Although Tracy Ainsworth had visited Australia’s Great Barrier Reef many times over the past decade, when she arrived in January 2020 it quickly became clear that something was different. During that trip, the massive reef system experienced its worst bleaching on record, the third such record-setting event in five years. Being there, she recalls, was a “whole-body experience.” Mucus fanning out from stressed corals has a distinctive odor, she notes, and spending so much time in the water, which was pushing 30 °C, was causing Ainsworth and her visiting students great anxiety. “I’ve spent a lot of time there, and [I’ve] never seen anyone have a panic attack because the water is so hot.” 

The corals, too, seemed stressed by the balmy waters. As the summer wore on, many grew pale or bone-white, a sign that they were losing their algal symbionts. “Normally there’s parts of the reef that aren’t badly affected,” particularly those living in deeper water, which typically provides some refuge from the heat, Ainsworth, a coral biologist at the University of New South Wales, tells The Scientist. But that year, she says, she saw the bleaching extend miles offshore. “The scale becomes quite confronting.”

But Ainsworth knew better than to count the corals out. During consecutive bleaching events in 2016 and 2017, she had seen colonies wrecked by the first event recover and fare better in the second. And she’s not the only one to have noticed that corals seem to “remember” past heat exposure and acclimate to subsequent warming. 

The historical use of satellites to study corals has given researchers some sense of how this “environmental memory” manifests at a broad level—with some reefs, or individual coral colonies on a reef, appearing to become more resilient than others from one event to the next. Lab studies have confirmed that sublethal levels of stress can indeed render corals more likely to survive high heat. It’s not clear, however, how long memory might last. An early study detailing work done in the 1990s suggests that corals might remember past exposures for a decade or more, but the timescale can also be quite short, limited to just a few weeks, supporting the argument that these changes reflect relatively rapid acclimation by long-lived coral colonies and not adaptation of corals over generations.

The speed at which reefs are degrading is pushing many scientists toward action even in the absence of complete information.

Increasingly, researchers are paying attention to the winners and losers of repetitive hardships to unpack the individual- and species-level differences in memory. “You see that variability across the reef, where some [corals] are totally pale and bleached and others look fully pigmented,” says Hollie Putnam, a molecular ecophysiologist at the University of Rhode Island. “I think that kind of natural variability sets up a really nice context to ask . . . mechanistic questions” about environmental memory. In recent years a panoply of new molecular tools has helped scientists determine how heat elicits changes in corals, and how exposures affect their algal symbionts, potentially preparing the animals for future harsh conditions.

A GIANT REBORN: The Florida Keys experienced record-breaking summer heat in 2014 and 2015, which led to back-to-back bleaching events at Cheeca Rocks, a shallow coral reef in the nearby national marine sanctuary. These mountainous star corals (Orbicella faveolata) were presumed dead by researchers in September 2014, but four years later, they had recovered, possibly surviving the 2015 heat thanks to the experiences of the previous year.
Derek Manzello, NOAA Coral Reef Watch

Already, scientists are hoping to use coral memory to improve conservation and restoration strategies before corals tip irreparably toward extinction. Coral biologists are exposing corals in the lab to mild thermal stress to “harden” the corals, or fortify them against future heat waves and bleaching events. Teams are also breeding corals to have a higher capacity for memory, or inoculating corals with heat-resistant symbionts thought to play a role in memory. The ultimate aim is to one day outplant these corals to rebuild reefs that are losing the battle against climate change.

See “Restored Corals Spawn Hope for Reefs Worldwide

Agencies including the National Oceanic and Atmospheric Administration (NOAA) and the National Academy of Sciences have recently produced reports touting the promise of these memory-based approaches for improving coral resilience, while the National Science Foundation (NSF) and restoration groups are awarding grants to probe the genetic and epigenetic drivers of coral memory. Florida International University environmental epigeneticist Jose Maria Eirin-Lopez, a recipient of several NSF grants, says that this support underscores the relevance of this field of inquiry. 

“Stress hardening has been highlighted as an intervention that can be incorporated into restoration, but we really don’t know quite enough to know how to do that efficiently,” he tells The Scientist. “It’s a glimmer of hope, and now we need to figure out how it’s working so we can actually implement this.”


Corals reefs worldwide are exposed to increasingly frequent and severe warming events that can lead to catastrophic bleaching when coral organisms eject their algal symbionts. But scientists have noted that some reefs bounce back after such warming events and often fare better during subsequent temperature increases, even if they are hotter or longer than previous ones. This phenomenon—whereby an organism modifies its response to past abiotic stimuli—has been termed “environmental memory,” and researchers are working to understand how and why certain corals may have a greater capacity for memory than others.

© CATHERINE DELPHIA; Adapted from Trends Ecol, 36:1011–23, 2021.
See full infographic: WEB | PDF

Taking it to the reefs 

The first evidence of environmental memory in corals came in 1994, when Barbara Brown, then a marine biologist at Newcastle University in the UK, noted that only the eastern sides of stony coral colonies (Coelastrea aspera, previously Goniastrea aspera) bleached during a high heat event at her field site in Phuket, Thailand. In a study published in Nature a few years later, she hypothesized that the western sides were more tolerant due to previous exposure to more sunlight, a suspicion she backed up in the lab using coral samples that she exposed to varying temperature and irradiance regimens.

In 2000, Brown returned to Thailand and rotated several corals so that the stress-tolerant west-facing sides now faced east, and vice versa. It would be a decade before the reef experienced another severe bleaching, but when it did, Brown found that the now-eastern sides of those corals, with their history of high sun exposure, fared better than the eastern sides of controls that had not been rotated, retaining four times as many symbionts. She and colleagues interpreted this as evidence that the corals had “retained a ‘memory’ of their previous high irradiance history despite living under lower irradiance for 10 years,” they wrote in the 2015 paper reporting the findings.

See “Ocean Heat Wave Wreaked Havoc on Great Barrier Reef

Since those early experiments, researchers have collected evidence of environmental memory in reef systems on a broader, albeit shorter, scale. Two satellite-based studies of the Great Barrier Reef—one published in Science in 2016 by Ainsworth’s group and another in Nature Climate Change in 2019 by James Cook University coral reef ecologist Terry Hughes and colleagues—found that prior exposure to heat stress could improve bleaching outcomes during subsequent warming events. 

Analyzing nearly three decades of sea surface temperature data, Ainsworth’s team found that about 75 percent of the time, bleaching events on the Great Barrier Reef come just a week or two after a period of warming below the bleaching threshold. This initial exposure to sublethal levels of heat proved protective, reducing coral mortality and symbiont loss by 50 percent compared with stress events in which the temperature vaulted over the bleaching threshold without any initial sublethal, priming exposure. 


Hughes, meanwhile, focused specifically on the length and magnitude of the warming during back-to-back bleaching events in 2016 and 2017—a crippling one-two punch to the massive reef system. Speaking to The Atlantic in 2018, Hughes noted that prior to 2015, there’d been roughly 2 billion corals in the Great Barrier Reef. By 2018, half were dead. 

However, Hughes’s team found that the severity of the damage didn’t strictly correlate with the severity of the heat. Prior to 2016, the 3,000 individual reefs that collectively make up the Great Barrier Reef hadn’t bleached in 14 years, and as a consequence, 2016’s was a particularly devastating event, the results showed. NOAA typically forecasts bleaching based on degree heating weeks (DHWs), or the number of weeks within a 12-week period that corals spend in water hotter than the mean daily maximum sea surface temperature for the month. Researchers consider bleaching likely after four DHWs, and severe bleaching and mortality after eight. But in 2016, Hughes says, those traditional thresholds for bleaching weren’t very reliable, “and in particular, they depend on history.” 

Eight DHWs increased the chance that reefs would suffer severe bleaching (defined as the bleaching of more than 30 percent of corals within the reef) by 90 percent—far more than NOAA had forecasted. Just a year later, though, another warming event elicited severe bleaching in only 50 percent of reefs, even though most reefs experienced greater and longer lasting heat. 

The study suggests there is a time limit to protective effects of prior exposure to heat, says Hughes. “There’s memory when bleaching events are one year apart, but when they’re 14 years apart, the behavior of the corals is very different.”

Scientists have also backed up Brown’s initial observations of environmental memory at the level of individual colonies. Derek Manzello, a coral ecologist and the federal coordinator for NOAA’s Coral Reef Watch, and his colleagues used cameras ferried about by divers to trace the fate of more than 4,000 colonies of 15 coral species in the Florida Keys during bleaching events in 2014 and 2015. Each species has its own way of responding to heat, Manzello says, but the team did find that the damage inflicted by the second event was less than that caused by the first, even though the number of DHWs was higher in 2015 than in 2014.

See “A Coral to Outlast Climate Change

Manzello’s team also noted that slow-growing, mounded corals are more resilient to heat damage than fast-growing, branching corals, findings that have been echoed in other studies. “The massive corals hang around because their physiology is more suited to stress,” he tells The Scientist. “They tend to have more lipids in their tissues, they tend to be better at feeding from the water column, things like that, whereas the ones that are really good at capturing light are also then really susceptible [to bleaching] when the temperature is high.” Whether or not memory also plays into the relative success of mounded corals compared to other types, he adds, remains to be seen.

Are Memories Inherited? 

As researchers continue to query the mechanisms of environmental memory and figure out how long it lasts, Hollie Putnam of the University of Rhode Island is researching the potential for corals to pass their memory on to subsequent generations. Using the branching coral Pocillopora damicornis, a simultaneous hermaphrodite that broods its young internally before releasing them as larvae into the water, Putnam and her colleagues exposed adult corals to six weeks of high temperature and low pH conditions, mimicking the simultaneous stressors of warming and ocean acidification (OA). They found that while these conditions caused the parent corals to partially bleach, their young were better prepared when subsequently exposed to the stressful conditions than were larvae from parents who hadn’t experienced the environmental stress. Specifically, larvae of heat- and OA-exposed corals grew less but increased their respiration, a proxy for metabolism. 

In their paper, the authors attributed this to “trans-generational acclimatization,” which they suggest may be an understudied mechanism by which corals can adapt to heat stress. More recent studies have similarly shown that corals’ algal symbionts too may be heritable. “Whenever I talk about this, I get really excited,” Putnam tells The Scientist, adding that corals most certainly have “untapped or unidentified potential to resist and to respond and to evolve.”

Molecular basis for memory

To explain observations of environmental memory in the field, scientists are studying corals in the lab to determine how individual coral colonies encode their experiences at the cellular and molecular levels. A 2014 study, for example, revealed more than 100 genetic loci where variation was associated with heat resistance in tabletop corals (Acropora hyacinthus) that were repeatedly exposed to heat. Researchers have also found that corals can elevate the expression of hundreds of genes over a period of hours to weeks following exposure to heat, effectively “frontloading” transcripts and proteins involved in apoptosis and responses to oxidative stress, heat shock, and the unfolded protein response. “That’s classic acclimation,” says Steve Palumbi, a marine geneticist at Stanford University who studies the transcriptional changes of corals following heat stress. “The environment sparks a change in transcription that then sparks a change in proteins that sparks a change in physiology.” 

Corals’ algal symbionts may also play a role in perpetuating memory—specifically, a shift in the dominant algal clade could make corals more heat resistant. Ross Cunning, a coral biologist at Chicago’s Shedd Aquarium, previously showed that the Caribbean coral Montastraea cavernosa bleached less often in the wild when its symbionts were dominated by the algal genus Durusdinium, which can survive temperatures higher than 30 °C that would kill other strains. M. cavernosa, along with two other coral species, also performed better on laboratory stress tests when inoculated with Durusdinium than when inoculated with less-heat-tolerant clades. Only a quarter of coral species can host multiple clades, but for those that can, Cunning says, “those symbiont associations might be the most important factor driving heat tolerance.” It reasons, then, that shifts in these communities in response to heat—likely due to heat-tolerant clades simply outcompeting less-hardy ones—might serve as one basis for environmental memory, he adds.

See “Corals Show Genetic Plasticity

Another possible mechanism underlying environmental memory in corals is epigenetic alterations, which could explain the changes in gene expression that researchers have observed. In 2018, researchers outplanted Acropora millepora corals in warmer or cooler parts of the Great Barrier Reef and found that methylation patterns differed between corals in more or less stressful habitats. Those exposed to more environmental stressors had upregulated genes involved in responding to their environment, and those in less stressful habitats had upregulated housekeeping genes. Outplanted corals whose methylation patterns more closely resembled those of local corals were more fit, based on weight gain and lipid, protein, and carbohydrate content, supporting methylation’s role in acclimation.

More recently, Cunning partnered with Eirin-Lopez on an experiment suggesting that symbionts can prompt epigenetic changes in their host. Specifically, when the researchers shifted the dominant symbiont in laboratory-housed M. cavernosa from the heat-sensitive Cladocopium to heat-tolerant Durusdinium, twice as many regions of the corals’ genomes were methylated. These included intergenic regions and areas in the introns of genes, as well as transposable elements that the authors speculate may play a role in transcriptional silencing, as they do in other organisms. “I think there’s this very complex relationship between the external environment and the internal environment, mediated by symbionts, that could drive epigenetic modifications in the coral,” Cunning says.


Corals in the wild have demonstrated an ability to “remember” their past exposure to heat stress, thereby helping them fare better during subsequent bleaching events. To understand this so-called environmental memory, scientists are studying the phenomenon in the laboratory and have uncovered multiple ways in which corals may remember previous exposure to heat stress. Corals might also be capable of passing on their memory to their offspring, although the mechanism for that specific process remains unknown.



Some coral species undergo a shift towards a more tolerant clade of symbiotic algae in response to heat stress.


Corals can “frontload” the expression of genes involved in apoptosis, the heat shock response, and oxidative stress, among others.


Exposure to heat can sometimes alter the amount of DNA methylation in a coral’s genome.


Preliminary evidence suggests that the larvae of corals that experience bleaching are better able to tolerate subsequent heat stress.

See full infographic: WEB | PDF

Can memory enhance restoration goals?

The ultimate goal, many coral researchers tell The Scientist, is to roll environmental memory into restoration initiatives. This may be done by stressing corals in a controlled way prior to outplanting them, by planting false memories of stress in the form of durable symbionts, or by breeding species with enhanced capacities to form and retain memory. Conservation and restoration scientists might one day include all three in their coral restoration plans.

Some of these efforts are just getting underway. Palumbi tells The Scientist that while he hasn’t yet outplanted any of his stress-hardened corals, “it’s something we’re hoping to do soon.” Meanwhile Erinn Muller, a coral restoration program manager at Florida’s Mote Marine Laboratory who successfully breeds and outplants thousands of corals each year, says that she recently provided corals to a colleague who is experimenting with outplanting artificially stress-primed individuals. “We are more than happy to provide the biological material for people to learn more about how [stress-priming] could potentially be useful,” she says.


Refining scientists’ understanding of environmental memory in coral is quite literally a race against time, experts say. The world has lost 14 percent of its corals since 2009, and the United Nations predicted that by 2034, all reefs will experience bleaching at least once a year. Ainsworth recalls a time when studying bleaching meant lab simulations, because natural bleaching simply wasn’t common enough. Now, she says, “we can’t work on the field bleaching fast enough to keep up with the events.”

See “Image of the Day: Stress-Resistant Corals

The speed at which reefs are degrading is pushing many scientists toward action even in the absence of complete information. “From the restoration side, there are a lot of studies that can be done with just stress exposure, without really looking into mechanisms,” says Serena Hackerott, a coral reef ecologist in Eirin-Lopez’s lab. Muller cautions that implementing new outplanting strategies will face challenges, as most communities based around reefs simply don’t have the infrastructure to churn out lab-reared, stress-hardened corals at scale or in a timely manner. “Whenever we have projects where we have to prep the corals months in advance, it’s always a nightmare,” Muller says. “If we can figure out how to do it and maintain flexibility, I mean, that’ll just be the key.”

Other researchers tell The Scientist that outplanting in general, even if the corals are stress-hardened, will be unable to compensate for coral losses. Muller is able to outplant roughly 30,000 coral colonies each year, for example, but Hughes notes that the Great Barrier Reef alone has lost more than a billion coral colonies across nearly 35 million hectares since 2016, and estimates that the cost of restoring those reefs is on the order of millions of dollars per hectare. The current literature on the topic of memory, he adds, is valuable for what it can teach researchers about coral biology, “but as a pathway to artificially re-growing coral reefs, I don’t believe [it’s] feasible.” 

However conservation efforts proceed, researchers agree that they need to happen quickly. “It’s on my mind all the time that the cost of doing nothing is going up and up,” says Palumbi. “We can treat this as a scientific exercise and want to know a bunch of things before we act, but at some point . . . if we don’t do something, we will lose coral reefs for a long time.” 

Editor’s Note (February 16): This story was updated to mention Trends Ecol, 36:1011–23, 2021 in the first infographic credit.