Photo: AP

With each passing year that the world fails to get a grip on its carbon emissions, the threat of climate change increases. For some, that makes geoengineering—a technique to artificially cool the planet—an increasingly enticing opportunity. But while scientists have shown that blasting tiny particles into the stratosphere could reflect enough sunlight back into space to keep the planet habitable for humans, there’s been little research on what it would mean for the world’s wildlife.

A new study published on Monday in Nature Ecology and Evolution gives us some our first clues. The study shows that while geoengineering could improve life for most plants and animals currently suffering the effects of climate change, it comes with two major catches. The first is that geoengineering itself would devastate the Amazon. The second, even more dire finding, is that once we start geoengineering the climate, we can’t stop because if we did, everything would go to hell in a hand basket.

The planet would rapidly warm at a rate much faster than species could learn to cope with through adaptation or by migrating to cooler climates. The results are a strong reminder that reducing carbon emissions is the only surefire way to make sure we don’t mess up the planet any further.

The study is one of the first of its kind, and came about due to a chance encounter at the American Association for the Advancement of Science conference three years ago. Alan Robock, a Rutgers climate modeler who focuses extensively on geoengineering, had just given a presentation on his work.

“Jessica Gurevitch, an ecologist, asked me about the ecological impacts,” he told Earther in an email. “I told her it was an important question, but nobody had looked into it yet, and we put together a team to do so.”


On the climate modeling side, Robock helped implement a geoengineering scenario where starting in 2020, the world starts injecting 5 teragrams of sulfate aerosols into the stratosphere every year to lower the planet’s temperature. That’s the equivalent a volcanic eruption roughly a quarter of the size of the 1991 Mt. Pinatubo eruption annually. They then compared it to a scenario where humanity cut its carbon emissions starting in midcentury, a trend that would still mean the world keeps warming, but at a slower rate than if emissions continued to increase.

To understand what geoengineering means for wildlife, the analysis looks at the climate velocity. The idea is that every plant and animal has an ideal climate, but the warming and precipitation shifts associated with climate change are causing that ideal climate to shift—generally poleward, or higher into the mountains. In order to keep pace with the ideal climate, species need to move with it. Big mammals, plants with wind-dispersed seeds, and birds that can cover large distances have the best chance of keeping pace, but small creatures or mostly stationary ones are in a tougher spot.

The findings show that keeping the planet from warming further via geoengineering would essentially allows species to shelter in place. In the case of highly climate-sensitive species like coral, geoengineering the climate could prevent them from going extinct. Though, of course, if we continue emitting carbon in that span, ocean acidification will still be a growing problem, so it’s not all roses. Geoengineering also doesn’t address all the other ways humans are stressing out ecosystems.


On land, species would also get a reprieve in most locations, with one notable exception. Cooling the planet by reflecting sunlight back into space would likely cause the Amazon basin, one of the most biodiverse parts of the world, to dry out. That would have a huge impact on ecosystems there, and set off a terrible chain reaction of more wildfires, increasing carbon emissions, and the need to do more geoengineering to keep things cool enough for everywhere else. Oh, and people living downwind would suffer bigly from terrible air quality.

Much of the Eurasian Arctic would also dry out, raising the risk of fires in the boreal forest as well.

More concerning still is what happens if policymakers suddenly decide to shut geoengineering down. The study looks at what would happen if we did that in 2070, and the results are absolutely horrifying.


Aerosols are really short-lived, which is why we would have to continually emit them to make geoengineering work. Suddenly stopping would result in a warming surge that would be much more dramatic than if climate change was just allowed to progress over the course of the preceding 50 years.

Precipitation patterns would also shift again, and the result for wildlife would likely be widespread die offs, and habitat fracturing. The climate velocity would be four times greater on land, and six times greater in the oceans, than recent climate change.

“We found that not only would climate velocities outpace the average dispersal speed of 93 percent of mammal species at geoengineering termination, the temperature and precipitation velocities on land move in different directions in many places, fracturing ecosystems in biodiversity-rich tropical oceans and in the Amazon,” Robock said.


So yeah, goodbye Amazon. And probably goodbye corals, mangroves, frogs and other amphibians, and many land mammals. The Arctic is also probably kaput (RIP sea unicorns).

The findings reinforce something we already know: geoengineering is an extremely risky proposition. And if humanity decides it’s a risk worth taking, then we essentially lock ourselves into doing it forever, or until we develop the technology to suck carbon out of the atmosphere at a scale large enough to stabilize the climate.

It also reinforces reality that the only way to reliably make our climate safe for wildlife and humans alike is to start rapidly reducing carbon emissions. And we better get started soon.