Wednesday, April 5, 2017

Falling Sea Level: The Critical Factor in 2016 Great Barrier Reef Bleaching!

It is puzzling why the recent 2017 publication in Nature, Global Warming And Recurrent Mass Bleaching Of Corals by Hughes et al. ignored the most critical factor affecting the 2016 severe bleaching along the northern Great Barrier Reef – the regional fall in sea level amplified by El Niño. Instead Hughes 2017 suggested the extensive bleaching was due to increased water temperatures induced by CO2 warming. 

Reef at Low Tide Around Lizard Island Great Barrier Reef

In contrast in Coral Mortality Induced by the 2015–2016 El-Niño in Indonesia: The Effect Of Rapid Sea Level Fall by Ampou 2017, Indonesian biologists had reported that a drop in sea level had bleached the upper 15 cm of the reefs before temperatures had reached NOAA’s Coral Reef Watch bleaching thresholds. As discussed by Ampou 2017, the drop in sea level had likely been experienced throughout much of the Coral Triangle including the northern Great Barrier Reef (GBR), and then accelerated during the El Niño. They speculated sea level fall also contributed to the bleaching during the 1998 El Niño. Consistent with the effects of sea level fall, other researchers reported bleaching in the GBR was greatest near the surface then declined rapidly with depth. Indeed if falling sea level was the main diver in 2016’s reef mortalities, and this can be tested, then most catastrophic assertions made by Hughes 2017 would be invalid.

Indeed the Great Barrier Reef had also experienced falling sea levels similar to those experienced by Indonesian reefs.  Visitors to Lizard Island had reported more extreme low tides and more exposed reefs as revealed in the photograph above, which is consistent with the extremely high mortality in the Lizard Island region during the 2016 El Niño. Of course reefs are often exposed to the air at low tide, but manage to survive if the exposure is short or during the night. However as seen in tide gauge data from Cairns just south of Lizard Island, since 2010 the average low tide had dropped by ~10 to 15 cm.  After previous decades of increasing sea level had permitted vertical coral growth and colonization of newly submerged coastline, that new growth was now being left high and dry during low tide. As a result shallow coral were increasingly vulnerable to deadly desiccation during more extreme sea level drops when warm waters slosh toward the Americas during an El Niño. 

Furthermore, an El Niño in the Coral Triangle not only causes a sudden sea level fall, but it also generates a drier high-pressure system with clear skies, so that this region is exposed to more intense solar irradiance. In addition, El Niño conditions reduce regional winds that drive reef-flushing currents and produce greater wave washing that could minimize desiccation during extreme low tides. And as one would predict, these conditions were exactly what were observed during El Niño 2016 around Lizard Island and throughout the northern GBR.

Aerial surveys, on which Hughes 2017 based their analyses, cannot discriminate between the various causes of bleaching. To determine the cause of coral mortality, careful examination of bleached coral by divers is required to distinguish whether bleached coral were the result of storms, crown-of-thorns attacks, disease, aerial exposure during low tides, or anomalously warmer ocean waters. Crown-of-thorns leave diagnostic gnawing marks, while storms produce anomalous rubble. Furthermore aerial surveys only measure the aerial extent of bleaching, but cannot determine the depth to which most bleaching was restricted due to sea level fall. To distinguish bleaching and mortality caused by low tide exposure, divers must measure the extent of tissue mortality and compare it with changes in sea level. For example, the Indonesian researchers found the extent of dead coral tissue was mostly relegated to the upper 15 cm of coral, which correlated with the degree of increased aerial exposure by recent low tides. Unfortunately Hughes et al never carried out, or never reported, such critical measurements.

However a before-and-after photograph presented in Hughes 2017 suggested the severe GBR bleaching they attributed to global warming primarily happened between February and late April. Their aerial surveys occurred between March 22 and April 17, 2016. And consistent with low tide bleaching, that is exactly the time frame that tide tables reveal reefs experienced two bouts of extreme low tides coinciding with the heat of the afternoon (March 7-11 & April 5-10). And such a combination of sun and low tide are known to be deadly.

A study of a September 2005 bleaching event on Pelorous and Orpheus Islands in the central GBR by Anthony 2007, Coral Mortality Following Extreme Low Tides And High Solar Radiation, had reported extreme deadly effects when extreme low tides coincided with high solar irradiance periods around midday. As in Indonesia, they also reported bleaching and mortality had occurred despite water temperatures that were “significantly lower than the threshold temperature for coral bleaching in this region (Berkelmans 2002), and therefore unlikely to represent a significant stress factor.” Along the reef crests and flats, “40 and 75% of colonies in the major coral taxa were either bleached or suffered partial mortality. In contrast, corals at wave exposed sites were largely unaffected (<1% of the corals were bleached), as periodic washing of any exposed coral by waves prevented desiccation. Surveys along a 1–9 m depth gradient indicated that high coral mortality was confined to the tidal zone.” [Emphasis mine]

The fortuitous timing of Ampou’s coral habitat mapping from 2014 to 2016 in Bunaken National Park (located at the northwest tip of Sulawesi, Indonesia) allowed researchers to estimate the time of coral mortality relative to sea level and temperature changes. Ampou reported that 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. By March 2016, Bunaken Island (North Sulawesi) displayed up to 85% mortality on reef flats” and that almost “all reef flats showed evidence of mortality, representing 30% of Bunaken reefs.” Based on the timing of reef deaths and changes in temperature they concluded, “the wide mortality we observed can not be simply explained by ocean warming due to El Niño.”  They concluded, “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.”

From the illustrations (below) of a generalized topography of a fringing or barrier reef, we can predict the effects of low sea level by examining where bleaching and mortality would occur within the whole reef system. Coral occupying the reef crests are most sensitive to drops in sea level and desiccation because they are first to be exposed to dangerous periods of aerial exposure and last to re-submerge. The inner reef flats are vulnerable to lower sea levels, as those shallow waters are more readily exposed at low tide because the reef crest prevents ocean waters from flooding the flats. If reefs flats are not exposed, the shallow waters that remain can heat up dangerously fast. Accordingly Anthony 2007 found 40 to 75%, and Ampou 2017 found 85% of the reef flats had bleached. In contrast coral in the fore reefs are the least vulnerable to desiccation and higher temperatures due to direct contact with the ocean, upwelling and wave washing. Accordingly Anthony 2007 reported <1% bleaching in the fore reefs.


Coral mortality due to a drop in sea level leaves other diagnostic telltale signs such as micro-atoll formation. As illustrated below in Fig. 4 from Goodwin 2008, during neap low tides (MLWN) sea water can still pass over the reef crest and flush the inner reef with relatively cooler outer ocean water. However during the low spring tides (MLWS), the reef crest is exposed and ocean water is prevented from reaching the reef flats. As mean sea level falls (MSL), coral on the crest and flats are increasingly exposed to the air for longer periods, and the upper layer of coral that had previously kept up with decades of rising sea level, are now exposed to increasing periods of desiccation and higher mortality.

There are over 43 species in the coral triangle that can be characterized as “keep-up” coral whose growth rates are much greater than average 20th century sea level rise. However their vertical growth is limited by the average low water level (HLC-Height of Living Coral in Fig. 4). Average low water level is calculated as the mean water level between low neap tides and lower low spring tides. (Due to the linear alignment of the sun, earth and moon and the resulting stronger gravitational pull during a full and new moon, spring tides result in both the highest high tides and lowest low tides. In contrast neap tides exert the least gravitation pull.  Spring tides typically happen twice a month, but usually no more than once a month will spring low tides coincide with the heat of the midday sun.)

When growing in deeper waters, a keep-up species like mounding Porites spp. grow at rates of 5 to 25 mm per year and form dome shaped colonies. However due to increased aerial exposure when growth reaches the surface, or due to exposure from sea level fall, the upper most surface dies from high air temperatures, higher UV damage and desiccation. This results in a flat-topped colony leading to the classic “micro-atoll” shape, with dead coral in the center surrounded by a ring of live coral, as exemplified by a Kiribati micro-atoll in the photograph below.


Micro-atoll patterns have been crucial for reconstructing past fluctuations in sea level on decadal to millennial timeframes. As Ampou 2017 observed in Bunaken NP, mortality due to a drop in sea level was mostly restricted to the upper 15 cm of coral, which leads to the formation of micro-atolls.  So before simply assuming climate-change-warming has induced mortality, micro-atoll formation and other associated patterns indicative of sea level change must be examined. A short discussion on how sea level changes can shape micro-atolls can be read here.

Due to its regional sensitivity to the sea level change that accompanies an El Niño, the northern Great Barrier Reef has an abundance of fossil micro-atolls that have allowed researchers to estimate El Niño activity and fluctuating sea levels over the past 4000 years. They estimated 4000 years ago low water neap tides were at least 0.7 meters higher than they are at present. Studies of micro-atolls in the Cook Islands further to the east in the southern Pacific, suggest that by 1000 AD during the Medieval Warm Period, average sea level had fallen, but remained about 0.45 meters higher than today. During the Little Ice Age sea level fell to 0.2 meters below current levels during the late 1700s and early 1800s, before recovering throughout the 1900s.

Hughes 2017 wanted to emphasize GBR bleaching as a “global-scale event” in keeping with his greenhouse gas/global warming attribution, but bleaching and mortality was patchy on both local and regional scales. And although Hughes presented their analyses as “a fundamental shift away from viewing bleaching events as individual disturbances to reefs,” the unusually high mortality around Lizard Island demands a closer examination of individual reef disturbances. The lack of mortality in 2016 across the southern and Central GBR, was explained as a result of the cooling effects of tropical storm Winston, but that does not explain why individual reefs in those regions have not bleached at all, while others bleached only once, and still others bleached twice or three times since 1998. Hughes’ shift away from examining what factors affected individual reefs will most likely obscure the most critical factors and yield false attributions.

Hughes reported the various proportions of areal bleaching as degrees of severity. But that frightened many in the public who confused bleaching with mortality, leading some misguided souls to blog the GBR was dead.  However bleaching without mortality is not a worrisome event no matter how extensive. Rates of mortality and recovery are more important indices of reef health. As discussed in the article The Coral Bleaching Debate: Is Bleaching the Legacy of a Marvelous Adaptation Mechanism or A Prelude to Extirpation?, all coral retain greater densities of symbiotic algae (symbionts) in the winter but reduce that density in the summer, which often leads to minor seasonal bleaching episodes that are usually temporary. Under those circumstances coral typically return to normal within weeks or months. Furthermore by ejecting their current symbionts, coral can acquire new symbionts that can promote greater resilience to changing environmental conditions. Although symbiont shifting and shuffling promotes adaptation to shifting ocean temperatures, symbiont shuffling cannot protect against extreme low tide desiccation, and dead desiccated coral can no longer adapt. Humans have little control over El Niños or low tides.

Hughes also contradicted past studies to mistakenly suggest that recurring bleaching in a given reef is evidence that corals are not adapting or acclimating. However bleaching happens for many reasons. Symbiont shuffling to better adapt to warmer waters does not guarantee adaptation to lower sea levels, cyclones or changes in salinity. Coral reefs deal with changing sea levels with rapid growth to keep-up as sea level rises, and then dying back when sea level falls. Decadal swings in regional sea level will likely cause decadal swings in bleaching and are not evidence of coral fragility.

Hughes 2017 modeled the 2016 GBR bleaching event as a function of surface ocean temperatures that surpass bleaching thresholds, although reefs will bleach below that threshold and will fail to bleach despite temperatures above that threshold. Despite the fact El Niños are well known to cause rapid sea level fall along the GBR, Hughes’ model never accounted for falling sea level. Nor did they account for past observations that falling sea levels induced bleaching when temperatures were below bleaching thresholds. More disturbing because sea level fall caused bleaching in various reefs, with some experiencing good water quality and others poor quality, Hughes asserted there was “no support for the hypothesis that good water quality confers resistance to bleaching.” However this contradicts an abundance of regional studies attributing increased coral disease and bleaching to high nutrient loading.

Woolridge 2013 have argued that coral eject their algal symbionts and bleach when temperature, light and nutrients increase to a level that accelerates the symbionts growth. Increased growth consequently reduces the amount of energy transferred to the coral, resulting in ejection of the slacking symbiont. Because increased nutrient loads can promote increased symbiont growth at relatively lower temperatures, higher nutrient loads can promote bleaching at lower temperatures.

Furthermore while coral’s symbiotic relationships allow them to recycle limited nutrients and out compete seaweeds, higher nutrient loads enable greater seaweed growth, which reduces corals’ competitive advantage. Furthermore seaweeds have been shown to harbor allelopathic chemicals that inhibit coral growth, as well as serving as reservoirs for bacteria that cause coral diseases. Higher nutrient loads induce more dissolved organic carbon that bacteria feed upon, allowing disease-causing bacteria to rapidly multiply. Higher nutrient loads also increase the survival of crown-of-thorns larvae, which then increases coral depredation and bleaching.

In a 2013 experimental study, Chronic Nutrient Enrichment Increases Prevalence And Severity Of Coral Disease And Bleaching, Vega-Thurber reported that higher nutrient loads caused a “twofold increase in both the prevalence and severity of disease compared with corals in unenriched control plots” as well as a “3.5-fold increase in bleaching frequency relative to control corals.”

Although Hughes 2017 suggests the pattern of recurring bleaching is simply a function of temperature and global warming, as illustrated in Hughes’ Figure “e” below, recurring bleaching is not a global phenomenon. (Black dots represent reefs that bleached during all 3 surveys: 1998, 2002, 2016; light gray represents reefs that bleached only once, and dark gray reefs bleached twice.) . In most cases the degree of recurring bleaching does not predict the recurrence of bleaching in nearby reefs despite similar ocean temperatures. Although an El Niño generates widespread bleaching, bleaching is still a regional issue affecting individual reefs differently. During an El Niño sea level rises in the eastern Pacific and falls in the western Pacific. Recurring bleaching in the Far North and Southern regions of the GBR are uncommon, while recurring GBR bleaching has been frequent between Cookstown and Townsville where temperatures have been quite variable. And in accord with prior research, the region between Cookstown and Townsville has suffered from lower water quality and higher nutrients loads, causing more frequent bleaching and greater crown-of-thorns attacks.

Reefs visited during 3 surveys and recurring bleaching

After perusing Hughes 2017, it was clear they had been led to incorrectly embrace the prevailing bias of  CO2-induced catastrophic bleaching because they failed to address the fall in sea level before and during the 2016 El Niño, and likewise they failed to address how weather created by El Niños promotes clear skies and increased solar heating. To add insult to injury, because sea level drops bleached reefs in both good water quality and bad, and bleaches reefs in both protected preserves and unprotected, Hughes 2017 presented a statistical argument that disparaged any significant value of ongoing conservation efforts to minimize bleaching by reducing nutrient loading and by protecting reefs from overfishing. By belittling or ignoring most critical factors affecting coral bleaching other than temperature, Hughes suggested our only recourse to protect reefs “ultimately requires urgent and rapid action to reduce global warming.”

And because such an apocryphal analysis was published in Nature and will undoubtedly mislead coral conservation policies,

I wept.