Context: This blog post is an introduction to climate stabilisation for people attending EAGxAustralasia (or considering whether to do so). As we're expecting ~50% first-timers, I figured an overview of this topic would be helpful , as it's not as widely discussed or self-explanatory as others.
Ice sheets, ocean currents and the planet's reflectivity may be destabilising faster than expected. Whether anyone tries to cool the planet in response, and who gets a say, is being worked out now. Dr Trish Lavery has explored how that call might get made, and she is speaking at EAGxAustralasia in Melbourne this November.
One way to decide which problems deserve more effort is to ask how large a problem is, how much extra work could achieve, and how few people are already working on it. By that test, climate change has often looked important but well funded. In October, Rethink Priorities applied the same test to one part of the problem and reached a different answer. Alongside cutting emissions, the world needs to work out how to stop several of the Earth's major systems from destabilising. This emerging field is called climate stabilisation. Rethink Priorities set out the case for funding this work in a review commissioned by Outlier Projects, which is now launching a fund in the area.
The UN Environment Programme (UNEP)'s latest report says holding warming to 1.5°C, even with a limited overshoot, is no longer possible. 2024 was the first calendar year above 1.5°C. The best remaining pathway peaks at 1.7 to 1.8°C around mid-century. Above 1.5°C, several of the Earth's large systems start to behave differently.
The planet is getting darker as ice and low cloud cover shrink, so it absorbs more heat, which melts more ice. The Greenland and West Antarctic ice sheets are losing mass at the top end of what models projected. In West Antarctica, warming ocean water is eating away at the floating ice shelves from below, and as they thin the glaciers behind them slide faster into the sea. The Atlantic ocean currents that carry warm water north may be slowing. Thawing permafrost and drying forests have started releasing carbon of their own, adding to the warming that released it. Each loop feeds the others, and the worry is that together they tip into runaway change that later emissions cuts can't pull back. Rethink Priorities found the evidence on all four pathways it examined points the same way. All four may be destabilising faster than mainstream models projected, and several show signs of self-reinforcing change that could continue even under deep emissions cuts.
UNEP also finds that even if every country delivered its current commitments in full, the world would warm by about 2.6°C this century. That's well inside the range where these systems can tip.
Harms that can't be undone deserve extra weight, because there's no chance to learn from them and try again. Rethink Priorities used deliberately conservative estimates, counting only direct deaths. Even so, it found each of the four risks on its own would justify funders in the effective altruism community committing US$263 to 445 million to research and governance. Rethink Priorities puts philanthropic spending on the regional approaches it examined, such as cloud brightening, at under US$10 million a year. That looks a lot like pandemic preparedness around 2017 or AI safety around 2015, when a few researchers were working on something most people assumed was being handled.
We've been changing the planet's energy balance and reflectivity for two centuries without meaning to, so climate change is already a kind of geoengineering. Sulphur pollution from shipping was hiding some of that warming by brightening clouds, and when rules cut it in 2020, part of the hidden warming came back. Scientists have long asked whether there are ways to push in the other direction while emissions come down. One option is to restore some of the reflectivity that's been lost.
The most studied idea is stratospheric aerosol injection, which copies what large volcanic eruptions do. Fine sulphate particles placed high in the atmosphere reflect a small fraction of sunlight back to space. Mount Pinatubo did this in 1991 and cooled the planet by about half a degree for more than a year. Modelling suggests a sustained program could offset one to two degrees of warming, and would act within years rather than decades. It's the subject of serious research, including a £57 million UK government program. The risks could be significant: changes to rainfall and monsoons, effects on the ozone layer, and a rapid rebound in temperature if a program ever stopped abruptly. Much of the work is about identifying what those risks are and whether they're small enough to be manageable, relative to the risks of inaction. None of the main sources here (Rethink Priorities, Centre for Future Generations and Giving Green) argues for deployment now. All three see cooling, if it's ever used, as something that would sit alongside cutting emissions rather than replace it.
Dr Trish Lavery is Strategic Foresight Lead at the independent, Brussels-based Centre for Future Generations and an honorary Associate Professor at Macquarie University. Her report looks at how a decision to cool the planet would actually get made, which is a question the science can't answer. Funded by the UK's Advanced Research and Invention Agency, it draws on more than 250 policymakers, scientists and citizens, plus 22 experts. She finds that existing analysis has been "scientifically sophisticated but politically simplistic".
Her four scenarios for 2036 are driven far more by values, cost-of-living politics, self-defence and rivalry than by climate models. In one, countries with programs refuse to ratify a UN restraint agreement. In another, a company-led test backed by host governments is sold as cheaper than cutting emissions. In a third, disinformation costs a research alliance public trust, and in the fourth, great powers treat the atmosphere as a security contest. She uses them to test whether today's arrangements would hold up, and says they aren't forecasts.
One of the report’s central findings is about who takes part. The climate researchers she interviewed worried about moral hazard, equity and future generations, but in defence, security and commercial circles those concerns rarely came up. As she puts it, "Declining to participate on principle does not stop the conversation. It leaves the conversation to others."
None of this means the answer is obvious. More than 600 academics have called for an international non-use agreement, African environment ministers have rejected the technology, and Vanuatu has called for a ban or non-use agreement. They worry that cooling could weaken the pressure to cut emissions, that its effects would fall unevenly, and that no one has the authority to decide for the whole world. These concerns are why the conversation matters, and why the people raising them need to be part of it.
Experience so far shows why communities need to be involved early. In 2021 a Harvard balloon test planned for northern Sweden was called off after the Saami Council objected that no one had asked the people who live there. In Australia, Southern Cross University scientists have trialled marine cloud brightening (spraying a fine seawater mist to make low clouds more reflective) over the Great Barrier Reef since 2020. They've worked closely with Traditional Owners and sought their free, prior and informed consent. Giving Green, which evaluates charities for our Environment Fund, makes the same point: research and governance have to go together, with local communities engaged early. Any serious research program will need First Nations communities, Traditional Owners and the countries most exposed to the effects to help set its terms from the start.
Effective altruism is about using evidence and careful reasoning to work out how to help others as much as possible, and revisiting those answers when the evidence changes. Giving What We Can's 2022 climate report argued that research on stratospheric aerosol injection could wait, because deployment looked many decades away. The new evidence suggests decisions about cooling may come sooner.
The risks climate stabilisation deals with are large, neglected and hard to undo. If the thinking on how to govern climate cooling is done now, it'll be ready before anyone has to decide in a hurry.
At EAGxAustralasia, on this and everything from global health to AI safety, you can hear the strongest version of an argument from the person who made it, push back, and meet the people doing the work.
Hear Dr Lavery at EAGxAustralasia in Melbourne, 27 to 29 November. Apply now: https://effectivealtruism.org.au/get-involved/eagx/
If you'd like to give to climate work now, the EAAE Environment Fund supports three climate charities working to cut emissions at scale, selected using Giving Green's research. Donations are tax deductible in Australia: https://effectivealtruism.org.au/eaa-environment-fund/
More to explore
