They've got a plan to fight global warming. It could alter the oceans.

Nova Scotia: In a quiet patch of forest in Nova Scotia, a company is building a machine designed to help slow global warming by transforming Earth's rivers and oceans into giant sponges that absorb carbon dioxide from the air.

When switched on later this year, the machine will grind up limestone inside a tall green silo and release the powder into the nearby West River Pictou, creating a chalky plume that should dissolve within minutes.

The effect could be potent, scientists say. Rivers contain carbon dioxide that is constantly escaping into the air, where it traps heat and warms the planet. But adding limestone converts some of that carbon dioxide into a stable molecule that instead stays underwater and washes into the sea, where it should remain trapped for thousands of years.

"The beauty of it is how simple the technology is," said Eddie Halfyard, a freshwater ecologist and co-founder of CarbonRun, the startup building the $400,000 limestone machine, with plans for many more. "We let the water do most of the work."

With the dangers of climate change growing and greenhouse gas emissions soaring, scientists and entrepreneurs are increasingly exploring ways to deliberately intervene in climate systems to cool the Earth. Overwhelmingly, scientists say nations must sharply cut the pollution from burning fossil fuels that is driving up global temperatures. But many also believe that some of the excess carbon in the atmosphere must also be pulled out in order to preserve a livable planet.

"The potential for ocean-based carbon removal is huge, and it's been really underexplored," said Nan Ransohoff, who heads Frontier, a $1 billion fund backed by tech giants such as Stripe and Alphabet that is investing heavily in strategies to take greenhouse gases out of the atmosphere.

Since the industrial age, oceans have naturally absorbed roughly one-third of the 1.7 trillion tons of carbon dioxide that humans have pumped into the atmosphere largely by burning coal, gas and oil. By speeding up that process, scientists believe even more carbon could be packed into those watery depths.

Proposals include creating floating forests of plankton or kelp that could inhale carbon dioxide from the air, or vacuuming carbon from the ocean and burying it on land.

One idea quickly moving into the mainstream is known as alkalinity enhancement, which involves adding limestone, magnesium oxide or another alkaline substance to rivers and oceans, changing their chemistry in a way that makes them soak up more carbon dioxide.

It also has potential as a business, with several startups attracting investment and conducting field trials in places such as Nova Scotia and Iceland. Their goal is to drive down costs enough that companies or governments might pay to offset their emissions by stashing carbon at sea.

On Monday, Frontier announced it would pay CarbonRun $25 million to add limestone to multiple rivers and remove an initial 55,442 tons of carbon dioxide from the atmosphere. That's equal to a year's worth of emissions from 13,000 cars. Scientists estimate that similar methods deployed in oceans could remove billions of tons per year: not enough to cool the planet single-handedly, but significant if societies also stop polluting.

Yet immense challenges loom.

CarbonRun has shown it can capture carbon in rivers, but it's much harder to prove that the same techniques would work amid the chaos and complexity of the high seas. And doing it on a grand scale would require excavating billions of tons of rocks and shipping them across the globe.

"It has to go from something that most people have never heard of to the largest industry the world has ever seen, in a really short time," said David Ho, an ocean scientist at the University of Hawaii at Manoa.

Toying with ocean chemistry also carries unknown risks. Some environmental groups worry that even early experiments with these techniques could threaten fish and other aquatic life.

But with some companies eager to turn a profit from engineering the ocean, scientists say it's imperative that experimentation takes place so the technology's benefits and perils are fully understood.

"They all have their problems; they all have consequences," Ken Buesseler, a senior scientist at the Woods Hole Oceanographic Institution in Massachusetts, said of the different ideas to slow global warming by engineering the ocean. "But I think we just have to also weigh these," he said, "against the consequences of doing nothing, which are to me catastrophic."

Altering marine environments to cool the planet has been contentious from the moment scientists first suggested it four decades ago.

One early proposal was to sprinkle iron into the sea to fertilize enormous meadows of plankton that would breathe in carbon and take it with them to the ocean floor when they died. The backlash was fierce, especially after entrepreneurs began conducting unauthorized iron tests in the Pacific.

Today, CarbonRun's founders are trying to avoid similar blowback by proving that adding limestone to rivers doesn't just take carbon out of the air -- it can also safely benefit local ecosystems.

It helps that the startup is using a technique invented long ago to deal with a different environmental problem: acid rain. In the 1970s and '80s, industrial pollution made rainfall more acidic, which poisoned lakes and streams around the world. Some of the hardest-hit countries, including Norway, Sweden and Canada, began adding limestone to their waterways to restore the pH balance and help fish populations recover. It worked.

A few years ago, two scientists in Nova Scotia, Shannon Sterling and Eddie Halfyard, realized that adding limestone also helped rivers sequester more carbon. They joined Luke Connell, a Toronto-based entrepreneur, to form CarbonRun.

"It's sort of like Ozempic," Connell joked, referring to the diabetes drug that was later shown to have significant weight-loss effects. "This is a way to restore rivers that also happens to help our climate problem."

The insight came as some major companies were willing to pay for carbon removal as a way of offsetting their emissions.

It's relatively straightforward for CarbonRun to show that adding limestone to rivers converts some carbon dioxide into a stable bicarbonate. Technicians sample river water above and below the limestone machines and can directly measure change, while accounting for a few other smaller complications.

The harder part is mining and moving the limestone cheaply: CarbonRun needs roughly 2 tons of rock for every ton of carbon it removes. If the company can solve that, there are hundreds of acidified rivers from Maine to Indonesia close to limestone deposits, potentially allowing for hundreds of millions of tons of carbon dioxide to be captured each year.

For now, CarbonRun has been holding extensive public meetings with communities and First Nations in Nova Scotia, understanding that public opposition could be one of their biggest obstacles. The fact that river liming has already helped revive the province's devastated fisheries is a selling point.

Donald Rutledge, 79, lives in Sheet Harbour and had watched the salmon, once so plentiful that they leaped out of the water, vanish in the 1980s. He was initially skeptical that adding limestone to rivers would help but became choked with emotion, he said, when the silvery fish returned.

"I can go fishing with my grandson. It means a lot," Rutledge said. "I'd love to see them build some more."

Some companies are eyeing a bigger prize: adding alkalinity to the oceans, which have the potential to lock away more carbon than all the world's rivers put together.

But oceans are bigger, more turbulent and much harder to treat.

Sprinkle in a bit of alkalinity and it quickly disperses across large distances or worse, gets dragged uselessly into the deep. Any resulting shift in the constant exchange of carbon dioxide between the air and the sea is hard to detect.

"Right now the biggest barrier to ocean alkalinity enhancement is proving that it works," said Jaime Palter, an oceanographer at the University of Rhode Island.

On a chilly August morning, Dariia Atamanchuk joined four other researchers on a 46-foot boat that held a jumble of sensors, equipment and a small seafaring drone known as the Blue Boat.

They headed into Halifax Harbour, where a carbon removal startup called Planetary Technologies has been releasing magnesium oxide into the water to increase the alkalinity at the surface. Atamanchuk is part of an independently funded scientific team at Dalhousie University trying to verify whether Planetary can safely do what it aspires to do: remove carbon dioxide from the air.

It's a painstaking effort. The crew stopped throughout the harbor to take samples, measuring variables like salinity, temperature and dissolved carbon dioxide in the water. They sent the drone to the foamy plume around the outfall pipes of a power plant, where Planetary is adding alkalinity. They plan to do this several times a month to inform a complex computer model overseen by Katja Fennel, a Dalhousie oceanographer, who is trying to calculate how much extra carbon is being transferred to the ocean.

It's still too early to know the results. "As a scientist, I'm always skeptical," Atamanchuk said. "But based on everything we've seen so far, I'm optimistic."

Even if the Dalhousie researchers prove the technique works in a protected harbor, others will have to test it elsewhere, in places with different circulation patterns, to have confidence in its efficacy. That includes the open ocean.

When Adam Subhas set off into the Atlantic last year, the first thing he did was turn the water rusty red, the color of fake movie blood.

Subhas and his team from the Woods Hole Oceanographic Institution poured a giant spiral of dye into the waters off Martha's Vineyard. For two days they tracked the plume with instruments as it bobbed and spread.

Now they want to do it again, only this time with 6,600 gallons of alkaline solution mixed in. After that, they want to try it with 66,000 gallons. They are hoping to see how well they can monitor alkalinity enhancement on the high seas, both its benefits and its potential side effects.

"It's definitely new; it's risky," Subhas said. "I don't have tenure, so it's also kind of risky for my career."

Their experience has also shown how hard it can be to convince governments and communities that tinkering with ocean chemistry is a good idea.

There is no comprehensive set of rules that governs ocean-based carbon removal in the United States or elsewhere. Projects are subject to laws, agreements and regulations -- if they exist -- in ways that vary from place to place.

Planetary, for instance, has permission to operate in Halifax Harbour because it is adding magnesium oxide through the outflow of a power plant; it needs to follow Canadian clean water laws and avoid raising the pH of the water above a certain level. International rules governing the open ocean are more complicated.

This regulatory patchwork might not be a big problem now, said David Santillo, a marine biologist at the University of Exeter and Greenpeace Research Laboratories. But as geoengineering grows, it could lead to disputes between countries. "It will always be very difficult to prove whether certain activity done in one area has caused an adverse effect that's detected somewhere else," Santillo said.

For Woods Hole's experiment, Subhas and his team need a permit from the U.S. Environmental Protection Agency. The agency gave tentative approval in May but when it invited public comment, the project was praised by scientists and condemned by pretty much everyone else.

"Your arrogance is astonishing," one person wrote. "Leave the ocean alone."

Many opponents focused on the alkaline chemical the scientists would be dumping: sodium hydroxide, also known as lye. It's caustic at high concentrations but common in soaps and cleaners. Cruise ships use it to remove pollutants from their engine exhaust. "It's essentially pure alkalinity," Subhas said, which means its effects should clearly show whether alkalinity enhancement works.

Subhas wants to mix the sodium hydroxide with fresh water, and the ocean would dilute it further. He said that would limit the ecological consequences, though not all experts agree.

"If this were to accidentally happen, and you drop 50% sodium hydroxide solution into the seawater, it would be considered a chemical spillage," said James Kerry, a marine biologist with the wildlife protection group OceanCare.

Certain types of ocean geoengineering, if tried at scale, are also bound to affect deep-sea life, said Lisa Levin, a marine ecologist at the Scripps Institution of Oceanography. "We owe it to future generations to have at least one part of the planet we haven't messed up," she said.

By August, the EPA still hadn't given final approval, and the research boat the scientists planned to use became unavailable. They postponed the experiment until next summer, a delay Subhas called "frustrating." He and his colleagues have met with fishermen and other community groups to try to assuage their concerns.

As he sees it, scientists are trying to study these climate solutions rigorously, which takes time. But for-profit companies are looking to move quickly. And they might not be as careful about getting the science right.

"That keeps me up at night," Subhas said.