Nobody planted the first one. In 1996, Ken Nedimyer was farming live rock off the Upper Florida Keys — slabs of limestone seeded on the seabed and sold to the aquarium trade — when he noticed something settling on his crop that he hadn't put there. Staghorn coral. Larvae had drifted in from somewhere, found bare hard substrate in clean water, and done what coral has done for a few hundred million years.
It was, in a sense, a complaint. Staghorn was growing on his rocks instead of on the reef. But it was also a demonstration: give this animal a clean surface and decent water and it will build. Nedimyer and his daughter, working the project through her 4-H club, turned that observation into an ocean-based coral farm in 2001.
In 2003, six staghorn corals from that farm were fixed onto a Keys reef site called the Wellwood. Those six branched, and branched again, and became several thousand colonies.
He founded the Coral Restoration Foundation in Key Largo in 2007, and ran it for ten years before retiring from the organisation in late 2017. What he left behind was a method — and the method is the reason this story has numbers in it at all.
Don't farm coral on the ground. That one decision is most of why this works.
The seafloor is where the trouble is. Sediment settles on tissue and smothers it. Algae competes for the same hard surface. Snails and worms that eat coral live down there and can walk to dinner. So the Coral Restoration Foundation grows its corals in mid-water instead, suspended from purpose-built structures — the Coral Tree, and its later variants, the Spiral Tree and the Mega Tree.
A tree is close to the simplest object that could work: a vertical trunk with horizontal arms, anchored to the bottom, held upright by a float. Coral fragments hang from the arms on monofilament line — as many as a hundred corals on a single tree. They sit in clean moving water, off the substrate, out of reach of most of what would eat them, turning slowly in the current so light reaches every side.
And because the whole structure is flexible and mostly line, it doesn't fight the ocean. A hurricane passes through and the tree bends with it rather than snapping. That single design choice is why the organisation now runs what it describes as the largest in-water coral nurseries in the world.
Here is an empty nursery tree in about six metres of water. Click the arms to hang coral fragments — or use the button, which does it for you — then run the months forward and watch what a fragment of staghorn does when nothing is eating it.
One tree holds up to a hundred. Fill it and scrub the timeline; the point of the exercise is how ordinary it looks. There is no clever chemistry here. There's just a piece of coral, hung in clean water, left alone.
Summary of this interactive: a nursery tree is a vertical trunk with horizontal arms, anchored to the seabed and held upright by a float. Coral fragments hang from the arms on fishing line, up to a hundred per tree, where they sit in clean moving water away from sediment, algae and predators. Advancing the month slider shows the fragments branching and thickening over roughly a year. Keyboard: focus this panel and press Enter or the space bar to hang a fragment; use the month slider below to advance time.
Branching corals already reproduce by breaking. A storm snaps an arm off a staghorn colony, the fragment lands somewhere it can attach, and it grows into a new colony that is genetically identical to the first. The nursery does not invent this. It just does it on purpose, with a clean cut, somewhere safe.
Which means a coral farm is a cloning operation — and that is exactly where it gets dangerous. Every fragment taken from one parent colony carries one genome. Propagate the fastest-growing, best-looking colony you have, fill your nursery with it, plant it across a reef, and you have built something that looks magnificent and can be erased by a single bad summer.
So the Foundation counts genotypes, not just corals. Staghorn and elkhorn are outplanted in cohorts of fifty to seventy fragments of a single genotype, with dozens of genetically distinct cohorts deployed at each restoration site. Boulder corals go out as plugs fixed straight onto the reef. Coral sperm and eggs go into cryogenic storage, so that a genome that dies in the water is not necessarily gone.
The reason for all of that bookkeeping is simple, and the next panel is the argument for it.
You have twenty-four slots and eight available genotypes. You can see how fast each one grows, because growth is easy to measure — you watch it.
You cannot see how much heat any of them can take. Nobody can, until the heat arrives. Allocate your slots, then run the summer.
Summary of this interactive: eight coral genotypes are available and twenty-four nursery slots must be allocated between them. Each genotype's growth rate is shown, but its heat tolerance is hidden and is randomised each time the summer is run. Concentrating all slots in the fastest-growing genotype risks losing every coral at once; spreading slots across many genotypes guarantees that some survive whatever the summer turns out to select for. This is why the Coral Restoration Foundation propagates dozens of genetically distinct cohorts rather than only its best performer.
It is worth being precise about what this work is for. Elkhorn and staghorn are not decorative species. They are the two corals that built the shallow Caribbean — fast-growing, structurally complex, the branching architecture that young fish hide inside and that breaks incoming waves before they reach the coastline. Take them out and the reef is not merely emptier; it is flatter, and a flat reef stops working as a reef.
Populations of both species have fallen by an estimated 92 to 97 percent since the 1970s — to disease, to storms, to warming, to everything at once. That collapse is why NOAA and its partners, the Foundation among them, launched Mission: Iconic Reefs, a first-of-its-kind effort to bring coral cover at seven ecologically and culturally significant sites in the Florida Keys National Marine Sanctuary from around two percent up to an average of twenty-five.
By the summer of 2023 the Foundation had been growing coral this way for sixteen years, and Mission: Iconic Reefs — launched in 2020 — was three summers into its work. It was working.
Foundation teams went out to Sombrero Reef, a restoration site off Marathon they had worked for more than a decade. The marine heatwave that summer was, by every account of it since, unprecedented. Corals under that kind of heat stress expel the symbiotic algae that feed them and give them their colour — they bleach, going bone-white, and if the heat does not relent they starve.
Dr Phanor Montoya-Maya, the Foundation's reef restoration program manager, reported what they found in four words: one hundred percent mortality.
Almost every coral in the Looe Key Nursery in the Lower Keys was lost as well. This was not a bad year at one site. In the Lower Keys, of two hundred transplanted staghorn corals, three were alive by November. At Dry Tortugas National Park, roughly 99.9 percent of more than five hundred transplanted staghorn corals were dead by that October. Statewide, fewer than 22 percent of staghorn and under 5 percent of elkhorn survived the event.
Climate change is our present reality. The impact on our reefs is undeniable.Dr R. Scott Winters, CEO, Coral Restoration Foundation
Then they had to choose, fast, with nowhere near enough tank space. Partners in the Mission: Iconic Reefs network pulled roughly five thousand corals out of the offshore nurseries and off parent colonies and moved them into land-based holding systems at Keys Marine Lab. Not the biggest corals. Not the best growers.
They evacuated for genetic diversity — grabbing one of this, one of that, protecting the widest possible spread of genomes rather than the largest possible amount of coral. They were not saving corals. They were saving a library.
The next panel gives you the same boat and the same clock.
Twenty-four corals are in the water. Twelve genotypes, two of each — some of them your largest, oldest, most productive colonies, marked as such. The land tanks have room for eight. The water is heating and you have about forty seconds before what's left in the nursery bleaches.
Tap corals to load them onto the boat. Choose quickly, and choose on purpose.
Summary of this interactive: twenty-four nursery corals representing twelve genotypes, two of each, must be triaged as a marine heatwave arrives. Land-based tank space holds only eight. Some corals are marked as large, established colonies, which is the tempting choice. The measure that matters is how many distinct genotypes reach the tanks, because a genome lost in the water is lost from the region entirely. This mirrors the real evacuation of roughly five thousand corals to Keys Marine Lab in 2023, which prioritised genetic diversity over volume.
The method changed because the ocean changed.
When NOAA and its partners went back over the wreckage of 2023, one pattern stood out: the slow, heavy corals had come through far better than the fast, branching ones. Brain, boulder and star corals — the massive species, the ones that take decades to build anything — were still standing where staghorn and elkhorn were gone.
So that is increasingly what gets grown. The Foundation has outplanted roughly ten thousand massive corals across seven reefs since the heatwave. It has changed how they go in, too, using applied nucleation: planting several species together in dense clusters rather than spreading a single species thinly over a wide area, an approach that has shown higher survivorship and faster growth. Gene banking has expanded alongside Mote Marine Laboratory and the Florida Aquarium. In February 2025, at the fifth anniversary of Mission: Iconic Reefs, what went back into the water were corals that had survived the bleaching.
Between October 2025 and June 2026 the Foundation returned nearly sixteen thousand corals of seventeen species to ten reef sites throughout the Florida Keys, more than thirteen thousand of them funded through the local Tourist Development Council. At Coralpalooza 2026, a hundred and ninety divers cleaned a hundred coral trees in the offshore nurseries — the unglamorous maintenance work that keeps algae off the lines.
It is a working organisation doing competent work. And the hardest question about it is still open.
In April 2025, a study in Nature Ecology & Evolution looked at 220 coral restoration projects worldwide and reached a conclusion that is hard to argue with: restoration cannot be scaled up to save reefs. Not "is not yet". Cannot.
The arithmetic is brutal. Between 30 and 40 percent of restoration projects fail outright — poor planning, unrealistic objectives, insufficient maintenance, persistent local pressures. Of the sites that succeed, 57 percent experienced severe bleaching within five years of being restored, and the authors project that 99.6 percent of restored reefs will be hit by bleaching events by 2100. Rehabilitating even a tenth of the world's degraded reef area would cost somewhere between one billion and 16.7 trillion US dollars depending on method. Meanwhile the median restored area in a systematic review of the field was about 100 square metres — against roughly 12,000 square kilometres of reef lost worldwide between 2009 and 2018.
Every serious analysis lands in the same place: restoration is a holding action against warming, not a treatment for it. One of the researchers behind the 2025 study put it directly — the feel-good narrative around coral restoration should not become a distraction from cutting emissions.
A 2022 study in PeerJ found a possible trade-off in microfragmented corals: more heavily cut fragments showed lower immune enzyme activity, and immune genes that ran opposite to growth. The authors are careful — four months of observation, edge tissue only, no disease challenge — but flag it as worth watching.
Research published in Conservation Biology in 2026 concluded that elkhorn and staghorn are now functionally extinct in Florida: too few remain to fulfil their ecological role, which disrupts the structure of the whole reef community around them.
And then, in the same breath, the other half of the ledger — from the same body of research. Without two decades of restoration work, the trained people, and the network of land-based biobanks that existed to receive five thousand evacuated corals in 2023, elkhorn and staghorn would in all likelihood have been wiped out of Florida altogether rather than reduced to remnants. The library survived because someone had spent twenty years building one.
Both things are true. The work did not save the reef, and the work is the only reason there is anything left to rebuild from. A researcher on the functional-extinction study, Jenna Dilworth, argues the strategy now has to change — that transplanting corals and hoping they survive a rapidly changing climate is no longer enough on its own. She is also blunt about the ceiling on all of it:
Climate change is the reason all this is happening, and if we don't address that, nothing we do is going to fix this.Jenna Dilworth, USC Dornsife
That is not a reason to stop growing coral. It is a reason to be honest about what growing coral is: buying time, and keeping the genomes alive, for a reef that can only actually be saved somewhere else — in the atmosphere.
This is independent editorial content — we are not affiliated with the Coral Restoration Foundation and receive nothing from the link below. If you'd like to support the work described, you can give directly.
Donate directly to Coral Restoration Foundation100% of donations go to the charity. This site is supported by advertising only.