Skip to content

Adaptation to Climate Change Impacts on Coral Reefs…By Sliding our Baselines.

Facebooktwitterredditpinterestlinkedin

It was early June 1998.  I was at a workshop at University of Miami, at which we finalized the methodology for what came to be known as AGRRA – the Atlantic and Gulf Rapid Reef Assessment.  AGRRA was the brainchild of Bob Ginsberg, noted reef geologist at University of Miami.  He thought it would be useful to have a rapid, easy-to-use method with which the science and diver community could survey reefs across the Caribbean and capture their current status.  Bob knew that we only value that which we know and care about, and we only know that which we have measured, counted, observed.  Further, we cannot tell if something as complicated as a coral reef is changing, unless we first measure the state it is in.  The problem, as is usually the case, is that funds are seldom plentiful and the Caribbean is a big place – so how to get a synoptic survey of the status of its reefs?  Using haphazard measurements by different people in different places would be next to useless.  Using sophisticated (expensive) methods to mount a comprehensive survey would be impossible.  And so… AGGRA.

We did not know, as we assembled in Miami from all over the Caribbean and places beyond, that the 1998 global mass bleaching event was just under way, and would prove to be a major wake-up call to the reef science community.  Over the following year, as I discussed in Our Dying Planet, reefs bleached under unusually high water temperatures and corals died.  But, I must stop digressing, because I really want to tell about one fragment of conversation from that workshop – a fragment that jarred me then, and has stuck with me for 15 years.  During one of the break-out sessions, a group of us were discussing the proposed method for monitoring coral abundance and condition.  As a fish expert, I was listening with interest, marveling at how much simpler it is to count and measure corals, which sit quietly waiting to be investigated.  Fish do not sit still, and counting and measuring them has provided many hours of experimentation, more hours of argumentation, and a number of robust “visual census” methods that are far more difficult to do well than most of us who use them are willing to admit.

!CS3

Counting fish on a coral reef.  Photo © Cindy Shaw

The procedure being proposed involved setting out a tape measure across the reef to create a 10 meter long ‘line transect’ – a line along which the investigator can count or measure corals.  (You do not have to count and measure corals along straight lines, but the line transect is a good way of focusing the mind, and getting the investigator to concentrate on what he or she is supposed to be doing, instead of gazing about, taking photos, and generally marveling in the wonder of the coral reef.)  Along this transect, the instruction was to select the first 10 coral colonies more than 10 cm in diameter, and take a series of measurements – diameter in two dimensions (length and width), height, amount of the colony, if any, that was dead, species, and so on.  And then came the fragment I remember….

A scientist from Mexico, I think from Veracruz, asked, “What should I do if none of the coral colonies along the transect are 10 cm in diameter?”

The discussion leader seemed momentarily confused, “I’m sorry, what do you mean?”

“Well, I think that on our reefs… they are not in very good condition, you know … and they have very few coral colonies still growing on them.  I think very few of them would be that big.  What do I do if I cannot find 10 colonies 10 cm in diameter?”

And then it hit me.  And the discussion leader.  And I dare say lots of the other participants.  Reefs differ, and some reefs would not even be recognized as reefs by some people.  Or, more directly, there are places where people have reefs they value which are in such degraded condition that people from more fortunate places would not even bother to call them reefs.  People living in Florida would just call them “hard bottom”, to distinguish them from sandy or silty substrata.  The interesting thing is that in places where reefs are already seriously degraded, we can find plenty of people who still value them, who would like to conserve them, and who long for effective ways to make them thrive.  I suspect the same is true for ravaged rainforests, and other damaged, degraded, eviscerated ecosystems as well.

Those individuals who sense the value in the natural world, those who think the loss of a species is something that matters, even when they cannot say how it might matter; those individuals do not give up trying until the situation is absolutely hopeless.  Indeed, they may continue their feeble efforts to conserve long after the task is hopeless.  To that Mexican scientist, that poor Veracruz reef, with its sparsely scattered, miserably few and small coral colonies, was still a reef, and still worth caring about.  That she had a snowball’s chance of actually bringing it back to health did not matter.  Faced with a degrading world, we shift our baselines, redefining as a way of avoiding the nasty reality of degradation.  What once would have been dismissed as “not a reef” is now seen as a coral reef worth trying to save and rehabilitate.  Since 1998 we have all been doing this recalibrating, quietly shifting our baselines, redefining what a coral reef is, so that degraded reefs remain reefs, and so that we can hide from the evidence of our destruction of the world.

On the one hand, we can celebrate this capacity of caring people to continue to try long after the case has become hopeless; on the other hand, our shifted baselines hide from us, and from those we talk to, the cosmic severity of the environmental problems we now face.

Hold that thought while I discuss two recent studies of human impacts on coral reefs, and bring us up to speed on coral reef decline.  Then I’ll come back to baseline shifting.

Since 1998, the status of coral reefs has got worse.  And AGRRA data are among the data that have been used to document just how much worse it has got.  In 2003, Toby Gardner and colleagues from the University of East Anglia in Norwich, UK, published an article in Science showing what I call a triple-D trend for Caribbean coral reefs – Downward, and Definitely Depressing.  They amassed data from numerous studies from all over the Caribbean to do this study, and showed that average coral cover had fallen from approximately 50% to about 10% in the space of 25 years.  Nobody has since suggested there was something wrong with their data or conclusions.

Fig 2 A Gardner at al coral cover

Figure 2A from Gardner et al, Science, 2003, showing how the percentage of surface area of Caribbean coral reefs that is covered by live coral has fallen from 1977 to 2002.  Two slightly different ways of estimating average coral cover are shown – the lines are closely overlapping.  Also shown (open circles) are the number of separate studies on which each estimate is based.

Four years later, John Bruno and Elizabeth Selig, of University of North Carolina at Chapel Hill, did a very similar study for the Indo-Pacific, publishing it in the journal Plos One (actually, they collected data from only part of the vast Indo-Pacific, but still a much larger region than the Caribbean).  They chose to provide their results as a long, narrow amalgamation of trends for each of a series of regions within their study area.

Figs 1 & 2 Bruno & Selig 2007

Figures 1 and 2 from the article by John Bruno and Elizabeth Selig.  The subregions of the Indo-Pacific from which they obtained data are shown in Figure 1 (right-hand map), and the percent of coral cover in 2002 or 2003 is shown for each in Figure 2A (graph at upper left).  The remaining sections of Figure 2 show the distribution of coral cover across reefs in each of four sub-regions in the early 1980s and again around 2003.  Distributions of coral cover vary among regions, but in every region there is a noticeable drop in coral cover over the ~20 year period.

Since John Bruno’s article, there have been a number of studies of reefs in particular locations all telling much the same story.  One of the most compelling was the report last year in the Proceedings of the National Academy of Sciences (USA), by Glenn De’ath and colleagues at Australian Institute of Marine Science.  They documented the loss in coral cover across the Great Barrier Reef during the past 27 years.  Based on a rigorous and comprehensive monitoring program in place since 1985 (which samples shallow portions of entire outer slopes of each reef sampled each year), they documented a 50.7% loss of coral cover (from an average of 28% to 13.8% over the 27 years).  Globally, the causes of decline vary from region to region and from year to year.  They include such stresses as overfishing and pollution – stresses that are widely present on coral reefs, and have been around almost since people ever came into contact with reefs.  They include outbreaks of the Crown-of-Thorns starfish, which themselves seem more prevalent since the 1960s than in the past, and may be partially a consequence of agricultural nutrification of coastal waters – the enriched waters, particularly following intense rain events and flooding, can favor larval survival leading to very much enhanced settlement of young starfish, and a few months later, large populations of hungry coral predators.  They include impacts of tropical storms, which may be becoming more intense due to our impacts on climate.  And they increasingly include two direct impacts on corals of our releases of CO2:  Warming leads to episodes of unusually warm water, and mass coral bleaching which can lead, as it did in 1998, to widespread death of coral.  And ocean acidification impedes the process of calcification (by making the process more energetically demanding), used by corals and other reef creatures to build their skeletons, and therefore to build or repair reef.

In recent weeks, scientists have published more evidence of how degraded coral reefs are becoming, and how powerful the impacts of climate change on reefs now are.  Dr. Sophie Dove and five colleagues from the University of Queensland took the superficially simple approach of putting coral reefs into the future to see how they would survive once climate had changed.  In a series of large 300 litre (75 gallon) plastic tubs plumbed for flowing seawater at the Heron Island Research Station, they assembled small patch reefs made up of corals and other critters collected from the nearby reef.

Some of these mesocosms – so-called because they are too big to be microcosms, and yet clearly not as big as the real world – were left to track current conditions of dissolved CO2 concentration and temperature for Heron Island waters.  Others were made to mimic the same daily and seasonal fluctuations, but for average temperatures and CO2 concentrations of 100 years ago.  Two other groups mimicked the same daily and seasonal fluctuations but around mean concentrations of CO2 and mean temperatures as expected for 2050 under two different GHG emissions trajectories – the A1F1, ‘business-as-usual’ path, and the B1, ‘keep temperature increases below 2oC’ path.   Water temperature and CO2 concentrations were controlled by a system that monitored, and adjusted as necessary, to keep CO2 to the required level and shift temperature according to the precise daily and seasonal patterns measured one year previous at a specific site at Heron Reef, the same site from which the corals for the small patch reefs had been collected.

Dove PNAS 2013 temperature profiles in tanks

The temperature records for the four sets of mesocosms during the course of the experiment.  All follow a daily and seasonal pattern of temperature change that mimics what happened out on the nearby reef one year previous.  But the average temperatures are off-set by the amounts appropriate for 100 years ago, today, or in 2050 under two different climate change scenarios.  Image © Dr. Sophie dove

The experiment ran for three months from late spring through summer, November 2011 to February 2012, and the results were published on line on 3rd September at Proceeding of the National Academy of ScienceHere is a brief summary.

Dr. Dove found that the patch reefs thrived under the conditions of 100 years ago, performed well, but not quite so well under present-day conditions, and degraded significantly under both scenarios for 2050.  Under the A1F1 future conditions, corals began to bleach during November, but bleaching did not occur in other treatments until the warmer days of February just before the experiment ended.  Corals living under conditions of 100 years ago did not bleach except for the temperature-sensitive Stylophora, and those experiencing today’s conditions showed only slight bleaching, as was happening out on the reef that February.  In the mesocosms representing 2050, colonies of Acropora, Seriatopora, Montipora, and Stylophora died and were overgrown by algae; colonies of Goniastrea, Porites, Lobophyllia, and Fungia, survived to the end of the experiment without shrinkage or overgrowth, although they exhibited severe bleaching for more than 2 months.  By the end of the experiment, the patch reefs exposed to the A1F1 future were pretty well dead, and those in the B1 future were not much better off.

Dove PNAS 2013 results for patch reefs

Photographs of typical mesocosms at times during the experiment.  Bleached corals are clearly evident in the two 2050 treatments, while the 100 years ago treatment looks lushest of all.  Photo © Dr. Sophie Dove.

During the experiment, rates of calcification were also measured.  These measurements revealed that the patch reef-wide rate of calcification was consistently negative in the A1F1 mesocosms, but was positive, though not much above zero in the B1 mesocosms.  Calcification rate was still higher in the mesocosms experiencing ambient conditions, and even greater in those subjected to conditions of 100 years ago.  (Remember that a coral reef has to maintain a positive net rate of calcification, because wave action, storm events, and the activities of a wide range of grinding, rasping, chewing bioeroders are all working to pulverize reef into sand – without positive rates of calcification, the reef just disappears.)

In setting out their conclusions, Dove remarks that the business-as-usual future to which we are heading is “very likely to have serious impacts on coral reef ecosystems over summer periods when temperatures and seawater CO2 concentrations are at their highest.  Under this scenario, carbonate substrates will tend to dissolve, and the community structure will bear little resemblance to the coral-dominated reefs of today”.  I translate this science-speak to mean that there will be lots of coral bleaching and mortality each summer, while the carbonate rock and sand of the reef slowly dissolves away, and the animals and plants able to survive will form a community that does not look a bit like a coral reef!

She goes on to note that aggressively reducing CO2 emissions (B1) will preserve some possibility of reef growth, and then she writes

“Perhaps of great interest is the observation that conditions [of 100 years ago] resulted in higher reef and coral calcification rates and reduced levels of coral mortality and bleaching, … suggesting that, despite the passage of more than 100 y, the [species of coral] included in the present study are better suited to conditions that occurred in the past than to those of the present day”.

Again, more science-speak – she means conditions have deteriorated over the past 100 years, and corals have not been able to adapt to the changes.  And then she says,

“This … has important implications for current speculations as to whether coral reef ecosystems will be able to adjust over the next 100 y to preserve the array of key ecological goods and services that typically are associated with their CaCO3 structures”,

which is a good scientist’s way of saying “this result tells me that corals will not be able to adapt to the greater changes that are coming this century, and coral reefs as we know them are down the tube.”

A couple of months earlier, Emma Kennedy of Exeter University, and 10 colleagues from Mexico, Germany, Israel, USA, Australia and UK published a modeling study of human impacts on coral reefs in Current Biology.  I commented previously on this article with respect to ocean acidification, but their study spanned the full range of human impacts on reefs and is relevant here.  By combining ecological models of Caribbean reefs with carbonate budgets, and subjecting the models to changes in temperature and CO2 concentration (and therefore pH) anticipated under various scenarios for the future, they were able to look at status of coral reefs towards the end of this century, and explore the effect on that status of different patterns of local management.  Their Figure 3 nicely summarizes the main results.

reef persistence Kennedy et al Current Biol 2013

Results of simulations assuming either weak local management (A,C,E,G) or strong local management sufficient to keep herbivores present and abundant (B,D,F,H), either a starting situation with low (10%) coral cover (A,B,E,F) or a better reef condition (20% coral cover)(C,D,G,H), and either a BAU approach to GHG mitigation (A,B,C,D) or an aggressive effort to reduce emissions (E,F,G,H).  Better local management buys time in all cases (the vertical blue bar, marking dates when carbonate budgets cross zero to become negative, shifts right), and good local management, better starting conditions, and strong effort to reduce emissions is the only future in which reefs persist beyond 2100.  Figure from Kennedy et al Current Biology 2013.

 

Under most sets of conditions, net calcification declines and reaches zero relatively early in the century.  Using aggressive efforts to reduce CO2 emissions delays the on-set of negative rates of calcification in all cases by a decade or more.  So does management of fisheries to ensure that large parrotfishes remain present on the reefs.  Their intensive grazing keeps algal species in check and provides ‘open’ space for new coral planulae to settle and grow.  In situations where reefs begin with relatively high coral cover (rather than beginning already degraded), and are managed for high parrotfish abundance, and in which CO2 emissions are kept low, the simulations show reefs growing and persisting beyond the end of the simulation (2080).

Emma Kennedy and her colleagues conclude their study with

“…our results suggest that local interventions are far from futile, and indeed are essential for assuring sustained ecosystem functioning.  Unfortunately, only three countries in the region have taken steps to protect herbivorous fish throughout their coastal zone (Belize, Bermuda, and Bonaire), so protection is usually confined to small no-take marine reserves.  We also provide unambiguous evidence that local efforts must be accompanied by rigorous global action to mitigate climate change.”

That’s a bit science-speaky; they are stating that their study suggests that if Caribbean countries wish to retain functional reef systems they must manage fisheries to ensure presence of parrotfishes and there must be aggressive global efforts to reduce CO2 emissions.  Without both of these, reefs are likely cooked.  They also note, politely, that only three countries in the region have enacted legislation that might keep parrotfishes abundant (fishery managers have known the value of this for at least a decade).

Putting the Kennedy and Dove studies together, along with the evidence of global reef decline over the last quarter century, it is abundantly clear that we have already lost a substantive proportion of the living coral that makes a limestone bench into a functional coral reef.  It’s also clear that the trend continues downward.  And the modeling and mesocosm studies show that without aggressive change to our emissions of CO2 the downward trend will continue through this century.  The suggestion that there will be anything around by 2050 resembling what we used to call coral reefs in the 1960s and 1970s seems naïve without a substantive change in our behavior.  To put this in context for readers who are more familiar with terrestrial ecosystems, the global rate of loss of reef coral cover is much faster than the rate of loss of rainforest area, and the trend is driven by a multiplicity of factors including ones related to CO2 emissions that can only be mitigated with global action.  We are losing our coral reefs far more rapidly than we are losing our rainforests.

And so we adapt.  We do not change our behavior – or at least we have not done so yet, and seem pretty resistant to changing it.  We still fish parrotfishes, and we still consume fossil fuels as fast as we can.  No, we adapt by shifting our baselines.  We quietly redefine what a coral reef is.  And what disturbs me is that it is the science community as much as everyone else that is engaged in this blatant self-deception.  Most of us are being very circumspect in saying what appears to be happening, and we obscure it further by cloaking it in science-speak.  No, we cannot know with certainty what the world’s reefs will be like in 2050, but we have very strong evidence that they are likely to be very badly, if not terminally degraded.  Yet we persist in seeing positive signs, reasons to remain optimistic.

OLYMPUS DIGITAL CAMERA

Acropora palmata used to be so abundant across the Caribbean, forming vast, nearly impenetrable ramparts, that the upper reef slope was formally named “the Acropora palmata zone” – now it occurs as scattered individual colonies, and is on the endangered species list.  Photo © Nick Hopgood.

We scientists jump to be first to criticize a colleague such as Roger Bradbury, of Australian National University, who had the temerity to write an op-ed piece in the New York Times in July 2012 in which he referred to coral reefs as “zombie ecosystems … on a trajectory to collapse within a human generation”, a scientist who at least tried to tell it like it is.  Maybe he was overly pessimistic, maybe not, but he was trying to tell the truth as he saw it.  Apparently reef scientists are expected to ensure they give people hope, by “not just preaching doom and gloom.”  I thought religion had the task of giving people hope, while science was about striving to discover the truth.  But no, science is now about shaping the message so it will give listeners hope for the future.  When you have lost 50% of coral cover in 27 years, when the trend shows no signs of slowing, when there is only 14% coral cover remaining (these percentages come from De’ath’s Great Barrier Reef study), just how many years do you wait before you decide to say that the Great Barrier Reef is actually disappearing?  Do you wait until there are only sparsely scattered, miserably few and small coral colonies left, or do you, even then, still consider it a reef, like that poor degraded reef in Veracruz, Mexico?

Increasingly, we find reputable scientists pointing to the exceptional places.  Places like the Line Islands, those reefs in the central Pacific about as far away from any human influences as a reef can be, reefs that still support large sharks, and provide opportunities for iconic photos and video.  Telling people about the magnificence of Kingman Reef does confirm that it is not yet too late to act, but it also helps fuel the false hope that all will be well after all, and I, personally, do not have to change my behavior at all.  Most readers do not pick up on the subtle idea that such places, by their very exceptionalism, point out how serious the situation is more generally.  Instead, they see pictures of lush coral reefs and read exciting tales of big fish and miles of living corals, and they assume statements that reefs generally are disappearing are probably exaggerated.  Other scientists continue to speak almost reverentially about the value of marine protected areas – build more, they help sustain coral reefs.  Yes, they can be of great value in restoring reef resilience, but only if they are well designed and well managed – and the great majority are not.  Building paper parks simply wastes valuable time and effort, and paper parks are being built as I write.  Meanwhile, the reefs that do get impacted by people degrade even further, and we passively watch them drift towards nothingness.  Watch as they dwindle away, and still we call them reefs.  Yes, Virginia, there really is such a thing as a coral reef.  I saw one once.  I’m sure you will see one some day.

2 thoughts on “Adaptation to Climate Change Impacts on Coral Reefs…By Sliding our Baselines.”

Comments are closed.