Coral bleaching following wintry weather

Abstract

Extensive coral bleaching Occurred intertidally in early August 2003 in the Capricorn Bunker group (Wistari Reef, Heron and One Tree Islands; Southern Great Barrier Reef). The affected intertidal coral had been exposed to unusually cold (minimum = 13.3degreesC; wet bulb temperature = 9degreesC) and dry winds (44% relative humidity) for 2 d during predawn low tides. Coral bleached in the upper 10 cm of their branches and had less than 0.2 x 10(6) cell cm(-2) as compared with over 2.5 x 10(6), Cell cm(-2) in nonbleached areas. Dark-adapted quantum yields did not differ between symbionts in bleached and nonbleached areas. Exposing symbionts to light, however, led to greater quenching of Photosystem 11 in symbionts in the bleached coral. Bleached areas of the affected colonies had died by September 2003, with areas that were essentially covered by more than 80% living coral decreasing to less than 10% visible living coral cover. By January 2004, coral began to recover, principally from areas of colonies that were not exposed during low tide (i.e., from below dead, upper regions). These data highlight the importance of understanding local weather patterns as well as the effects of longer term trends in global climate.

Full text

265
Limnol. Oceanogr., 50(1), 2005, 265–271
q
2005, by the American Society of Limnology and Oceanography, Inc.
Coral bleaching following wintry weather
Ove Hoegh-Guldberg
1
and Maoz Fine
Centre for Marine Studies, University of Queensland, St Lucia 4072, Queensland, Australia
William Skirving
NOAA/NESDIS/ORA/ORAD—E/RA3, NOAA Science Center, Room 601, 5200 Auth Road, Camp Springs, Maryland
20746-4304
Ron Johnstone and Sophie Dove
Centre for Marine Studies, University of Queensland, St Lucia 4072, Queensland, Australia
Alan Strong
NOAA/NESDIS/ORA/ORAD—E/RA3, NOAA Science Center, Room 601, 5200 Auth Road, Camp Springs, Maryland
20746-4304
Abstract
Extensive coral bleaching occurred intertidally in early August 2003 in the Capricorn Bunker group (Wistari
Reef, Heron and One Tree Islands; southern Great Barrier Reef). The affected intertidal coral had been exposed to
unusually cold (minimum
5
13.3
8
C; wet bulb temperature
5
9
8
C) and dry winds (44% relative humidity) for 2 d
during predawn low tides. Coral bleached in the upper 10 cm of their branches and had less than 0.2
3
10
6
cell
cm
22
as compared with over 2.5
3
10
6
cell cm
22
in nonbleached areas. Dark-adapted quantum yields did not differ
between symbionts in bleached and nonbleached areas. Exposing symbionts to light, however, led to greaterquench-
ing of Photosystem II in symbionts in the bleached coral. Bleached areas of the affected colonies had died by
September 2003, with areas that were essentially covered by more than 80% living coral decreasing to less than
10% visible living coral cover. By January 2004, coral began to recover, principally from areas of colonies that
were not exposed during low tide (i.e., from below dead, upper regions). These data highlight the importance of
understanding local weather patterns as well as the effects of longer term trends in global climate.
The majority of Scleractinian coral live in a mutualistic
endosymbiosis with single-celled dinoflagellate algae of the
genus Symbiodinium. Together, these two organisms are re-
sponsible for a major component of the structure (frame-
work) and function (primary productivity) of tropical reef
systems. In recent years, the abundance of coral colonies on
reefs worldwide has been rapidly declining under the pres-
sure of human-derived stresses (Bryant et al. 1998; Wilkin-
son 2000; Hughes et al. 2003). Among these stresses, cli-
mate change has assumed a major importance as ocean
temperatures have warmed and major periods of symbiotic
dysfunction, called mass bleaching events, have been trig-
gered. These events are undocumented prior to the 1970s
but have been expanding since then in frequency, severity,
and geographic scale (Hoegh-Guldberg 1999). Since their
advent, there have been six major periods of bleaching
across the planet. The frequency in some areas is even high-
er. Mass coral bleaching events, which have been reported
as the worst yet in each case, have affected coral populations
on the Great Barrier Reef twice since 1997 (1998, 2002). In
1
Corresponding author (oveh@uq.edu.au).
Acknowledgments
We thank Efrat Fine for laboratory assistance; Mark Davy, Paul
Fisher, Sascha Thyer, and the Stanford Australia Program class of
2003 for field assistance; and David Logan, Michael Phillips, and
Collette Bagnato for collecting meteorological and field data.
each case, coral bleaching has spread to more than 50% of
the Great Barrier Reef Marine Park (Berkelmans and Oliver
1999; Berkelmans 2002; Dennis 2002).
Understanding how variability in the physical environ-
ment affects coral reefs is a priority if we are to truly un-
derstand the ramifications of the changing global climate.
Mass coral bleaching events are triggered by periods in
which sea temperatures rise above the long-term averages
for a particular region. Plant–animal endosymbioses are very
sensitive to changes in temperature, which result in an in-
creased sensitivity of the dinoflagellate symbiont to photo-
inhibition (Iglesias-Prieto et al. 1992; Warner et al. 1996;
Jones et al. 1998), cellular damage, and eventually disinte-
gration. The mechanism that underlies the earliest stages of
coral bleaching is very similar to that seen during thermal
stress in higher plants (Jones et al. 1998; Salvucci and
Crafts-Brandnera 2004), leading to the conclusion that coral
bleaching is at least in part a result of thermal stress occur-
ring within the photosynthetic processes of the dinoflagel-
lates. Warm seas, often only 1–2
8
C above the long-term av-
erages, can be detected by satellite measurements and used
to predict (with more than 95% accuracy) mass coral bleach-
ing events several weeks in advance (Strong et al. 2000).
Reduced temperatures also intensify photoinhibition in
higher plants (Aro et al. 1990; Lyons 1973; Greer and Laing
1991; Long et al. 1994) in a similar way to that that occurs
at elevated temperatures. Reduced temperatures lead to a re-

266 Hoegh-Guldberg et al.
duction in the rate at which the quenching of Photosystem
II (PSII) develops (Krause 1992), leading to an accumulation
of active oxygen species and cellular damage. Cold temper-
atures have also been observed to trigger the loss of dino-
flagellate symbionts from anemones and coral (Shinn 1966;
Coles and Jokiel 1977; Steen and Muscatine 1987). In a
previous study, Saxby et al. (2003) established that cold
stress creates similar physiological symptoms in coral to
those seen when they are heat stressed. Saxby et al. (2003)
demonstrated that water temperatures of 12
8
C for 12 h or
more led to the complete loss of photosynthetic efficiency
by PSII and death of exposed coral. Exposure of coral to
14
8
C revealed a light-dependent response, in which (as with
elevated temperature) thermal stress in low light had little
effect while coral exposed to 14
8
C and full sunlight were
heavily impacted. In the latter case, photosynthetic efficiency
was reduced and the coral became bleached in 24 h. These
observations match those of Jones et al. (1998) and other
research groups for elevated thermal stress, with the en-
hancement of the effect by light giving insight into where
the damage due to low or high thermal stress occurred.
This paper investigates cold stress on reefs that manifested
itself as a mass bleaching event in the intertidal areas of the
Capricorn–Bunker group (southern Great Barrier Reef) in
the austral winter of 2003. The measurements made during
this event confirm the conclusions of Saxby et al. (2003) and
previous workers. In addition, this study highlights the im-
portance of understanding the impact of the variability in
weather patterns as well as the overall shift that has been
occurring due to climate change.
Materials and methods
Heron Island is a platform reef (approximately 8 km
2
)
located within the Capricorn–Bunker group of reefs on the
southern Great Barrier Reef (23
8
26
9
60
0
S, 151
8
55
9
0
0
E). Large
areas of the intertidal flat at Heron Island were found to be
bleached in the austral winter, being first observed on 5 Au-
gust 2003. In addition to the exploration of environmental
data, measurements were made of the condition of the coral
and their fate over 5 months following the event.
Changes in coral cover—With the advent of bleaching in
the reef flat at Heron Island in early August 2003, the Cap-
ricorn–Bunker group of reefs was surveyed from the air to
assess the scale of the bleaching event.
Two 50-m permanent belt transects were set to follow the
progression and patterns of bleaching and recovery on the
reef flat on the northern side of Heron Island. Permanent
belt transects were established that were oriented north to
south (perpendicular to the shore line) or east to west (par-
allel to the shore line). These sites were photographed using
a digital camera (Nikon Coolpix 5000) over a measuring
tape. Photography was carried out during high tide to have
sufficient distance between the camera and the studied sub-
strate and in order to minimize observer impact on the sur-
veyed reef. Each exposure covered 1 m
2
of substrate. The
pictures were analyzed using the point-intercept method by
placing a transparency marked with 16 randomly marked
points over each image on the computer screen. The sub-
strate underlying the points was scored according to three
categories: healthy, bleached, or dead coral. We then calcu-
lated percentage of each of the categories for each survey
date. The same procedure was repeated during September
and October 2003 (6 and 10 weeks after the first set of pho-
tography, respectively) and during early February 2004 (6
months after the first set of measurements).
Environmental data—Temperature data for Heron Island
was obtained from three sources.
1. Average sea surface temperature for the period 30 July–
4 August (1600 h) of the Capricorn Islands group was re-
trieved from NOAA-16 AVHRR Satellite (downloaded at the
Australian Institute of Marine Science), pixel size
5
1 km.
2. A set of data loggers that were deployed at Heron Island
reef crest. One logger was positioned approximately 10 m
from the low-tide mark on the southern side of Heron Island.
A second logger was deployed midway across the intertidal
flat and a third was deployed off the reef crest in 5-m depth
at low tide. In each set of data loggers, there were sensors
for temperature, light, and pressure (depth/tides).
3. A log of weather observations by the Heron Island Re-
search Station staff, performed twice daily. These included
meteorological parameters, such as sea temperature, air tem-
perature, humidity, wind speed, precipitation, cloudiness,
and tides. These parameters have been logged since 1967.
Physiological measurements—Twelve fragments, approx-
imately 5 cm long, from four colonies of Acropora aspera
were sampled for area density of symbiotic dinoflagellates
on 5 August 2003. They were taken as follows: Four were
taken from the upper 10 cm (bleached section) of the colo-
nies, four from 10 cm under the bleached part, and four from
20 cm under the bleached section. In the lab, the tissue of
each fragment was removed using an air gun. The resulting
slurry was homogenized and diluted to a total volume of 50
ml. A volume of 10
m
l was sampled out twice from every
test tube and put onto a hemocytometer (upper and lower
fields). Two hemocytometer fields were photographed using
a fluorescent microscope (Olympus BX5) and a digital cam-
era. Counting of the fluorescent symbiotic dinoflagellates
was performed from the digital images using Blob Analysis
in MATROX 2.1 software (Matrox Electronic Systems). The
software counts the number of fluorescent objects in the pho-
to under predefined conditions of size and area. The number
of cells was calculated for the total 50-ml volume.
Surface area was measured by dipping each fragment in
hot wax (65
8
C) for 10 s. It was then cooled to room tem-
perature to allow the wax to solidify. After weighing the
fragments, they were dipped for 10 s in wax again at the
same temperature and weighed again after it solidified. The
net weight (weight 2
2
weight 1) was multiplied by a co-
efficient of 0.038, which was obtained by using the same
coating method on cylinders with known surface area and
calibrating against a regression coefficient. The number of
cells was then calculated per square centimeter coral surface
area.
Photosynthetic efficiency—The portable Diving-Pulsed
Amplitude Modulated (PAM) Fluorometer (Walz Gmbh)

267Coral bleaching and wintry weather
Fig. 1. Changes in coral cover over the period August 2003–
January 2004.
Fig. 2. Photograph taken on 24 January 2004 of areas severely bleached in early August 2003. Regrowth of coral out of crevices is
evident, as indicated by arrows. Transect tape is shown running down middle of photograph.
was used to explore the photosynthetic capacity of dinofla-
gellates inhabiting the tissue of A. aspera colonies during
the bleaching event. The PAM light meter was precalibrated
against a quantum sensor of a Li-Cor LI-189 light meter. In
each measurement, the tip of the PAM main optical fiber
was placed on the coral surface. In situ measurements were
performed on 5 August, right after the initial observation of
bleaching and on 26 August (3 weeks after the onset of
bleaching). Both measurements were undertaken at 2200 h
(low tide; total darkness). Dark-adapted maximal quantum
yield (Fv/Fm) was measured for three sections of each of
five branches (note: branches were taken from five separate
colonies): (a) upper 10 cm (bleached section) of the colonies,
(b) 10 cm under the bleached section, and (c) 20 cm under
the bleached section. Four saturation pulses were done on
each of these sections. Quantum yields were also measured
after incubating branches in light (1 min continuous actinic
light before saturation pulse; 1,000
m
mol m
22
s
21
). This was
done on five bleached and five nonbleached fragments of A.
aspera. At the end of these measurements, rapid light curves
of photosynthesis versus irradiance were done on five
bleached and five nonbleached fragments of A. aspera. Rap-
id light curves were done by measuring the quantum yield
of symbionts after illuminating branches from each of the
colonies for 10 s at each one of a series of eight irradiances.
Results
Inspection of bleaching at Heron Island revealed that most
of the visual impact on coral was restricted to the upper
portions of colonies, which had been presumably exposed at
low tide (for photographs, see Hoegh-Guldberg and Fine
2004). Affected areas stretched as far as the eye could see
along the reef crest at Heron Island. Aerial inspection by
helicopter revealed extensive bleaching around the entire
reef crest at Heron and the intertidal areas of neighboring
Wistari Reef and One Tree Island.
Changes in coral cover—Monthly surveys done after first
observing mass bleaching in the intertidal areas of Heron

268 Hoegh-Guldberg et al.
Fig. 3. (A) Average sea surface temperature (30 July–4 August
2003). The coolest water of the Great Barrier Reef during this pe-
riod is along the southern coast. (B) Average sea surface tempera-
ture of the Capricorn (30 August–4 September 2003). Note the cool
water engulfing Heron Island. Data sets in both A and B from
NOAA-16 AVHRR data downloaded at the Australian Institute of
Marine Science.
Fig. 4. Data retrieved from loggers placed in the intertidal areas
of Heron Island. Water temperatures for loggers positioned (A)
within 10 m of the shoreline of Heron Island on the intertidal flat.
Missing data in July are due to logger failure. Arrows indicate the
extremely cold days of 31 July and 1 August 2003. (B) Midway
across the intertidal flat, and (C) off the reef crest in 5-m depth at
low tide.
Island revealed that the affected areas of bleached coral
mostly died (Fig. 1). By September, areas that had greater
than 80% coral cover prior to the bleaching event (due main-
ly to one species, A. aspera) had less than 30% living coral
cover. By October, living coral represented less than 12%.
By early January 2004, coral cover had begun to recover
(17.8%), with coral tissue growing out of regions that were
shaded as well as not being exposed at low tide (Fig. 2).
Environmental data—Waters along the inshore region of
the southern Great Barrier Reef were cool by comparison
with other regions within the Reef and ranged between 19
8
C
and 21
8
C. Average sea surface temperature (30 July–4 Au-
gust 2003) retrieved from the NOAA-16 AVHRR dataset
revealed that the coolest water for the Great Barrier Reef lay
along the southern coastal regions. A tongue of relatively
cool water (21
8
C) extended out from the coast to the seaward
side of the platform reef on Heron Island (Fig. 3A,B). This
patch of cool water bathed Wistari and Heron Island but did
not reach as far as One Tree Island, which was approxi-
mately 0.5–1.0
8
C warmer.
The satellite data were validated by data loggers that were
located in intertidal and reef crest areas associated with
bleached areas (Fig. 4A–C). Sea temperatures just off the
reef crest (5 m; Fig. 4C) averaged 21.0
8
C
6
0.02
8
C, with a
maximum value of 22.7
8
C and a minimum of 18.1
8
C. These
values match the values reported within the satellite data
(Fig. 3). The two loggers based in the intertidal also revealed
similar average temperatures of 21.0
8
C
6
0.06
8
C and 21.0
8
C
1
0.05
8
C for inshore and offshore intertidal locations, re-
spectively (Fig. 4A,B). The range of values was much larger
for both than sea temperatures measured by the logger lo-
cated off the reef crest. In this case, minimum values of
14.4
8
C and 14.5
8
C and maximum values of 26.3
8
C and
27.0
8
C were recorded for inshore and offshore intertidal
sites, respectively. Importantly, the minimum values for sea
temperature off the reef crest occurred 3–4 d before bleach-

269Coral bleaching and wintry weather
Fig. 5. Detail of (A) water temperature and (D) light data taken
from data loggers positioned midway across the reef crest at Heron
Island (29 July–5 August). Meteorological data collected by Heron
Island Research Station for the same period is also shown. (B) Wind
speed (left axis, solid line) and tidal height (right axis, solid high-
frequency sinusoidal line), and (C) relative humidity (left axis, solid
line) and air temperature (right axis, dashed line). Accompanying
data and comments for records are shown in Table 1.
Fig. 6. Symbiont density as a function of distance along branch-
es of Acropora aspera (shown) on 5 August 2003. Upper portions
(left hand in this diagram) were visibly bleached relative to lower
portions (right hand in this diagram).
Table 1. Daily weather data collected by Heron Island Research Station over the period 26 July–5 August.
Date Rain (mm) T
max
(
8
C) T
min
(
8
C) Remarks from research personnel
26 Jul 03
27 Jul 03
28 Jul 03
29 Jul 03
30 Jul 03
31 Jul 03
0
0
0
0
0
0
18.3
19.3
18.6
20.4
21.3
20.4
15.4
14.7
15.4
18.3
17.6
14.4
Little cumulus to the north
No cloud
Cumulus and alto cumulus on horizon
Scattered cumulus and stratus
Cold, minute amount of cumulus
Freezing, minute amount of cumulus
1 Aug 03
2 Aug 03
3 Aug 03
4 Aug 03
5 Aug 03
0
0
0
6.4
1
17.4
18.8
20.7
21.1
23.4
13.3
14.9
18.2
17.1
19
Freezing, tiny amount of cumulus
Freezing, tiny amount of cumulus
Freezing, mostly cumulus
Varied cloud cover, rain surrounding
Mostly alto cumulus
ing was first reported on the intertidal regions of Heron Is-
land.
Detailed examination of dates revealed that the coldest
days (Fig. 5A) coincided with predawn low tides and were
typified by high winds (Fig. 5B), low humidity and air tem-
peratures (Fig. 5C) on cloudless days (Table 1, Fig. 5D).
Significantly, this is the only time in the history of daily
weather records on Heron Island (from 1967 to 2004) in
which observers used the word ‘‘freezing’’ to describe the
conditions (Table 1). Air temperature ranged from 13.3
8
Cto
23.4
8
C over the period 26 July–5 August 2003. From these
data, it is possible to calculate a wind chill of 9
8
C for coral
exposed at low tide during these conditions (based on the
coldest day, 1 August, in which air temperatures dropped to
13.3
8
C and wind speed increased to 0.040 km h
21
). This
would have been accentuated by the very low humidities
(44%) that accompanied these conditions.
Physiological condition of cold-stressed coral—Bleaching
was most pronounced in affected colonies in the upper 10
cm of the branch tips. Symbiont density decreased in the
upper 10 cm to less than 0.2
3
10
6
cell cm
22
and increased
away from the affected zone to over 2.5
3
10
6
cell cm
22
(Fig. 6). The lower values were similar to the values ob-
tained by Dove (2004) for control A. aspera. Dark-adapted
photosynthetic efficiency did not correlate with the extent of
bleaching down the branches (Fig. 7A). A major difference
in quenching behavior was revealed when quantum efficien-
cy was measured at successively higher light levels (Fig.
7B). This confirmed the observation of low-light–adapted
yields in the bleached regions as compared with those mea-
sured in the nonbleached regions (Fig. 7C). In this case,
symbionts remaining in the bleached areas quenched the
photosynthetic efficiency of PSII by a much larger amount

270 Hoegh-Guldberg et al.
Fig. 7. Physiological state of symbionts in Acropora aspera as a function of distance down the branches. (A) Dark-adapted yield at
three different sections of A. aspera branches measured soon after 2200 h on 5 August 2003. (B) Light-adapted yield (after 1 min of 1,000
m
mol photons m
22
s
21
) for bleached and unbleached A. aspera. (C) Quantum yield of PSII of symbionts in A. aspera as a function of light
level on 5 August 2003. (D) Measurements repeated for symbionts in A. aspera, 3 weeks after the onset of bleaching (26 August 2003).
Symbols indicate different positions on the colony: top section, 10 cm down from tip and 20 cm down from the tip are shown.
as light levels increased. Three weeks later, this behaviorwas
much less pronounced (Fig. 7D).
Discussion
Sea temperature has a major influence over survivorship
of reef-building coral and their symbionts and is considered
to be a major determinant of their latitudinal distribution
(Kleypas et al. 1999). While coral can sustain seasonal var-
iability in sea temperature that may be more than 12
8
C
(Kleypas et al. 1999), small excursions above geographically
associated maxima or thresholds leads to syndromes asso-
ciated with stress (Harvell 1999; Hoegh-Guldberg 1999).
One of these syndromes, coral bleaching, has increased enor-
mously in magnitude and frequency over the past 30 years.
A combination of elevated sea temperature and exposure
time predicts mass coral bleaching with great certainty
(Hoegh-Guldberg 1999; Strong et al. 2000; Hoegh-Guldberg
2001), with high values leading to mass mortality events.
Estimates of mortality from mass coral-bleaching events
range from zero to an almost total loss of reef-building coral
from affected areas. An average of 17.7% of the living coral
in six major coral reef regions of the world was killed during
one of the warmest years on record, 1998. The range of
mortality estimates is perhaps the most interesting detail hid-
den within the average. While some regions (e.g., Australia
and Papua New Guinea) lost small amounts (3%), regions
like the Arabian Gulf and the Wider Indian Ocean lost 33%
and 46%, respectively during the single event in 1998.
Cold temperatures can also be a problem for reef-building
coral. While some coral are adapted for colder conditions
(e.g., down to 11.5
8
C in the case of the temperate hermatypic
coral Plesiastrea versipora; Kevin and Hudson 1979), down-
ward excursions have a similar impact on reef-building coral
in laboratory experiments. The massive coral Montastrea an-
nularis is killed by water temperatures of 14
8
C for more than
9 h (Mayer 1914), while many coral species appear able to
tolerate 15
8
C for short periods (Roberts et al. 1982). Saxby
et al. (2003) revealed similar responses from the hardy in-
tertidal coral Montipora digitata at Heron Island. In this
case, 16
8
C and 12
8
C lead to a progressive reduction of pho-
tosynthetic efficiency (Fv/Fm) of algal symbionts in M. dig-
itata over6hofexposure to these temperatures in the light.
In both cases, incubation in the dark over the next 12 h did
not lead to recovery of Fv/Fm, suggesting chronic photo-
damage had occurred. In subsequent experiments, Saxby et
al. (2003) showed that coral became bleached and did not
recover when exposed to temperatures as low as 12
8
C for
over 12 h.
The lower threshold of M. digitata from the Heron Island
in the laboratory was 12–14
8
C for exposure times of 1–2 d
(Saxby et al. 2003). In the light of these conclusions for what
is considered to be a fairly hardy coral, the observation of
a bleaching event on 5 August is not surprising. Coral were
exposed at low tide to reduced temperatures at night (down
to 13.3
8
C) with 0.030–0.035 km h
21
, low humidities, and
clear sunny conditions during the day. Affected areas of the
colonies (upper 10 cm) were exposed to air at night time
during these periods. Combining the effect of high winds
and low humidities, coral tissue was exposed to surface tem-
peratures of 9
8
C (wet bulb temperature) for several hours.
Based on the observations of Saxby et al. (2003), exposure
to the sunlit conditions of the intertidal while experiencing
these low temperatures would have led to chronic photo-

271Coral bleaching and wintry weather
inhibition as light levels increased. This damage appears to
have been repaired by 5 August, as measurements of the
quantum efficiency of PSII under darkness revealed similar
values for symbiotic dinoflagellates in bleached and un-
bleached coral (Fig. 7). The quantum efficiency of PSII in
symbionts left in bleached areas was considerably lower
when measured after incubation in the photosynthetically ac-
tive radiation (1,000
m
mol m
22
s
21
). In this case, symbionts
left in the bleached areas exhibited extensive quenching
when exposed to the light. One explanation for this is that
light levels within the skeletons were much higher due to
greater reflection of light within the polyp calices after the
majority of symbionts had left (Roberto Iglesius-Prieto,
UNAM, pers. comm.). This highlights the important obser-
vation that light stress (and hence the effect of anything that
affects the speed at which excitations are processed) will
have proportionately higher effects on coral that are partially
or fully bleached.
This paper highlights the importance of downward as well
as upward excursions of air and water temperatures for reef-
building coral and their symbionts. While day-to-day vari-
ability in local weather patterns is hard to attribute to climate
change, global climate-change models do project complex
changes to local weather patterns that may involve localized
downward trends in local sea and air temperatures (IPCC
2001). Understanding these changes may be as important as
appreciating the direct effects of an upwardly moving tem-
perature signal on important ecosystems such as coral reefs.
References
A
RO
, E. M., T. H
UNDAL
,I.C
ARLBERG
,
AND
B. A
NDERSSON
. 1990.
In vitro studies on light-induced inhibition of photosystem II
and D1-protein degradation at low temperatures. Biochim. Bio-
phys. Acta 1019: 269–275.
B
ERKELMANS
, R. 2002. Time-integrated thermal bleaching thresh-
olds of reefs and their variation on the Great Barrier Reef. Mar.
Ecol. Prog. Ser. 229: 73–82.
,
AND
J. K. O
LIVER
. 1999. Large-scale bleaching of corals
on the Great Barrier Reef. Coral Reefs 18: 55–60.
B
RYANT
, D., L. B
URKE
,J.M
C
M
ANUS
,
AND
M. S
PALDING
. 1998.
Reefs at risk: A map-based indicator of threats to the worlds
coral reefs. WRI/ICLARM/WCMC/UNEP. World Resources
Institute. 56.
C
OLES
,S.L.,
AND
P. L. J
OKIEL
. 1977. Effects of temperature on
photosynthesis and respiration in hermatypic corals. Mar. Biol.
43: 209–216.
D
ENNIS
, C. 2002. Reef under threat from bleaching outbreak. Nature
415: 947.
D
OVE
, S. 2004. Scleractinian corals with photoprotective pigments
are hypersensitive to thermal stress. Mar. Ecol. Prog. Ser. 272:
99–116.
G
REER
, D. H.,
AND
W. A. L
AING
. 1991. Low-temperature and light
induced photoinhibition of photosynthesis in kiwi fruit leaves.
Acta Hortic. 297: 315–321.
H
ARVELL
, C. D., C. E. M
ITCHELL
,J.R.W
ARD
,S.A
LTIZER
,A.P.
D
OBSON
,R.S.O
STFELD
,
AND
M. D. S
AMUEL
. 2002. Climate
warming and disease risks for terrestrial and marine biota. Sci-
ence 296: 2158–2162.
H
OEGH
-G
ULDBERG
, O. 1999. Climate change, coral bleaching and
the future of the world’s coral reefs. Mar. Freshw. Res. 50:
839–866.
. 2001. The future of coral reefs: Integrating climate model
projections and the recent behaviour of corals and their dino-
flagellates. Proceeding of the Ninth International coral reef
symposium, October 23–27, 2000, Bali, Indonesia. 2: 1105–
1110.
,
AND
M. F
INE
. 2004. Cold weather causes coral bleaching.
Coral Reefs, 23: (10.1007/s00338-004-0401-2).
H
UGHES
,T.P.,
AND OTHERS
. 2003. Climate change, human impacts,
and the resilience of coral reefs. Science 301: 929–933.
I
GLESIAS
-P
RIETO
, R., J. L. M
ATTA
,W.A.R
OBINS
,
AND
R. K.
T
RENCH
. 1992. Photosynthetic response to elevated-tempera-
ture in the symbiotic dinoflagellate Symbiodinium microadria-
ticum in culture. Proceed. Natl. Acad. Sci. USA 89: 302–305.
IPCC. 2001. Climate change 2001, synthesis report. Intergovern-
mental Panel on Climate Change.
J
ONES
, R. J., O. H
OEGH
-G
ULDBERG
,A.W.D.L
ARKUM
,
AND
U.
S
CHREIBER
. 1998. Temperature-induced bleaching of corals be-
gins with impairment of the CO2 fixation mechanism in zoo-
xanthellae. Plant Cell Environ. 21: 1219–1230.
K
EVIN
,K.M.,
AND
R. C. L. H
UDSON
. 1979. The role of zooxan-
thallae in the hermatypic coral Plesiastrea urvillei from cold
waters. J. Exp. Mar. Biol. Ecol. 36: 1–23.
K
LEYPAS
, J. A., J. M
C
M
ANUS
,
AND
L. M
ENEZ
. 1999. Using envi-
ronmental data to define reef habitat: Where do we draw the
line? Am. Zool. 39: 146–159.
K
RAUSE
, G. H. 1992. Effects of temperature on energy-dependent
fluorescence quenching in chloroplasts. Photosynthetica 27:
249–252.
L
ONG
, S. P., S. H
UMPHRIES
,
AND
P. G. F
ALKOWSKI
. 1994. Photo-
inhibition of photosynthesis in nature. Annu. Rev. Plant. Phy-
siol. Plant. Mol. Biol. 45: 633–662.
L
YONS
, J. M. 1973. Chilling injury in plants. Annu. Rev. Plant.
Physiol. 24: 445–466.
M
AYER
, A. G. 1914. The effects of temperature on tropical marine
animals. Pap Tort Lab, Carn Inst Wash 6: 1–24.
R
OBERTS
, H. H., L. J. R
OUSE
,J
R
., N. D. W
ALKER
,
AND
J. H. H
UD
-
SON
. 1982. Coldwater stress in Florida Bay and Northern Ba-
hamas: A product of winter cold-air outbreaks. J. Sediment.
Petrol. 52: 145–155.
S
ALVUCCI
,M.E.,
AND
S. J. C
RAFTS
-B
RANDNER
. 2004. Inhibition of
photosynthesis by heat stress: The activation state of Rubisco
as a limiting factor in photosynthesis. Physiol. Plantarum 120:
179–186.
S
AXBY
, T., W. C. D
ENNISON
,
AND
O. H
OEGH
-G
ULDBERG
. 2003. Pho-
tosynthetic responses of the coral Montipora digitata cold tem-
perature stress. Mar. Ecol. Prog. Ser. 248: 85–97.
S
HINN
, E. A. 1966. Coral growth-rate, an environmental indicator.
J. Paleontol. 40: 233–240.
S
TEEN
,R.G.,
AND
L. M
USCATINE
. 1987. Low temperature evokes
rapid exocytosis of symbiotic algae by a sea anemone. Biol.
Bull. (Woods Hole) 172: 246–263.
S
TRONG
, A. E., K. K. G
JOVIG
,
AND
E. K
EARNS
. 2000. Sea surface
temperature signals from satellites—an update. Geophys. Res.
Lett. 27: 1667–1670.
W
ARNER
, M. E., W. K. F
ITT
,
AND
G. W. S
CHMIDT
. 1996. The effects
of elevated temperature on the photosynthetic efficiency of zo-
oxanthellae in hospite from four different species of reef coral:
A novel approach. Plant Cell Environ. 19: 291–299.
W
ILKINSON
,C.[
ED
.]. 2000. Status of coral reefs of the world: 2000.
Global Coral Reef Monitoring Network, Australian Institute of
Marine Science.
Received: 5 May 2004
Accepted: 1 September 2004
Amended: 26 September 2004