When John Bruno sees shark fins circling his boat, he’s happy. They tell him that he’s sitting on top of pristine coral reefs, like the ones he swam among as a child. “When you drop down, there’ll be four to six sharks circling you, bumping you, checking you out,” he says. They’ll be accompanied by six-foot goliath groupers, three-foot snappers, barracuda, and more. Such reefs were the norm for the Caribbean in the 1970s. Now, they’re vanishingly rare. “In 99 percent of the reefs, the predators are absent,” says Bruno. “I once went 10 years without seeing a shark.”

That’s a problem. These large predators both reflect and safeguard the health of coral reefs. If they’re fished out, the rippling consequences can be devastating, leading to fewer fish and sicklier corals. And since those changes happened decades ago, they’ve influenced our perceptions of what coral reefs should look like. We think of the kaleidoscopic realms of Pixar movies or aquarium tanks, but those are reefs that have already been badly depleted. Pristine ones are worlds where predators abound, and colorful prey cower within the coral. “It’s like the difference between the English countryside and the African Serengeti,” says Bruno.

“Coral reefs reef around the world should have way more fish than they have right now,” says Abel Valdivia from the University of North Carolina Chapel Hill. The question is: how much more? “We don’t know the answer for sure, because most studies have looked at reefs after we fished them out.”

To understand just how far they’ve fallen—and thus, how much work we’ll need to do to restore them—Valdivia and Bruno traveled to 39 reefs across the Caribbean, from degraded and barren ones, to protected sites that are still teeming with predators. In each place, they measured temperature, depth, coral cover, wave exposure, and more. They noted human influence by recording nearby population sizes, electrical consumption, amount of cultivated land, proximity to markets, and more. And they counted fish, big and small, predator and prey.

They found huge variations. Compared to the most impoverished site, the most pristine one had 9 times as many fish by weight, and almost 90 times as many predators. The team showed that all the other factors they studied—human influence, environmental conditions, and more—could account for around half of that variation. (“For ecology, that’s amazing,” says Valdivia.) And using their data, they could reasonably predict how many fish a particular reef should theoretically be able to hold, if you removed human influence.

Some, like the Gardens of the Queen in Cuba and the Dry Tortugas in Florida, are already performing as well as expected. The majority are not. Some reefs contain between 10 and 30 percent of their maximum capacity of sharks, groupers, barracudas, jacks, tarpon, and other large predators. Even reefs within marine reserves, where fishing is forbidden, had, on average, just a third of the predators they ought to harbor. Partly, that’s because the fishing bans aren’t being enforced properly. It’s also because large predators move a lot; if they swim out of protected waters, they get hooked.

But while others might see the tragedy of past decades, Bruno’s team sees the potential of future ones. Their work shows that certain reefs will recover more dramatically than others. Some places would never have supported many large predators and probably never will, because they don’t have the right conditions or don’t contain enough prey. Others would be havens for hunters, if only those uber-hunters—humans—would leave them alone. And the team believes that they can now distinguish the latter from the former, and identify “super sites” where conservation measures could produce the most dramatic effects.

That’s a huge advance, says Isabelle Côté from Simon Fraser University. “This can guide conservation efforts towards sites that have the potential to show spectacular results,” she says. “This could have been another doom-and-gloom study, but to me, it’s good news. It suggests that in spite of all the headlines about the imminent demise of coral reefs, there is still enough quality habitat to support large numbers of predatory fishes, if we can get fishing under control.”

“People have been an important part of coastal environments throughout history,” adds Stephanie Green from Stanford University. Instead of aiming for pristine conditions, management plans for developed regions may have the goal of “the sustainable extraction of marine resources to support livelihoods and economies.” It’s more about reducing human influence than removing it. "This work gives us a glimpse into what these systems would look like without people, which helps us evaluate how well management actions are working."

Bruno agrees, and notes that fully restored reefs would benefit local communities. If protected areas are large enough, and properly enforced, they create steady pools of fish that can spillover into fish-able areas. They would also create “bonanza effects on tourism”. “It would be fantastic to have a half-dozen of these reefs around the Caribbean,” Bruno says. “People have no idea what these ecosystems are meant to look like.”

On land, conservationists are also running several re-wilding initiatives to restore large animals to the habitats they used to roam. But Bruno thinks that marine biologists can achieve the same effect without having to physically relocate sharks or grouper. They just have to leave reefs alone. “It’s really hard to drive fish extinct,” he says. “When we stop killing them, they bounce back pretty reliably. It takes a couple of decades, but they do.”