Short-term changes of fish assemblages observed in the near-pristine reefs of the Phoenix Islands

RESEARCH PAPER

Short-term changes of fish assemblages observedin the near-pristine reefs of the Phoenix Islands

Sangeeta Mangubhai • Ayron M. Strauch •

David O. Obura • Gregory Stone •

Randi D. Rotjan

Received: 14 August 2012 / Accepted: 20 September 2013 / Published online: 24 November 2013

� Springer Science+Business Media Dordrecht 2013

Abstract Climate change-related disturbances are

increasingly recognized as critical threats to biodiver-

sity and species abundance. On coral reefs, climate

disturbances have known consequences for reef fishes,

but it is often difficult to isolate the effect of coral

bleaching from preceding or simultaneous distur-

bances such as fishing, pollution, and habitat loss. In

this study, pre-bleaching surveys of fish family

assemblages in the remote Phoenix Islands in 2002

are compared to post-bleaching in 2005, following

severe thermal stress. Post-bleaching, total coral cover

decreased substantially, as did the combined abun-

dance of all fish families. Yet, changes in abundance

for specific fish families were not uniform, and varied

greatly from site to site. Of the 13 fish families

examined, 3 exhibited significant changes in abun-

dance from 2002 to 2005, regardless of site (Carang-

idae, Chaetodontidae, and serranid subfamily

Epinephelinae). For these families, we explored

whether changes in abundance were related to island

type (island vs atoll) and/or declining coral cover

(percent change). Carangidae on islands experienced

larger changes in abundance than those on atolls,

though declines in abundance over time were not

associated with changes in live coral cover. In

contrast, for Chaetodontidae, declines in abundance

over time were most dramatic on atolls, and were also

associated with changes in live coral cover. The

remoteness of the Phoenix Islands excludes many

typical local anthropogenic stressors as drivers of

short-term changes; observed changes are instead

more likely attributed to natural variation in fish

populations, or associated with coral loss following

the 2002–2003 major thermal stress event.

Keywords Climate change � Disturbance �Coral reef � Kiribati � Chaetodontidae

Introduction

Disturbance plays an important role in determining

species diversity and community structure (Petraitis

et al. 1989; Pickett and White 1986; Sousa 1984).

Climate change-related disturbances are increasingly

recognized as critical threats to biodiversity (Thomas

S. Mangubhai

IUCN Oceania Regional Office, 5 Ma’afu Street,

Suva, Fiji

S. Mangubhai � A. M. Strauch � D. O. Obura �G. Stone � R. D. Rotjan (&)

John H. Prescott Marine Laboratory, New England

Aquarium, One Central Wharf, Boston, MA 02110, USA

e-mail: [email protected]

D. O. Obura

CORDIO East Africa, P.O. Box 10135, Bamburi,

Mombasa 80101, Kenya

G. Stone

Global Marine Division, Conservation International,

2011 Crystal Drive, Arlington, VA 22202, USA

123

Rev Fish Biol Fisheries (2014) 24:505–518

DOI 10.1007/s11160-013-9327-5

et al. 2004; Balmford et al. 2005; Smale and Wernberg

2013); such disturbances are wide-ranging and include

acute as well as chronic events with varying frequency

and intensity (e.g. Hoegh-Guldberg et al. 2007).

Predicting the more subtle impacts of disturbance

is difficult, especially in highly diverse ecosystems

that are changing prior to the measurement of

historical baseline states. Such is the case for coral

reefs, which are among the most diverse and complex

ecosystems on the planet and for which little historical

data of pre-anthropogenic disturbance conditions are

available (though see Jackson 2001; Pandolfi et al.

2003).

Tropical reefs, created by coral animals, support

diverse communities of reef fishes, with over 4,000

tropical fish species worldwide (Allen 2007; Lieske

and Myers 1994). Declines in reef fish abundance have

been well documented on nearshore reefs, largely due

to direct causes such as overfishing, habitat loss, and

pollution (reviewed by Jones and Syms 1998; Wilson

et al. 2006). Similarly, an increasing number of studies

have shown that rising temperatures, increased storm

frequency and intensity, and habitat loss due to climate

change have substantial impacts on reef fishes (Mun-

day et al. 2008; Pratchett et al. 2008, 2011; Chong-

Seng et al. 2012). However, on most reefs, multiple

disturbances may occur simultaneously, thus con-

founding the influence of any one disturbance. Climate

change disturbance events are generally coupled with

local anthropogenic impacts (acute and/or chronic).

As a result, the majority of climate-related disturbance

reef research has inevitably occurred against a back-

drop of concurrent local anthropogenic stressors,

revealing the combined impacts of these stressors

instead of solely isolating the impacts of climatolog-

ical disturbances (but see Christensen et al. 1996;

Sandin et al. 2008).

Recent reviews and workshops have assessed

current knowledge gaps and highlighted the immedi-

ate need to explore reef fish responses to climate

change disturbances, with heavy emphasis on under-

standing how fish community dynamics and diversity

change as a result of habitat loss or changing habitat

composition (Munday et al. 2008; Wilson et al.

2010a). For example, fish diversity and abundance

decline with coral loss (Wilson et al. 2006; Graham

et al. 2006; Jones et al. 2004). Examining the

taxonomic and functional group responses to bleach-

ing is an active topic of research, and fishes do not

have a universal response to bleaching and/or coral

loss. For example, some groups, such as herbivores

that benefit from conversion of coral to algal cover in

the short term, have in some cases been shown to

increase following coral loss (Lindahl et al. 2001)

while fishes with a known coral-dependency cannot

tolerate coral mortality (e.g. Bonin et al. 2009). A

recent review also highlighted the need for tropical

field observations of response to climate change

events to complement the majority of studies that

have thus far focused on lab manipulations (Wernberg

et al. 2012a, b).

The Phoenix Islands, located just south of the

equator in the Central Pacific in the Republic of

Kiribati, are ideally located for examining the natural

variation of fish assemblages over time, as well as the

potential impacts of coral bleaching. Phoenix Islands

reefs are protected from local anthropogenic stressors

such as fishing, coastal development, and pollution.

These islands are largely uninhabited, and are among

the most remote reefs on earth (3 days by sea from the

closest port). Kanton is the only inhabited atoll, and

between 2000 and 2012 has hosted a caretaker

population of fewer than 50 residents. Orona was

settled briefly from 2001 to 2003 under a Kiribati

government trial resettlement scheme, while the

remaining atolls have been uninhabited since the

1960s.

Expeditions to the Phoenix Islands in 2000 and

2002 found the reefs to be in excellent condition with

coral cover averaging 45.1 and 58.1 %, respectively,

with a maximum cover of 100 % live coral (Obura

et al. 2011; Obura and Stone 2002) and 516 species of

fishes (Allen and Bailey 2011). However, during the

expedition in 2002 the early stages of coral bleaching

were observed on one of the atolls in the Phoenix

Islands, and over the following 10 months a sea

surface temperature (SST) ‘hotspot’ developed and

remained over the central Pacific from June 2002 to

March 2003 (Alling et al. 2007). Data from in situ

temperature loggers showed the highest maximum

SST was recorded in November 2002, being about

0.5–1 �C warmer than the following two years. The

hotspot did not fully dissipate and remained for an

extended period of 21 Degree Heating Weeks (DHW).

Post-bleaching surveys in 2005 documented live coral

cover, which was significantly reduced to 12.1 %

overall in the Phoenix Islands (Obura and Mangubhai

2011a). This level of coral loss is consistent with other

506 Rev Fish Biol Fisheries (2014) 24:505–518

123

studies where bleaching has had substantial negative

impact on the abundance and composition of coral

communities (e.g. Vargas-Angel et al. 2011). Given

their remoteness and protection from local anthropo-

genic stressors, the reefs are considered relatively

pristine, with highly abundant fish populations (Obura

et al. 2011; Obura and Stone 2002), which provides an

opportunity to investigate the impacts of bleaching,

decoupled from other concurrent stressors.

The remarkable abundance of fishes across all

trophic groups that are largely overfished elsewhere

makes the Phoenix Islands an ideal study site

compared to more developed and/or exploited Central

Pacific reefs (Allen and Bailey 2011). The lack of local

anthropogenic impacts in the Phoenix Islands also

allows for examination of how natural topographic

features contribute to relative changes in benthic and

fish assemblages. Studies examining how coral cover

differs from site to site often note high between-site

variability, but it is often difficult to decouple the

influence of anthropogenic versus natural features of

each site. For example, remote and undeveloped

lagoonal islands may be differently sensitive to

thermal stress compared to non-lagoonal islands

(Obura and Mangubhai 2011), given their differences

in water quality and nutrient characteristics. However,

different fishes might be predicted to respond differ-

ently to thermal stress and resulting loss of coral

habitat, depending on their trophic mode and flexibil-

ity, mobility, lifespan, and other natural history

characteristics (e.g. Wilson et al. 2006; Pratchett

et al. 2008). The extreme remoteness of the archipel-

ago, coupled with the severity of this thermal event,

make the Phoenix Islands a relatively unique location

for exploring reef fish response to climate change

disturbance in the field.

In this study, we examined the short-term

(*3 year) impacts of severe thermal stress on fish

assemblages in the Phoenix Islands. We compared fish

family abundances pre- and post-bleaching by site,

and explored several hypotheses to explain why the

abundances of some families differed, including site

location, island type, and hard coral cover. We tested

the hypothesis that fish decline differed between

lagoonal atolls and non-lagoonal islands, and that fish

decline was correlated to hard coral cover decline. For

fish abundance data, the family level is the highest

taxonomic resolution available for 2002 and 2005.

Methods

Study site

The Phoenix Islands in the Republic of Kiribati are

comprised of eight islands (Nikumaroro, Kanton,

Orona, Enderbury, Manra, Rawaki, McKean and

Birnie) and two shallow submerged reefs (Carondelet

and Winslow), all located within the equatorial region

of the central Pacific Ocean (2.5–5�S, 174.8–170.1�W)

(Fig. 1). The three largest islands (Nikumaroro, Kan-

ton, and Orona) are atolls with lagoons that connect to

the sea. The remaining islands have no interior lagoon

that interacts with the sea (Fig. 1).

In 2002, a total of 451 fish species were recorded in

the Phoenix Islands, including 212 new records for the

island group, for a total of 516 total known species in

217 genera and 67 families (Allen and Bailey 2011). In

2002, nine permanent monitoring sites were estab-

lished around the Phoenix Islands in an effort to

standardize sampling locations over time (Fig. 1).

Eight of nine sites were chosen on the leeward sides of

islands and atolls, for both consistency and accessi-

bility. These sites were visited in 2002 (pre-bleach-

ing), and again in 2005 (post-bleaching).

Hard coral cover analysis

The reef platform of the Phoenix Islands is typically

\100 m wide and within a similar distance from the

shoreline. Full descriptions of each island are available

from Obura et al. (2011). Study sites were marked by

GPS position and navigated using distinct coastal and

underwater features. Within sites, coral communities

were surveyed by haphazardly-chosen sampling loca-

tions. Benthic cover was sampled using photo quadrats

collected at a single depth (12–15 m) on the forereef,

along the reef edge. High resolution digital images

(5–8 megapixels) were collected with a digital camera

held *60 cm above the substratum with the image

plane parallel to the surface. To analyze images, 25

points were chosen from an even 5 9 5 grid and coral

cover was determined at each point. Four images were

aggregated (100 points) as a single replicate (Obura

and Mangubhai 2011), reflecting a compromise

between the size of a sample unit (4 m2) and grain/

resolution of photoquadrats (25 pts per image). In

2002, five replicates per site were used, while in 2005,

Rev Fish Biol Fisheries (2014) 24:505–518 507

123

ten replicates were used per site. Hard coral cover

surveys were analyzed between years and sites with a

two-way analysis of variance (ANOVA). More

detailed analyses of benthic change can be found in

Obura and Mangubhai (2011). Both coral and fish

surveys were done at the same sites shown in Fig. 1.

Family-level fish surveys

At each site, 6–12 circular point counts were

conducted with a 7 m radius (area totaling

*150 m2 each) to measure fish abundance (as in

Wilson et al. 2010b). Data were gathered mainly at

the fish family level, with some notes subfamilies

(genus and species were not quantified in point

counts), by the same observer in both 2002 and

2005. Family-level resolution was chosen for long-

term monitoring given the needs of Kiribati at the

time, consistent with guidance under the Global

Coral Reef Monitoring Network (Salvat 2002;

English et al. 1997). Point counts were separated

by at least 15–20 m underwater. In each replicate,

the abundances of fishes in 13 key fish families were

Fig. 1 Map of the Kiribati

Phoenix Islands showing the

location of the permanent

monitoring study sites

examined in this study. The

small black square

represents the boundary of

the Phoenix Islands

Protected Area, created in

2006. Black circles on each

island map denote sampling

sites. Note the changing

scale bars on each island

graph. Maps created by

K. Lagueux


123

recorded—Acanthuridae, Balistidae, Carangidae, Chae-

todontidae, serranid subfamily Epinephelinae, Haemu-

lidae, Labridae, Lethrinidae, Lutjanidae, Pomacanthi-

dae, Scaridae and Sphyraenidae; and elasmobranchs.

One family was counted at a time, focusing on the

faster moving and shyer species first (e.g. Carangidae,

Scaridae), and then counting the more sedentary

fishes (e.g. serranid subfamily Epinephelinae) that

tended to hide in the reef matrix. Other serranids

were not included in surveys. Faster swimming,

highly mobile fish that traversed each transect were

also included. The time taken to complete a count

varied depending on the number of fish present, but

ranged from 3 to 10 min. To reduce positive skew

due to either rare occurrences or dense schools of

mobile and transient fishes, abundances were

log(x ? 1) transformed. In order to estimate the

within-site variation, we considered point counts as

the statistical unit of replication.

A two-way ANOVA with site as a random factor

was used to examine year and site effects within fish

families on log(x ? 1) transformed data. Changes

across all fish families across sites were also explored

using a two-way ANOVA, looking at the mean

abundance of all fishes per site from 2002 and 2005.

We conducted post hoc power analyses assuming an

effect size of 0.8, an alpha of 0.05, and inputted the

calculated standard deviation for each taxonomic

group. For all between-year comparisons, there was

enough power to detect a difference ([0.8) given our

sample size.

Exploring hypotheses for changes in fish

abundance

To explore hypotheses that might explain temporal

changes in fish abundance, we examined the effects of

island type (lagoonal atoll versus non-lagoonal island)

and fish family abundance for the 3 families that

showed significant year effects (Table 1); we used a

fully-fixed two-way ANOVA model on log(x ? 1)

transformed data to satisfy the assumptions of nor-

mality. For the same fish families, we also examined

how fish abundance changed as a result of percent

change in hard coral cover using a linear regression on

log-transformed data.

Table 1 Two-way ANOVA of transformed fish family abun-

dances, exploring site (random), year (fixed), and site*year

interaction effects

Fish Family Effect SS F df p

Carangidae Site 14.262 6.74 8 0.0070

Year 5.288 20.02 1 0.0020

Site*Year 2.116 1.19 8 0.3116

Elasmobranchs Site 0.886 2.37 8 0.1215

Year 0.005 0.11 1 0.7545

Site*Year 0.373 2.61 8 0.0116

Epinephelinae Site 1.060 1.61 8 0.2568

Year 0.561 6.84 1 0.0304

Site*Year 0.657 1.19 8 0.3116

Lutjanidae Site 9.462 3.13 8 0.0637

Year 0.302 0.80 1 0.3961

Site*Year 3.026 3.14 8 0.0030

Chaetodontidae Site 3.039 2.92 8 0.0752

Year 2.641 20.43 1 0.0019

Site*Year 1.039 1.94 8 0.0601

Acanthuridae Site 2.549 0.88 8 0.5701

Year 0.263 0.73 1 0.4176

Site*Year 2.901 4.73 8 \ 0.0001

Scaridae Site 3.852 0.78 8 0.6332

Year 0.003 0.01 1 0.9422

Site*Year 4.939 4.68 8 \ 0.0001

Balistidae Site 9.193 2.71 8 0.0903

Year 1.227 2.91 1 0.1260

Site*Year 3.397 3.38 8 0.0016

Labridae Site 0.524 0.70 8 0.6839

Year 0.342 3.70 1 0.0903

Site*Year 0.818 1.87 8 0.07

Lethrinidae Site 0.332 0.58 8 0.7710

Year 0.233 3.26 1 0.1078

Site*Year 0.573 1.01 8 0.4360

Pomacanthidae Site 2.119 2.48 8 0.1100

Year 0.110 1.04 1 0.3382

Site*Year 0.854 1.65 8 0.1174

Haemulidae Site 0.052 1.00 8 0.500

Year 0.019 2.91 1 0.1258

Site*Year 0.052 1.70 8 0.1056

Sphyraenidae Site 0.257 2.10 8 0.1567

Year 0.006 0.04 1 0.8491

Site*Year 0.122 1.01 8 0.4340

Haemulidae were absent in all surveys in 2005

Italics text denotes significant differences


123


123

Results

Benthic response to thermal stress

In 2002, the highest mean coral cover was found at

Weird Eddie (83 %), Lone Palm (77 %), and Satellite

Beach (71 %), while Algae Corner had the lowest mean

coral cover (10 %), as reported in Obura and Mangubhai

(2011). Following substantial coral mortality in 2005,

the highest mean coral cover was recorded at Lone Palm

(43 %) and Deepwater (38 %), representing almost half

of their 2002 means. The greatest coral cover losses

were recorded on the three atolls: on Kanton, Weird

Eddie and Satellite Beach lost 97 and 87 % cover,

respectively; on Nikumaroro cover declined by 80 % at

Amelia’s Lost Causeway and 97 % on Windward Wing,

and on Orona, Dolphin Ledge lost 92 % live coral cover.

There was a significant year effect (F = 444.9,

df = 1,1, p \ 0.001), site effect (F = 20.07, df = 8,8,

p \ 0.001) and year*site interaction (F = 11.47,

df = 8,8, p \ 0.001) comparing hard coral cover over

time (two-way ANOVA, F = 44.06, df = 17, 149,

p \ 0.001), which explained most of the variation in the

data (R2 = 0.82). At all sample sites, there was a

significant decrease in hard coral cover (Fig. 2 insets).

Differences in fish family abundance between sites

We found significant site effects for Carangidae, and

Chaetodontidae (Table 1) and significant year effects for

Carangidae, Chaetodontidae, and Epinephelinae

(Table 1). Five families showed significant site by year

interactions that did not correspond with significant site

or year effects alone. The most abundant fish family in

both years was Acanthuridae; but the average abundance

across all sites did not change pre- and post- bleaching

(t = 0.17, df = 131, p = 0.86); in 2002 abundance

averaged 23.7 fish per 154 m2; in 2005, 24.5 fish per

154 m2 for each point count. A two-way ANOVA

revealed a significant year effect, whereby the overall

abundance of all fish combined significantly declined

from 2002 to 2005 (F = 8.765, df = 1,1 p = 0.0038).

However, there was also significant site-to-site variabil-

ity (Fig. 3; F = 3.484, df = 8,8, p = 0.0014) and a

significant year by site interaction (F = 3.346, df = 8,8,

p = 0.002), since not all sites showed a decline in mean

abundance of all fish combined. This difference was

maintained across sites (Fig. 3; paired t = 3.41, df = 8,

p = 0.0093), with all sites declining in fish abundance

between 2002 and 2005 except for Windward Wing.

Exploring possible causes for changes in fish

abundance using targeted fish families

To test hypotheses that might explain the differences

in fish abundances over time, we explored the effect of

island type (atoll versus island), and change in percent

coral cover in the subset of three fish families that

changed over time regardless of site (Fig. 4; Table 2).

For Epinephelinae and all fish, island type did

not explain the differences between years as there

were no island type or island type * year interactions

(Table 2A). However, there was a significant island

type effect and/or a significant island type*year

interaction for Carangidae and Chaetodontidae

(Table 2A). Correspondingly, abundances of Chae-

todontidae varied significantly with declines in coral

cover (Table 2B; Fig. 4). Although the decline of live

coral cover was observed at all sites (Fig. 2, insets), no

other significant relationships between fish family

abundance and coral decline were found.

Fig. 3 Mean abundance of all fish families shown by site,

between years from 2002 and 2005

Fig. 2 Mean abundance of fish families (±standard error) per

transect (154 m2) organized by permanent monitoring sites in

the Phoenix Islands in 2002 (black bars) and 2005 (white bars).

O Orona Atoll, N Nikumaroro Island, M Manra Atoll, K Kanton

Atoll, E Enderbury Island, R Rawaki Island. Exposure is

denoted as L or W for Leeward or Windward respectively,

Habitat is denoted as A or I for atoll or island. Results from

Haemulidae and Sphyraenidae families were not presented

because of their low density. Upper right inset depicts mean

(±standard error) percent hard coral cover for 2002 in black

(n = 10 transects), and for 2005 in white (n = 5 transects)

b


123

Discussion

This paper aims to describe family-level changes in fish

abundance over time from a relatively isolated and near-

pristine reef; it is among a growing number of studies

that use isolated reefs as benchmarks for benthic

degradation and recovery, and resulting changes in

community structure (e.g. Halford and Caley 2009;

Chong-Seng et al. 2012; Wernberg et al. 2012a, b;

Gilmour et al. 2013). This paper contributes several

Fig. 4 Exploration of factors (habitat and coral cover) that may

have influenced post-bleaching fish abundance (per transect; or

per 154 m2) for families that experienced a significant change

between 2002 (black bars) and 2005 (white bars), and for all fish

(representing 13 families)


123

confirming insights into how fish families on near-

pristine reefs change in the short-term, following a

severe coral bleaching event. Firstly, there is dramatic

variation in site-level response. Because the Phoenix

Islands are isolated from local anthropogenic impacts,

site-to-site variation is likely indicative of naturally-

varying factors such as nutrient availability, benthic

structure, oceanographic features, recruitment dynam-

ics and/or population connectivity. Despite the baseline

differences pre-bleaching (in 2002), all sites experi-

enced declines in coral cover in 2005, suggesting that

despite this variation, no sites were immune to severe

thermal stress. In contrast, mean total fish abundance

varied from site-to-site, and significant site-specific

family-level changes in fish community composition

were observed. The documentation of these site- and

family-specific differences in the Phoenix Islands is the

main contribution of this paper.

Reef response over time

Consistent with a large and global body of literature

dedicated to the consequences of increased ocean

temperatures (e.g. Hoegh-Guldberg et al. 2007), corals

of the Phoenix Islands bleached and experienced

subsequent mortality following extreme thermal stress

(Fig. 2 insets) (Alling et al. 2007; Obura et al. 2011;

Obura and Mangubhai 2011). It is important to note,

however, that this thermal stress was among the largest

reported for any reef worldwide, at any time, peaking at

21 degree heating weeks (Obura and Mangubhai 2011).

Extreme and discrete thermal stress events are increas-

ing in frequency and magnitude (Kerr 2011), and have

been shown to cause ecological changes such as

diversity loss via shifting species distributions (Smale

and Wernberg 2013). However, even with such thermal

severity, we observed site-to-site variation in coral

cover decline (Fig. 2 insets; see also Table 3 and Fig. 3

in Obura and Mangubhai 2011), which is consistent

with previous studies indicating site- or species-level

variation in response to severe bleaching due to

differences in reef habitat (Grimsditch et al. 2010),

species resilience (Pratchett et al. 2010), and/or

repeated thermal exposure (Williams et al. 2010). Such

variation suggests that while extreme stress events can

cause ecological change, such change is not necessarily

Table 2 Exploring hypotheses for changes in fish family abundance between 2002 and 2005

Fish family A. Habitat (Island vs. Atoll) B. Coral Cover (% change)

Effect SS F df p SS F df P

Carangidae Habitat 1.514 4.67 1 0.033

Year 4.346 13.4 1 0.0004 Model 0.003 5.19 1 0.072

Habitat*Year 0.082 0.25 1 0.616

Error 41.85 Error 0.006 5

Chaetodontidae Habitat 0.238 2.79 1 0.097

Year 0.790 9.27 1 0.003 Model 1.451 6.45 1 0.039

Habitat*Year 0.632 7.42 1 0.007


Epinephelinae Habitat 0.0001 0.003 1 0.959

Year 0.208 2.79 1 0.097 Model 0.046 0.379 1 0.379

Habitat*Year 0.024 0.33 1 0.568


All Fish Habitat 0.343 1.44 1 0.230

Year 0.773 3.25 1 0.072 Model 0.001 0.003 1 0.956

Habitat*Year 0.858 3.61 1 0.058


(A) Two-way ANOVA of log-transformed fish abundances exploring habitat (island versus atoll) and year effects for each family

identified as having a significant decrease over time. (B) Linear regression of log-transformed changes in fish abundance and coral

cover (% change over time)

For both (A) and (B), all fish includes all fish families surveyed


123

uniform and is highly influenced by interacting local

factors.

In this study, we found that overall, total fish

abundance and live coral cover both declined from

2002 to 2005 and family- and site- specific changes in

the fish community over time were observed. Similar to

other bleaching events and non-structural reef distur-

bances (e.g. Pratchett et al. 2010), the 2002 to 2003

bleaching event in the Kiribati Phoenix Islands caused

massive tissue mortality but left coral skeletons largely

intact. Thus, at least for the short-term, topographic

complexity was not immediately compromised, which

may have maintained higher organismal diversity and

abundance compared to disturbances where topographic

complexity and heterogeneity were immediately

reduced (Sano et al. 1987; Graham et al. 2006). Along

these lines, we found that many families of fish did not

significantly change in abundance despite the severe

thermal event and corresponding coral loss (Fig. 2;

Table 1) Over longer periods of time, coral mortality

leaves the reef vulnerable to erosion, facilitating the loss

of habitat structure (Sheppard et al. 2002; Graham et al.

2006), which may be a driver of reef fish decline over the

longer term (Wilson et al. 2006; Garpe et al. 2006).

Changes in fish family abundance over time

We examined 13 fish families across 9 sites in 2002

and 2005, and found marked variability in abundance

across all. Three of the fish families declined, regard-

less of site (Carangidae, Chaetodontidae, and serranid

subfamily Epinephelinae). Despite major coral mor-

tality, Lethrinidae, Labridae, Sphyrinidae, and Pom-

acanthidae did not differ in abundance across sites or

years. Such consistent abundances may be tied to their

natural history; for example, species within Lethrin-

idae often prefer sandy and/or rubble habitats and are

often omnivorous, displaying trophic flexibility that

may allow stable abundances even in variable envi-

ronments. The most abundant fish family pre- and

post-bleaching was Acanthuridae, of which many

species are widely regarded as important herbivores

that contribute to overall reef resilience (e.g. Burkepile

and Hay 2008). Our findings are consistent with a

previous study that also found little/no change in

relative Acanthurid abundance over time (Halford and

Caley 2009). We found a decline in Chaetodontidae

over time, which is consistent with post-bleaching

decline observed at other locations (Bozec et al. 2005;

Wilson et al. 2008; Halford and Caley 2009; Pratchett

et al. 2011).

Examining hypotheses to explain differences

in fish family abundance

We explored two hypotheses that might explain the

significant declines of Carangidae, Chaetodontidae, and

subfamily Epinephelinae abundance post-bleaching:

the impact of island type and/or changes in coral cover.

For Epinephelinae, neither of these hypotheses ade-

quately explained the observed changes in abundance at

a family level. These fish families may have a more

nuanced response to bleaching than has previously been

appreciated, perhaps related to other habitat character-

istics, or their trophic, species-specific, reproductive, or

behavioral characteristics—all of which could be

examined with finer taxonomic resolution.

Changes in the abundance of fishes in the family

Chaetodontidae over time were most dramatic on atolls,

and were also associated with changes in live coral

cover. Atolls may have a significant impact on coral

cover, since they have been shown to be more sensitive

to thermal stress and bleaching severity as compared to

non-lagoonal islands (Obura and Mangubhai 2011).

Similarly, leeward sites near lagoonal waters also

experienced high mortality, especially on Kanton and

Nikumaroro. The highest mortality rates occur closest

to the mouth and decrease with distance, suggesting that

these atoll lagoons influence water quality and nutrient

characteristics, which may have downstream impacts

on coral resilience, recovery, and growth. Pre-bleach-

ing, there was a high density of corals in shallow

lagoons, particularly of large table Acropora colonies.

Anecdotally, butterflyfishes commonly found in

2002—including the obligate corallivore Chaetodon

trifascialis—were only rarely sighted in 2005 post-

bleaching. The combination of habitat and nutritional

specificity likely explains our finding that Chaetodonti-

dae abundance differed by island type.

In addition to location considerations, changes in

Chaetodontid abundance over time were also associ-

ated with declines in live coral cover. Chaetodontidae

feed primarily on living tissue from scleractinian and

alcyonacean corals (Anderson et al. 1981); conse-

quently, their abundance and performance are usually

tied to live coral cover (Findley and Findley 1985;

Williams 1986; Pratchett et al. 2004; Bozec et al.

2005). Pre-bleaching, we found that Chaetodontidae


123

abundance was high, likely reflective of both coral

community composition and/or coral nutritional value.

Post-bleaching, we found a decline in Chaetodontidae

abundance across sites, consistent with previous studies

that found obligate coral-feeding fish declined substan-

tially following a severe bleaching event (Graham et al.

2006) or a severe decrease in live coral cover (Bozec

et al. 2005). Chaetodontids comprise the highest

number of species of obligate corallivores for any fish

family (Cole et al. 2008; Rotjan and Lewis 2008), and

our findings are consistent with a recent study that

flagged obligate corallivores as being especially sensi-

tive to climate vulnerability, displaying the highest risk

of local extinction compared to other functional groups

(Graham et al. 2011).

In contrast to Chaetodontidae, changes in the

abundance of fishes in the family Carangidae over

time were most dramatic on islands, and were not

associated with changes in live coral cover. Many

species in the family Carangidae are known to be

highly mobile, and are not known to be among the fish

families that are tightly coral-associated (Pratchett

et al. 2008). Variation in fish counts has been

acknowledged in the literature (e.g. McClanahan

et al. 2007), and such variation is likely to be

augmented in highly mobile species. Our data also

show high site-to-site variation in Carangidae abun-

dance (Fig. 2), suggesting that there may be more

nuanced site-level drivers of Carangidae abundance.

Other potential drivers of changing fish abundance

include nutrient availability, other aspects of benthic

structure, oceanographic features, recruitment dynam-

ics and/or population connectivity. We found a variety

of interactions between site and year for other fish

families that could be explained by some of these

alternative drivers. Furthermore, the significant time

effect that we found for Epinephelinae is indicative of

other factors at play, given that neither island type nor

changes in hard coral cover explained the observed

differences over time. Given the opportunistic nature

of this dataset, a comprehensive analysis of each factor

was not possible, but should be a priority for the future.

Taxonomic approach

We focused primarily on fish abundance and diversity

organized by taxonomic family, as the initial purpose

for monitoring was to establish long term monitoring

stations for national purposes consistent with guidance

under the Global Coral Reef Monitoring Network

(Salvat 2002; English et al. 1997). This ‘‘higher

taxon’’ approach is considered sufficient for some

systems such as birds, intertidal molluscs, plants, and

insects (e.g. Gaston and Williams 1993; Kallimanis

et al. 2012; Gladstone and Alexander 2005). We found

this approach to be adequate in determining broad

patterns in response to bleaching in a logistically-

constrained location, and our results are similar to

other recent studies assessing the influence of benthic

condition on reef fish assemblages (reviewed by

Pratchett et al. 2011). However, there are clear

limitations to family level data (Green and Bellwood

2009; Obura and Grimsditch 2009). For example, it is

known that high species diversity and abundance

within functional groups helps to reduce the impact of

disturbance at the community level (Schmitz et al.

2000). The species diversity within family is known

for the Phoenix Islands (Allen and Bailey 2011), and

assessing the ecological, species-specific contribu-

tions to reef resilience and/or recovery is a priority for

future studies (Wilson et al. 2010a).

Abundance, diversity, and biomass changes

in response to bleaching

Assessing changes in fish diversity or biomass in

response to disturbance is a growing area of research

and is far from straightforward. For example, Wilson

et al. (2009) found that reef fish species richness was

maintained in 9 out of 10 sites on the Great Barrier

Reef, despite frequent disturbances and subsequent

coral declines of 46–96 %. Thus, patchy and/or severe

changes to the benthos do not necessitate immediate

losses in fish species richness (Wilson et al. 2009) or

uniform family-level loss of abundance (this study).

Despite a non-uniform response, there is an abundant

literature on how habitat degradation drives fish

community changes (e.g. Feary et al. 2009; Graham

et al. 2008; Wilson et al. 2008; Pratchett et al. 2011;

Chong-Seng et al. 2012) or local human influence (e.g.

Wilson et al. 2006; DeMartini et al. 2008). Taken

together, it seems that fishing pressure, coastal devel-

opment, and/or habitat loss are all important contrib-

utors to overall reef fish decline; the major

contribution of this study is to demonstrate how, at

least in the short-term, site-to-site variation in habitat

degradation can impact fish family abundance

changes, even in the absence of other pressures.


123

Long-term impacts

While this study examined the response of fish families

over the short-term (3-years) following severe bleach-

ing in the absence of local human influence, long-term

impacts remain to be investigated. There are several

delayed ecosystem responses that may help to promote

or prevent reef recovery, and the fish community

composition may change as a result (Graham et al.

2008). For example, coral-associated juvenile fish may

suffer consequences from bleaching with a resulting lag

in adult fish decline; previous studies have shown that

coral mortality can decrease settlement rates for up to

65 % of reef fishes (Jones et al. 2004). Similarly,

McCormick et al. (2010) found that 10 times as many

fish settled to healthy coral compared to sub-lethally

bleached coral. Thus, following coral mortality, long-

term recruitment may be hindered even though few

immediate effects are observed in the adult population.

Long-term recovery may also be hindered by repeated

or ongoing disturbances, with resulting consequences

for the community of coral-dependent fishes (Berumen

and Pratchett 2006).

The Phoenix Islands present a unique opportunity

to assess global disturbance events in the absence of

local human impacts. During the time of this study

(2002–2005), the Phoenix Islands were not yet a

marine protected area and species-specific monitoring

was not yet implemented. However, the reefs were

considered a ‘‘de facto’’ marine protected area due to

its extreme isolation. We are reasonably confident that

any change in Phoenix Islands reef fish assemblages

between 2002 and 2005 was likely due directly to

severe thermal stress, or indirectly due to the cascad-

ing impact of thermal stress on related habitat.

However, the possibility remains that changes in fish

abundances (or lack of) were completely independent,

and instead reflect natural variation. The isolation and

protection of the Phoenix Islands should hopefully

shed light on the natural variation of fish families

during intervals without major thermal stress.

The establishment of the Phoenix Islands Protected

Area (PIPA) so soon post-bleaching will ensure that

the future trajectory of PIPA reefs should be unim-

paired by fishing-induced loss of fish diversity,

abundance or biomass. However, continued thermal

stress and associated habitat loss will continue to be

important drivers of PIPA reef status. A 2010 bleach-

ing event in the U.S. Phoenix Islands (Vargas-Angel

et al. 2011) and corresponding high temperature event

in the Kiribati Phoenix Islands (Mangubhai et al.

2012) will likely impact the recovery trajectory of the

Phoenix Islands, but will also heighten the importance

of this region for the opportunity to examine long-term

reef response to repeated disturbance in the absence of

local human stressors.

Acknowledgments The authors would like to thank Stuart

Sandin and Les Kaufman for their thoughtful comments on this

manuscript. We are grateful to Kerry Lagueux for producing the

maps depicted in Fig. 1, and to Tania Lemos Eskin for drawing

the fish sketches used in Table 2 and Fig. 3. This study is part of

a larger effort examining reef recovery and resilience processes

on isolated atoll reefs in the central Pacific; this effort has been

funded and supported by the New England Aquarium, the

National Geographic Society, Conservation International

(Global Conservation Fund), the Government of Kiribati, the

Akiko Shiraki Dynner Fund for Ocean Exploration and

Conservation, and the Nai’a (Rob Barrel and Cat Holloway).

Findings from this manuscript were presented at the World

Fisheries Congress in 2012.

References

Allen GR (2007) Conservation hotspots of biodiversity and

endemism for Indo-Pacific coral reef fishes. Aquatic Con-

serv Mar Fresh Eco 18:541–556

Allen GR, Bailey S (2011) Reef fishes of the Phoenix Islands,

Central Pacific Ocean. Atoll Res Bull 589:83–118

Alling A, Doherty O, Logan H, Feldman L, Dustan P (2007)

Catastrophic coral mortality in the remote central Pacific

Ocean: Kirabati (sic) Phoenix Islands. Atoll Res Bull 551:

1–18

Anderson GRV, Ehrlich AH, Ehrlich PR, Roughgarden JD,

Russell BC, Talbot FH (1981) Community Struct Coral

Reef Fishes. Am Nat 117(4):476–495

Balmford A, Bennun L, Ten Brink B (2005) The convention on

biological diversity’s 2010 target. Science 307:212–213

Berumen ML, Pratchett MS (2006) Recovery without resilience:

persistent disturbance and long-term sifts in the structure of

fish and coral communities at tiahura Reef, Moorea. Coral

Reefs 25:647–653

Bonin MC, Munday PL, McCormick MI, Srinivasan M, Jones GP

(2009) Coral-dwelling fishes resistant to bleaching but not to

mortality of host corals. Mar Ecol Prog Ser 394:215–222

Bozec YM, Doledec S, Kulbicki M (2005) An analysis of fish-

habitat associations on disturbed coral reefs: chaetodontic

fishes in New Caledonia. J Fish Biol 66:966–982

Burkepile DE, Hay ME (2008) Herbivore species richness

and feeding complementarity affect community structure

and function on a coral reef. Proc Natl Acad Sci 105:

16201–16206

Chong-Seng KM, Mannering TD, Pratchett MS, Bellwood DR,

Graham NAJ (2012) The influence of coral reef benthic

condition on associated fish assemblages. PLoS ONE

7:e42167


123

Christensen N, Bartuska A, Brown J, Carpenter S, Da C, Francis

R, Franklin JF, MacMahon JA, Noss RF, Parsons DJ,

Peterson CH, Turner MG, Woodmansee RG (1996) The

report of the ecological society of America committee on

the scientific basis for ecosystem management. Ecol App

9:1266–1277

Cole AJ, Pratchett MS, Jones GP (2008) Diversity and func-

tional importance of coral-feeding fishes on tropical coral

reefs. Fish Fish 9(3):286–307

DeMartini EE, Friedlander AM, Sandin SA, Sala E (2008)

Differences in fish-assemblage structure between fished

and unfished atolls in the northern Line Islands, central

Pacific. Mar Ecol Prog Ser 365:199–215

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

tropical marine resources 2nd edn. Australian Institute of

Marine Sciences ISBN 0 642 2594 4

Feary DA, McCormick MI, Jones GP (2009) Growth of reef

fishes in response to live coral cover. J Exp Mar Biol Ecol.

doi:10.1016/j.jembe.2009.03.002

Findley JS, Findley MT (1985) A search for pattern in butterf-

lyfish communities. Am Nat 126:800–816

Garpe KC, Yahya SAS, Lindahl U, Ohman MC (2006) Long-

term effects of the 1998 coral bleaching even on reef fish

assemblages. Mar Ecol Prog Ser 315:237–247

Gaston KJ, Williams PH (1993) Mapping the world’s species -

the higher taxon approach. Biodivers Lett 1:2–8

Gilmour JP, Smith LD, Heyward AJ, Baird AH, Pratchett MS

(2013) Recovery of an isolated coral reef system following

severe disturbance. Science 340:69–70

Gladstone W, Alexander T (2005) A test of the higher-taxon

approach in the identification of candidate sites for marine

reserves. Biodivers Conserv 14:3151–3168

Graham NAJ, Wilson SK, Jennings S, Polunin NVC, Bijoux JP,

Robinson J (2006) Dynamic fragility of oceanic coral reef

ecosystems. Proc Nat Acad Sci 103:8425–8429

Graham NAJ, McClanahan TR, MacNeil MA, Wilson SK,

Polunin NVC et al (2008) Climate warming, marine pro-

tected areas and the ocean-scale integrity of coral reef

ecosystems. PLoS ONE 3(8):e3039. doi:10.1371/journal.

pone.0003039

Graham NAJ, Chabanet P, Evans RD, Jennins S, Letourner Y,

MacNeil MA, McClanahan TR, Ohman MC, Polunin

NVC, Wilson SK (2011) Extinction vulnerability of coral

reef fishes. Ecol Lett 14:341–348

Green AL, Bellwood DR (2009) Monitoring functional groups

of herbivorous reef fishes as indicators of coral reef resil-

ience – A practical guide for coral reef managers in the

Asia Pacific region. IUCN working group on climate

change and coral reefs, IUCN, Gland, Switzerland, 70 pp

Grimsditch G, Mwaura JM, Kilonzo J, Amiyo N (2010) The

effects of habitat on coral bleaching responses in Kenya.

Ambio 39:195–304

Halford AR, Caley MJ (2009) Towards an understanding of

resilience in isolated coral reefs. Glob Change Biol 15:

3031–3045

Hoegh-Guldberg O, Mumby PJ, Hooten AJ, Steneck RS,

Greenfield P, Gomez E, Harvell CD, Sale PF, Edwards AJ,

Caldeira K, Knowlton N, Eakin CM, Iglesias-Prieto R,

Muthiga NA, Bradbury RH, Dubi A, Hatziolos ME (2007)

Coral reefs under rapid climate change and ocean acidifi-

cation. Science 318(5857):1737–1742

Jackson JBC (2001) What was natural in the coastal oceans?

Proc Nat Acad Sci 98(10):5411–5418

Jones GP, Syms C (1998) Disturbance, habitat structure and the

ecology of fishes on coral reefs. Aust J Ecol 23:287–297

Jones GP, McCormick MI, Srinivasan M, Eagle JV (2004) Coral

decline threatens fish biodiversity in marine reserves. Proc

Nat Acad Sci 101:8251–8253

Kallimanis AS, Mazaris AD, Tsakanikas D, Dimopoulos P, Pantis

JD, Sgardelis SP (2012) Efficient biodiversity monitoring:

which taxonomic level to study? Ecol Indic 15:100–104

Kerr RA (2011) Humans are driving extreme weather; time to

prepare. Science 334:1040

Lieske E, Myers R (1994) Coral reef fishes: Indo-Pacific and

Caribbean. Harper Collins, London

Lindahl U, Ohman MC, Schelten CK (2001) The 1997/1998

mass mortality of corals: effects on fish communities on a

Tanzanian coral reef. Mar Poll Bull 42(2):127–131

Mangubhai S, Rotjan RD, Obura DO (2012) Phoenix Islands

Protected Area 2012 Expedition Report. New England

Aquarium, Boston, MA, USA, 44 p

McClanahan TR, Graham NAJ, Maina J, Chabanet P, Brugge-

mann JH, Polunin NVC (2007) Influence of instantaneous

variation on estimates of coral reef fish populations and

communities. Mar Ecol Prog Ser 340:221–234

McCormick MI, Moore JAY, Munday PL (2010) Influence of

habitat degradation on fish replenishment. Coral Reefs

29:537–546

Munday PL, Jones GP, Pratchett MS, Williams AJ (2008) Cli-

mate change and the future for coral reef fishes. Fish Fish

9:261–285

Obura DO, Grimsditch G (2009) Resilience assessment of coral

reefs – Assessment protocol for coral reefs, focusing on

coral bleaching and thermal stress. IUCN working group

on climate change and coral reefs, IUCN, Gland, Swit-

zerland, 70 pp

Obura DO, Mangubhai S (2011) Coral mortality associated with

thermal fluctuations in the Phoenix Islands, 2002-2005.

Coral Reefs 30:607–619

Obura DO, Stone GS (2002) Phoenix Islands: Summary of

marine and terrestrial assessments conducted in the

Republic of Kiribati: June 5-July 10, 2002. New England

Aquarium, Boston

Obura D, Stone G, Mangubhai S, Bailey S, Yoshinaga A, Hol-

loway C, Barrel R (2011) Baseline marine biological sur-

veys of the Phoenix Islands, July 2000. Atoll Res Bull

589:1–62

Pandolfi JM, Bradbury RH, Sala E, Hughes TP, Bjorndal KA,

Cooke RG, McArdle D, McClenachan L, Newman M,

Paredes G, Warner RR, Jackson JBC (2003) Global tra-

jectories of the long-term decline of coral reef ecosystems.

Science 301:955–958

Petraitis PS, Latham RE, Niesenbaum RA (1989) The mainte-

nance of species diversity by disturbance. Q Rev Biol

64:393–418

Pickett STA, White PS (eds) (1986) The ecology of natural

disturbance and patch dynamics. Academic Press, Ontario

Pratchett MS, Wilson SK, Berumen ML, McCormick MI (2004)

Sublethal effects of coral bleaching on an obligate coral

feeding butterflyfish. Coral Reefs 23(3):352–356

Pratchett MS, Munday PL, Wilson SK, Graham NAJ, Cinner JE,

Bellwood DR, Jones GP, Polunin NVC, McClanahan TR


123

http://dx.doi.org/10.1016/j.jembe.2009.03.002

http://dx.doi.org/10.1371/journal.pone.0003039

http://dx.doi.org/10.1371/journal.pone.0003039

(2008) Effects of climate-induced coral bleaching on coral-

reef fishes: ecological and economic consequences. Oce-

anogr Mar Biol Ann Rev 46:251–296

Pratchett MS, Trapon N, Berumen ML, Chong-Seng K (2010)

Recent disturbances augment community shifts in coral

assemblages in Moorea, French Polynesia. Coral Reefs

online first. doi:10.1007/s00338-00010-00678-00332

Pratchett MS, Hoey AS, Wilson SK, Messmer V, Graham NAJ

(2011) Changes in the biodiversity and functioning of reef

fish assemblages following coral bleaching and coral loss.

Diversity 3:424–452

Rotjan RD, Lewis SM (2008) Impact of coral predators on

tropical reefs. Mar Ecol Prog Ser 367:73–91

Salvat B (2002) Status of southeast and central Pacific coral

reefs ‘Polynesia Mana Node’: Cook Islands, French Poly-

nesia, Kiribati, Niue, Tokelau, Tonga, Wallis and Futuna.

In: Wilkinson C (ed) Status of coral reefs of the world:

2002. Australian Institute of marine Science, Townsville,

pp 203–216

Sandin SA, Smith JE, deMartini EE, Dinsdale EA, Donner SD,

Friedlander AM, Konotchick T, Malay M, Maragos JE,

Obura D, Pantos O, Paulay G, Ritchie M, Rohwer F,

Schroeder RE, Walsh S, Jackson JBC, Knowlton N, Sala E

(2008) Baslines and degredation of coral reefs in the

Northern Line Islands. PLoS ONE 3(2):w1548

Sano M, Shimizu M, Nose Y (1987) Long-term effects of

destruction of hermatypic corals by Acanthaster planci

infestation on reef fish communities at Iriomote Island,

Japan. Mar Ecol Prog Ser 37:191–199

Schmitz OJ, Hamback PA, Beckerman AP (2000) Trophic

cascades in terrestrial systems: a review of the effects of

carnivore removals on plants. Am Nat 155:141–153

Sheppard CRC, Spalding MD, Bradshaw C, Wilson SK (2002)

Erosion vs. recovery of coral reefs after 1998 El Nino:

Chagos reefs. Ambio 31:40–48

Smale DA, Wernberg T (2013) Extreme climatic event drives

range contraction of a habitat-forming species. Proc Roy

Soc B 280:20122829

Sousa WP (1984) The role of disturbance in natural communi-

ties. Ann Rev Ecol Sys 5:353–391

Thomas CD, Cameron A, Green RE (2004) Extinction risk from

climate change. Nature 416:389–395

Vargas-Angel B, Looney EE, Vetter OJ, Coccagna EF (2011)

Severe, widespread El Nino associated coral bleaching in

the US Phoenix Islands. Bull Mar Sci 87:623–638

Wernberg T, Smale DA, Thomsen MS (2012a) A decade of cli-

mate change experiments on marine organisms: procedures,

patterns, and problems. Glob Change Biol 18:1491–1498

Wernberg T, Smale DA, Tuya F, Thomsen MS, Langlois TJ,

deBettignes T, Bennett S, Rousseaux CS (2012b) An

extreme climatic event alters marine ecosystem structure in

a global biodiversity hotspot. Nat Clim Chang 3:78–82

Williams DM (1986) Temporal variation in the structure of reef

slope fish communities (central great barrier reef): short-

term effects of Acanthaster planci infestation. Mar Ecol

Prog Ser 28:157–164

Williams GJ, Knapp IS, Maragos JE, Davy SK (2010) Modeling

patterns of coral bleaching at a remote Central Pacific atoll.

Mar Poll Bull 60:1467–1476

Wilson SK, Graham NAJ, Pratchett MS, Jones GP, Polunin

NVC (2006) Multiple disturbances and the global degre-

dation of coral reefs: are reef fishes at risk or resilient? Glob

Change Biol 12:2220–2234

Wilson SK, Burgess S, Cheal A, Emslie M, Fisher R, Miller I,

Polunin NVC, Sweatman HPA (2008) Habitat utilisation

by coral reef fish: implications for specialists vs. generalists

in a changing environment. J Anim Ecol 77:220–228

Wilson SK, Dolman AM, Cheal AJ, Emslie MJ, Pratchett MS,

Sweatman HPA (2009) Maintenance of fish diversity on

disturbed coral reefs. Coral Reefs 28:3–14

Wilson SK, Adjeroud M, Bellwood DR, Berument ML, Booth

D, Bozec Y-M, Chabanet P, Cheal AJ, Cinner J, Dep-

czynski M, Feary DA, Gagliano M, Graham NAJ, Halford

AR, Halpern BS, Harbone AR, Hoey AS, Holbrook SJ,

Jones GP, Kulbiki M, Letourneur Y, De Loma TL,

McClanahan T, McCormick MI, Meekan MG, Mumby PJ,

Munday PL, Ohman MC, Pratchett MS, Riegl B, Sano M,

Schmitt RJ, Syms C (2010a) Crucial knowledge gaps in

current understanding of climate change impacts on coral

reef fishes. J Exp Biol 213:894–900

Wilson SK, Fisher R, Pratchett MS, Graham NAJ, Dulvy NK,

Turner RA, Cakacaka A, Polunin NVC (2010b) Habitat

degredation and fishing effects on the size structure of coral

reef fish communities. Ecol App 20:442–451


123

http://dx.doi.org/10.1007/s00338-00010-00678-00332

Short-term changes of fish assemblages observed in the near-pristine reefs of the Phoenix Islands

Documents