Sharks, among the fiercest predators in the ocean, are also some of the most vulnerable. Three-quarters of open-ocean shark and ray species are threatened with extinction, primarily because of overfishing.

For years, scientists have warned that these plummeting numbers could be catastrophic. Because sharks are top predators that help keep the food web in check, it would logically follow that they are keystone species—species that have a disproportionately large effect on their ecosystems. Without keystone species, ecosystems change or even disappear. (Here’s why keystone species are so important.)

For sharks, however, this idea has long been mostly theoretical. The fish and their marine habitats are tough to study, requiring special gear just to get in the water and observe them. The more than 500 known shark species also vary in size, prey on different animals, and live in dramatically different environments, where it’s almost impossible to pinpoint the influence of one species.

These complexities have bedeviled past efforts to study how sharks affect their ecosystems from the top down. For example, a 2007 study seemed to provide clear evidence that a loss of great white sharks in the North Atlantic led to an over-abundance of cownose rays, which in turn decimated populations of scallops, clams, and oysters. But scientists have since questioned whether the bivalve decline may have been influenced by some other factor—such as boating traffic.

“We want simple, but it’s not going to be,” says Michael Heithaus, a marine ecologist at Florida International University and founder of the Shark Bay Ecosystem Research Project, which has studied the bay, an 8,800-square-mile refuge in western Australia, for more than two decades. (Read about the most fascinating shark discoveries of the past 10 years.)

Shark Bay research has shown that sharks support their ecosystems not necessarily as hunters, but as regulators. By keeping their environments stable and resilient, sharks could help slow climate change and dampen the effects of extreme weather events, such as heat waves and hurricanes.

Untangling an ecosystem

Shark Bay is an ideal place to study how sharks interact with their environments. Its tiger sharks are seasonal, allowing researchers to observe how other animals behave when they’re present and when they’re gone. The UNESCO World Heritage Site also has healthy populations of predators and prey and dense meadows of temperate seagrass, which slow water currents, keep the water clear, and provide habitat and food for many species. Seagrasses also trap carbon dioxide, a potent greenhouse gas, and stabilize the sediments on the ocean floor that store the gas. Once carbon dioxide gets stuck in the sediments, it can stay down there for tens of thousands of years.

“I don’t know of another place like it on Earth,” says Rob Nowicki, an affiliate research scientist at Mote Marine Laboratory who has worked with Heithaus in Shark Bay. “It’s the perfect setting to test this stuff on a really large scale.”

Since Heithaus, a National Geographic Explorer, launched the project in 1997, researchers have been collecting data on all the elements of the Shark Bay ecosystem, providing the groundwork for untangling their intricate relationships.

One important discovery, published in 2012, showed that tiger sharks exert control over Shark Bay’s inhabitants, such as dugongs (relatives of manatees) and sea turtles, by essentially scaring the herbivores away into tropical seagrass meadows, which are less effective carbon sinks than temperate seagrasses. In other words, losing tropical seagrasses isn’t as detrimental to the environment as losing temperate ones.

But in places where sharks have declined and sea turtles have been protected, such as the Caribbean and Indonesia, the reptiles have begun to overgraze on seagrass, which could hamper efforts to lessen global climate change, according to Trisha Atwood, who leads the Aquatic Ecology and Global Change Lab at Utah State University.

“Over the past two decades, we’ve realized that ecosystems like seagrass are actually some of our best carbon stores in the world,” Atwood says. “They can sequester carbon much faster than any terrestrial forest.” (Read how the Amazon rainforest is now emitting carbon dioxide instead of trapping it.)

Building on the 2012 research, in 2015, Atwood found that Shark Bay tiger sharks also keep herbivores from disturbing the carbon-filled sediments beneath temperate seagrasses.

“We’re not suggesting that we shouldn’t be protecting sea turtles,” she says. “What we’re suggesting is that you need to protect the sharks so that they can help regulate their eating.”

Recovering from extreme weather

Shark Bay has also offered insight into how sharks might make ecosystems more resilient to the effects of climate change.

In 2011, an extreme marine heat wave hit Shark Bay and destroyed about 90 percent of its temperate seagrass, leaving what remained particularly vulnerable. Knowing the seagrass would need plenty of time to recover, researchers took it as an opportunity.

“We wanted to know what happens in a bizarro world of Shark Bay where the tiger sharks had been fished out,” Nowicki says. “Would the dugongs be able to come in and finish the job that the heat wave had started?”

Nowicki and colleagues divided the bay’s seagrass beds into plots and simulated how they would fare both with and without sharks present. They found that the seagrass beds that were not protected by tiger sharks spiraled out to the point of collapse. Meanwhile, the plots patrolled by them were more stable, as the sharks provided the seagrass more time to recover. (Read how seagrasses fight ocean acidification in the Chesapeake Bay.)

The study shows that tiger sharks are a true keystone species in Shark Bay, Nowicki says—and they probably are elsewhere too.

“It raises a giant red flag that having robust populations of big sharks can be important for the stability of ecosystems,” says Heithaus, a study co-author. “As ecosystems take more punches, we probably want to err on the side of having intact predator populations rather than cross our fingers and hope it’s okay.”

Keeping it together

Shark Bay has offered a glimpse into the important role sharks can play as predators, but Heithaus says the question now is how applicable that model is across the world. “You can’t just study the sharks themselves—you have to study all aspects of the ecosystem.”

For instance, Heithaus says there’s also evidence that baby bull sharks, which can swim in fresh water, deposit vital nutrients upstream in the Everglades—but it remains to be seen how much this infusion of nutrients matters.

Nowicki likens shark diversity in an ecosystem to the supports that hold up a bridge, with the loss of each species cutting down another pillar holding the structure together.

“The problem is we don’t know when that is going to result in a collapse,” he says. But with predators like sharks “we need to care—because we stand to lose more by losing them.”