Marine Pollution Bulletin 122 (2017) 259–262

Contents lists available at ScienceDirect

Marine Pollution Bulletin

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

Linking fish species traits to environmental conditions in the Jakarta Bay- MARK
Pulau Seribu coral reef system
Daniel F.R. Cleary
Department of Biology, CESAM, Universidade de Aveiro, Campus Universitário de Santiago, 3810-193 Aveiro, Portugal

A R T I C L E I N F O A B S T R A C T

Keywords: Coral reefs around the globe have been subjected to a wide range of stressors. In the present study, fish species
Indonesia were recorded across a pronounced in-to-offshore gradient in the Jakarta Bay-Pulau Seribu reef system. In ad-
Java dition to this, fish species traits were obtained from FishBase. RLQ analysis revealed a significant association
Ordination between fish species traits and environmental variables. Fish species associated with perturbed, inshore waters
RLQ
were resilient to disturbance, had higher mortality rates, higher growth rates and mainly consumed animals. In
contrast, fish species associated with less perturbed, mid- and offshore waters had greater life expectancy, higher
age at maturity, greater life span, greater generation time and mainly fed on plants or plants and animals.
Eutrophication, pollution and physical destruction of coral substrate in inshore waters has thus selected for a low
biomass and depauperate fish community characterised by fast growing and short lived species.

Coral reefs are one of the most diverse marine ecosystems. They also conditions in JBTI. Inshore, JBTI has been shown to have poor to very
provide a wide range of ecosystem goods and services including coastal poor water transparency, heavy nutrient loads and depauperate coral
protection, tourism and food (Rees et al., 1999; Hughes et al., 2003; assemblages. Offshore, coral cover and diversity is much higher and
Green and Bellwood, 2009). Despite this, coral reefs have been subject water transparency and quality better than inshore (DeVantier et al.,
to substantial unsustainable exploitation including overfishing and 1998; Cleary et al., 2008, 2014). Mean annual precipitation is
coral mining. Their proximity to the coast has also subjected them to 1700 mm yr− 1 with a ‘wet’ season during the northwest monsoon
land-based pollution including untreated sewage, agricultural and in- (November–March) and a ‘dry’ season during the southeast monsoon
dustrial runoff and increased sedimentation following the loss of coastal (May–September) (Rees et al., 1999).
forests and mangroves (Jackson et al., 2001; Williams et al., 2000; In the present study, fish species were recorded and habitat and
Green and Bellwood, 2009; Farhan and Lim, 2012). Inshore waters in water quality parameters measured in order to link fish species traits to
particular have over time become eutrophied, which has led to poor environmental conditions using RLQ analysis, a three-table ordination
water visibility and quality. method that directly relates species traits to environmental attributes
An area that exemplifies all the problems faced by coral reef systems (Dolédec et al., 1996). Scores of species in RLQ axes can, furthermore,
is the Jakarta Bay-Pulau Seribu system (hereafter referred to as JBTI). be used to define functional groups, which can aid conservation man-
The reefs are relatively well studied (Cleary et al., 2006; Rachello- agement. Changes in disturbance regimes can thus be directly related to
Dolmen and Cleary, 2007; Cleary et al., 2014; Cleary et al., 2016) and a changes in species traits instead of being indirectly related via taxo-
prime example of a still intact albeit perturbed coral reef system ad- nomic composition (Ribera et al., 2001).
jacent to a megacity, namely Jakarta, the capital of Indonesia. Jakarta One can ask what traits species of disturbed inshore reefs share and
is a major conurbation housing more than > 26 million inhabitants what traits species of the less disturbed offshore reefs share. Identifying
(Demographia, 2012). Sewage and storm water are transported over a the sets of traits associated with both types of reefs can help us to un-
2000 km2 catchment area to the bay, which is defined by two flanking derstand how disturbance affects the coral reef fish community. If
delta systems with considerable sediment input (Rees et al., 1999; species at multiple sites are affected by the same perturbation then
Williams et al., 2000; Rinawati et al. 2012). River discharge in JBTI species that survive should possess traits promoting persistence
contains (excessive) nutrients, pollutants, heavy metals and household whereas species that disappear or decline in abundance should possess
litter (Willoughby, 1986; Uneputty and Evans, 1997; Willoughby et al., traits that promote susceptibility. The identification of these traits can
1997; Hosono et al., 2011; Baum et al., 2015). Previous studies have help in identifying how the perturbation affects communities and can
reported on the pronounced in-to-offshore gradient in environmental be used to develop effective conservation strategies (Lips et al., 2003).

E-mail addresses: cleary@ua.pt, dfrcleary@gmail.com.

http://dx.doi.org/10.1016/j.marpolbul.2017.06.054
Received 8 December 2016; Received in revised form 19 June 2017; Accepted 19 June 2017
Available online 26 June 2017
0025-326X/ © 2017 Elsevier Ltd. All rights reserved.
D.F.R. Cleary Marine Pollution Bulletin 122 (2017) 259–262

In particular, understanding the relationship between sets of traits and variable defined as high, medium, low and very low in response to
environmental conditions can help us to understand which traits are disturbances. Data for certain variables were not available for all spe-
associated with increased sensitivity to disturbance (Ribera et al., cies. Generic mean values were then used instead in order to avoid
2001). The aim of the present study was to test for significant asso- missing values.
ciations between fish species traits and environmental variables in re- Species traits were directly linked to environmental variables with a
lation to the in-to-offshore disturbance gradient in JBTI. three-table ordination method known as RLQ analysis, (Dolédec et al.,
Sampling for the present study occurred from September 7th – 23rd 1996; Ribera et al., 2001) using the ADE4 software package (http://
2005 mainly on the northwestern sides of surveyed reefs (see Cleary pbil.univ-lyon1.fr/ADE-4/) within R (http://www.r-project.org/). See
et al., 2016 for a map and table of the reefs sampled). Fishes were vi- Rachello-Dolmen and Cleary (2007) and Cleary et al. (2007) for de-
sually assessed by scuba diving along six transects (30 m long) at each tailed descriptions of the method. In summary, R is a table of en-
of two studied depths (3 and 5 m). Individuals observed within 5 m on vironmental variables at a set of sites, L is a table of the species com-
either side of the transect were identified to species, if possible, and position at corresponding sites and Q is a table of traits for the species in
recorded. Individuals that left the transect area and re-entered were not table L. Prior to RLQ analysis, three separate tables were constructed
counted again (Hoey and Bellwood, 2008; Dickens et al., 2011). In using correspondence analysis (CA) for species composition and prin-
total, 28 sites were surveyed. For the purposes of this study, the two cipal components analysis (PCA) for environmental and traits variables.
depth transects were pooled per site. During sampling, the cover of R was subsequently linked with L using row weights (site scores) ob-
coral and other benthic cover categories was estimated along each tained with CA and Q was linked with L using column weights (species
transect using the line intercept transect survey method (English et al., scores) obtained with CA. RLQ combines these three analyses by max-
1997). This included the cover of Acropora corals, branching corals, imising the covariation between environmental variables and species
encrusting corals, foliose corals, massive corals, submassive corals, traits whereby site scores in the R table constrain site scores in the L
mushroom corals, turf algae, macroalgae, dead coral, rubble and sand. table and species scores in the Q table constrain species scores in the L
In line with earlier studies of JBTI (DeVantier et al., 1998; Cleary et al., table. The relationship between the R and Q tables was tested for sig-
2006; Cleary et al., 2008), sample sites were assigned to specific zones nificance with a permutation test using 1000 random permutations in
along an in-to-offshore gradient, namely: an inshore zone (zone 1) to which the inertia was compared with the total inertia in the dataset.
the south of −5.97° latitude, a midshore zone (zone 2) between The constrained ordination resulting from the three tables was finally
− 5.77° and − 5.97° latitude, and an offshore zone (zone 3) to the north compared with each of the unconstrained separate analyses whereby
of − 5.77° latitude. The offshore zone is largely contained within the the percentage of the variation calculated by the RLQ analysis was
Pulau Seribu National Marine Park, the first marine park established in subsequently compared to the variation explained by the separate
Indonesia (Farhan and Lim, 2012). Raw fish counts per site and values analyses.
for benthic cover are provided as supplementary tables to Cleary et al. There was a highly significant (Permutation test; P < 0.001) as-
(2016). sociation between fish species traits and environmental variables. The
In addition to assessing the benthic substrate, water quality was also first two axes of the RLQ analysis explained 86% and 8% respectively of
quantified using an aqua Troll water quality instrument (In situ Inc., the total variance in the matrix crossing the environmental variables
Fort Collins, CO, USA) to measure temperature (Tmp), pH, dissolved and traits. The first two RLQ axes also accounted for 82% of the var-
oxygen (DO; Clark DO μgs l− 1) and salinity (Sal: parts per thousand iation in the separate analysis of environmental variables, 69% of the
[ppt] or ‰). This was supplemented with water transparency, mea- variation in trait variables and 23% of the variation in species com-
sured once in each site using a Secchi disk, and satellite imagery to position. Less variation is generally explained in the species composi-
measure four water quality variables. These variables were: coloured tion table because RLQ analysis rearranges species and site scores to
dissolved organic matter index (CDOM), near-surface chlorophyll-a maximise covariance between species traits and environmental vari-
concentration (Chla), sea surface temperature (SST) and remote sensing ables. A detailed examination of the covariation of environmental
reflectance at 645 nm (Rrs_645). They were assessed following pre- variables and species traits is presented in Fig. 1 showing the first two
viously described methods (Polónia et al., 2015). Rrs_645 is a proxy for RLQ axes.
total suspended sediments (Miller and Mckee, 2004; Chen et al., 2007) Along axis 1 (eigenvalue: 0.179; covariance: 0.423), fish species
and CDOM largely consists of humic and fulvic substances from runoff associated with inshore waters, characterised by higher Chlorophyll-a
containing decaying plant material from areas with high vegetation values, higher sand cover and higher SST among other things (see
productivity or originating from mangroves and seagrasses (Carder Cleary et al., 2016), were resilient, had a higher mortality rate, higher
et al., 1999; Martin, 2004; Richardson and LeDrew, 2006). Given that growth rate (K), consumed animals and thus had a higher mean trophic
satellite data accuracy is often higher with long-term averaging (Patt level. In contrast, species associated with mid- and offshore waters
et al., 2003), a single mean value per sample site was generated for the characterised by higher transparency, live branching and Acropora
previously mentioned parameters for the years 2002–2005. coral cover and CDOM, among other things, (Cleary et al., 2016) had a
Data on selected fish traits were obtained from the species ecology greater life expectancy, higher age at maturity, greater life span, greater
matrix of the Indian Ocean in the FishBase database (Froese and Pauly, generation time and fed mainly on plants or plants and animals. Along
2016). The fish species traits used in the present study are the popu- the second axis (eigenvalue: 0.017; covariance: 0.128), fish species
lation means of: maximum length (Lmax; cm), L infinity (Linf; cm), the associated with sites with higher turf algae cover were longer (Lm, Linf,
von Bertalanffy growth function K (K; 1/year), life span (Years), natural Lmx and Lopt) than fish species associated with sites with higher dead
mortality (M; 1/year), generation time (Gen; years), age at first ma- coral, encrusting coral, foliose coral and massive coral cover.
turity (AgeM; years), length at maturity (Lm; cm), length at maximum Previous studies of JBTI showed that there was a pronounced in-to-
yield (Lopt; cm), resilience (Res), trophic level (Trop) and dummy offshore gradient in environmental conditions and concomitant pro-
variables for the food items animals, plants and both. Detailed de- nounced variation in the composition of a number of taxa including
scriptions of the variables can be found at http://fishbase.us/manual/ fishes, corals and sponges along the gradient (Cleary et al., 2008; Cleary
English/Key%20Facts.htm. In summary, L infinity (Linf) is the length et al., 2014; Cleary et al., 2016). Significant associations between en-
that a fish would reach if it were to grow indefinitely, the von Berta- vironmental variables and coral species traits have also been previously
lanffy growth function K is the rate at which asymptotic length is ap- demonstrated using RLQ analysis (Rachello-Dolmen and Cleary, 2007).
proached and generation time is the average age of parents at the time In the present study, this also holds true for fish species. Fish species in
their young are born. Length at maximum yield is the length class with perturbed inshore waters are thus relatively fast growing and short-
the highest biomass. Resilience (or productivity) is a categorical lived compared to their offshore counterparts. This includes species

260
D.F.R. Cleary Marine Pollution Bulletin 122 (2017) 259–262

Fig. 1. Results of the RLQ analysis indicating asso-

ciations along the first two axes between a) environ-
mental variables and b) species traits. Closer points
indicate closer associations between either a) en-
vironmental variables or b) species traits. Positions of
points relative to the origin indicate relative con-
tributions to RLQ axes, and similar positions relative
to the origin in both plots indicate close associations
between environmental variables and species traits.
Environmental variables: Acropora coral cover (Acr),
branching coral cover (Cr.br), encrusting coral cover
(Cr.en), foliose coral cover (Cr.fo), massive coral cover
(Cr.ma), submassive coral cover (Cr.su), mushroom
coral cover (Cr.mu), turf algae cover (Al.tu), macro-
algae cover (Al.ma), dead coral cover (Dead.cr),
rubble cover (Rub), sand cover (Snd), temperature
(Tmp), pH, dissolved oxygen (DO), salinity (Sal), co-
loured dissolved organic matter index (CDOM), near-
surface chlorophyll-a concentration (Chla), sea surface
temperature (SST) and remote sensing reflectance at 645 nm (Rrs). Species traits: maximum length (Lmx), L infinity (Linf; cm), length at maturity (Lm; cm), length at maximum yield
(Lopt), the von Bertalanffy growth function K (K), life span (Yrs), natural mortality (Mort), generation time (Gen), age at first maturity (AgeM), resilience (Res), trophic level (Trop) and
food items Animals (Anim), Plants (Plan) and both (Both).

found inshore such as Canthigaster compresus, Cephalopholis boenak, (PPO-LIPI, Jakarta, Indonesia) and Naturalis Biodiversity Center
Cheilodipterus quinquelineatus and Siganus canaliculatus. Species found (Leiden, the Netherlands). The research by Naturalis was part of the
inshore tended to occur across the whole of JBTI suggesting a gen- project ‘Climate change and Indonesian coral reef biotas’ within the
eralistic life style and tolerance to a range of environmental conditions. theme ‘Biodiversity in Relation to Global Change’ of the Council for
Mid- and offshore fish communities in contrast consisted of relatively Earth and Life Sciences of the Netherlands Organisation for Scientific
longer-living species and slower growing species such as Abudefduf Research (ALW-NWO grant 852.000.50). Thanks to LIPI Indonesia for
sexfasciatus and Hemiglyphidodon plagiometopon that were completely granting research permits and the staff of PPO-LIPI for logistical sup-
absent inshore (Cleary et al., 2016). In addition to fish communities in port.
these environments being less diverse and less abundant, they thus also
differ in traits. References
Inshore fish assemblages were also characterised by the prevalence
of animal feeding and thus a higher mean tropic level compared to mid- Barboni, D., Harrison, S.P., Bartlein, P.J., Jalut, G., New, M., Prentice, I.C., Sanchez-Goñi,
and offshore communities. This result confirms data on fish biomass M.F., Spessa, A., Davis, B., Stevenson, A.C., 2004. Relationships between plant traits
and climate in the Mediterranean region: a pollen data analysis. J. Veg. Sci. 15,
collected in JBTI in 2011 (Polónia et al., in press) where herbivorous 635–646.
fish biomass was much lower in sites proximate to the coast. Vincent Baum, G., Januar, H.I., Ferse, S.C.A., Kunzmann, A., 2015. Local and regional impacts of
et al. (2011) also showed that overall fish herbivore biomass increased pollution on coral reefs along the Thousand Islands north of the megacity Jakarta,
Indonesia. PLoS One 10, e0138271.
as a function of distance from shore. Importantly, herbivorous fishes are Carder, K.L., Chen, F.R., Lee, Z.P., Hawes, S., Kamykowski, D., 1999. Semi-analytic
recognised as an important determinant of coral reef resilience due to MODIS algorithms for chlorophyll a and absorption with bio-optical domains based
their ability to limit algal growth and thereby promote coral recruit- on nitrate-depletion temperatures. J. Geophys. Res. 104, 5403–5421.
Chen, Z., Hu, C., Muller-Karger, F., 2007. Monitoring turbidity in Tampa Bay using
ment (Hughes et al., 2007; Ledlie et al., 2007; Green and Bellwood, MODIS/Aqua 250-M imagery. Remote Sens. Environ. 109, 207–220.
2009). Likewise, In Tahiti, Legendre et al. (1997) found fish herbivores Cleary, D.F.R., Suharsono, Hoeksema, B.W., 2006. Coral diversity across a disturbance
to be positively associated with live coral and calcareous algae sub- gradient in the Pulau Seribu reef complex off Jakarta, Indonesia. Biodivers. Conserv.
15, 3653–3674.
strates. Sessile invertebrate grazers and carnivores in contrast were
Cleary, D.F.R., Boyle, T.J.B., Setyawati, T., Angraeti, C.D., Menken, S.B.J., 2007. Bird
overrepresented on coral debris, turf and dead coral. species richness, composition and ecological traits associated with unlogged and
Although the in-to-offshore shift in composition and environmental logged Bornean hill forest. Ecol. Appl. 17, 1184–1197.
conditions is pronounced in JBTI, even the offshore sites were per- Cleary, D.F.R., DeVantier, L., Giyanto, Vail L., Manto, P., de Voogd, N.J., Rachello-
Dolmen, P.G., Tuti, Y., Budiyanto, A., Wolstenholme, J., Hoeksema, B.W., Suharsono,
turbed (Cleary et al., 2008; Cleary et al., 2014). Large fishes in parti- 2008. Relating variation in species composition to environmental variables: a multi-
cular were scarce or absent. This includes sharks but also large par- taxon study in an Indonesian coral reef complex. Aquat. Sci. 70, 419–431.
rotfishes (Scaridae). DeVantier et al. (1998) already described the Cleary, D.F.R., Polónia, A.R.M., Renema, W., Hoeksema, B.W., Wolstenholme, J., Tuti, Y.,
de Voogd, N.J., 2014. Coral reefs next to a major conurbation: a study of temporal
conspicuous absence of these fishes throughout JBTI during a 1995 change (1985–2011) in coral cover and composition in the reefs of Jakarta Indonesia.
survey. Their absence can be attributed to environmental degradation Mar. Ecol. Prog. Ser. 501, 89–98.
and overexploitation of fish stocks. For example, coral cover declined Cleary, D.F., Polónia, A.R., Renema, W., Hoeksema, B.W., Rachello-Dolmen, P.G.,
Moolenbeek, R.G., Budiyanto, A., Yahmantoro, Tuti Y., Giyanto, Draisma S.G.,
precipitously offshore from 1985 to 1995 (Cleary et al., 2014). Prud'homme van Reine, W.F., Hariyanto, R., Gittenberger, A., Rikoh, M.S., de Voogd,
Erdmann (1998) also suggested that fishery collapse occurred between N.J., 2016. Variation in the composition of corals, fishes, sponges, echinoderms, as-
1985 and 1995. Importantly, variation in the prevalence of species cidians, molluscs, foraminifera and macroalgae across a pronounced in-to-offshore
environmental gradient in the Jakarta Bay-Thousand Islands coral reef complex. Mar.
traits along environmental gradients should reflect variation in the re- Pollut. Bull. 110, 701–717. http://dx.doi.org/10.1016/j.marpolbul.2016.04.042.
lative importance of adaptive mechanisms (Barboni et al., 2004). The Demographia, 2012. World urban areas (World Agglomerations): 8th annual edition,
eutrophication, pollution and physical destruction of coral substrate has April 2012. http://www.demographia.com/db-worldua.pdf.
DeVantier, L., Suharsono, Budiyanto A., Tuti, J., Imanto, P., Ledesma, R., 1998. Status of
thus selected for a low biomass and depauperate fish community
coral communities of Pulau Seribu, 1985–1995. In: Soemodijhardjo, S. (Ed.),
(Cleary et al., 2016) characterised by disturbance-resilient, fast growing Contending with Global Change. Proceedings: Coral Reef Evaluation Workshop Pulau
and short lived species that feed predominantly on animals. Seribu, Jakarta, Indonesia. 11–20 September Vol. 1995. UNESCO, Jakarta, Indonesia,
pp. 1–26.
Dickens, L.C., Goatley, C.H.R., Tanner, J.K., Bellwood, D.R., 2011. Quantifying relative
Acknowledgements diver effects in underwater visual censuses. PLoS One 6 (4), e18965.
Dolédec, S., Chessel, D., ter Braak, C.J.F., Champely, S., 1996. Matching species traits to
environmental variables: a new three-table ordination method. Environ. Ecol. Stat. 3,
This publication is a result of the Thousand Islands Expedition, 143–166.
September 2005, organized by the Research Centre for Oceanography English, S., Wilkinson, C., Baker, V., 1997. Survey Manual for Tropical Marine Resources,

261
D.F.R. Cleary Marine Pollution Bulletin 122 (2017) 259–262

2nd edition. Australian Institute of Marine Science, Townsville (390 pp). population declines in Central America. Conserv. Biol. 17, 1078–1088.
Erdmann, M.V., 1998. Destructive fishing practice in Kepulauan Seribu Archipelago. In: Martin, S., 2004. An Introduction to Ocean Remote Sensing. Cambridge University Press.
Soemodijhardjo, S. (Ed.), Contending with Global Change. Proceedings: Coral Reef Miller, R., Mckee, B., 2004. Using MODIS Terra 250 m imagery to map concentrations of
Evaluation Workshop Pulau Seribu, Jakarta, Indonesia. 11–20 September Vol. 1995. total suspended matter in coastal waters. Remote Sens. Environ. 93, 259–266.
UNESCO, Jakarta, Indonesia, pp. 84–89. Patt, F., Barnes, R., Eplee Jr., R., Franz, B., Robinson, W., Feldman, G., Bailey, S., Gales, J.,
Farhan, A.R., Lim, S., 2012. Vulnerability assessment of ecological conditions in Seribu Werdell, P., Wang, M., et al., 2003. Algorithm updates for the fourth SeaWIFS data
Islands, Indonesia. Ocean Coast. Manag. 65, 1–14. reprocessing. NASA Tech. Memo. 206892, 22–74.
Froese, R., Pauly, D., 2016. FishBase. World Wide Web electronic publication. www. Polónia, A.R.M., Cleary, D.F.R., de Voogd, N.J., Renema, W., Hoeksema, B.W., Martins,
fishbase.org (version (10/2016)). A., Gomes, N.C.M., 2015. Importance of space and environment to explain the var-
Green, A.L., Bellwood, D.R., 2009. Monitoring Functional Groups of Herbivorous Reef iation in the composition of coral reef taxa in the Spermonde Archipelago, Indonesia:
Fishes as Indicators of Coral Reef Resilience – A Practical Guide for Coral Reef a multitaxon study. Sci. Total Environ. 537, 139–151.
Managers in the Asia Pacific Region. IUCN Working Group on Climate Change and Polónia, A.R.M., Cleary, D.F.R., Duine, A.A., van Dijk, J., de Voogd, N.J., 2017.
Coral Reefs. IUCN, Gland, Switzerland. Assessment of fish biomass along an in- to offshore gradient in the Jakarta Bay-Pulau
Hoey, A.S., Bellwood, D.R., 2008. Cross-shelf variation in the role of parrotfishes on the Seribu reef complex. J. Mar. Biol. Assoc. U. K (In press).
great barrier reef. Coral Reefs 27, 37–47. Rachello-Dolmen, P.G., Cleary, D.F.R., 2007. Relating coral species traits to environ-
Hosono, T., Su, C.C., Delinom, R., Umezawa, Y., Toyota, T., Kaneko, S., Taniguchi, M., mental conditions in the Jakarta Bay/Pulau Seribu reef system. Estuar. Coast. Shelf
2011. Decline in heavy metal contamination in marine sediments in Jakarta Bay, Sci. 73, 816–826.
Indonesia due to increasing environmental regulations. Estuar. Coast. Shelf Sci. 92, Rees, J.G., Setiapermana, D., Sharp, V.A., Weeks, J.M., Williams, T.M., 1999. Evaluation
297–306. of the impacts of land-based contaminants on the benthic faunas of Jakarta Bay,
Hughes, T.P., Baird, A.H., Bellwood, D.R., Card, M., Connolly, S.R., Folke, C., Grosberg, Indonesia. Oceanol. Acta 22, 627–640.
R., Hoegh-Guldberg, O., Jackson, J.B.C., Kleypas, J., Lough, J.M., Marshall, P., Ribera, I., Dolédec, S., Downie, I.S., Foster, G.N., 2001. Effect of land disturbance and
Nyström, M., Palumbi, S.R., Pandolfi, J.M., Rosen, B., Roughgarden, J., 2003. Climate stress on species traits of ground beetle assemblages. Ecology 82, 1112–1129.
change, human impacts, and the resilience of coral reefs. Science 301, 929–933. Richardson, L.L., LeDrew, E.F., 2006. Remote sensing of aquatic coastal ecosystem pro-
Hughes, T.P., Rodrigues, M.J., Bellwood, D.R., Ceccarelli, D., Hoegh-Guldberg, O., cesses. In: Science and Management Applications 9. Springer, Netherlands. http://dx.
McCook, L., Moltschaniwskyj, N., Pratchett, M.S., Steneck, R.S., Willis, B., 2007. doi.org/10.1007/1-4020-3968-9.
Phase shifts, herbivory, and the resilience of coral reefs to climate change. Curr. Biol. Uneputty, P.A., Evans, S.M., 1997. Accumulation of beach litter on islands of the Pulau
17, 1–6. Seribu Archipelago, Indonesia. Mar. Pollut. Bull. 34, 652–655.
Jackson, J.B.C., Kirby, M.X., Berger, W.H., Bjorndal, K.A., Botsford, L.W., Bourque, B.J., Vincent, I.V., Hincksman, C.M., Tibbetts, I.R., Harris, A., 2011. Biomass and abundance of
Bradbury, R.H., Cooke, R., Erlandson, J., Estes, J.A., Hughes, T.P., Kidwell, S., Lange, herbivorous fishes on Coral reefs off Andavadoaka, western Madagascar. West. Indian
C.B., Lenihan, H.S., Pandolfi, J.M., Peterson, C.H., Steneck, R.S., Tegner, M.J., Ocean J. Mar. Sci. 10, 83–99.
Warner, R.R., 2001. Historical overfishing and the recent collapse of coastal eco- Williams, T.M., Rees, J.G., Setiapermana, D., 2000. Metals and trace organic compounds
systems. Science 293, 629–638. in sediments and waters of Jakarta Bay and the Pulau Seribu Complex, Indonesia.
Ledlie, M.H., Graham, N.A.J., Blythell, J.C., Wilson, S.K., Jennings, S., Polunin, N.V.C., Mar. Pollut. Bull. 40, 277–285.
Hardcastle, J., 2007. Phase shifts and the role of herbivory in resilience of coral reefs. Willoughby, N.G., 1986. Man-made litter on the shores of the Thousand Island archipe-
Coral Reefs 26, 641–653. lago, Java. Mar. Pollut. Bull. 17, 224–228.
Legendre, P., Galzin, R., Harmelin-Vivien, M.L., 1997. Relating behavior to habitat: so- Willoughby, N.G., Sangkoyo, H., Lakaseru, B.O., 1997. Beach litter: an increasing and
lutions to the fourth-corner problem. Ecology 78, 547–562. changing problem for Indonesia. Mar. Pollut. Bull. 34, 469–478.
Lips, K.R., Reeve, J.D., Witters, L.R., 2003. Ecological traits predicting amphibian

262