacific, however, there thermore, the relatively high density of grazing sea urchins inshore
are more LBF species and they occupy a greater variety of habitats may have a profound effect on the macroalgae. Additionally, the inshore
(Hohenegger, 1994; Renema, 2008, 2010). Within LBF, certain taxa, par- islands lack lagoons and the shallower reefs provide less variation in
ticularly members of the genus Calcarina tolerate both poor water qual- habitat than the reefs around the deeper channel in the center of the
ity and benthic habitats dominated by algae as opposed to corals study area, i.e., the northern part of zone 2 and the southern part of
(Renema, 2010). Within this group, some species are particularly sensi- zone 3. The greatest number of macroalgal species were found around
tive to substrate type, including the presence of corals and the absence this channel. Interestingly, known indicator taxa of eutrophication,
of sand and others to water quality (Renema, 2010). Compared to such as Ulva and Chaetomorpha (Barile, 2004; Teichberg et al.,
other reef systems in the region, reefs in JBTI are dominated by Calcarina 2010) were only found in zone 3, but always entangled in Sargassum
species that tolerate poor water quality. Furthermore, taxa that tolerate plants and at inhabited islands with perhaps locally elevated nutri-
algal rather than rubble substrates dominated as well. In this study, we ent levels. The species B. pennata, which can be classified as a turf
confirm that water quality appears to be an important driver of LBF alga was, in contrast, restricted to inshore islands and may be consid-
assemblage composition. ered a eutrophication indicator species. Other taxa that were com-
Very few macroalgae were recorded in the highly perturbed inshore mon in inshore islands consisted of taxa belonging to species-rich
reefs of Jakarta Bay. Eutrophication tends to shift the dominant primary genera that were also found mid- and offshore, e.g., Dictyota,
producers in coastal marine environments from slow growing Lobophora and Padina.
Fig. 9. Variation in site scores of the first PCO axis on a map of the study area for a) ascidians, b) molluscs, c) LBF and d) algae. Open symbols indicate positive values along the first axis and
shaded symbols negative values. The size of the symbol is proportional to the score. Codes for the midshore sites closest to shore are given: LaB, Bok, Dap and DaB. For molluscs additional
extra codes are shown for other mid- and offshores sites: PaS, PaB, PaK, SeD, Put and SeB.
In contrast to the other taxa, habitat variables were the most impor- offshore sites. Reef sites close to the islands of Pari (PaS) and Semak
tant predictors of echinoderms (14% versus 8%) and molluscs (9% versus Daun (SeD), for example, both contained assemblages of echinoderms
8%). Both of these groups, however, only consisted of a subset of all taxa and molluscs resembling those of inshore reefs. Both of these islands
and are not truly representative of the full taxon diversity. Nevertheless, also have areas of aquaculture in the surrounding water (pers. obs.
in French Polynesia, Adjeroud (1997) found that encrusting coralline DFR Cleary). It is possible that both of these groups respond to the
algal and macroalgal cover, depth and turbidity were significant predic- local perturbation related to the aquaculture activities.
tors of mollusc composition. In our study water transparency, CDOM, The purely spatial component was an important explanatory com-
sand and rubble cover were significant predictors of variation in mollusc ponent (N 10%) for corals, fishes, sponges and LBF. It should, however,
composition. Adjeroud (1997) also found that conductivity, organic be noted that despite the importance of this component, the explained
carbon, exposure and carbonates were significant predictors of echino- variation may be related to unmeasured spatially structured environ-
derm composition. In our study, significant predictors were CDOM, free- mental parameters. Sponges are an important component of coral
living mushroom and branching coral cover, sand cover and SST. In reefs. Sponges compete with corals and other benthic taxa for available
Moorea, sea urchins and damselfishes interacted to determine coral re- substrate and are therefore important components of reef dynamics
generation with much higher density and diversity of post-disturbance (VanVeghel et al., 1996; Aerts and van Soest, 1997; de Voogd et al.,
corals in damselfish territories. Sea urchins in Moorea (D. setosum, 2004). The number of species encountered was highest offshore, where-
Echinometra mathaei and E. calamaris) in contrast caused widespread as considerably fewer species were recorded inshore, particularly with-
destruction of dead standing coral skeletons (Done et al., 1991). Al- in the inner bay. The inshore sites tended to be very shallow, and
though we only collected presence/absence data on echinoderm assem- seldom exceeded 10 m in depth. It has been shown that the diversity
blages, during sampling we did notice very high densities of sea urchin of coral reef-associated sponges increases with increasing depth
species, particularly D. setosum, in inshore reefs. It is likely that these ur- (Fromont et al., 2006; de Voogd et al., 2006). However, although coral
chins have contributed to the highly simplified habitat structure preva- cover was also adversely affected by disturbance including riverine
lent in inshore reefs due to their bioeroding activity. In addition to the transport of sediments and nutrients inshore in the Berau region (East
above, echinoderms and molluscs both deviated from the general pat- Borneo), these specific environmental conditions seemed to have a pos-
tern of clear compositional differences between inshore and mid- to itive effect on sponges, as they were more abundant, larger and diverse
Table 2
Results of stepwise model building for constrained ordination for corals, fishes, sponges, echinoderms, ascidians, molluscs, LBF and algae. Type: substrate: substrate predictors; water qual-
ity: water quality predictors; spatial: spatial predictors; variable: explanatory variable; AIC: Akaike information criterion; Perm: number of permutations.
Taxon Type Variable AIC F Perm P
Corals Substrate Turf algae cover −17.00 1.54 99 0.030

Rubble cover −16.91 1.63 99 0.010
Sand cover −12.39 6.12 99 0.010
Water quality CDOM −16.82 1.42 199 0.045
Temperature −16.54 1.65 99 0.010
Dissolved oxygen −16.53 1.66 99 0.010
Chlorophyll-a −16.47 1.71 99 0.010
Transparency −16.39 1.77 99 0.010
Spatial PCNM2 −16.16 1.34 299 0.057
PCNM5 −15.99 1.49 99 0.030
PCNM10 −15.60 1.83 99 0.010
PCNM1 −10.57 6.72 99 0.010
Fishes Substrate Rubble cover −27.97 2.15 99 0.020
Sand cover −22.89 7.57 99 0.010
Temperature −28.27 1.5 199 0.050
Dissolved oxygen −28.13 1.61 99 0.030
SST −27.45 2.17 99 0.020
Transparency −27.86 1.83 99 0.010
Chlorophyll-a −27.44 2.17 99 0.010
Spatial PCNM5 −26.23 1.47 399 0.050
PCNM10 −25.93 1.73 199 0.045
PCNM14 −25.94 1.72 99 0.010
PCNM1 −21.13 6.35 99 0.010
Sponges Substrate Encrusting coral cover −14.49 1.38 599 0.066
Sand cover −11.73 4.11 99 0.010
Water quality Transparency −14.98 1.96 99 0.010
CDOM −14.77 2.15 99 0.010
SST −14.66 2.25 99 0.010
PCNM9 −14.13 1.59 199 0.035
Spatial PCNM3 −14.1 1.62 99 0.03
PCNM2 −13.6 2.06 99 0.01
PCNM1 −11.13 4.36 99 0.01
Echinoderms Substrate Branching coral cover −19.54 1.81 199 0.060
Sand cover −19.44 1.88 99 0.045
Mushroom coral cover −18.90 2.37 99 0.020
Acropora −18.85 2.42 99 0.020
Water quality SST −19.07 1.76 199 0.030
CDOM −18.12 2.68 99 0.010
Spatial PCNM1 −17.26 4.61 99 0.010
Ascidians Substrate Foliose coral cover −5.04 1.53 299 0.030
Macroalgae cover −5.03 1.54 99 0.030
Sand cover −3.17 3.24 99 0.010
Water quality Salinity −5.912 1.57 199 0.030
Temperature −5.802 1.67 199 0.030
Chlorophyll-a −5.395 2.03 99 0.010
Spatial PCNM3 −5.425 1.35 999 0.083
PCNM1 −3.139 3.57 99 0.010
Molluscs Substrate Rubble cover −2.81 1.56 99 0.030
Sand cover −2.55 1.81 99 0.010
Water quality Transparency −2.95 1.55 99 0.010
CDOM −2.542 1.94 99 0.010
Spatial PCNM3 −2.373 1.43 99 0.03
PCNM1 −2.364 1.44 99 0.03
LBF Substrate Sand cover −24.95 3.52 99 0.020
Encrusting coral cover −23.84 4.68 99 0.020
Water quality Dissolved oxygen −23.84 2.12 599 0.062
Rrs_645 −23.32 2.56 199 0.05
Chlorophyll-a −21.43 4.23 99 0.03
Temperature −22.72 3.07 99 0.02
SST −22.03 3.69 99 0.02
Spatial PCNM2 −23.85 1.81 99 0.100
PCNM9 −23.77 1.88 99 0.097
PCNM5 −23.80 1.85 99 0.087
PCNM11 −22.90 2.61 19 0.050
PCNM1 −22.58 2.88 19 0.030
Algae Substrate Mushroom coral cover −16.63 1.74 99 0.020
Sand cover −15.19 3.14 99 0.010
Transparency −16.51 1.56 99 0.030
SST −16.49 1.58 99 0.030
Chlorophyll-a −16.3 1.74 99 0.020
Spatial PCNM2 −16.85 1.47 199 0.035
PCNM1 −13.7 4.62 99 0.01
inshore (de Voogd et al., 2009). As mentioned previously, the environ- adverse effects of excessive UV radiation, a phenomenon which has
ment of the inshore zone was characterised by high turbidity and very also been linked to coral bleaching (Gleason and Wellington, 1993).
low live coral cover with the dominant substrate composed of silt,
sand and rubble. The inshore reefs are probably degraded to such an ex- 5. Conclusion
tent that most sponge species cannot cope. There were, however, some
sponge species found in inshore sites including Amphimedon cf. Water quality, substrate and spatial variables were significant pre-
paraviridis, C. joubini, S. inconstans, Chalinula hooperi, D. arenaria and dictors of the variation in composition of a wide range of coral reef
S. ceylonensis. Spheciospongia inconstans is a massive cake forming taxa. There were, however, differences in the relative importance of
sponge, the early growth stages of which can excavate coral limestone. these variables with water quality variables more important predictors
The species may appear as thick encrustations and can also grow very of the compositional variation of corals, fishes, sponges, LBF, ascidians
well on soft bottom substrate. Also the Indo-Pacific bath sponge and algae whereas substrate variables were more important predictors
S. ceylonensis has a preference for soft bottom substrate, particularly in of echinoderm and mollusc composition. Remotely sensed water quality
shallow water environments. Other sponge species often associated variables also tended to be more important predictors of variation in
with perturbed environments such as the moon sponge Cinachyrella composition than locally measured variables. This highlights the impor-
australienis and Paratatetilla bacca were absent from the inshore sites. tance of these variables as predictors of both water quality and the com-
Sponge species associated with offshore sites included H. erectus, Ircinia position of coral reef taxa.
ramosa, S. carteri and C. cervicornis. These sponges are common through- Supplementary data to this article can be found online at http://dx.
out the Indonesian archipelago and usually occur across a wide range of doi.org/10.1016/j.marpolbul.2016.04.042.
environmental conditions (de Voogd et al., 2006, 2009).
In general, satellite-derived water quality variables outperformed Acknowledgements
locally measured water quality variables in line with a previous study
of a set of coral reef taxa in the Spermonde archipelago (Polónia et al., This publication is a result of the Thousand Islands Expedition, Sep-
2015). Chlorophyll-a and CDOM concentrations in particular proved im- tember 2005, organized by the Research Centre for Oceanography (PPO-
portant predictors. Chlorophyll-a concentration and/or transparency LIPI, Jakarta, Indonesia) and Naturalis Biodiversity Center (Leiden, the
were the most significant explanatory water quality variables of varia- Netherlands). The research by Naturalis was part of the project ‘Climate
tion in the composition of corals, fishes, ascidians and algae whereas change and Indonesian coral reef biotas’ within the theme ‘Biodiversity
CDOM was the most significant explanatory water quality variable of in Relation to Global Change’ of the Council for Earth and Life Sciences of
variation in the composition of echinoderms and molluscs (Table 2). the Netherlands Organisation for Scientific Research (ALW-NWO grant
SST was the most significant explanatory water quality variable of vari- 852.000.50). DFRC was supported by the Portuguese Foundation for Sci-
ation in the composition of sponges and LBF. Chlorophyll-a concentra- ence and Technology (FCT) under grant PTDC/AAC-AMB/115304/2009
tions were much higher in the waters surrounding inshore reefs than (LESS CORAL). WFP'hvR and SGAD were supported by NWO-WOTRO
waters surrounding mid- and offshore reefs (Appendices 4 and 5 in grant R 85-381. We are grateful to LIPI for granting research permits.
the online version at http://dx.doi.org/10.1016/j.marpolbul.2016.04. We thank staff of PPO-LIPI for logistical support.
042.). Chlorophyll-a is an important eutrophication indicator (Boyer
et al., 2009) and the elevated levels found inshore highlight the dilapi-
References
dated state of the Jakarta Bay environment.
The importance of CDOM as a predictor of variation in composition is Adjeroud, M., 1997. Long-term changes of epibenthic macrofauna communities in
inline with previous studies that have pointed to dissolved organic mat- a closed lagoon (Taiaro Atoll, French Polynesia): 1972-1994. Hydrobiologia 356
(1–3), 11–19.
ter as an important factor in structuring coastal communities (Ayoub Aerts, L.A.M., van Soest, R.W.M., 1997. Quantification of sponge/coral interactions in
et al., 2009; Baña et al., 2014). High chlorophyll-a concentrations also a physically stressed reef community, NE Colombia. Mar. Ecol. Prog. Ser. 148,
appear to enhance dissolved organic matter fluxes, which provide an 125–134.
Ayoub, L., Hallock, P., Coble, P., 2009. Colored dissolved organic material increases resil-
important source of food for LBF and corals (Renema and Troelstra,
iency of coral reefs by controlling exposure to UVR. Proc 11th Int Coral Reef Symp
2001; Eberwein and Mackensen, 2006; Houlbreque and Ferrier-Pagès, 1, pp. 579–583.
2009). Our study, however, contrasts with previous studies Ayoub, L.M., Hallock, P., Coble, P.G., Bell, S.S., 2012. MAA-like absorbing substances in
Florida Keys phytoplankton vary with distance from shore and CDOM: implications
(e.g., Polónia et al., 2015) where chlorophyll-a and CDOM concentra-
for coral reefs. J. Exp. Mar. Biol. Ecol. 420, 91–98.
tions were higher inshore than offshore. In contrast, the highest Baird, A.H., Pratchett, M.S., Hoey, A.S., Herdiana, Y., Campbell, S.J., 2013. Acanthaster planci
CDOM concentrations in our study were measured offshore as can be is a major cause of coral mortality in Indonesia. Coral Reefs 32, 803–812.
seen in the association of CDOM with offshore sites (Supplementary Baña, Z., Ayo, B., Marrasé, C., Gasol, J.M., Iriberri, J., 2014. Changes in bacterial metab-
olism as a response to dissolved organic matter modification during protozoan
Figs. 1 and 2 in the online version at http://dx.doi.org/10.1016/j. grazing in coastal Cantabrian and Mediterranean waters. Environ. Microbiol.
marpolbul.2016.04.042.). However, this anomaly is probably related to 16, 498–511.
the particular environment of Jakarta Bay and the inshore reefs. High Barile, P.J., 2004. Evidence of anthropogenic nitrogen enrichment of the littoral waters of
East Central Florida. J. Coast. Res. 20, 1237–1245.
CDOM concentrations are normally found in inshore coastal areas Baum, G., Januar, H.I., Ferse, S.C.A., Kunzmann, A., 2015. Local and regional impacts of pol-
with intact shorelines, i.e., relatively undisturbed coastlines with still in- lution on coral reefs along the Thousand Islands north of the megacity Jakarta,
tact coastal mangroves, forests and seagrass beds (Ayoub et al., 2009, Indonesia. PLoS ONE 10, e0138271.
Becking, L.E., Cleary, D.F.R., de Voogd, N.J., Renema, W., de Beer, M., van Soest, R.W.M.,
2012). Destruction of coastal forests and wetlands has, however, led Hoeksema, B.W., 2006. Beta-diversity of tropical marine assemblages in the
to a reduction in natural sources of CDOM to coastal waters. The Spermonde Archipelago, Indonesia. Mar. Ecol. 27, 76–88.
coastal environment of Jakarta Bay is highly modified, consisting Bellwood, D.R., Hughes, T.P., Folke, C., Nyström, M., 2004. Confronting the coral reef crisis.
Nature 429, 827–833.
mainly of built-up and agricultural land. Most of the original forests
Berumen, M.L., Pratchett, M.S., 2006. Recovery without resilience: persistent disturbance
and mangroves have been destroyed, which can be seen as the and long-term shifts in the structure of fish and coral communities at Tiahura Reef,
cause of the low CDOM values inshore. A close examination of re- Moorea. Coral Reefs 25, 647–653.
Borcard, D., Legendre, P., 2002. All-scale spatial analysis of ecological data by
mote sensing images of CDOM plumes offshore appeared to show
means of principal coordinates of neighbour matrices. Ecol. Model. 153,
that they originated from the still intact coastline of Borneo (pers. 51–68.
obs. ARM Polónia). Importantly, CDOM has been shown to protect Boyer, J.N., Kelble, C.R., Ortner, P.B., Rudnick, D.T., 2009. Phytoplankton bloom status:
marine organisms from ultraviolet (UV) radiation and shorter wave- chlorophyll a biomass as an indicator of water quality condition in the southern estu-
aries of Florida, USA. Ecol. Indic. 9, S56eS67.
lengths of PAR (Shank et al., 2010; Ayoub et al., 2012). The coastal Brown, B.E., Suharsono, 1990. Damage and recovery of coral reefs affected by El Niño re-
forests of Borneo may thus be protecting JBTI coral reefs from the lated seawater warming in the Thousand Islands, Indonesia. Coral Reefs 8, 163–170.
Bruno, J.F., Precht, W.F., Vroom, P.S., Aronson, R.B., 2014. Coral reef baselines: how much Hoeksema, B.W., 1991. Control of bleaching in mushroom coral populations (Scleractinia:
macroalgae is natural? Mar. Pollut. Bull. 80, 24–29. Fungiidae) in the Java Sea: stress tolerance and interference by life history strategy.
Burkepile, D.E., Allgeier, J.E., Shantz, A.A., Pritchard, C.E., Lemoine, N.P., Bhatti, L.H., Mar. Ecol. Prog. Ser. 74, 225–237.
Layman, C.A., 2013. Nutrient supply from fishes facilitates macroalgae and sup- Hoeksema, B.W., 2012a. Distribution patterns of mushroom corals (Scleractinia:
presses corals in a Caribbean coral reef ecosystem. Sci. Rep. 3, 1493. Fungiidae) across the Spermonde Shelf, South Sulawesi. Raffles Bull. Zool. 60, 183–212.
Carder, K.L., Chen, F.R., Lee, Z.P., Hawes, S., Kamykowski, D., 1999. Semi-analytic MODIS Hoeksema, B.W., 2012b. Evolutionary trends in onshore–offshore distribution patterns of
algorithms for chlorophyll a and absorption with bio-optical domains based on mushroom coral species (Scleractinia: Fungiidae). Contrib. Zool. 81, 199–221.
nitrate-depletion temperatures. J. Geophys. Res. 104, 5403–5421. Hoey, A.S., Bellwood, D.R., 2008. Cross-shelf variation in the role of parrotfishes on the
Carpenter, K.E., Abrar, M., Aeby, G., Aronson, R.B., Banks, S., Bruckner, A., et al., 2008. One- Great Barrier Reef. Coral Reefs 27, 37–47.
third of reef-building coral face elevated extinction risk from climate change and local Hohenegger, J., 1994. Distribution of living larger Foraminifera NW of Sesoko-Jima, Okina-
impacts. Science 321, 560–563. wa Japan. PSZN I Mar. Ecol. 15, 291–334.
Chen, K.S., Hsieh, H.J., Keshavmurthy, S., Leung, J.K.L., Lien, I.T., et al., 2011. Latitudinal gra- Hosono, T., Su, C.C., Delinom, R., Umezawa, Y., Toyota, T., Kaneko, S., Taniguchi, M., 2011.
dient of morphological variations in Zebra Coral Oulastrea crispata (Scleractinia: Decline in heavy metal contamination in marine sediments in Jakarta Bay, Indonesia
Faviidae) in the West Pacific. Zool. Stud. 50, 43–52. due to increasing environmental regulations. Estuar. Coast. Shelf Sci. 92, 297–306.
Chen, Z., Hu, C., Muller-Karger, F., 2007. Monitoring turbidity in Tampa Bay using MODIS/ Houlbreque, F., Ferrier-Pagès, C., 2009. Heterotrophy in tropical scleractinian corals. Biol.
Aqua 250-M imagery. Remote Sens. Environ. 109, 207–220. Rev. 84, 1–17.
Cleary, D.F.R., Suharsono, Hoeksema, B.W., 2006. Coral diversity across a disturbance gra- Hughes, T.P., Graham, N.A., Jackson, J.B., Mumby, P.J., Steneck, R.S., 2010. Rising to the
dient in the Pulau Seribu reef complex off Jakarta, Indonesia. Biodivers. Conserv. 15, challenge of sustaining coral reef resilience. Trends Ecol. Evol. 25, 633–642.
3653–3674. Jackson, J.B.C., Buss, L.W., 1975. Allelopathy and spatial competition among coral reef in-
Cleary, D.F.R., Becking, L.E., de Voogd, N.J., Renema, W., de Beer, M., van Soest, R.W.M., vertebrates. Proc. Natl. Acad. Sci. U. S. A. 72, 5160–5516.
Hoeksema, B.W., 2005. Variation in the diversity and composition of benthic taxa as Johan, O., Bengen, D.G., Zamani, N.P., Sweet, M.J., 2015. The distribution and abundance of
a function of distance offshore, depth and exposure in the Spermonde Archipelago, Black Band Disease and White Syndrome in Kepulauan Seribu, Indonesia. Hayati
Indonesia. Estuar. Coast. Shelf Sci. 65, 557–570. J. Biosci. 22, 105–112.
Cleary, D.F.R., DeVantier, L., Giyanto, V.L., Manto, P., de Voogd, N.J., Rachello-Dolmen, P.G., Lecchini, D., Adjeroud, M., Pratchett, M.S., Cadoret, L., Galzin, R., 2003. Spatial structure of
Tuti, Y., Budiyanto, A., Wolstenholme, J., Hoeksema, B.W., Suharsono, 2008. Relating coral reef fish communities in the Ryukyu Islands, southern Japan. Oceanol. Acta 26,
variation in species composition to environmental variables, a multi-taxon study in 537–547.
an Indonesian coral reef complex. Aquat. Sci. 70, 419–431. Legendre, P., Gallagher, E., 2001. Ecologically meaningful transformations for ordination
Cleary, D.F.R., Mooers, A.Ø., Eichhorn, K.A.O., van Tol, J., de Jong, R., Menken, S.B.J., 2004. of species data. Oecologia 129, 271–280.
Diversity and community composition of butterflies and odonates in an ENSO- Lewis, A.R., 1997. Effects of experimental coral disturbance on the structure of fish com-
induced fire affected habitat mosaic: a case study from East Kalimantan, Indonesia. munities on large patch reefs. Mar. Ecol. Prog. Ser. 161, 37–50.
Oikos 105, 426–446. Lirman, D., 2014. Competition between macroalgae and corals: effects of herbivore exclu-
Cleary, D.F.R., Polónia, A.R.M., Renema, W., Hoeksema, B.W., Wolstenholme, J., Tuti, Y., de sion and increased algal biomass on coral survivorship and growth. Coral Reefs 19,
Voogd, N.J., 2014. Coral reefs next to a major conurbation: a study of temporal change 392–399.
(1985–2011) in coral cover and composition in the reefs of Jakarta, Indonesia. Mar. Littler, M.M., Littler, D.S., 1984. Models of tropical reef biogenesis: the contribution of
Ecol. Prog. Ser. 501, 89–98. macroalgae. In: Round, F.E., Chapman, D.J. (Eds.), Progress in Phycological Research
DeVantier, L., Suharsono, Budiyanto, A., Tuti, Y., Imanto, P., Ledesma, R., 1998. Status of 3. Biopress, Bristol, England, pp. 232–364.
coral communities of Pulau Seribu (Indonesia). In: Soemodihardjo, S. (Ed.), Lubchenco, J., Gaines, S.D., 1981. A unified approach to marine plant herbivore interac-
Contending with global change 10Proceedings, Coral Reef Evaluation Workshop, tions. I. Populations and communities. Annu. Rev. Ecol. Evol. Syst. 2, 405–437.
Pulau Seribu, Jakarta, Indonesia, 1995. UNESCO, Jakarta, pp. 1–24. Maduppa, H.H., Subhan, B., Suparyani, E., Siregar, A.M., Arafat, D., Tarigan, S.A., Alimuddin,
Dickens, L.C., Goatley, C.H.R., Tanner, J.K., Bellwood, D.R., 2011. Quantifying relative diver K.D., Rahmawati, F., Bramandito, A., 2013. Dynamics of fish diversity across an
effects in underwater visual censuses. PLoS ONE 6 (4), e18965. environmental gradient in the Seribu Islands reefs off Jakarta. Biodiversitas 14,
Dominici-Arosemena, A., Wolff, M., 2005. Reef fish community structure in Bocas del Toro 17–24.
(Caribbean, Panama): gradients in habitat complexity and exposure. Caribb. J. Sci. 41, Makarenkov, V., Legendre, P., 2002. Nonlinear redundancy analysis and canonical corre-
613–637. spondence analysis based on polynomial regression. Ecology 834, 1146–1161.
Done, T.J., Dayton, P.K., Dayton, A.E., Steger, R., 1991. Regional and local variability in re- Martin, S., 2004. An Introduction to Ocean Remote Sensing. Cambridge University Press.
covery of shallow coral communities — Moorea, French-Polynesia and central McClanahan, T.R., 1997. Primary succession of coral-reef algae: differing patterns on
Great-Barrier-Reef. Coral Reefs 9, 183–192. fished versus unfished reefs. J. Exp. Mar. Biol. Ecol. 218 (1), 77–102.
Duarte, C.M., 1995. Submerged aquatic vegetation in relation to different nutrient re- van der Meij, S.E.T., Moolenbeek, R.G., Hoeksema, B.W., 2009. Decline of the Jakarta Bay
gimes. Ophelia 41, 87–112. molluscan fauna linked to human impact. Mar. Pollut. Bull. 59, 101–107.
Eberwein, A., Mackensen, A., 2006. Regional primary productivity differences off Morocco van der Meij, S.E.T., Suharsono, Hoeksema, B.W., 2010. Long-term changes in coral assem-
(NW-Africa) recorded by modern benthic foraminifera and their stable carbon isoto- blages under natural and anthropogenic stress in Jakarta Bay (1920–2005). Mar.
pic composition. Deep-Sea Res. I Oceanogr. Res. Pap. 53, 1379–1405. Pollut. Bull. 60, 1442–1454.
Emslie, M.J., Cheal, A.J., Sweatman, H., Delean, S., 2008. Recovery from disturbance of coral Menge, B.A., Farrell, T.M., 1989. Community structure and interaction webs in shallow
and reef fish communities on the Great Barrier Reef, Australia. Mar. Ecol. Prog. Ser. marine hard-bottom communities: tests of an environmental stress model. Adv.
371, 177–190. http://dx.doi.org/10.3354/meps07657. Ecol. Res. 19, 189–262.
English, S., Wilkinson, C., Baker, V., 1997. Survey Manual for Tropical Marine Resources. Menge, B.A., Lubchenco, J., 1981. Community organization in temperate and tropical
second ed. Australian Institute of Marine Science, Townsville (390 pp.). rocky intertidal habitats: prey refuges in relation to consumer pressure gradients.
van der Ent, E., Hoeksema, B.W., de Voogd, N.J., 2015. Abundance and genetic varia- Ecol. Monogr. 51, 429–450.
tion of the coral-killing cyanobacteriosponge Terpios hoshinota in the Spermonde Miller, R., Mckee, B., 2004. Using MODIS Terra 250 m imagery to map concentrations of
Archipelago, SW Sulawesi, Indonesia. J. Mar. Biol. Assoc. UK http://dx.doi.org/10. total suspended matter in coastal waters. Remote Sens. Environ. 93, 259–266.
1017/S002531541500034X. Moll, H., Suharsono, 1986. Distribution, diversity and abundance of reef corals in Jakarta
Farhan, A.R., Lim, S., 2012. Vulnerability assessment of ecological conditions in Seribu Bay and Kepulauan Seribu. UNESCO Rep. Mar. Sci. 40, 112–125.
Islands, Indonesia. Ocean Coast. Manag. 65, 1–14. Mora, C., Chittaro, P.M., Sale, P.F., Kritzer, J.P., Ludsin, S.A., 2003. Patterns and processes in
Fox, R.J., Bellwood, D.R., 2007. Quantifying herbivory across a coral reef depth gradient. reef fish diversity. Nature 421, 933–936.
Mar. Ecol. Prog. Ser. 339, 49–59. Oksanen, J., Kindt, R., Legendre, P., O'hara, B., Simpson, G.L., Solymos, P., Stevens, M., Wagner,
Fromont, J., Vanderklift, M.A., Kendrick, G.A., 2006. Marine sponges of the Dampier Archi- H., 2009. Vegan: Community Ecology Package. R Package Version 1.15–2. 2009.
pelago, Western Australia: patterns of species distributions, abundance and diversity. Patt, F., Barnes, R., Eplee Jr., R., Franz, B., Robinson, W., Feldman, G., Bailey, S., Gales, J.,
Biodivers. Conserv. 15, 3731–3750. Werdell, P., Wang, M., et al., 2003. Algorithm updates for the fourth SeaWIFS data
Gittenberger, A., Draisma, S.G.A., Arbi, U., Langenberg, V., Erftemeijer, P., Tuti, Y., Hoeksema, reprocessing. NASA Tech Memo 206892, pp. 22–74.
B.W., 2014. Coral reef organisms as bioregion indicators off Halmahera, Moluccas, Plass-Johnson, J.G., Schwieder, H., Heiden, J., Weiand, L., Wild, C., Jompa, J., Ferse, S.C.A.,
Indonesia. Aquat. Conserv. Mar. Freshwat. Ecosyst. http://dx.doi.org/10.1002/aqc.2495. Teichberg, M., 2015. A recent outbreak of crown-of-thorns starfish (Acanthaster planci)
Gleason, D.F., Wellington, G.M., 1993. Ultraviolet radiation and coral bleaching. Nature in the Spermonde Archipelago, Indonesia. Reg. Environ. Chang. 15, 1157–1162.
365, 836–838. http://dx.doi.org/10.1038/365836a0. Polónia, A.R.M., Cleary, D.F.R., de Voogd, N.J., Renema, W., Hoeksema, B.W., Martins, A.,
Gleeson, S.K., Wilson, D.S., 1986. Equilibrium diet: optimal foraging and prey coexistence. Gomes, N.C.M., 2015. Importance of space and environment to explain the variation
Oikos 46, 139–144. in the composition of coral reef taxa in the Spermonde Archipelago, Indonesia: a
Goatley, C.H., Bellwood, D.R., 2013. Ecological consequences of sediment on high-energy multitaxon study. Sci. Total Environ. 537, 139–151.
coral reefs. PLoS One 8, e77737. Pratchett, M.S., Caballes, C.F., Rivera-Posada, J.A., Sweatman, H.P.A., 2014. Limits to under-
Hallock, P., Lidz, B.H., Cockey-Burkhard, E.M., Donnelly, K.B., 2003. Foraminifera as standing and managing outbreaks of crown-of-thorns starfish (Acanthaster spp.).
bioindicators in coral reef assessment and monitoring: the FORAM Index. Environ. Oceanogr. Mar. Biol. Annu. Rev. 52, 33–200.
Monit. Assess. 81, 221–238. R Core Team, 2013. R: A Language and Environment for Statistical Computing. (Vienna,
Hay, M.E., 1981a. The functional morphology of turf-forming seaweeds persistence in Austria. http://www.R-project.org).
stressful marine habitats. Ecology 62, 739–750. Rachello-Dolmen, P.G., Cleary, D.F.R., 2007. Relating coral species traits to environmental
Hay, M.E., 1981b. Herbivory, algal distribution, and the maintenance of between habitat conditions in the Jakarta Bay/Pulau Seribu reef system, Indonesia. Estuar. Coast. Shelf
diversity on a tropical fringing reef. Am. Nat. 118, 520–540. Sci. 73, 816–826.
Hay, M.E., 1981c. Spatial patterns of grazing intensity on a Caribbean barrier reef: herbiv- Raffaelli, D.G., Raven, J.A., Poole, L.J., 1998. Ecological impacts of green macroalgal blooms.
ory and algal distribution. Aquat. Bot. 11, 97–109. Oceanogr. Mar. Biol. Annu. Rev. 36, 97–125.
Rees, J.G., Setiapermana, D., Sharp, V.A., Weeks, J.M., Williams, T.M., 1999. Evaluation of spp. Glob. Chang. Biol. 16, 2624–2637. http://dx.doi.org/10.1111/j.1365-2486.2009.
the impacts of land-based contaminants on the benthic faunas of Jakarta Bay, 02108.x.
Indonesia. Oceanol. Acta 22, 627–640. Tuomisto, H., Ruokolainen, K., Yli-Halla, M., 2003. Dispersal, environment, and floristic
Renema, W., 2006. Large benthic foraminifera from the deep photic zone of a mixed variation of western Amazonian forests. Science 299, 241–244.
siliciclastic-carbonate shelf off East Kalimantan, Indonesia. Mar. Micropaleontol. 58, Umbgrove, J.H.F., 1939. Madreporaria from the bay of Batavia. Zool. Meded. 22, 1–64.
73–82. Uneputty, P.A., Evans, S.M., 1997. Accumulation of beach litter on islands of the Pulau
Renema, W., 2008. Habitat selective factors influencing the distribution of larger benthic Seribu Archipelago, Indonesia. Mar. Pollut. Bull. 34, 652–655.
foraminiferal assemblages over the Kepulauan Seribu. Mar. Micropaleontol. 68, Uthicke, S., Nobes, K., 2008. Benthic Foraminifera as ecological indicators for water quality
286–298. on the Great Barrier Reef. Estuar. Coast. Shelf Sci. 78, 763–773.
Renema, W., 2010. Is increased calcarinid (foraminifera) abundance indicating a larger role VanVeghel, M.L.J., Cleary, D.F.R., Bak, R.P.M., 1996. Interspecific interactions and compet-
for macro-algae in Indonesian Plio-Pleistocene coral reefs? Coral Reefs 29, 165–173. itive ability of the polymorphic reef-building coral Montastrea annularis. Bull. Mar.
Renema, W., Troelstra, S., 2001. Larger foraminifera distribution on a mesotrophic carbon- Sci. 58, 792–803.
ate shelf in SW Sulawesi (Indonesia). Palaeogeogr. Palaeoclimatol. Palaeoecol. 175, Veron, J.E.N., 2000. Corals of the World. vol. 1–3. Australian Institute of Marine Science,
125–146. Townsville.
Richardson, L.L., LeDrew, E.F., 2006. Remote Sensing of Aquatic Coastal Ecosystem Pro- Verstappen, H.T., 1988. Old and new observations on coastal changes of Jakarta Bay, an
cesses. Science and Management Applications. 9. Springer, Netherlands. http://dx. example of trends in urban stress on coastal environments. J. Coast. Res. 4, 573–587.
doi.org/10.1007/1-4020-3968-9. de Voogd, N.J., Cleary, D.F.R., 2008. An analysis of sponge diversity and distribution at
Rinawati, K.T., Koike, H., Kurumisawa, R., Ito, M., Sakurai, S., Togo, A., Saha, M., Arifin, Z., three taxonomic levels in the Thousand Islands/Jakarta bay reef complex, West-
Takada, H., 2012. Distribution, source identification, and historical trends of organic Java, Indonesia. Mar. Ecol. 29, 205–215.
micropollutants in coastal sediment in Jakarta Bay, Indonesia. J. Hazard. Mater. 217, de Voogd, N.J., Becking, L.E., Cleary, D.F.R., 2009. Sponge community composition in the
208–216. Derawan Islands, NE Kalimantan, Indonesia. Mar. Ecol. Prog. Ser. 396, 219–230.
Roberts, C.M., McClean, C.J., Veron, J.E.N., Hawkins, J.P., Allen, G.R., McAllister, D.E., de Voogd, N.J., Becking, L.E., Noor, A., Hoeksema, B.W., van Soest, R.M.W., 2004. Sponge in-
Mittermeier, C.G., Schueler, F.W., Spalding, M., Wells, F., Vynne, C., Werner, T.B., teractions with spatial competitors in SW Sulawesi. In: Pansini, M., Pronzato, R.,
2002. Marine biodiversity hotspots and conservation priorities for tropical reef. Sci- Bavestrello, G., Manconi, R. (Eds.), Sponge science in the new millennium. Boll Mus
ence 295, 1280–1284. http://dx.doi.org/10.1126/science.1067728. Inst Biol Univ Gen 68, pp. 253–261.
Ross, S.W., Dalton, D.A., Kramer, S., Christensen, B.L., 2001. Physiological (antioxidant) re- de Voogd, N.J., Cleary, D.F.R., Dekker, F., 2013. The coral-killing sponge Terpios hoshinota
sponses of estuarine fishes to variability in dissolved oxygen. Comp. Biochem. Physiol. invades Indonesia. Coral Reefs 32, 755.
C Toxicol. Pharmacol. 130, 289–303. de Voogd, N.J., Cleary, D.F.R., Hoeksema, B.W., Noor, A., van Soest, R.W.M., 2006. Sponge
Schaffelke, B., Mellors, J., Duke, N.C., 2005. Water quality in the Great Barrier Reef region: beta diversity in the Spermonde Archipelago, SW Sulawesi, Indonesia. Mar. Ecol.
responses of mangrove, seagrass and macroalgal communities. Mar. Pollut. Bull. 51, Prog. Ser. 309, 131–142.
279–296. Walther, G.-R., Post, E., Convey, P., Menzel, A., Parmesan, C., Beebee, T.J.C., Fromentin, J.-M.,
Scoffin, T.P., Tudhope, A.W., 1985. Sedimentary environments of the central region of the Hoegh-Guldberg, O., Bairlein, F., 2002. Ecological responses to recent climate change.
Great Barrier Reef of Australia. Coral Reefs 4, 81–93. Nature 416, 389–395.
Shank, G.C., Zepp, R.G., Vahatalo, A., Lee, R., Bartels, E., 2010. Photobleaching kinetics of Williams, T.M., Rees, J.G., Setiapermana, D., 2000. Metals and trace organic compounds in
chromophoric dissolved organic matter derived from mangrove leaf litter and float- sediments and waters of Jakarta Bay and the Pulau Seribu Complex, Indonesia. Mar.
ing sargassum colonies. Mar. Chem. 119, 162–171. Pollut. Bull. 40, 277–285.
Syms, C., Jones, G.P., 2000. Disturbance, habitat structure, and the dynamics of a coral-reef Willoughby, N.G., 1986. Man-made litter on the shores of the Thousand Island archipela-
fish community. Ecology 81, 2714–2729. go, Java. Mar. Pollut. Bull. 17, 224–228.
Teichberg, M., FOX, S.E., OlsenL, Y.S., Valiela, I., Martinetto, P., Iribarne, O., Muto, E.Y., Petti, Willoughby, N.G., Sangkoyo, H., Lakaseru, B.O., 1997. Beach litter: an increasing and
M.A.V., Corbisier, T.N., Soto-Jiménez, M., Páez-Osuna, F., Castro, P., Freitas, H., Zitelli, changing problem for Indonesia. Mar. Pollut. Bull. 34, 469–478.
A., Cardinaletti, M., Tagliapietra, D., 2010. Eutrophication and macroalgal blooms in Yamashiro, H., 2000. Variation and plasticity of skeletal color in the Zebra Coral Oulastrea
temperate and tropical coastal waters: nutrient enrichment experiments with Ulva crispata. Zool. Sci. 17, 827–831.

37

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

37

Uploaded by

Copyright:

Available Formats

Marine Pollution Bulletin 110 (2016) 701–717

Contents lists available at ScienceDirect

Marine Pollution Bulletin

journal homepage: www.elsevier.com/locate/marpolbul

Variation in the composition of corals, ﬁshes, sponges, echinoderms,

1. Introduction herbivory as top-down mechanism and an increasing supply of nu-

AyB Ayer Besar Island 1 106.775 −6.001 33 90 – – 8 2 7 7 6 4 8

Taxon Type Variable AIC F Perm P

Corals Substrate Turf algae cover −17.00 1.54 99 0.030

You might also like