Global Change Biology

Global Change Biology (2015) 21, 165–172, doi: 10.1111/gcb.12667

Coral–algal phase shifts alter fish communities and

reduce fisheries production
C A M E R O N H . A I N S W O R T H 1 and P E T E R . J . M U M B Y 2
1
College of Marine Science, University of South Florida, 140 7th, St. Petersburg, FL 33701, USA, 2Marine Spatial Ecology Lab,
School of Biological Sciences, The University of Queensland, St. Lucia, Qld 4072, Australia

Abstract
Anthropogenic stress has been shown to reduce coral coverage in ecosystems all over the world. A phase shift
towards an algae-dominated system may accompany coral loss. In this case, the composition of the reef-associated
fish assemblage will change and human communities relying on reef fisheries for income and food security may be
negatively impacted. We present a case study based on the Raja Ampat Archipelago in Eastern Indonesia. Using a
dynamic food web model, we simulate the loss of coral reefs with accompanied transition towards an algae-domi-
nated state and quantify the likely change in fish populations and fisheries productivity. One set of simulations
represents extreme scenarios, including 100% loss of coral. In this experiment, ecosystem changes are driven by coral
loss itself and a degree of habitat dependency by reef fish is assumed. An alternative simulation is presented without
assumed habitat dependency, where changes to the ecosystem are driven by historical observations of reef fish
communities when coral is lost. The coral–algal phase shift results in reduced biodiversity and ecosystem maturity.
Relative increases in the biomass of small-bodied fish species mean higher productivity on reefs overall, but much
reduced landings of traditionally targeted species.
Keywords: acidification, bleaching, coral disease, coral reefs, Ecopath with Ecosim, ecosystem-based management, Raja Ampat,
reef fish

Received 18 January 2014; revised version received 12 June 2014 and accepted 13 June 2014

the most biodiverse regions on Earth possessing over

Introduction
75% of the world’s coral species and almost a thousand
Coral reefs are assailed by an array of anthropogenic species of reef fish (McKenna et al., 2002; Donnelly
stressors rapidly intensifying with the growth of et al., 2003; Halim & Mous, 2006). Although the human
human populations and the expansion of human indus- population is low, the coral reef ecosystem faces
try. Climate change, including global warming and numerous threats including overexploitation, destruc-
ocean acidification, and indirect impacts like disease tive fishing practices (blast and cyanide fishing), coral
and corallivore outbreaks may come to rival overex- mining, crown-of-thorn (Acanthaster planci) outbreaks
ploitation and pollution as the major drivers of coral and land-based pollution (Pet-Soede & Erdmann, 1998;
loss (Hoegh-Guldberg et al., 2007; Frieler et al., 2012). Kaczmarsky et al., 2005).
As we project these effects to intensify, we may expect It is difficult to estimate the amount of coral loss that
further loss of coral structure and reduced abundance has occurred historically in Raja Ampat. Blast fishing is
of numerous reef-associated species (Wilson et al., present, but estimates of the frequency vary widely
2006), many of which support reef fisheries (Pratchett (e.g. Erdmann & Pet, 2002; McKenna et al., 2002; Don-
et al., 2011). Tropical reef fisheries are an indispensable nelly et al., 2003), while the effects on coral reefs of dis-
source of income in many communities of the develop- ease and environmental stress remain poorly studied.
ing world and critical for food security (McManus, As few empirical data are available with which to base
1997; Bell et al., 2011). In this article, we attempt to the rate of loss, we engage here in simulation modelling
quantify in broad terms the likely impact that coral loss that represents a wide range of possible futures.
followed by a phase transition to an algae-dominated We present simulations for a range of coral loss, up
state may have on the reef fish assemblage and on fish- to 100%, from the Raja Ampat ecosystem. We refer to
eries productivity. these as ‘conjectural’ simulations because the effect of
We select as a case study the Raja Ampat archipelago coral loss lies outside of the observed historical range
in Eastern Indonesia. Still relatively pristine, it is among for any such large area (Raja Ampat model encom-
passes 45 000 km2; Ainsworth et al., 2008b). However,
Correspondence: Cameron Ainsworth, tel. 727-553-3373, similar losses of coral have been observed at the scale
fax 727-553-1189, e-mail: [email protected] of individual reefs after coral bleaching events

© 2014 The Authors Global Change Boilogy Published by John Wiley & Sons Ltd. 165
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use,
distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes.
166 C . H . A I N S W O R T H et al.

(Edwards et al., 2001; Mumby et al., 2001) and corre- changes in the vulnerability of prey to predator by some third
sponding fish population impacts have been studied mediating species group (e.g. Cox et al., 2002) or affect the
(Sano, 2004; Pratchett et al., 2008). We assume some productivity of a group according to the biomass of a mediat-
degree of habitat dependency for reef fish species and ing group (e.g. Okey et al., 2004). Ainsworth et al. (2008a,b)
developed four mediation functions for the Raja Ampat mod-
drive ecosystem changes by removing coral from the
els that affect prey vulnerability according to the biomass of a
model. We also assume space-limited growth by mac-
mediating group. These describe tuna facilitating small pela-
roalgae. gic predation by birds, coral protection of fish and inverte-
A second set of simulations, more firmly grounded in brates, cleaner wrasse symbiosis with large reef fish and sea
empirical data, drives changes in the ecosystem not by grass/mangrove protection of juvenile reef fish (Fig. 1). The
coral loss directly, but by changes in the fish assem- protection effect from coral is particularly relevant to this
blage that result from coral loss (as documented by study as it establishes the response of reef-associated fish to
Wilson et al., 2006). In this case, model predictions are coral loss (an assumption relaxed under the Wilson simula-
made concerning the wider fish community and we tion). The function is modelled so that the vulnerability of the
present overall ecosystem impacts as the combination prey species changes in inverse linear proportion to coral
of data from Wilson’s observations and predictions biomass. All predators are affected equally (effectively we
assume a similar mode of attack). The vulnerabilities are free
made by the model. For comparison, it is worth noting
to increase to a maximum of 29 the baseline value during
that the assemblage changes presented by Wilson are
periods of low coral biomass and can decrease to near 1
driven, on average, by a 33.4% loss of coral relative to during periods of high coral biomass. Note that in high
the initial (2012) coral biomass value (i.e. from 100% complexity reefs, certain size classes of prey may be
down to 66.6%). Thus, the Wilson simulation corre- disproportionately affected by coral loss (Rogers et al., 2014).
sponds to a moderate change in coral reef status rela- These four functions are applied to appropriate species in
tive to the conjectural simulations. the model (Table 1). In this study, we have added a fifth
affecting benthic algal productivity (Fig. 2). As coral biomass
declines, productivity of benthic algae increases. Adding this
Materials and methods mediation effect is necessary to capture space-limited growth
in algae as moderated by competitive exclusion by coral.
EwE models Without the mediation effect, algal growth in EwE is moder-
ated only by herbivory. Note that increased productivity is
Working with nongovernmental and academic partners, Ains-
realistic with coral decline if fleshy macroalgae are replaced
worth et al. (2008a,b) constructed a suite of ecosystem models
by fast-turnover benthic turfs (Hatcher, 1988) – our benthic
using Ecopath with Ecosim (EwE: Christensen & Pauly, 1992;
algae group implicitly includes both. Although the potential
Walters et al., 1997) and Ecospace (Walters et al., 1998) repre-
for algal phase shifts are well documented (Done, 1992; Nors-
senting various regions in Raja Ampat. Development of the
om et al., 2009) and an important assumption in our model-
tr€
models utilized field information from dive transects, fish
ling methodology, Carassou et al., 2013 note that macroalgae
stomach sampling, community interviews and coastal sur-
density is related to coral coverage only in degraded reef sys-
veys. The models were tuned to historical catch and biomass
tems. However, due to the large spatial domain of the model
data (1990–2006; including illegal removals, Varkey et al., 2010
(45 000 km2; Ainsworth et al., 2008b), we represent here the
and anecdotal biomass information Ainsworth et al., 2008c)
net effect in a mosaic of degraded and healthy reefs.
and used to reconstruct the history of exploitation in the
region and to answer practical management and conservation
questions posed by the Indonesian Ministry of Marine Affairs Conjectural simulations
and Fisheries (Departemen Kelautan dan Perikanan, DKP). A
thorough technical description of the model, including all We force the biomass of coral to decline from 0% to 100% in
input data, assumptions, fits to observational data and diag- eleven 20-year simulations (2012–2032). Results are presented
nostic testing is available in Ainsworth et al., 2008b located at at the end state of the simulations (in 2032). For each of the
http://www.fisheries.ubc.ca/node/3755. Applications of the simulations, we estimated fisheries productivity in kg
model include Varkey et al., 2012; Pitcher & Ainsworth, 2010 C km2 yr1 based on an annual catch rate and assuming a
and Ainsworth et al., 2008a. 1 : 20 dry to wet weight conversion ratio (Cushing et al., 1958)
We employ the present-day Raja Ampat model (assumed and Redfield element proportions C : N : P = 106 : 16 : 1
representative of 2012), representing the entire archipelago (Redfield, 1934). To calculate the productivity and biomass of
bounded at the north-west coordinate 129°120 E, 0°120 N and the ecosystem under the most extreme coral decline scenario
the south-east coordinate 130°300 E, 2°420 S. This model is (100%), we generated a new Ecopath model based on the
described in Ainsworth et al. (2008b), so we will present only end state of the simulation utilizing the .eii file input/output
the information most relevant to this study. The most impor- procedure available in EwE (Christensen et al., 2005). Ecosys-
tant parameter is the degree of species’ dependency on coral tem productivity is determined as the sumproduct of biomas-
reefs, which we represent by use of EwE’s mediation functions ses and production-per-unit-biomass (P/B) across species
(Christensen et al., 2005). These can be used to represent groups. A weighted average of biomass/production (B/P)

© 2014 The Authors Global Change Boilogy Published by John Wiley & Sons Ltd. 21, 165–172
 E F F E C T O F C O R A L – A L G A L P H A S E S H I F T S O N F I S H 167

(A) (B)

(C) (D)

Fig. 1 Ecosim mediation functions. Vulnerability of prey vs. mediating group biomass: (A) tuna facilitating small pelagic predation
by birds; (B) reef-building coral protection of reef fish and invertebrates; (C) cleaner wrasse symbiosis with large reef-associated fish;
(D) sea grass and mangrove protection of juvenile reef fish. x- and y-axes are relative to model baseline values. Reproduced from
Ainsworth et al., 2008b.

Table 1 Assignment of mediation functions (Med.) in Raja Ampat EwE model. A, B, C and D are defined as in Fig. 1. Reproduced
from Ainsworth et al., 2008b

Prey group Med. Prey group Med. Prey group Med. Prey group Med.

Ad groupers B, C Ad coral trout B, C Juv small reef assoc. B, D Ad eroding grazers B, C

Sub groupers B Juv coral trout B, D Ad large planktivore B, C Juv eroding grazers B, D
Juv groupers B, D Ad small pelagic A Juv large planktivore B, D Ad scraping grazers B, C
Ad snappers B, C Juv small pelagic A Ad small planktivore B Juv scraping grazers B, D
Sub snappers B Ad large reef assoc. B, C Juv small planktivore B, D Penaeid shrimps D
Juv snappers B, D Juv large reef assoc. B, D Ad anchovy A Shrimps and prawns D
Ad Napoleon wrasse B, C Ad medium reef assoc. B, C Juv anchovy A Octopus B
Sub Napoleon wrasse B Juv medium reef assoc. B, D Ad macroalgal browsing B, C Small crabs B
Juv Napoleon wrasse B, D Ad small reef assoc. B Juv macroalgal browsing B, D

(units: yr) is used as an indicator of ecosystem maturity after coral decline projected by the model. The coral decline
Odum (1969). amounts to an 8.3% loss over the 20 year simulation. This rate
represents the effects of coral mining, blast fishing, cyanide
fishing and corallivory by bioeroding fish and crown-of-thorns
Wilson simulation starfish and was set by Ainsworth et al., 2008a to reflect trends
For the Wilson simulation, we force biomass of seven fish in Raja Ampat (McKenna et al., 2002). It is similar to the 7.4%
groups. Biomass change is based on Fig. 2 in Wilson et al. decline observed in the Indo-Pacific over the same period by
(2006), which provides biomass change at the species level rel- Bruno & Selig (2007). The programmed loss of coral in the
ative to the change in coral cover. Wilson’s species are aggre- model ensures that trophodynamic effects impacting species
gated (averaged) to the level of EwE functional groups and whose biomass is not forced are realistically portrayed. Based
the absolute change in biomass is determined relative to the on this technique, the following biomass changes from Wilson

© 2014 The Authors Global Change Boilogy Published by John Wiley & Sons Ltd. 21, 165–172
168 C . H . A I N S W O R T H et al.

(A)

(B)

Fig. 2 Mediation function boosts algal productivity when coral

biomass is low.

are inferred: butterflyfish 7.6%, cleaner wrasse 4.8%, large

reef-associated fish +1.0%, medium reef-associated fish 1.5%,
small reef-associated fish 5.0%, scraping grazers +1.8% and
small planktivores 2.5%. We assume a linear increase or
decrease in species group biomass leading to these values in
(C)
the final year of a 20-year simulation (relative to initialization
biomass). Biodiversity is assessed using two methods, the
Shannon index (Shannon & Weaver, 1949), which measures
evenness, and the Q90 index (Ainsworth & Pitcher, 2006),
which measures evenness and richness. All simulations run
from 2012 to 2032 and we assume that current (2012) fishing
mortalities are maintained into the future.

Results
Detailed model results for conjectural and Wilson simu-
lations are provided in the supplement. Tables S1
through S6 provide biomass, catch and catch value
results; Tables S7 and S8 show changes in ecosystem
trophic level; Table S9 shows mixed trophic impacts
from Ecopath indicating net trophic effects for
impacted-impacting group combinations (see Ula- Fig. 3 Simulation results. (A) Transition from coral-dominated
nowicz & Puccia, 1990 and Christensen et al., 2005). ecosystem (left) to algae-dominated ecosystem (right). Data rep-
resent end-points of 20-year simulations. Results are from the
Table S10 shows changes in production rates for the
conjectural simulations (0%, 10% . . . 100% coral loss scenarios).
conjectural simulation. Figures S1 through S6 show
Grey lines: biomass changes relative to initialization (2012) bio-
Shannon and Q90 biodiversity trajectories. Figure S7
mass; black hatches: reef fisheries productivity in kg
shows average ecosystem biomass/production. C km2 yr1. (B) Biomass changes per trophic level under the
100% coral loss scenario (all species groups). Error bars show
the range for species groups within these trophic levels. (C)
Conjectural simulations
Fisheries catch by fleet under the 100% coral loss scenario rela-
Biomass trajectories for the conjectural simulations tive to 0% loss scenario. Catch for the year 2032 is compared.
are presented in Fig. 3A. When coral is eliminated,
the ecosystem shifts towards algal dominance. Her- (small reef fish 97%, medium reef fish 61%, large
bivorous fish increase 14% and urchins increase 117% planktivores 78%) (see Table S1). The total produc-
under the extreme coral loss scenario (relative to the tion rate of reef-associated groups increases by about
0% loss scenario). However, the reef fish groups’ 40% as high-turnover smaller species replace slow-
biomasses decrease on average by 46%, with some growing larger species (Table S10). This finding is
reef-dependent groups showing severe depletions corroborated as mean ecosystem trophic level also

© 2014 The Authors Global Change Boilogy Published by John Wiley & Sons Ltd. 21, 165–172
 E F F E C T O F C O R A L – A L G A L P H A S E S H I F T S O N F I S H 169

drops from 1.62 to 1.56 between 2012 and 2032 loss scenario. The only fisheries that clearly benefit
(Table S8), reflecting a major structural change. Lar- from the loss of coral structure are those targeting
ger species tend to be higher trophic level, and we shrimp and other benthic invertebrates (Fig. 3C). Bio-
see a skewing of the trophic pyramid towards smal- mass increases in both the penaeid shrimp (+64%) and
ler-bodied consumers (Fig. 3B). In the 100% coral loss nonfished shrimp groups (+42%) (Table S1). The mixed
scenario, Shannon biodiversity decreases and then trophic impacts routine (Ulanowicz & Puccia, 1990)
recovers incompletely (Figure S1), Q90 biodiversity suggests that this is due to a decrease in predation mor-
decreases steadily (Figure S2). tality by carnivorous macrobenthos, which is a result of
When corals are removed, total ecosystem biomass coral loss (Table S9). A small increase in the availability
increases 18.5% (Table S1). With trophic chains short- of detritus may also contribute.
ened, fewer trophic steps result in less energy lost
through thermodynamic inefficiencies. However, the
Wilson simulation
increased biomass does not occur in groups targeted by
fisheries. Annual reef fish landings decline by 39%, Forcing EwE with Wilson et al. biomass changes in reef
from 152 to 93 kg C km2 yr1 [contrast this against fish leads to a 10–30% decline in biomass over 20 years
Rogers et al. (2014) who estimated a 55% decline in in fished groups like large pelagic fish, groupers and
predator productivity under similar conditions in the snappers (Table S4). There is a steady decrease in eco-
Caribbean]. Reef fish biomass has been halved from 373 system biodiversity from 2012 to 2032 according to the
to 202 kg C km2 (Table S1). Midtrophic level species, Shannon index, but the Q90 index decreases and then
which previously acted as a conduit for vertical energy recovers (Figures S3 and S4). This disagreement
flow, are reduced in biomass (small pelagic fish 45%, between the metrics implies that evenness has been
small reef fish 97%, anchovy 43%) (Table S1). This is impacted, but ecosystem biomass eventually recovers
from a combination of top-down and bottom-up effects (although not necessarily in the same groups as those
(Table S9). Macroalgal grazers increase in biomass that declined). Considering biodiversity within the reef
because of the abundance of algae in coral loss scenar- fish assemblage, both biodiversity metrics indicate
ios, yet they do not eat algae exclusively. A small frac- steady decline (Figures S5 and S6). As in the conjectural
tion of their diet includes forage species (e.g. about 19% simulations, there is an increase in shrimp biomass
of predation mortality on small pelagic fish is due to (Fig. 4). Penaeid shrimp and nonfished shrimp increase
this group) so there is increased predation from macro- 7% and 6% respectively (Table S4), while the shrimp
algal grazers. There is also a decrease in availability of trawl fleet benefits from a 9% increase in the annual
small herbivorous zooplankton, an important prey item catch rate (Table S6).
for forage fish. This is due to a trophic cascade connect-
ing herbivorous zooplankton to carnivorous zooplank-
Discussion
ton to reef-associated fish (Table S9).
Structural changes are further evidenced by a The Raja Ampat ecosystem is exceptionally biodiverse
decrease in ecosystem maturity (Figure S7), with the and may serve as a sensitive test site for coral loss stud-
greatest reduction in B/P occurring in the 100% coral ies. In other ways, the ecosystem is typical of coral reef

Fig. 4 Ecosystem biomass changes in Raja Ampat from 2012 to 2032 using Wilson et al., 2006 biomass forcing.

© 2014 The Authors Global Change Boilogy Published by John Wiley & Sons Ltd. 21, 165–172
170 C . H . A I N S W O R T H et al.

areas: it endures a multitude of threats with environ- Nonlinear or threshold effects may actually worsen
mental and human-related stressors on the rise. We this problem at low coral densities (Pratchett et al.,
have simulated in this article both a typical pattern of 2014). We conclude that human communities relying
coral loss, similar to the recent world case studies sum- on this depauperate ecosystem would likely be
marized by Wilson et al. (2006), and more dramatic required to abandon traditional target species in favour
losses as may become common in years to come. of less valuable but more abundant species. As the EwE
The conjectural simulations, in which up to 100% of model we have employed does not consider opportu-
coral biomass was removed, result in fundamental nistic fishing behaviour or market effects, it is difficult
shifts in ecosystem structure and function. The ecosys- to estimate the change in fisheries profitability. Despite
tem transitions from a coral-dominated state to an great abundance, these high-turnover species of fish
algal-dominated state, although these algae are primar- and invertebrates are likely to fluctuate with environ-
ily envisaged to be short turfs rather than fleshy macro- mental variability more than the longer lived species
algae, and do not necessarily pose a threat to ecosystem that are the traditional mainstay of fisheries. This could
recovery (sensu Arnold et al., 2010). This is consistent carry implications for the consistency of fisheries bene-
with evidence from the Indo-Pacific (Mumby et al., fits and food security.
2013). The conjectural simulations and the Wilson simula-
The model predicts that transition towards algal tion agree that loss of coral results in large decreases in
dominance results in increases in herbivorous species reef-associated fauna, and that smaller reef-associated
and decreases in reef-associated fauna and high tro- fish species are particularly impacted. This shift in size
phic level piscivorous species, the main targets of fish- structure has been noticed empirically (Graham et al.,
eries (Sano, 2004; Pratchett et al., 2008). Evidence for a 2007; Ledlie et al., 2007). Also, the conjectural simula-
herbivore numerical response may be ambiguous in tions and the Wilson simulation agree that an increase
the Indo-Pacific (Wismer et al., 2009; Carassou et al., in shrimp biomass and shrimp landings is likely when
2013; Heenan & Williams, 2013) although some stud- coral is depleted. This effect can be traced to reduced
ies found indications of such (Cheal et al., 2008; Gil- predation on shrimp by carnivorous macrobenthos,
mour et al., 2013). Moreover, these relationships are which are dependent on reefs, and a greater availability
common in the Caribbean (Williams et al., 2001; Mum- of detritus.
by et al., 2005, 2006, Newman et al., 2006; Carpenter, Qualitative agreement between the Wilson and con-
1990) and this lends credence to the model’s behav- jectural simulations within the range of historically
iour. Food limitation in herbivorous fish contributes observed coral declines in this region lends credibility
to the numerical response (Ainsworth et al., 2008b). A to the more extreme coral loss scenarios ventured by
nonlinear effect may occur at low population sizes the conjectural simulations. However, one conspicuous
where herbivore fish response is decoupled from algal disagreement is that the conjectural simulations predict
density (Hern andez-Landa et al., 2014), but this is a decrease in small midtrophic level fish with implica-
unlikely to affect our results as herbivore fish biomass tions for vertical flow of energy to the upper food web.
remains relatively high in Raja Ampat (Ainsworth This behaviour is not present in the Wilson simulation.
et al., 2008b). Differentiation within the herbivore The (inputted) fish abundance data drawn from Wilson
guild is missed by our model due to species aggrega- et al. (2006) does in fact include a 5% decrease in small
tion, so it is difficult to infer which functional roles reef-associated fish (one of the largest changes observed
(see Heenan & Williams, 2013) remain present after by those authors), but our simulation does not predict a
the phase shift and therefore implications for reef similar decrease in small pelagic fish. Rather, it predicts
resiliency (Cheal et al., 2008). a small increase in those groups leaving the pelagic for-
The shift in biomass towards lower trophic level spe- age assemblage intact. One possible explanation is that
cies indicates that trophic chains are shortened overall greater losses of coral are required to elicit this effect
and the food web is simplified. The model predicts that (more akin to the extreme scenarios tested in the conjec-
less energy passes to the upper food web and through tural simulations). An alternative explanation is that
fewer conduits. This is indicated by a decrease in the the conjectural simulations, which assumed a uniform
average trophic level, reduced biomass in forage spe- effect of coral loss on different size categories of
cies (caused by simultaneous top-down and bottom-up reef-associated fish, lost nuanced differences that were
effects), and reduced ecosystem biodiversity – a finding more properly represented by the Wilson simulation.
supported by observation (Jones et al., 2004; but see This study uses a combination of modelling the
Cheal et al., 2008). Finally, ecosystem maturity potential effects of major coral loss (conjectural simula-
decreases as long-lived species are replaced with high- tions) with more empirically grounded simulations of
turnover species. the consequences of a modest decline in coral cover

© 2014 The Authors Global Change Boilogy Published by John Wiley & Sons Ltd. 21, 165–172
 E F F E C T O F C O R A L – A L G A L P H A S E S H I F T S O N F I S H 171

(the Wilson et al. simulations). In both cases, the pro- Edwards AJ, Clark S, Zahir H, Rajasuriya A, Naseer A, Rubens J (2001) Coral bleach-
ing and mortality on artificial and natural reefs in Maldives in 1998: sea surface
ductivity of reef fisheries declines severely. This is a temperature anomalies and initial recovery. Marine Pollution Bulletin, 42, 7–15.
concern for food security given the high dependence on Erdmann MV, Pet JS (2002) A rapid marine survey of the northern Raja Ampat
fish protein that often exists in tropical coastal areas Islands (Eastern Indonesia). Report from Henry Foundation/The Nature Conser-
vancy/NRM/EPIQ, 36 pp, The Nature Conservancy. Available at: joanne_wil-
(Burke et al., 2011). [email protected].
Frieler K, Meinshausen M, Golly A, Mengel M, Lebek K, Donner SD, Hoegh-Guld-
berg O (2012) Limiting global warming to 2C is unlikely to save most coral reefs.
Acknowledgements Nature Climate Change, 3, 165–170.
Gilmour JP, Smith LD, Heyward AJ, Baird AH, Pratchett MS (2013) Recovery of an
Funding for this article was provided in part by the National isolated coral reef system following severe disturbance. Science, 5, 69–71.
Oceanic and Atmospheric Administration through their support Graham NA, Wilson SK, Jennings S, Polunin NV, Robinson J, Bijoux JP, Daw TM
of the Marine Resource Assessment program at USF. We thank (2007) Lag effects in the impacts of mass coral bleaching on coral reef fish, fisher-
Peter Sale for comments on an early draft of this article. ies, and ecosystems. Conservation Biology, 21, 1291–1300.
Halim A, Mous P (2006) Community perceptions of marine protected area manage-
ment in Indonesia. A report to National Oceanic and Atmospheric Administration
(NOAA). NA04NOS4630288. Available at: [email protected].
References
Hatcher BG (1988) Coral reef primary productivity: a beggar’s banquet. Trends in
Ainsworth CH, Pitcher TJ (2006) Modifying kempton’s species diversity index for use Ecology and Evolution, 3, 106–111.
with ecosystem simulation models. Ecological Indicators, 6, 623–630. Heenan A, Williams ID (2013) Monitoring herbivorous fishes as indicators of coral
Ainsworth CH, Varkey DA, Pitcher TJ (2008a) Ecosystem simulations supporting eco- reef resilience in American Samoa. PloS One, 8, e79604.
system based fisheries management in the coral triangle, Indonesia. Ecological Hernandez-Landa RC, Acosta-Gonzalez G, N un ~ ez-Lara E, Arias-Gonzalez JE (2014)
Modelling, 214, 361–374. Spatial distribution of surgeonfish and parrotfish in the north sector of the
Ainsworth CH, Varkey DA, Pitcher TJ (2008b) Ecosystem simulation models of Raja Mesoamerican Barrier Reef System. Marine Ecology, doi: 10.1111/maec.12152.
Ampat, Indonesia, in support of ecosystem based fisheries management. In: Eco- Hoegh-Guldberg O, Mumby PJ, Hooten AJ et al. (2007) Coral reefs under rapid cli-
logical and Economic Analyses of Marine Ecosystems in the Bird’s Head Seascape, Papua, mate change and ocean acidification. Science, 318, 1737–1742.
Indonesia: II (eds Bailey M, Pitcher TJ), pp. 3–124. Fisheries Centre Research Jones GP, McCormick MI, Srinivasan M, Eagle JV (2004) Coral decline threatens fish
Reports 16(1), pp 186, Fisheries Centre, University of British Columbia, Vancouver, biodiversity in marine reserves. Proceedings of the National Academy of Sciences, 101,
Canada. 8251–8253.
Ainsworth CH, Pitcher TJ, Rotinsulu C (2008c) Evidence of fishery depletions and Kaczmarsky L, Draud M, Williams EH (2005) Is there a relationship between proxim-
shifting cognitive baselines in Eastern Indonesia. Biological Conservation, 141, 848– ity to sewage effluent and the prevalence of coral disease? Caribbean Journal of Sci-
859. ence, 41, 124–137.
Arnold SN, Steneck RS, Mumby PJ (2010) Running the gauntlet: inhibitory effects of Ledlie MH, Graham NAJ, Bythell JC, Wilson SK, Jennings S, Polunin NVC, Hardcas-
algal turfs on the process of coral recruitment. Marine Ecology Progress Series, 414, tle J (2007) Phase shifts and the role of herbivory in the resilience of coral reefs.
91–105. Coral Reefs, 26, 641–653.
Bell JD, Andrew NL, Batty MJ et al. (2011) Adapting tropical Pacific fisheries and McKenna SA, Boli P, Allen GR (2002) Condition of coral reefs at the Raja Ampat
aquaculture to climate change: management measures, policies and investments. Islands, Papua Province, Indonesia. In: A Marine Rapid Assessment of the Raja Ampat
In: Vulnerability of Tropical Pacific Fisheries and Aquaculture to Climate Change (eds Islands, Papua Province, Indonesia, Bulletin of the Rapid Assessment Program, Vol.
Bell JD, Johnson JE, Hobday AJ), pp. 803–876. Secretariat of the Pacific Commu- 22 (eds McKenna SA, Allen GR, Suryadi S), pp. 66–78. Conservation International,
nity, Noumea, New Caledonia. Washington, DC.
Bruno JF, Selig ER (2007) Regional decline of coral cover in the Indo-Pacific: timing, McManus JW (1997) Tropical marine fisheries and the future of coral reefs: a brief
extent, and subregional comparisons. PLoS ONE, 2, e711. review with emphasis on Southeast Asia. Coral Reefs, 16, S121–S127.
Burke L, Reytar K, Spalding M, Perry AL (2011) Reefs at Risk Revisited. World Mumby PJ, Chisholm JRM, Edwards AJ et al. (2001) Unprecedented bleaching-
Resources Institute, Washington, DC. induced mortality in Porites spp. at Rangiroa Atoll, French Polynesia. Marine Biol-
Carassou L, Leopold M, Guillemot N, Wantiez L, Kulbicki M (2013) Does herbivorous ogy, 139, 183–189.
fish protection really improve coral reef resilience? A case study from New Cale- Mumby PJ, Foster NL, Glynn Fahy EA (2005) Patch dynamics of coral reef macroal-
donia (South Pacific). PLoS ONE, 8, e60564. gae under chronic and acute disturbance. Coral Reefs, 24, 681–692.
Carpenter RC (1990) Mass mortality of Diadema antillarium. Effects on population Mumby PJ, Dahlgren CP, Harborne AR et al. (2006) Fishing, trophic cascades, and the
densities and grazing intensity of parrotfishes and surgeonfishes. Marine Biology, process of grazing on coral reefs. Science, 311, 98–101.
104, 79–86. Mumby PJ, Bejarano S, Golbuu Y, Steneck RS, Arnold SN, van Woesik R, Fiedlander
Cheal AJ, Wilson SK, Emslie MJ, Dolman AM, Sweatman H (2008) Responses of reef AM (2013) Empirical relationships among resilience indicators on Micronesian
fish communities to coral declines on the Great Barrier Reef. Marine Ecology Pro- reefs. Coral Reefs, 32, 213–226.
gress Series, 372, 211–223. Newman MJH, Paredes GA, Sala E, Jackon JBC (2006) Structure of Caribbean coral
Christensen V, Pauly D (1992) ECOPATH II - A software for balancing steady-state reef communities across a large gradient of fish biomass. Ecology Letters, 9, 1216–
models and calculating network characteristics. Ecological Modeling, 61, 169–185. 1227.
Christensen V, Walters CJ, Pauly D (2005) Ecopath with Ecosim: A User’s Guide. Fisher- Norstr€
om AV, Nystr€om M, Lokrantz J, Folke C (2009) Alternative states on coral
ies Centre, University of British Columbia, Vancouver. November 2005 edition. reefs: beyond coral-macroalgal phase shifts. Marine Ecology Progress Series, 376,
Available at: www.ecopath.org. 295–306.
Cox SP, Essington TE, Kitchell JF, Martell SJ, Walters CJ, Boggs C, Kaplan I (2002) Odum EP (1969) The strategy of ecosystem development. Science, 104, 262–270.
Reconstructing ecosystem dynamics in the central Pacific Ocean, 1952–1998 II. The Okey T, Vargo GA, Mackinson S et al. (2004) Simulating community effects of sea
trophic impacts of fishing and effects on tuna dynamics. Canadian Journal of Fisher- floor shading by plankton blooms over the West Florida Shelf. Ecological Modeling,
ies and Aquatic Science, 59, 1736–1747. 172, 339–359.
Cushing DH, Humpfrey GF, Banse K, Laevastu T (1958) Report of the committee on Pet-Soede L, Erdmann MV (1998) An overview and comparison of destructive fishing
terms and equivalents. Rapports et Proces-verbaux des Reunions du Conseil interna- practices in Indonesia. SPC Live Reef Fish Information Bulletin, 4, 28–36.
tional pour l0 Exploration de la Mer, 144, 15–16. Pitcher TJ, Ainsworth CH (2010) Resilience to change in two coastal communities:
Done TJ (1992) Phase shifts in coral reef communities and their ecological signifi- using the maximum dexterity fleet. Marine Policy, 34, 810–814.
cance. Hydrobiologia, 247, 121–132. Pratchett MS, Munday PL, Wilson SK et al. (2008) Effects of climate-induced coral
Donnelly R, Neville D, Mous PJ (eds.) (2003) Report on a rapid ecological assessment bleaching on coral-reef fishes. Oceanography and Marine Biology: An Annual Review,
of the Raja Ampat Islands, Papua, Eastern Indonesia, held October 30–November 46, 251–296.
22, 2002. Final Draft, November 2003, 246 pp, Available at: [email protected].

© 2014 The Authors Global Change Boilogy Published by John Wiley & Sons Ltd. 21, 165–172
172 C . H . A I N S W O R T H et al.

Pratchett MS, Munday PL, Graham NAJ et al. (2011) Vulnerability of coastal fisheries Varkey DA, Ainsworth CH, Pitcher TJ (2012) Modelling reef fish population
in the tropical Pacific to climate change. In: Vulnerability of Tropical Pacific Fisheries responses to fisheries restrictions in marine protected areas in the coral triangle.
and Aquaculture to Climate Change (eds Bell JD, Johnson JE, Hobday AJ), pp. 493– Journal of Marine Biology, 2012, Article ID 721483. doi: 10.1155/2012/721483.
576. Secretariat of the Pacific Community, Noumea, New Caledonia. Walters C, Christensen V, Pauly D (1997) Structuring dynamic models of exploited
Pratchett MS, Hoey AS, Wilson SK (2014) Reef degradation and the loss of critical ecosystems from trophic mass-balance assessments. Reviews in Fish Biology and
ecosystem goods and services provided by coral reef fishes. Current Opinion in Fisheries, 7, 139–172.
Environmental Sustainability, 7, 37–43. Walters C, Pauly D, Christensen V (1998) Ecospace: prediction of mesoscale spatial
Redfield AC (1934) On the proportions of organic derivates in sea water and their patterns in trophic relationships of exploited ecosystems, with emphasis on the
relation to the composition of plankton. In: James Johnstone Memorial Volume (ed. impacts of marine protected areas. Ecosystems, 2, 539–554.
Daniel RJ), pp. 176–192. University of Liverpool, Liverpool, UK. Williams ID, Polunin NVC, Hendrick VJ (2001) Limits to grazing by herbivorous
Rogers A, Blanchard JL, Mumby PJ (2014) Vulnerabilty of coral reef fisheries to a fishes and the impact of low coral cover on macroalgal abundance on a coral reef
loss of structural complexity. Current Biology, 24, 1000–1005. in Belize. Marine Ecology Progress Series, 222, 187–196.
Sano M (2004) Short-term effects of a mass coral bleaching event on a reef fish assem- Wilson SK, Graham NAJ, Pratchett MS, Jones GP, Polunin NVC (2006) Multiple dis-
blage at Iriomote Island, Japan. Fisheries Science, 70, 41–46. turbances and the global degradation of coral reefs: are reef fishes at risk or resil-
Shannon CE, Weaver W (1949) The Mathematical Theory of Communication. University ient? Global Change Biology, 12, 2220–2234.
of Illinois Press, Urbana, IL. Wismer S, Hoey AS, Bellwood DR (2009) Cross-shelf benthic community structure on
Ulanowicz RE, Puccia CJ (1990) Mixed trophic impacts in ecosystems. Coenoses, 5, the Great Barrier Reef: relationships between macroalgal cover and herbivore bio-
7–16. mass. Marine Ecology Progress Series, 376, 45–54.
Varkey DA, Ainsworth CH, Pitcher TJ, Goram Y, Sumaila R (2010) Illegal, unreported
and unregulated fisheries catch in Raja Ampat Regency, Eastern Indonesia. Marine
Policy, 34, 228–236.

Supporting Information

Additional Supporting Information may be found in the online version of this article:

Table S1. Conjectural simulation results (biomass by species group, all simulations).
Table S2. Conjectural simulation results (catch and value rates by species group, selected simulations).
Table S3. Conjectural simulation results (catch and value rates by fishery, selected simulations).
Table S4. Wilson simulation results (biomass by functional group).
Table S5. Wilson simulation results (catch and value rates by functional group).
Table S6. Wilson simulation results (catch and value by fishery).
Table S7. Trophic level at initialization (2012) and end state (2032) for Wilson simulation.
Table S8. Trophic level at initialization (2012) and end state (2032) for conjectural simulations.
Table S9. Mixed trophic impacts.
Table S10. Production rate in 2032 under 100% coral loss scenario.
Figure S1. Shannon biodiversity for conjectural simulations (0%, 50% and 100% coral loss scenarios).
Figure S2. Q90 biodiversity for conjectural simulations (0%, 50% and 100% coral loss scenarios).
Figure S3. Shannon biodiversity for Wilson simulation (all groups).
Figure S4. Q90 biodiversity for Wilson simulation (all groups).
Figure S5. Shannon biodiversity for Wilson simulation (reef fish groups).
Figure S6. Q90 biodiversity for Wilson simulations (reef fish groups).
Figure S7. Average group biomass/production ratios in the ecosystem.

© 2014 The Authors Global Change Boilogy Published by John Wiley & Sons Ltd. 21, 165–172