Podcast

What Is Stratospheric Aerosol Injection (SAI)?

SAI is the most studied of all sunlight reflection methods. What would it take to use SAI to cool the planet? And who could do it?

: :

On June 15th, 1991, the densely populated island of Luzon in the Philippines awoke to an explosion that would turn out to be the second largest volcanic eruption of the 20th century. Mount Pinatubo had erupted, releasing a huge cloud of volcanic ash, hundreds of kilometers across and 40 kilometers high. As satellites tracked the ash cloud spread around the globe several times over, atmospheric scientists noted that over the next year, the Earth’s global temperature had decreased by as much as half a degree Celsius. The eruption had added around 17 million tons of sulfur dioxide into the stratosphere, a layer of the atmosphere between 10 and 50 kilometers above the surface. And this sulfur had gone on to form countless tiny aerosol particles. In the lower atmosphere, these particles would have been washed out in days, but because the stratosphere is dry and stable, these particles lasted for several years, reflecting light, and cooling the Earth.

Could the climate cooling effect of this eruption be replicated as a way to help tackle climate change? In this episode, we focus on the basics of the sunlight reflection method known as Stratospheric Aerosol Injection, or SAI, an SRM idea that looks like it could offer a practical means of halting or even reversing global warming within a few years. What is SAI? What would it take to cool the planet? And who could do it?

Featuring Dr. Daniele Visioni, an Assistant Professor of Earth and Atmospheric Sciences at Cornell University and Dr. Joshua Horton, a Senior Program Fellow at the John F. Kennedy School of Government at Harvard University.

Transcript

Dr. Pete Irvine: [00:00:00] Welcome to Climate Reflections, the SRM360 podcast, where we discuss sunlight reflection methods, or SRM, ideas to reduce the impacts of climate change by reflecting sunlight away from the Earth. I’m your host, Dr. Pete Irvine, and I’m a climate scientist who has studied SRM since 2009.

On June 15th, 1991, the densely populated island of Luzon in the Philippines awoke to an explosion that would turn out to be the second largest volcanic eruption of the 20th century. Mount Pinatubo had erupted, releasing a huge cloud of volcanic ash, hundreds of kilometers across and 40 kilometers high. The eruption killed more than 800 people and cost more than $700 million in damages. But the eruption had impacts far beyond Luzon and the Philippines. As satellites tracked the ash cloud spread around the globe several times over, atmospheric scientists noted that over the next year, the Earth’s global [00:01:00] temperature had decreased by as much as half a degree Celsius. The eruption had added around 17 million tons of sulfur dioxide into the stratosphere, a layer of the atmosphere between 10 and 50 kilometers above the surface. And this sulfur had gone on to form countless tiny aerosol particles. In the lower atmosphere, these particles would have been washed out in days, but because the stratosphere is dry and stable, these particles lasted for several years, reflecting light, and cooling the Earth.

Could the climate cooling effect of this eruption be replicated as a way to help tackle climate change? In this episode, we focus on the basics of the sunlight reflection method known as Stratospheric Aerosol Injection, or SAI, an SRM idea that looks like it could offer a practical means of halting or even reversing global warming within a few years.

What is SAI? What would it take to cool the planet? And who could do it? To address these fundamental questions about SAI, I spoke to two SRM experts. Dr. Daniele Visioni, an Assistant [00:02:00] Professor of Earth and Atmospheric Sciences at Cornell University and Dr. Joshua Horton, a Senior Program Fellow at the John F. Kennedy School of Government at Harvard University. First, let’s turn to Dr. Daniele Visioni, who uses climate models to study the effects of aerosols on the climate for a technical overview of SAI and how it could cool the planet.

Dr. Daniele Visioni: SAI is the idea that as the current energy balance of the planet, energy in minus energy out, is out of whack – there is a bit too much energy being trapped into the system that warms the planet – SAI is the idea that we can affect this energy balance and make sure that there isn’t that much future warming by preventing some solar radiation from reaching the surface throughout the use of aerosols dispersed in the stratosphere, which is a part of the atmosphere where there’s no clouds, there’s no water, everything is pretty still, and so if you put stuff [00:03:00] up there, it stays for much, much longer than it would close to the surface.

Dr. Pete Irvine: And how would it be done?

Dr. Daniele Visioni: To do it, you need to find ways to reach the stratosphere and release material into the stratosphere. One of the most discussed materials that can be used is sulfate, mainly because that’s what we can observe after volcanic eruption actually producing aerosols and reflecting sunlight. And so, in this case, you would probably use what we call aerosol precursor so you wouldn’t actually release aerosols, but you would release some kind of gas like SO2 that then chemically reacts with other elements in the stratosphere and produces these aerosols.

Dr. Pete Irvine: So you would release these materials into the stratosphere, which would turn into tiny aerosol particles and reflect sunlight. But how would you get this stuff up there?

Dr. Daniele Visioni: So fundamentally, you need ways to reach the stratosphere, which is pretty high up. It’s not a part of the atmosphere we [00:04:00] normally go to so it’s around 16, 18, 20 kilometers at low latitudes. It can be 10, 12, 13 kilometers at a high latitude. And so we need instruments that would carry payload all the way there.

Dr. Pete Irvine: Daniele brings up a key challenge. Implementing this strategy to cool the Earth would require delivering material to a very high altitude. To get into the tropical stratosphere, which would be needed to produce an even global cooling effect, would require getting to around 60,000 feet up, or 18 kilometers. That’s about twice the height that a commercial airliner flies.

What would be needed to reach these altitudes, and who has the technological capacity to do it? I asked Dr. Josh Horton about what it would take to lower global temperatures with SAI.

Dr. Joshua Horton: To do one of these big planetary scale deployments, we don’t have a fleet of aircraft on standby to go begin these operations. We need to have an aircraft can fly at very high altitudes, higher than most aircraft do,[00:05:00] certainly higher than commercial aircraft do, and that can carry big payloads, lots of, say, sulfur. And, uh, they need very powerful engines to fly that high for long enough to disperse a lot of material and we don’t really have off the shelf engines and aircrafts that fit those requirements. So that’s bottleneck, one might say, in terms of what it would take to get from where we are today to an imagined future where we’re able to say go and we launched the aircraft to do this.

There are a handful of companies in the world, big companies mostly in the global north, that have the expertise to design and build these aircraft and these engines in particular. They’re mostly in the West. There are some in China, there’re some in Russia, there’s one in India, one in Brazil, they are very much concentrated though, in those countries, especially in the US and Western Europe. These [00:06:00] companies are familiar to many people. It’s GE, it’s Pratt & Whitney, it’s Rolls Royce. These are familiar companies, but that are huge aerospace companies that have the capacity to sort of fund and orchestrate long-term decadal design programs to start from scratch and to build the kind of engines and aircraft that we’re talking about.

So if you think about what’s required, who can do this kind of large-scale geoengineering, it’s really not any old person or even any old billionaire. It’s governments who have access and cooperation, probably ongoing engagements with these particular aerospace companies. So you can think about these companies as being almost appendages of these governments. They’re not really appendages, but they’re not going to go off on their own and launch geoengineering, launch SAI, because they have immense amounts of money at stake. The amount of money they might make from doing this, plunking into tens of billions a year, that pales in [00:07:00] comparison to the kinds of money they’re making from selling bombers and attack aircraft and rockets and all this kind of stuff, which is really their market. So, they would never jeopardize that kind of income for the sake of disregarding the wishes of their home government and going off on some sort of quixotic quest to geoengineer the planet.

Dr. Pete Irvine: It’s probably worth stressing at this point, SAI is not currently happening, and we don’t have the aircraft needed to reach the high altitudes needed. That said, this is something we may want to develop over the coming decades. The Paris Climate Agreement set out to limit warming to below 2 degrees Celsius above preindustrial levels, and ideally below 1.5 degrees, but if countries follow through on their promised cuts to fossil fuel emissions, we will miss that goal, and we’ll likely see warming of around 2.5 degrees Celsius. Could SAI make up for that difference? Helping us to limit warming to 1.5 degrees while countries decarbonize their economies. We asked Daniele about the [00:08:00] cooling potential of SAI.

Dr. Daniele Visioni: So we do know that, yes, a certain amount of aerosols in the stratosphere would, at the global level, absolutely reduce the energy imbalance and cool the planet.

Now, whether we’re on track of the Paris Agreement or not, that’s another matter. We are certain that if we had a way to get to the stratosphere and release the aerosols, the aerosol would have a cooling effect on the climate. What we are not certain of is exactly how many aerosols we would need. There would be very dense technical questions about how would you evaluate how much cooling you’re doing, but we are also certain that, um, if we put some amount of aerosols, they would cool.

Dr. Pete Irvine: Daniele went on to explain how the locations where aerosols would be released would affect how the Earth cools.

Dr. Daniele Visioni: If you put the aerosols over the North Pole, they’re not going to get to the South [00:09:00] Pole. So you’re essentially just cooling the North Pole or thereabout. If you put the aerosols in the tropics, they mostly stay in the tropics for a long while, which means that they cool the tropics a lot and they’d cool, they’d pretty much cool very little high latitudes, so the poles. And we can observe the stratospheric circulation and we know where it goes, especially on an average over a decade. And so we know where the aerosols would go. And where the aerosols would go would lead to decisions about where to cool.

Dr. Pete Irvine: Climate model simulations of SAI show that it cannot undo the effects of climate change. While it looks like SAI could be deployed to offset future warming in most places, it would not be possible to restore rainfall patterns. SAI might reduce overall changes to rainfall, but some regions would see greater rainfall changes and may suffer as a result.

Cloud model simulations also show that choices about where aerosol particles are released in the stratosphere would affect rainfall patterns. For example, if aerosols were only added to one [00:10:00] hemisphere, this would produce large shifts in tropical rainfall and the monsoons. That would be a very bad idea, but could be avoided by cooling each hemisphere evenly.

SAI wouldn’t just affect the climate, it has a number of side effects, increasing acid rain and impacting the ozone layer. We asked Daniele what would the risks and tradeoffs of a sensible deployment of SAI be, and how would it compare to the risks of climate change alone?

Dr. Daniele Visioni: We are yet to find anything in which the risks coming from SAI would trump those from climate change. As in from a purely physical point of view, and from an idealized deployment, I think that all of the research that has been done in the last 20 years points to the fact that less warming in the future is always better. The problem is that we don’t live in a perfect idealized world.

When it comes to geopolitical risks, I think that it’s still very much up for debate. When we simulate this in our climate models, we don’t need to [00:11:00] ask countries for their permission. We don’t need to think about airbases and planes, we just put the stuff in the stratosphere and then go from there. Of course, if you imagine this happening in the real world, there’s tons of obstacles between the idealized version and the real one. Would countries agree and would they agree on amounts and on location and, and all of that? So this is sort of the first main problem in a way. There’s also some research pointing out that different countries would benefit differently from different amounts of cooling.

So there’s clearly there a first problem that opens up with how do you work together? Do you need a global cooperation, because we’re not very good at that? Can you just have a few powerful countries decide, but then will these countries do what’s best for most of the people of the world? And what do you do if one country says, look, the perfect version is not going to work, but we’re still going to do it because we think it works best for us? [00:12:00]

So that’s sort of one of the main problems. The other one I would say is that when we think about climate change and about SAI in our climate models, we can run our experiments tens of times, hundreds of times, and we can always say things kind of in a very definitive way. In the real world, things are more complicated. So this is another big, big, big issue on a way that goes in between the idealized version and the real-world version of SAI.

Dr. Pete Irvine: So in the ideal world of a climate model, a sensible deployment of SAI could cool the planet, alleviating some of the risks brought about by warming. But as Daniele said, we do not live in this ideal world.

To bring us back to the real world, let’s turn to Josh and look at what it would take to deploy SAI and how much that would cost.

Dr. Joshua Horton: What does it take to, to do SAI? That depends on what you want to do with stratospheric aerosol injection. You could do a full scale, large scale, big planetary intervention that would [00:13:00] last decades that would seek to say lower temperatures by a full degree Celsius.

On the other hand, you could do just a stunt. You could have a country or a company or a person, in theory, go out, loft some material a single time knowing that it’s not going to lead to long lasting effects but as a way to raise attention about the issue and there’s a wide variety of, uh, sort of scenarios you can imagine in between those two ends of the spectrum.

So for the really small, stunt-like deployments, one could imagine a country doing that today. Not on a continuous basis, not a whole fleet, just a single aircraft or a couple of aircrafts. That, I imagine, could cost less than a billion dollars, although nobody really knows, but it seems to be more practicable to do that than the really big-scale deployments that we usually think about, which is this idea of a planetary effort that would take decades and a fleet of aircraft, say hundreds of aircraft, [00:14:00] launching from multiple locations across the world, almost certainly for both the Northern and Southern Hemisphere. The best estimates are that that would cost in the tens of billion dollars a year.

Now of course there are downsides to SAI too. So it’s, it’s not all that it’s just, you spend 10 billion a year and you’ve solved things, you haven’t. You’ve addressed some problems, you may have also caused some other problems and you’ve created controversy, certainly. So there’s a lot of caveats attached to that.

Dr. Pete Irvine: Could you explain what some of these problems or controversies would be?

Dr. Joshua Horton: One big ticket item would be the possibility or probability that countries would expect to be compensated if they believed that they were damaged by some aspect of a geoengineering or SAI deployment. They may, a country may claim that a monsoon was worse than it would have been ordinarily, caused more damage than would have been expected without SAI and therefore they ought to [00:15:00] be paid some amount of money by the international community or the countries that are out there implementing SAI. So, there are indirect costs that could be much larger than the relatively small-scale costs involved, that appear to be involved, in doing SAI just in a direct sense.

Dr. Pete Irvine: To recap, Josh thinks it’s unlikely that companies or individuals could implement SAI at large scale. Developing an SAI program to cool the planet would require the development of new aircraft, or some other technology, capable of delivering lots of material to the stratosphere. The direct cost of doing this would be in the range of tens of billions of dollars per year and would likely involve technologies restricted for national security reasons. This means that it would be beyond the means of individuals, companies, or even small countries. While tens of billions of dollars is a lot of money, it pales in comparison to the investments needed to decarbonize the economy. By mid-century, these are estimated to be in excess of 9 trillion per year.

Turning back to Josh, I asked, if a powerful [00:16:00] country or group of countries decided to develop and deploy large scale SAI, how quickly could they do it?

Dr. Joshua Horton: You know, there’s no hard or fast rules, but my understanding is that the development cycle for these big aerospace companies is at least a decade.

And so, what needs to be developed is first and foremost engines that are powerful enough to carry stuff up to the stratosphere. So an engine program to develop it and get it off the ground and tested and of course, the confidence in these engines, I think would take at least 10 years, perhaps 20 years. That could be done in parallel with the program to design and develop the aircraft itself to which the engines would be attached, which isn’t as difficult a job, technically speaking, but still we don’t have aircraft that are that high capacity, that are designed to fly that high for that long. And that would take, again, perhaps a decade to develop the aircraft in parallel with the engines. So let’s say perhaps two decades to be a bit more realistic. I’m not saying it’s going [00:17:00] to happen in twenty years, but from what I understand, a twenty-year time period, if there were really determined action on the part of big governments to put the money out and to sort of organize things on a, almost a war footing, you could have a fleet in the hundreds, say, within 20, 25 years. So, not tomorrow, but by mid-century, certainly.

Dr. Pete Irvine: 20 to 25 years is a long time. But as Josh explains, there might be some steps that could be taken towards deployment while these new aircraft are being developed.

Dr. Joshua Horton: Over the interim period, there are a whole set of things that one could do to sort of ramp up a deployment.

Aircraft being retrofitted and beefed up to be sort of workable, although not a long-term solution. So it doesn’t mean you’d have to start only in 2050. You could sort of get there using the sort of technical fixes and workarounds that would be suboptimal but could get you there. But in terms of looking out, you know, that point of distance in terms of what we need to do [00:18:00] today, if we want to ensure that we might be able to do that in 25 years, well, I think that’s a good question.

I mean, I hesitate to say that we should start developing engines or aircraft at this stage, because I think we simply, that seems awfully premature to me to be honest. The science isn’t there. There’s no, nothing like a consensus amongst scientists, let alone stakeholders. So I hesitate to endorse the kinds of, you know, engineering and technical development programs that would be required in order to sort of have that fleet available at mid-century. So that’s a bit of a quandary, I suppose. I guess what that speaks to is the urgent need to do a lot more basic science really, really soon so that if things turn out to look as promising as they sometimes do, then we can begin to set those wheels in motion, knowing that it’ll take a decade or two to sort of be able to have a fleet on hand to do this in a sort of a safe, [00:19:00] responsible way.

Dr. Pete Irvine: Given that SAI deployment would have these global impacts, should governments be discussing this now?

Dr. Joshua Horton: I think it’s premature for formal talks among governments at this stage, there’s just not sufficient awareness. There’s too much controversy and no government wants to really sit down and talk about this at length because they don’t really know what they’re talking about and there’s, uh, a lot of controversies and downsides to doing that.

So I, I don’t think that public formal sitting around a meeting table discussion with government officials at this point would be very useful. I think it would be impossible to get it done, for some good reasons but I think now is the time for informal discussions. Discussions amongst those who maybe aren’t currently in government but have phone numbers of people in government who’d be very curious to learn about their conversations with counterparts in other countries.

So it’s kind of a set of discussions that, again, formal officials don’t take part in, but people they trust, [00:20:00] experts, whether in universities or industry or in NGOs. Sort of speak informally, but seriously, and try to build the ground, and again, sort of build common understandings of where different countries see things going, how they understand the risks, the possibilities, the potential benefits.

I think informal, non-governmental, but-plug-in-to-government conversations are entirely appropriate today. We may be a little way from that being doable at this stage, but we’re much closer to that than to sort of formal government to government talks.

Dr. Pete Irvine: What would you say to government officials about the potential role that SAI or other sunlight reflection methods could play in the fight against climate change?

Dr. Joshua Horton: The only way to solve the climate problem is to decarbonize both the economy and the atmosphere ultimately. That’s the solution. SAI, SRM, these aren’t really [00:21:00] solutions, these are addressing symptoms of the problem. It could be very beneficial, but they’re really just addressing the worst outcomes. So, um, it’s a band aid.

It could be a very effective band aid, but it’s not ultimately going to solve the problem. So you really can’t substitute one for the other. The only way to really solve it is by decarbonizing.

Dr. Pete Irvine: That’s it for this episode of Climate Reflections. Thanks for listening. We’ll be returning again and again to stratospheric aerosol injection in future episodes, as there’s a lot more to say about this idea, including much more on its potential environmental risks and the geopolitical issues it raises.

Thanks for sticking with us to the end. This is a new podcast and we’re hoping to build our audience. So if you enjoyed it, please do share it on social media or recommend it to a friend. And if you have a question about SRM, or just want to find out more, go to our website, SRM360.org. We answer questions from the audience in our monthly news roundup podcast. So you might hear your question answered there. [00:22:00] You can find a transcript of today’s episode with links to sources on our website, so please check that out.

Ask us a question!

Ask a Question Form

Citation

SRM360 (2025) – "What Is Stratospheric Aerosol Injection (SAI)?" [Podcast]. Published online at SRM360.org. Retrieved from: 'http://srm360.org/podcast/what-is-stratospheric-aerosol-injection-sai/' [Online Resource]

Reuse this work freely

The content produced by SRM360 is open access under the Creative Commons BY license. You are free to use, distribute, and reproduce these in any medium, provided that SRM360 and the authors are credited.

The sources used by SRM360 are subject to the licence terms of the original third party. We will always indicate the original sources in our content, so please review the licence of any third-party sources before use and redistribution.