Eighty-four percent. Stop there for a second. That is the share of the world's coral reef ecosystems impacted by bleaching-level heat stress between January 2023 and March 2025, according to the International Coral Reef Initiative. In 1998, the first global bleaching event hit 21% of reefs. In the third event, from 2014 to 2017, it was 68%. The trajectory is not a trend line. It is an acceleration curve, and we are somewhere near the vertical.
The numbers from the Great Barrier Reef alone should make you put your phone down. The Australian Institute of Marine Science surveyed 124 reefs between August 2024 and May 2025 and found that hard coral cover in the Southern Great Barrier Reef dropped 30.6% in a single year, falling from 38.9% to 26.9%. That is the largest annual regional decline ever recorded. In the north, cover fell 24.8%. One in ten surveyed reefs showed any increase at all. The GBR has now experienced six mass bleaching events since 2016. The summer of 2025 brought the sixth.
This is bigger than you think, and not just ecologically. Coral reefs cover less than 1% of the ocean floor and shelter roughly 25% of all marine species. An ecosystem that occupies a sliver of the seafloor does the structural work of a continent. Losing it is not like losing a beautiful place. It is like pulling a keystone out of an arch and expecting the arch to stand.
This Is Not Bleaching. It Is a Duration Problem.
Here is the part the headlines almost always miss. Bleaching, on its own, is not death. When water temperatures rise, corals expel the photosynthetic algae living in their tissues. They turn white. But if temperatures fall quickly enough, the algae return and the coral recovers. The ocean has always had temperature swings. Corals evolved to handle them.
What they did not evolve to handle is sustained heat. A 2025 paper in Communications Earth and Environment identified a critical threshold: once annual bleaching rates exceed 7.9%, coral reef ecosystems undergo significant degradation. We are not approaching that threshold. We have blown past it. NOAA's bleaching alert scale, which previously topped out at Level 2, had to be expanded to Level 5 specifically to describe what is happening now. Level 5 means risk of over 80% of all corals on a reef dying from prolonged bleaching. NOAA scientists compare Alert Level 5 to a Category 5 hurricane in terms of reef impact.
A 2025 paper published in Proceedings of the Royal Society B tracked 112 individual Goniopora coral colonies at One Tree Reef in the southern Great Barrier Reef through the 2024 event. By October, 75% of tagged colonies were dead. Many were compounded by an outbreak of black band disease, a rare necrotic wasting condition that took hold precisely because bleaching had left corals immunologically vulnerable. This is how the biology actually works: thermal stress does not just bleach corals. It opens doors to secondary killers that finish the job.
The compounding is the thing that should terrify you. Recovery from a bleaching event historically requires roughly a decade of relatively stable temperatures. The interval between bleaching events is now measured in years, sometimes months. Corals are not getting the recovery window. The GBR gains from the early 2020s, the result of years of painstaking work and some favorable La Niña conditions, were largely reversed in a single summer. The universe does not care about your timeline, and neither do marine heatwaves.
The Off-Ramp Is Real, But It Is Conditional
This is where I want to be careful, because this is where science communication usually goes one of two directions: false despair or false hope. Both are epistemically wrong, and both are useless.
The despair case is seductive. Live coral cover has been cut roughly in half since the 1950s. The IPCC projected that reefs would decline 70-90% at 1.5°C of warming above pre-industrial levels, jumping to 99% at 2°C. We crossed 1.5°C in 2024. But projection is not prophecy, and the science is simultaneously documenting something genuinely interesting: coral populations carry heritable variation in heat tolerance, and researchers are learning to use it.
A 2024 study in Nature Communications, led by Newcastle University's CoralAssist Lab, demonstrated that selectively breeding adult corals for high heat tolerance increased the tolerance of offspring, with improvements observable in a single generation. A 2025 paper in Proceedings of the Royal Society B found that coral larvae with just one parent from a warmer reef population exhibited more than 2.2-fold higher survival under extreme heat stress compared to larvae bred entirely from cooler-water parents. This is not science fiction. These are peer-reviewed, replicated findings. The biology of thermal adaptation in corals is real.
NOAA's Mission: Iconic Reefs program is already moving. The agency awarded $16 million to the University of Miami to breed and outplant heat-resistant corals across seven key sites in the Florida Keys, targeting an increase in coral cover from 2% to 25%. Mote Marine Laboratory and the Coral Restoration Foundation are scaling similar programs. The researchers themselves are consistent on the caveat: selective breeding is not a substitute for emissions reduction. It is a bridge. It buys time. As study lead author Liam Lachs put it, "rapid reductions of global greenhouse gas emissions are an absolute requirement" to give corals the thermal headroom to adapt.
That is the conditional. The off-ramp requires both lanes: aggressive intervention science and the willingness of a civilization to stop pumping heat into the oceans. The biology can be made more resilient. What it cannot do is outrun physics indefinitely.
Coral reefs took tens of millions of years to build the structural complexity we are measuring in centimeters of annual decline. If Earth's history were compressed into 24 hours, the Great Barrier Reef in its current form would appear in the last few seconds. We are making decades-long decisions about geological timescales. The science is not ambiguous about what happens next if we do nothing. What remains open, genuinely open, is whether we are the kind of species that acts on evidence before the last branching coral bleaches white.
