August 26, 2023: Stratospheric Aerosol Injection
Good evening. Last week, I discussed the science around nuclear winter, the idea that a large nuclear exchange could ignite firestorms and release large quantities of dust into the atmosphere, blocking sunlight to the surface, reducing plant photosynthetic potential, and causing famine. I noted some deficiencies in the scientific case for nuclear winter, but there is merit in the basic idea—atmospheric aerosols would reflect sunlight—and the phenomenon has been observed with volcanic eruptions.
However, at present the leading atmospheric concern is not cooling, but rather global warming caused by the injection of carbon dioxide. And so it naturally occurs to some researchers that if the process of stratospheric aerosol injection (SAI), observed with volcanic winter and hypothesized with nuclear winter, were to be conducted in a controlled manner, the warming of the greenhouse effect could be offset.
Before we proceed with SAI, a taxonomy of concepts is in order. “Geoengineering”, also called “climate geoengineering”, refers to any number of methods to offset global warming though planetary-scale engineering projects. In theory, geoengineering could be conducted for reasons other than climate change mitigation, but there are not many proposals to do so.
Climate geoengineering generally falls into two main categories: carbon dioxide removal and solar radiation management. Approaches to carbon dioxide removal include reforestation and tropical reforestation, ocean iron fertilization, soil carbon sequestration, biochar, carbon-negative building materials, enhanced weathering, mineral carbonization, and ocean alkalinity enhancement. Approaches to solar radiation management include SAI, which is today’s main subject, marine cloud brightening, cirrus cloud thinning, cool roofs, arctic ice management, and orbital sunshades. This is not a comprehensive list. Each technique has its costs, limitations, and risks.
To my knowledge, the first SAI proposal dates to 1974, from the Soviet climatologist Mikhail Budyko. Budyko pioneered such work in climatology as understanding of the albedo effect of polar ice and snowball earth. However, SAI is best associated with Paul Crutzen, whose proposal is outlined in a 2006 paper. Crutzen was also a coauthor, along with John Burks, of the 1983 paper, “The Atmosphere after a Nuclear War: Twilight at Noon”, a forerunner of the soon-to-follow TTAPS paper that did so much to set the public understanding of nuclear winter.
The first question we might ask is, would this work? The International Panel on Climate Change concluded,
The efficacy of a number of SRM strategies was assessed, and there is medium confidence that stratospheric aerosol SRM is scalable to counter the RF from increasing GHGs at least up to approximately 4 W m–2; however, the required injection rate of aerosol precursors remains very uncertain.
How much is 4 watts per square meter? A common scenario used in climate modeling is the IPCC’s RCP 8.5 (representation concentration pathway, 8.5 watts per square meter), which is sometimes called the “business as usual” scenario for warming, but it is now understood to be implausibly pessimistic. Therefore, it is reasonable to expect that SAI could offset at least half, and possibly all, warming expected from elevated CO2 concentration.
The IPCC goes on to explain that SAI could lead to a decrease in precipitation, ozone depletion, and that if SAI were undertaken and then terminated (for reasons that are unclear) without reducing CO2 emissions, than global temperatures would quickly revert to what they would have been without SAI. While SAI would mitigate warming, it would not reduce atmospheric CO2 concentrations and therefore would not address ocean acidification.
The next question is, what would this cost? Using RCP 8.5, Niemeier and Timmreck estimate that 45±7 million tons of sulfate must be injected per year by 2100 to maintain 2020 temperatures. However, as noted above, RCP 8.5 is unduly pessimistic, and so the actual value is probably lower. Niemeier and Timmreck model a 5.5 W/m^2 temperature reduction and assess that the cooling efficiency of sulfates would be cut in half under their target.
Now some cost estimates. This paper estimates $18 billion/year for 1 degree Celsius of warming mitigated. The relationship between radiative forcing and warming isn’t exactly linear—there is an estimated 3.2-5.4 °C warming under RCP 8.5, 2.0-3.7 °C under RPC 6, 1.7-3.2 °C under RCP 4.5, and 0.9-2.3 °C under RPC 2.6—but it is close. That works out to as much as as $194 billion/year to mitigate all warming under RCP 8.5 under the factor of 2 efficiency penalty, and this figure is probably a major overestimate (again; RCP 8.5 is probably too pessimistic; I took the high range of warming under RCP 8.5; and do we really need to mitigate all warming?). The social cost of carbon is typically estimated at between $50-200/ton CO2, and world CO2 emissions are around 40 billion tons per year, so that gives us annual damages from CO2 emissions at $2-8 trillion.
This paper estimates the cost at $1500/ton of material injected. Using the 45±7 million tons/year of material estimated by Niemeier and Timmreck, the cost would be $57-78 billion per year. The paper also notes that aircraft do not exist yet that would fulfill the mission requirements for SAI, but there is no reason to doubt they could be developed. It might cost $2.35 billion to develop the aircraft.
This paper gives a cost of $2-8 billion to deliver 5 million tons of material. Under the estimates of Niemeier and Nimmreck, that would be $18-72 billion/year to deliver 45 million tons.
This paper quotes $10 billion/year for a reduction of radiative forcing of 2 W/m^2. Taking the efficiency penalty, the cost would be less than $100 billion under RCP 8.5.
Put a different way, the carbon mitigation costs of SAI would probably be less than $2/ton, albeit it is not complete mitigation because it doesn’t address ocean acidification. Nevertheless, SAI is less expensive than most other options.
In fairness, the 45 million ton figure is a case of false precision. Just how effectively aerosols cool the climate is a subject of ongoing research. The lifetime of aerosols in the atmosphere is also uncertain. These issues are also present in nuclear winter theories.
The third question we might ask is, what are the risks? There are quite a few issues. It is well-established that SAI reduces rainfall, but how much is unclear. Some aerosols, such as sulfate, contribute to ozone loss, and so it is proposed to use non-sulfate aerosols; calcite (limestone) injection may even mitigate ozone loss. Skies might be whitened and brightened. Stratospheric temperature change might affect surface circulation and weather patterns in ways that are unclear. There could be a small increase in acid rain. Possible effects on ecosystems are unknown. On crop yields, there would be a cost of reduced sunlight for photosynthesis, and a benefit from reduced thermal stress and CO2 fertilization, and so whether the effect on yields is a benefit or a cost is unclear. Solar photovoltaic production might take a 2-5% hit, whereas a 5.9% hit to concentrated solar is estimated. All in all, these risks should be taken seriously, but I don’t see any showstoppers.
Before SAI can be deployed at scale, we have to address these uncertainties: what it will cost, what kind of particles should be delivered, how should the delivery be done, and what are the environmental ramifications. The only way to determine these things is to conduct experiments. A field experiment was done in Russia in 2009. There is a proposed experiment SCoPEx. But what has been done so far is far from adequate, and 17 years after Paul Crutzen’s proposal, the lack of progress is inexplicable.
While there could be risks from full-scale deployment of SAI, there is no plausible risk from field trials. This doesn’t stop critics from coming up with objections to experiments. Such objections include that covert deployment could be done, that there could be technological lock-in, that SAI could be weaponized, that SAI deployment could lie on a slippery slope, and that there is a blurry line between experiments and full-scale deployment. This highly readable paper from Wake Smith and Claire Henly illustrates the weaknesses of these objections.
But in my mind, the most perplexing objection is the “moral hazard” complaint, outlined here. The moral hazard argument is that, if a geoengineering solution is available for greenhouse gas emissions, then it will reduce the impetus for more conventional methods of mitigation. Perhaps it will, but why is this a problem? Under what criteria is mitigation a more moral solution than adaptation or geoengineering? We know full well that with or without geoengineering, mitigation alone will not be sufficient to keep global temperatures below a level that most climate hawks say is necessary to avoid severe damage.
In a world where the damages from CO2 emissions are raised by artificially restricting the set of acceptable solutions, then (it is hoped by climate hawks) the world will be compelled to spend more on mitigation as a result of catastrophic damage. This also means fewer resources will be available for health care, defense, research and development, and many other competing, worthy social priorities. I fail to see how this is a “moral” solution.
I can’t conclude this post without addressing some other, often unspoken concerns about SAI and geoengineering in general. Some researchers such as these seize upon uncertainties around SAI and blow them up into something that can only be characterized as fearmongering. Some of our usual suspects, such as the Bulletin of the Atomic Scientists and the Union of Concerned Scientists, oppose SAI on specious grounds. Here’s Time Magazine with a sensationalist headline, “Why Billionaires are Obsessed With Blocking Out the Sun”. And SAI gets wrapped up in stupid things like chemtrails conspiracy theories. More seriously, SAI and other geoengineering schemes are inconsistent with some ethics of the environmental movement, as explained in this 1996 paper by Dale Jamieson.
SAI is not a complete solution for climate change, and I know of no serious researchers who think it is. But it is a potentially valuable solution, and there is no good reason to oppose the field trials necessary to determine if it really a good solution.
Quick Hits
The blog Marginal Revolution, run by Tyler Cowen and Alex Tabarrok, is 20 years old. I don’t read MR as much as I used to, but I still read it from time to time and find it to be chock full of interesting content. MR is one of the inspirations for the work I do here.
Yevgeny Prigozhin, the Wagner warlord who staged an abortive coup in Russia two months ago, is presumably dead in a plane crash under suspicious circumstances. What exactly happened remains unclear. The Institute for the Study of War argues that the crash was probably deliberate and probably ordered by Putin. U.S. intelligence suggests that the crash was not caused by a surface-to-air missile; maybe it was caused by an on-board bomb. A Council on Foreign Relations piece examines some of the theories. If deliberate and caused by the government, the crash is another example of the mafia culture that permeates Russian leadership. In addition to Prigozhin and other Wagner leaders, several innocent people were killed.
The Lookout gives a rundown of where the Ukrainian counteroffensive stands right now.
Francisco Pires of Tom’s Hardware reports that LK-99, the lead apatite material that is suspected of being a room-temperature superconductor, is still under investigation, and the claims of Lee Sukbae and Kim Ji-Hoon are not as dead as media reports might have one believe.
The Plain Bagel reports on Evergrande, the Chinese real estate firm, broader problems in the Chinese real estate market, and suggestions that Evergrande’s bankruptcy is some kind of “Lehman Brothers” moment are not on-point.
Last Sunday’s gospel reading in the Catholic Church was Matthew 15:21-28. In this passage, Jesus heals the daughter of a Canaanite woman in Tyre, which is in modern-day Lebanon. Jews at the time would have looked down on Canaanites, and although Jesus did not share these negative stereotypes, he uses them as a test of the woman’s faith. How should Christians today think about ethnic differences, and is nationalism compatible with Christian values? This is a topic that I have wanted to write about for a long time, but due to the difficulty and sensitivity of it, I have not yet attempted to do so.
The New York Times has an article titled “A Summer Rite in Spain: Coping With the British Tourist Invasion”. Anti-tourism is a form of nativism that is more palatable to the political left than anti-immigration, and I have seen this thinking from many people who regard themselves as tolerant. This dissertation from Eva Zetterberg Patterson explains anti-tourism as follows.
The importance of tourism as a road to national identity (British and American) is the concern of Buzard's The Beaten Track. Of particular interest to him is what he calls "anti-tourism", a derogatory view of tourism. Like Culler, Buzard considers this snobbish attitude to tourism to be, in fact, inherent to the cultural practice of tourism. He explains that "[s]nobbish 'anti-tourism,' an element of modern tourism from the start, has offered an important, even exemplary way of regarding one's cultural experiences as authentic and unique, setting them against a backdrop of always assumed tourist vulgarity, repitition, and ignorance."
Bob Barker died today at the age of 99. Barker was known as the long-time host of The Price is Right, and also as an animal rights advocate. He repeatedly exhorted viewers to get their pets spayed or neutered, and he supported many animal rights groups, including controversially the Sea Shepherd Society, which has been labeled an eco-terrorist organization. Animal rights and animal welfare is another topic that I have long wanted to write about, but from which I have been deterred by the difficulty and sensitivity of the subject.
Great post as always. I don’t usually feel I have anything sufficiently interesting to say to comment on you posts but I always appreciate reading them.