BGBiogeosciencesBGBiogeosciences1726-4189Copernicus PublicationsGöttingen, Germany10.5194/bg-14-817-2017Coral mortality induced by the 2015–2016 El-Niño in Indonesia: the effect of rapid sea level fallAmpouEghbert ElvanJohanOfriMenkesChristophe E.NiñoFernandoBirolFlorenceOuillonSylvainAndréfouëtSergeserge.andrefouet@ird.frUMR9220 ENTROPIE, Institut de Recherche pour le Développement, Université de la Réunion, CNRS, B.P.A5, 98848, Noumea, New CaledoniaInstitute for Marine Research and Observation, SEACORM/INDESO center, Jl. Baru Perancak, Negara-Jembrana, Bali 82251, IndonesiaLaboratoire d'Etudes en Géophysique et Océanographie Spatiales, Université de Toulouse, CNRS, IRD, CNES, UPS, 14 avenue Edouard-Belin, 31400 Toulouse, FranceResearch and Development Institute for Ornamental Fish Culture, Jl. Perikanan No. 13, Pancoran Mas, Kota Depok, Jawa Barat 16436, IndonesiaLaboratoire d'Océanographie et du Climat: Expérimentations et Approches Numériques, Sorbonne Universités, UPMC Université Paris 06, IPSL, UMR CNRS/IRD/MNHN, B.P.A5-98848, Noumea, New CaledoniaSerge Andréfouët (serge.andrefouet@ird.fr)24February20171448178266September20169September201613January20173February2017This work is licensed under a Creative Commons Attribution 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by/3.0/This article is available from https://bg.copernicus.org/articles/14/817/2017/bg-14-817-2017.htmlThe full text article is available as a PDF file from https://bg.copernicus.org/articles/14/817/2017/bg-14-817-2017.pdf
The 2015–2016 El-Niño and related ocean
warming has generated significant coral bleaching and mortality worldwide.
In Indonesia, the first signs of bleaching were reported in April 2016. However,
this El Niño has impacted Indonesian coral reefs since 2015 through a
different process than temperature-induced bleaching. In September 2015,
altimetry data show that sea level was at its lowest in the past 12 years,
affecting corals living in the bathymetric range exposed to unusual
emersion. In March 2016, Bunaken Island (North Sulawesi) displayed up to
85 % mortality on reef flats dominated by Porites, Heliopora
and Goniastrea corals with differential mortality rates by coral
genus. Almost all reef flats showed evidence of mortality, representing
30 % of Bunaken reefs. For reef flat communities which were living at a
depth close to the pre-El Niño mean low sea level, the fall induced
substantial mortality likely by higher daily aerial exposure, at least during
low tide periods. Altimetry data were used to map sea level fall throughout
Indonesia, suggesting that similar mortality could be widespread for shallow
reef flat communities, which accounts for a vast percent of the total extent
of coral reefs in Indonesia. The altimetry historical records also suggest
that such an event was not unique in the past two decades, therefore rapid sea
level fall could be more important in the dynamics and resilience of
Indonesian reef flat communities than previously thought. The clear link
between mortality and sea level fall also calls for a refinement of the
hierarchy of El Niño impacts and their consequences on coral reefs.
Introduction
El Niño-Southern Oscillation (ENSO) is the most important coupled
ocean-atmosphere phenomenon impacting climate variability at global and
inter-annual time scales (McPhaden, 2007). The consequences on coral reefs
have been well-documented, especially since the 1997–1998 massive coral
bleaching event, which reached planetary dimensions
(Hoegh-Guldberg, 1999). In short, El Niño increases
temperature in several coral reef regions and induces zooxanthellae
expulsion from the coral polyp, resulting in a coral colony looking white,
hence “bleaching”. If the situation persists the coral colony eventually
dies. Coral bleaching intensity has been related to different temperature
thresholds, other environmental factors and stressors, and type of
zooxanthellae and corals (Baker
et al., 2008). Bleaching episodes due to ocean warming were recorded during
the strong 1982–1983 El Niño in Australia
(Glynn, 2000) and have since been
reported worldwide in several instances
(Guest
et al., 2012; Wouthuyzen et al., 2015). The last bleaching episode has
occurred in 2015–2016 during what occurs to be the strongest El Niño
event on record (Schiermeier, 2015). Bleaching events were often global in
the past, including Indonesia
(Suharsono,
1990; Guest et al., 2012; Wouthuyzen et al., 2015). The last reports for
Indonesia in 2016 are still under analysis, and Reef Check survey locations
are presented at http://reefcheck.or.id/bleaching-indonesia-peringatan/. Thus, it is assumed
that coral bleaching induced by ocean warming will be the main culprit if
post-El Niño surveys report coral mortalities.
While in Bunaken National Park in 23 February–5 March 2016
for a biodiversity survey, we noticed recent mortalities on the upper part
of many massive colonies on several reef flats. This prompted a systematic
investigation of the phenomenon's spatial distribution. We report here
observations on what appears to be the first significant impact of the
2015–2016 El Niño on Indonesia reefs. Unlike what is expected during
such a strong event, the mortality was not related to warm water
induced-bleaching, but could be tracked to rapid sea level variations. Coral
mortality data around Bunaken Island are provided, and we investigate
various altimetry and sea level anomaly data sets to explain mortality. The
clear link between mortality and sea level fall calls for a refinement of
the hierarchy of El Niño impacts and their sequences on coral reefs.
Bunaken reef flats. (a) Close-up of one Heliopora coerula
colony with clear tissue mortality on the upper part of the colonies;
(b) same for a Porites lutea colony; (c) reef flat Porites
colonies observed at low spring tide in May 2014. Even partially above water
a few hours per month in similar conditions, the entire colonies were alive.
(d) A living Heliopora coerula (blue coral) community in 2015 in a
keep-up position relative to mean low sea level, with almost all the space
occupied by corals. In that case, a 15 cm sea level fall will impact most of
the reef flat. (e–h) Before–after comparison of coral status for colonies
visible in (c). In (e), healthy Porites lutea (yellow and pink massive
corals) reef flat colonies in May 2014, observed at low spring tide. The
upper part of colonies is above water, yet healthy; (f) same colonies in
February 2016. The white lines visualize tissue mortality limit. Large
Porites colonies (P1, P2) at low tide levels in 2014 are affected,
while lower colonies (P3) are not. (g) P1 colony in 2014. (h) Viewed from
another angle, the P1 colony in February 2016. (i) Reef flat community with
scattered Heliopora colonies in February 2016, with tissue
mortality and algal turf overgrowth.
Material and methods
Bunaken National Park (BNP) is located at the northwest tip of Sulawesi,
Indonesia. The location is at the core of the epicenter of marine
biodiversity, the so-called Coral Triangle, a vast area spanning Malaysia to
Solomon Island, where the number of marine species is maximum (Hoeksema,
2007). BNP includes several islands with Bunaken Island (1.62379∘ N,
124.76114∘ E), one of the most studied Indonesian reef sites.
The Bunaken Island is surrounded by a simple fringing reef system, comprising
reef flats, several small enclosed lagoons and forereefs. The tide regime is
semidiurnal, but with marked diurnal inequalities (Ray et al., 2005), and with a
maximum spring tidal range that can reach 2.52 m. The Bunaken Island is
generally exposed to southwest wind from May to October, resulting in calm
seas due to the short fetch between mainland and the island, and to
northwest wind from November to February, which can be strong at time and
generate large waves breaking on the west and north shores.
Two previous BNP surveys for habitat mapping, in May–June 2014 and May–June
2015, did not show any significant signs of widespread mortalities on reef
flats. Different species of corals were frequently exposed above water level
at low spring tide, yet they were entirely alive (Fig. 1). Microatolls were
present. They have not been studied in Bunaken NP, but as in
other sites, their growth is likely constrained by a Mean Low Water (MLW)
between Mean Low Water Neaps (MLWN) and Mean Low Water Springs (MLWS)
(Smithers and Woodroffre, 2000; Goodwin and Harvey, 2008). Several reef
flats were characterized by compact communities of massive and semi-massive
colonies that could be described as keep-up communities, limited in their
vertical growth by the MLW (in agreement with the terminology of Holocene
reefs provided by Neumann and Macintyre, 1985).
In contrast with the 2015 observations, in late February 2016, during a
coral biodiversity census survey, we noticed the widespread occurrences of
dead massive corals and we performed a systematic investigation on the
spatial distribution of the phenomenon. Using the habitat map created by
Ampou (2016), all coral habitat polygons present on reef flats around Bunaken Island were visually surveyed and recent mortality was recorded
(presence/absence). Geographic coordinates of the presence of mortality were
compiled to map its extent. In practice, when mortality was observed on a
habitat polygon, the entire polygon was flagged as positive. Then, in
different locations around the island, mortality was measured on six reef
flat locations characterized by high coral cover and different dominant
massive coral species, principally Porites lutea and the octocoral
Heliopora coerulea, using six 10 m long Line Intercept Transect
(LIT) (English et al., 1997). We recorded the percent cover of
live and dead tissue for each coral and summed the total. We also recorded
the species/genus for each coral, and substrate categories between colonies.
We did not keep track of the number of colonies present on each transect.
A clear sharp horizontal limit of tissue mortality was present in impacted
colonies. The distribution of dead tissue between colonies and among
colonies (Fig. 1) suggested that mortality was related to sea level
variations, with increased aerial exposure time during the last few months.
In order to test this hypothesis, we needed to identify sea level variation
data. For this, long-term data from a tide-gauge or a pressure sensor are
ideal but these were not available for Bunaken. Tide-gauge data are scarce
in Indonesia but fortunately there are two tide-gauges in the north of
Sulawesi in the city of Bitung, east of Bunaken (latitude 1.430∘ N and
longitude 125.200∘ E) and on the other side of Sulawesi, compared to Bunaken. Thus,
while tide-gauge data are available in the region, they are not exactly on
Bunaken, but can help visualize the range of conditions found in Bunaken.
Bitung data were retrieved from the Sea Level Center in Hawaii (SLCH),
specifically at http://uhslc.soest.hawaii.edu/thredds/uhslc_quality_daily.html?dataset=RQD033A. The Sea Surface Height
(SSH) provided is referenced, for Bitung, against a GPS station located at
Bako (http://www.igs.org/igsnetwork/network_by_site.php?site=bako) which is itself referenced against the WGS84
ellipsoid. Hence, raw Bitung SSH does not represent absolute depth above the
Bitung seafloor. SLCH provides high quality data (available till early 2015)
that have been controlled for most outliers and errors, and lower quality
data that include the most recent coverage embodied in our period of interest
(2015–2016). The Bitung tide-gauge stopped recording in many instances for
reasons unknown to us, hence the records present many, irregularly spaced gaps.
In addition to the Bitung tide-gauge data, different sea level anomaly
products were investigated, based on their temporal coverage and spatial
resolution. First, we used gridded altimetry data in terms of Absolute
Dynamic Topography (ADT), from the Archiving, Validation and Interpretation
of Satellite Oceanographic Data (AVISO) center at the spatial resolution of
1/4∘. ADT provides the sea level with respect to the
geoid. Data are available from 1993 to 2016, allowing a long-term comparison
of the sea level trends. The mean ADT over the period was extracted for a
small area next to Bunaken Island (1.5–1.7∘ N;
124.5–124.8∘ E), a larger area (3 by 3 degrees around the smaller
area) centered on Bunaken Island and including the north of Sulawesi and
Tomini Bay in the south, and for Indonesia in its entirety.
(-14.9–10.0∘ S, 94.9–140.0∘ E). The difference between the minimum value
(observed in September 2015) and the 2005–2014 mean or the 1993–2016 mean
periods were also computed. In addition, we also retrieved ADT data
corresponding to the Bitung tide-gauge location to compare altimetry sea
level anomalies with in situ data. The selected retrieved location
is the closest available from Bitung (1.375∘ N and longitude 125.125∘ E). To
compute sea level anomalies, we considered only the periods of time covered
by both data sets in order to use a common baseline.
Second, to extract geophysical information from higher spatial resolution
altimeter data, we used the along-track measurements from SARAL/AltiKa
Geophysical Data Records (GDRs) that were distributed by the AVISO service
(http://www.aviso.altimetry.fr/fr/). This data set was chosen because the
new Ka-band instrument from SARAL has a finer spatial resolution and enables
a better observation of coastal zones (Verron et al., 2015). Data extend
from March 2013 (cycle 1 of the satellite mission) to May 2016 (cycle 33),
with a repeat period of 35 days. Over this period, we use all altimeter
observations located between 10∘ S–10∘ N and
105–140∘ E. Two tracks (No.535 and No.578)
intersect the north of Sulawesi and contain sampling points just off Bunaken Island. The data analysis is done in terms of sea level anomalies
(SLA) computed from the 1 Hz altimeter measurements and geophysical
corrections provided in GDRs products. The SLA data processing and editing
are described in details in Birol and Niño (2015). The 1 Hz SLA data
have a spatial resolution of ∼ 7 km along the satellite
tracks. In order to quantify the spatial variations of the regional sea
level change in March 2013–May 2016, a linear trend model is applied (using
a simple linear regression) to the individual SLA time series observed at
the different points along the altimeter tracks that cross the area of
interest. The trend is the slope of the regression (in cm year-1). The
resulting 3–year sea level trend values can be represented on a map.
Mortality rates (mean ± standard deviation, n=6) of all
corals for the six reef flat sites. The three dominant species were
Porites lutea, Heliopora coerulea, and Goniastrea minuta. Several species and genus were found only once. Standard deviation
is not shown when only one measurement per type of coral could be achieved
(i.e., one colony per site).
Coral PoritesHelioporaGoniastreaAcroporaGalaxeaCyphastreaMontiporaPoritesLobophylliaPocilloporaMeanluteacylindricaSite144±3652±2442±4046239±1618±8100±010057354±5100±0100±010010058420±171002510055561±1329±1867100±085652±2370±846±5110047Mean44428245891001004210010058ResultsMortality rates per dominant coral genus
For all colonies found on the six stations, dead tissues were found on the
top and upper-flank of the colonies, with the lower part of the colonies
remaining healthy (Fig. 1). Mortality was not limited to microatoll-shaped
colonies only. Round massive colonies were also impacted. On microatolls and
other colonies that may have lived close to MLW, the width of dead tissue
appeared to be around a maximum of 15 cm. Dead tissues were systematically
covered by turf algae, with cyanobacteria in some cases, suggesting that the
stressor responsible for the mortality occurred a few months earlier. There
were no obvious preferential directions in tissue damage at the colony surface,
as previously reported for intertidal reef flat corals in
Thailand (Brown et al., 1994).
The six surveyed reef flat locations were dominated by H. coerulea
and P. lutea, while other genus and species occurred less
frequently (Table 1). When taking into account all genus, up to 85 % of
the colonies were dead (Site 5). The average mortality was around 58 % with all
sites included (Fig. 2). When it was present, Goniastrea minuta
colonies were the most impacted, with a 82 % mortality on average (Fig. 2).
Highest mortalities were found on keep-up communities relative to sea
level (Fig. 1).
Top: Bunaken location in the north of Sulawesi, the large island
in grey. Middle: Close-up of the island of Bunaken. The yellow area shows where
coral mortality occurred around Bunaken reef flats, with the position of six
sampling stations. Dark areas between the yellow mask and the land are
seagrass beds. Blue-cyan areas are slopes and reef flats without mortality.
Bottom: Mortality rates for the six sites for two dominant species
Porites lutea and Heliopora coerulea. The latter is not
found on Sites 3 and 4. The number of colonies ranged between 10 and 30 per
transect, depending on the size of the colonies.
Map of occurrences of mortality
The survey around the island revealed presence of mortality all around the
island except the north reef flats, where corals were scarce and encrusting
(Fig. 2). The same coral genus as listed in table 1 were impacted, but
mortality levels differed depending on colony heights. When colonies were
clearly below the present minimum sea level, they remained healthy (Fig. 1).
Locations of positive observations show that mortality has occurred mostly
along the crest, which is expected to be the most vulnerable during sea
level fall (Fig. 2). The spatial envelop of mortality occurrences is shown
on the Bunaken map in Fig. 2. The survey and generalization using the habitat
map suggests that nearly 163 ha, or 30 % of the entire reef system,
has been impacted by some mortality. However, this does not mean that 30 %
of the reef has died.
Comparison between tide-gauge and altimetry data
We found a good correlation (Pearson r=0.83) between sea level anomalies
from altimetry (ADT) and from tide-gauge. The two time series are compared in
Fig. 3 to confirm the agreement. The Bitung sea-level data reveal the type of
sea-level variations that likely occurred around Bunaken, although patterns
may not be exactly the same considering the distances between sites. Figure 3
shows the daily mean sea level from the available Bitung data (that can be
compared to sea level as provided by altimetry), and the daily lowest level
(which can not be directly measured by altimetry). This graph suggests what
the range of sea level variations happening in Bunaken before El Niño
likely was, due to normal tide fluctuations. The daily lowest value (blue
curve in middle and lower panels in Fig. 3) exhibited a ∼ 40 cm
variation from neap tide to spring tide. In 2014, and 2015, we witnessed
Porites corals that had the upper part of the colonies well above
the sea level, and without signs of mortality during spring tide conditions
(Fig. 1). Hence, the upper limit of coral survival is somewhere around 20 cm
above the spring tide, the lowest level for the end of the period shown in
Fig. 3. In other words, the limit of coral survival is close to the mean of
the daily lowest level curve. If this mean value changes through time, the
limit of mortality also changes dynamically. The ∼ 15 cm fall that we observed on altimetry
data around Bunaken and on most of eastern Indonesia changed the values of the lowest levels for a short time (several weeks), but these changes lasted long enough so that coral tissues were damaged by excessive UV and air exposure. During a few weeks in
August–September 2015, this fall resulted in a shift of the mean low level
towards the pre-El Niño lowest levels shown in Fig. 3 (lower panel).
Sea-level data from the Bitung (east North Sulawesi) tide-gauge,
referenced against Bako GPS station. On top, sea level anomalies measured by
the Bitung tide-gauge station (low-quality data), and overlaid on altimetry
ADT anomaly data for the 1993–2016 period. Note the gaps in the tide-gauge
time series. Middle: Bitung tide-gauge sea level variations (high-quality
data, shown here from 1986 till early 2015) with daily mean and daily lowest
values. Bottom, a close-up for the 2008–2015 period.
Absolute Dynamic Topography time series
The ADT time series (Fig. 4) shows a significant sea level fall congruent
with El Niño periods, at all spatial scales, although the pattern is not
as pronounced at Indonesia-scale (Fig. 4). The 1997–1998 and the 2015–2016
years display the highest falls. The September 2015 value is the local
minima, considering the last 10 years (Fig. 4). The 8 cm fall in September
2015 compared to the previous 4 years, and the 15 cm fall compared to the
1993–2016 mean (Fig. 4) is consistent with the pattern of mortality
following a maximum of ∼ 15 cm width on the top of the
impacted micro-atolls and colonies that were living close to the mean low
sea level before the event (Fig. 1).
Time series of ADT, minus the mean over the 1993–2016 period, for Bunaken Island (top), North Sulawesi (middle), and Indonesia (bottom). The
corresponding spatial domains are shown in Fig. 6. El Niño periods
(http://www.cpc.ncep.noaa.gov/products/analysis_monitoring/ensostuff/ensoyears.shtml)
are depicted with grey shadings. The
September 2015 minimum corresponds to an 8 cm fall compared to the minima then
four previous years, and a 14 cm fall compared to the 1993–2016 mean. The
1998 El Niño displays the largest sea level fall.
Sea level anomaly trends
SARAL/AltiKa data in March 2013–May 2016 are
shown in Fig. 5 for a small area that includes Bunaken Island (at the top)
and a larger box (at the bottom) covering part of the western equatorial
Pacific Ocean and Coral Triangle. A substantial sea level fall is observed
around Bunaken Island, with values ranging from 4 to 8 cm year-1 (12
to 24 cm accumulated over 3 years, Fig. 5). Further analysis of the
individual sea level time series indicates that the overall trend is
explained, and accelerated, by the fall due to El Niño (not shown). This
result agrees with findings from Luu et al. (2015) around Malaysia and can
be extended to much of the Coral Triangle. Fig. 5 shows that this phenomenon
is consistent over a large part of Indonesia and the warm water pool, where
strong differences in sea level variations (up to -15 cm year-1) are
observed between Asia and Micronesia, north of 5∘ N and east of
130∘ E.
Top: Map of along-track SLA trend (in cm year-1), 2013–2016,
for the north Sulawesi area. The position of Bunaken Island is shown (BNK).
Bottom: Map of along-track SLA trend (1 Hz), 2013–2016, for Indonesia. The
domain on the top panel is the rectangle in the Indonesia map of the bottom panel.
Discussion
A common ground exists between this study and the use of massive corals to
reconstruct sea level. Reconstructions of paleo-sea levels, whether it is
induced by tectonic events or not, is a science that takes advantage of the
shape of modern or fossil micro-atolls (Meltzner et al., 2006). However, we
stress to the reader that this study is not about reconstructing sea levels using dead
corals. Rather, we explained coral mortality using sea level data, primarily
from altimetry data. The agreement between altimetry and tide-gauge data
(Fig. 3) confirms that altimetry data are suitable to monitor sea level
variation close to a coast. More specifically, this confirms the value of
using altimetry observations to help identifying the cause of shallow coral
mortality, even without any other local in situ source of sea level
data, as in Bunaken.
Interestingly, we found that sea level fall appeared to be responsible for
coral mortality, while most recent climate change literature is generally
focused on the present and future effects of sea level rise (Hopley, 2011).
Geological records and present-time observation have already demonstrated
that sea level variation is a driver of coral community changes. Sea level
rise can have antagonistic effects: on the one hand, it can provide new
growing space for corals. On the other hand, higher depth may enhance wave
propagation and increase physical breakage in areas that were previously
sheltered. If sea level rise is fast, corals may not keep-up and the reef
may drown relative to the new sea level. As such, sea level rise is
seen as one of the three main climate change threats for coral reefs. This
study reminds the reader that the processes can be much more variable at ecological
time scales.
We aimed to document the spatial scale and the cause of an ecological
event that could be easily overlooked when documenting the 2016 El Niño
impact on Indonesian coral reefs. Many studies have emphasized the role of
hydrodynamics and sea level on the status and mortality of coral communities
growing on reef flats (e.g., Anthony and Kerswell, 2007; Hopley, 2011; Lowe
et al., 2016). Here we emphasize, with altimetry data, used for one of the first times for a reef flat study (Tartinville and Rancher, 2000), that the 2015–2016
El Niño has generated such mortality, well before any ocean
warming-induced bleaching. The exact time of the mortality remains unknown,
but it is likely congruent to the lowest level in September 2015. The aspect
of the colonies in February 2016, with algal turf covering the dead part
(Fig. 1), is also consistent with a lowest sea level occurring a few months
earlier. Fig. 1 shows corals that were fine in May 2015 even when
exposed to aerial exposure during low spring tide, without wave or wind, for
several hours, during several days of spring tide. Thus, we assume the
mortality was due to several weeks of lower water, including spring tide
periods, which are compatible with the temporal resolution of the altimetry
observations. The aerial exposure could have led to tissue heating,
desiccation, photosystem or other cell function damage (Brown, 1997). It is
possible that colonies could have looked bleached during that period (Brown et
al., 1994). Lack of wind-induced waves in the September period also prevented
wave washing and water mixing which could have limited the damage (Anthony
and Kerswell, 2007).
The various satellite Sea Surface Temperature (SST) products for coral
bleaching warning, available at http://coralreefwatch.noaa.gov/,
do not suggest any bleaching risk in the Bunaken region before June 2016,
hence the wide mortality we observed can not be simply explained by ocean
warming due to El Niño. We also verified on
http://earthquake.usgs.gov/
that between the May 2015 habitat mapping survey and the February 2016 coral
survey, no tectonic movement could generate such a 15 cm–uplift, with an
upward shift of coral colonies relative to sea level, as it has been reported
in different places in the past, including in Sumatra, Indonesia after the
2004 Sumatra Earthquake (Meltzner et al., 2006). An uplift of that magnitude
would be related to a significant event, but there are no reports higher
than a 6.3 magnitude earthquake (16 September 2015, origin
1.884∘ N 126.429∘ E) in the north Sulawesi area for
that period.
Altimetry data have been seldom used to study coral reef processes, even in
a sea level rise era that may affect coral reef communities and islands.
They have been useful to assess the physical environment (wave, tide,
circulation, lagoon water renewal) around islands and reefs (e.g.,
Tartinville and Rancher, 2000; Andréfouët et al., 2001, 2012; Burrage et
al., 2003; Gallop et al., 2014), or explain
larval connectivity and offshore physical transport between reefs (e.g.,
Christie et al., 2010), but this is the first time to our knowledge that
altimetry data, including the new SARAL/AltiKa data, are related to a coral
ecology event. Different measures of sea level and sea level anomalies
confirmed an anomalous situation following the development of the 2015–2016
El Niño, resulting in lower sea level regionally averaging 8 cm in the
north of Sulawesi compared to the previous 4 years (Fig. 3–6). Mortality
patterns on coral colonies strongly suggest that sea level fall is
responsible for the coral die-off that could reach 80 % of reef flat
colonies that were in a keep-up position relative to, usually, rising
sea levels in this region (Fenoglio-Marc et al., 2012).
Top: Map of the 2005–2014 Absolute Dynamic Topography (ADT, in
centimeters) average over Indonesia. Middle: Map of the September 2015 ADT
mean value over Indonesia. The two squares indicate the domain just around Bunaken Island (arrow on top panel) and the north Sulawesi domain used for
the ADT time series presented in Fig. 4. Bottom: Map of correlation
between ADT and the Nino3.4 index (1993–2016, monthly average minus seasonal
cycle).
While mortality due to sea level fall was characterized opportunistically in
Bunaken NP, the impact remains unquantified elsewhere. However, we speculate
that similar events have occurred throughout the Indonesian seas when
considering ADT values for this region (Fig. 6). The stretch of reefs and
islands between South Sumatra, the south of Java, the Flores Sea and Timor;
and the domain centered by the island of Seram and comprised of east
Sulawesi, West Papua and the Banda Sea could have been particularly impacted
by sea level fall. These areas have substantial reef flat presence (e.g., for
the Lesser Sunda region comprised between Bali, Maluku and Timor islands, see
maps in Torres-Pulliza et al., 2013).
Specifically for Bunaken NP, the event we have witnessed helps to explain
long-term observations of reef flat dynamics and resilience. Indeed, our
surveys along with historical and very high-resolution satellite imagery show the fast colonization of reef flats by Heliopora coerula and by carpets of branching Montipora around Bunaken Island in the years 2004–2012, a period congruent to substantial rising sea levels (Fig. 3)
(Fenoglio-Marc et al., 2012). Rising seas have allowed these corals,
especially fast growing and opportunistic ones like H. coerula
(Babcock, 1990; Yasuda et
al., 2012) to cover previously bare reef flats by taking advantage of the
additional accommodation space. A similar process occurred in the Heron Island
reef flats in Australia, with an artificially induced sea level rise due to
local engineering work (Scopélitis et al., 2011). In Bunaken, and
probably elsewhere in Indonesia and the Coral Triangle, the 2015–2016 El
Niño event counter-balances this period of coral growth with rising
seas.
The ADT time series (Fig. 4) suggests that similar low level situations have
probably previously occurred, and almost certainly at least in 1997–1998,
the highest anomaly on altimetry record. Reef flat coral mortality reported
in the Coral Triangle as the consequences of bleaching in these years is
thus most likely also the consequences of sea level fall. The discrimination
between thermal and sea level fall-induced mortality could be difficult to
pinpoint on reef flats if surveys occurred several months after the
thermally-induced bleaching. In Bunaken NP, mortality due to sea level fall
preceded the first occurrences of bleaching in Indonesia by nearly 7 months,
as reported in April 2016. The real impact of sea level fall could have been
largely underestimated during all El Niño episodes and especially in
Asia. The implications for coral reef monitoring in the Coral Triangle are
substantial. Surveys that may have started in April 2016 may have been confused and
assigned reef flat mortalities to coral bleaching. In future years,
monitoring SLA may be as important as monitoring SST. While there are
several SST-indices specifically used as early warning signals for potential
coral bleaching (Teneva et al., 2012), there are no sea level indices
specific for coral reef flats. However, several ENSO indices can help
tracking the likelihood of similar events for Indonesia. The high
correlation between the NINO3.4 index and ADT over the
1993–2016 period (monthly mean minus seasonal baseline, Fig. 6) shows this
potential. Other indices, such as the Southern Oscillation Index (SOI,
computed as the pressure difference between Darwin and Tahiti), or the
equatorial SOI (defined by the pressure difference between the
Indonesia-SLP, standardized anomalies of sea level, and the equatorial
Eastern Pacific SLP) appear to be even more suitable over Indonesia and the
Coral Triangle to develop suitable early warning signals related to sea
level variations.
Conclusions
This study reports coral mortality in Indonesia after an El Niño-induced
sea level fall. The fact that sea level fall, or extremely low tides,
induces coral mortality is not new, but this study demonstrates that through
rapid sea level fall, the 2015–2016 El Niño has impacted Indonesian
shallow coral reefs well before high sea surface temperature could
trigger any coral bleaching. Sea level fall appears as a major mortality
factor for Bunaken Island in North Sulawesi, and altimetry suggests similar
impact throughout Indonesia. Our findings confirm that El Niño impacts
are multiple and the different processes need to be understood for an
accurate diagnostic of the vulnerability of Indonesian coral reefs to
climate disturbances. This study also illustrates how to monitor local sea
level to interpret changes in a particular coastal location. For Indonesia
coral reefs, in addition to sea level fall, depending on the ENSO situation
further changes can be expected due to coral bleaching, diseases, predator
outbreaks, storms and sea level rise (Baird et al., 2013; Johan et al.,
2014). Considering the amount of services that shallow coral reefs offer, in
coastal protection, food security and tourism, the tools presented here
offer valuable information to infer the proper diagnostic after
climate-induced disturbances.
Data availability
Altimetry and tide-gauge data are publicly available from
http://www.aviso.altimetry.fr/en/data.html and
http://uhslc.soest.hawaii.edu/thredds/uhslc_quality_daily.html?dataset=RQD033A
(Caldwell et al., 2015). The authors are not supposed to redistribute these
data. Biological data (coral mortality) are provided in the paper (Table 1
and Fig. 2).
The authors declare that they have no conflict of interest.
Acknowledgements
This study was possible with the support of the Infrastructure Development
of Space Oceanography (INDESO) project in Indonesia, and its Coral Reef
Monitoring Application. Fieldwork on Bunaken Island was authorized by the
research permit 4B/TKPIPA/E5/Dit.KI/IV/2016 delivered by the Ministry of
Research, Technology and Higher Education of the Republic of Indonesia to
SA. This is ENTROPIE contribution 178.
Edited by: G. Herndl
Reviewed by: two anonymous referees
References
Ampou, E. E.: Caractérisation de la résilience des communautés
benthiques récifales par analyse d'images à très haute
résolution multi-sources: le cas du Parc National de Bunaken,
Indonésie, PhD Dissertation; Université Paul Sabatier Toulouse,
France, 2016.
Andréfouët, S., Pages, J., and Tartinville, B.: Water renewal time for
classification of atoll lagoons in the Tuamotu Archipelago (French
Polynesia), Coral Reefs, 20, 399–408, 2001.
Andréfouët, S., Ardhuin, F., Queffeulou, P., and Le Gendre, R.:
Island shadow effects and the wave climate of the Western Tuamotu
Archipelago (French Polynesia) inferred from altimetry and numerical model
data, Mar. Poll. Bull., 65, 415–424, 2012.
Anthony, K. R. N. and Kerswell, A. P.: Coral mortality following extreme low
tides and high solar radiation, Mar. Biol., 151, 1623–1631, 2007.Babcock, R.: Reproduction and development of the blue coral
Heliopora coerulea (Alcyonaria: Coenothecalia), Mar. Biol., 104,
475–481, 1990.Baird, A. H., Pratchett, M. S., Hoey, A. S., Herdiana, Y., and Campbell, S. J.:
Acanthaster planci is a major cause of coral mortality in
Indonesia, Coral Reefs, 32, 803–812, 2013.
Baker, A. C., Glynn, P. W., and Riegl, B.: Climate change and coral reef
bleaching: An ecological assessment of long-term impacts, recovery trends
and future outlook, Estuar. Coast. Shelf S., 80, 435–471, 2008.Birol, F. and Niño, F.: Ku- and Ka-band Altimeter Data in the
Northwestern Mediterranean Sea: Impact on the Observation of the Coastal
Ocean Variability, Mar. Geod., 38, 313–327,
10.1080/01490419.2015.1034814, 2015.
Brown, B. E.: Adaptations of reef corals to physical environmental stress,
Adv. Mar. Biol., 31, 221–299, 1997.
Brown, B. E., Dunne, R. P., Scoffin, T. P., and LeTessier, M. D. A.: Solar
damage in intertidal corals, Mar. Ecol. Prog. Ser., 105, 219–230,
1994.Burrage, D. M., Steinberg, C. R., Mason, L. B., and Bode, L.: Tidal corrections
for TOPEX altimetry in the Coral Sea and Great Barrier Reef Lagoon:
Comparisons with long-term tide gauge records, J. Geophys. Res., 108,
3241, 10.1029/2000JC000441, 2003.Caldwell, P. C., Merrfield, M. A., and Thompson, P. R.: Sea level measured by
tide gauges from global oceans – the Joint Archive for Sea Level holdings
(NCEI Accession 0019568), Version 5.5, NOAA National Centers for
Environmental Information, Dataset, 10.7289/V5V40S7W, 2015.Christie, M. R., Tissot, B. N., Albins, M. A., Beets, J. P., Jia, Y., Ortiz,
D. M., Thompson, S. E., and Hixon, M. A.: Larval Connectivity in an Effective
Network of Marine Protected Areas, PlosOne, 5, e15715, 10.1371/journal.pone.0015715, 2010.
English, S., Wilkinson, C., and Baker, V.: Survey Manual for Tropical Marine
Resources, 2nd edition, ASEAN. ed. Australia Marine Science Project, Living
Coastal Resources, Australian Institute of Marine Science, Townsville,
Australia, 1997.
Fenoglio-Marc, L., Schone, T., Illigner, J., Becker, M., Manurung, P., and
Khafid: Sea level change and vertical motion from satellite altimetry, tide
gauges and GPS in the Indonesian region, Mar. Geod., 35, 137–150, 2012.
Gallop, S. L., Young, I. R., Ranasinghe, R., Durrant, T. H., and Haigh, I. D.:
The large-scale influence of the Great Barrier Reef matrix on wave
attenuation, Coral Reefs, 33, 1167–1178, 2014.
Glynn, P. W.: El Nino-Southern Oscillation mass mortalities of reef corals; a
model of high temperature marine extinctions?, Geol. Soc. Spec. Publ., 178,
117–133, 2000.
Goodwin, I. and Harvey, N.: Subtropical sea-level history from coral
microatolls in the Southern Cook Islands, since 300 AD, Mar. Geol., 253,
14–25, 2008.Guest, J. R., Baird, A. H., Maynard, J. A., Muttaqin, E., Edwards, A. J.,
Campbell, S. J., Yewdall, K., Affendi, Y. A., and Chou, L. M.: Contrasting
patterns of coral bleaching susceptibility in 2010 suggest an adaptive
response to thermal stress, PlosOne, 7, e33353, 10.1371/journal.pone.0033353, 2012.
Hoegh-Guldberg, O.: Climate change, coral bleaching and the future of the
world's coral reefs, Mar. Freshwater Res., 50, 839–866, 1999.
Hoeksema, B.: Delineation of the Indo-Malayan centre of maximum marine
biodiversity: The Coral Triangle, in: Biogeography, Time,
and Place: Distributions, Barriers, and Islands, edited by: Renema, W.,
Springer, 117–178, 2007.
Hopley, D.: Climate change: impact of sea level rise on reef flat zonation
and productivity, in: Encyclopedia of modern coral reefs. Structure, form
and process, edited by: Hopley, D., Springer, Berlin, 2011.Johan, O., Bengen, D. G., Zamani, N. P., Suharsono, Smith, D., Lusiastuti, A. M., and
Sweet, M. J.: Microbial community of black band disease on infection,
healthy, and dead part of scleractinian Montipora sp. colony at
Seribu Islands, Indonesia, Indonesian Aquaculture Journal, 9, 1–11, 2014.Lowe, R., Pivan, X., Falter, J., Symonds, G., and Gruber, R.: Rising sea
levels will reduce extreme temperature variations in tide-dominated reef
habitats, Science Advances, 2, e1600825, 10.1126/sciadv.1600825, 2016.Luu, Q. H., Tkalich, P., and Tay, T. W.: Sea level trend and variability around
Peninsular Malaysia, Ocean Sci., 11, 617–628, 10.5194/os-11-617-2015, 2015.
McPhaden, M. J.: El Niño and La Niña: Physical Mechanisms and Climate
Impacts, in: The Impact of Environmental Variability on Ecological Systems,
edited by Vasseur, D. A. and McCann, K. S., Springer, New York, 1–16, 2007.Meltzner, A. J., Sieh, K., Abrams, M., Agnew, D. C., Hudnut, K. W., Avouac,
J. P., and Natawidjaja, D. H.: Uplift and subsidence associated with the
great Aceh-Andaman earthquake of 2004, J. Geophys. Res.-Sol.
Ea., 111, B02407, 10.1029/2005JB003891, 2006.
Neumann, A. C. and Macintyre, I. G.: Reef response to sea level rise: keep-up, catch-up
or give-up?, Proc. of the 5th Int. Coral Reef Congress, Tahiti, 3-105-110,
1985.Ray, R., Egbert, G., and Erofeeva, S.: A brief overview of tides in the
Indonesian Seas, Oceanography, 18, 74–79, 2005.
Schiermeier, Q.: Hunting the Godzilla El Niño, Nature, 526, 490–491,
2015.
Scopélitis, J., Andréfouët, S., Phinn, S., Done, T., and
Chabanet, P.: Coral colonisation of a shallow reef flat in response to
rising sea level: quantification from 35 years of remote sensing data at
Heron Island, Australia, Coral Reefs, 30, 951–965, 2011.
Smithers, S. G. and Woodroffe, C. D.: Microatolls as sea-level indicators on
a mid-ocean atoll, Mar. Geol., 168, 61–78, 2000.
Suharsono: Ecological and physiological implications of coral bleaching at
Pari Island, Thousand Islands, Indonesia (Electronic Thesis or
Dissertation), University of Newcastle Upon Tyne, 1990.
Tartinville, B. and Rancher, J.: Wave-induced flow over Mururoa atoll reef,
J. Coastal Res., 16, 776–781, 2000.
Teneva, L., Karnauskas, M., Logan, C., Bianucci, L., Currie, J., and
Kleypas, J.: Predicting coral bleaching hotspots: The role of regional
variability in thermal stress and potential adaptation rates, Coral Reefs,
31, 1–12, 2012.
Torres-Pulliza, D., Wilson, J., Darmawan, A., Campbell, S., and Andréfouët, S.:
Ecoregional scale seagrass mapping: A tool to support resilient MPA network
design in the Coral Triangle, Ocean Coastal Management, 80, 55–64, 2013.
Verron, J., Sengenes, P., Lambin, J., Noubel, J., Steunou, N., Guillot, A.,
Picot, N., Coutin-Faye, S., Sharma, R., Gairola, R. M., Raghava Murthy,
D. V. A., Richman, J. G., Griffin, D., Pascual, A., Rémy, F., and Gupta,
P. K.: The SARAL/AltiKa Altimetry Satellite Mission, Mar. Geod., 38, 2–21,
2015.
Wouthuyzen, S., Abrar, M., and Lorwens, J.: Pengungkapan Kejadian Pemutihan
Karang Tahun 2010 di Perairan Indonesia Melalui Analisis Suhu Permukaan
Laut, Oseanologi Dan Limnol. Indones., 1, 305–327, 2015.Yasuda, N., Abe, M., Takino, T., Kimura, M., Lian, C., Nagai, S., Nakano,
Y., and Nadaoka, K.: Large-scale mono-clonal structure in the north
peripheral population of blue coral, Heliopora coerulea, Mar.
Genomics, 7, 33–35, 2012.