Dead before arrival: The governance of stratospheric aerosol injection

By Aaron Tang | March 15, 2023

Clouds in the stratosphere (Photo by Kaushik Panchal on Unsplash )

In the wake of six destructive hurricanes between 1954 and 1955, and billions of dollars in damages along the East Coast, the United States Government began conducting field tests with the aim of weakening tropical cyclones. Called Stormfury, the project was based on the idea that seeding clouds with silver iodide could reduce cyclone intensity. Science eventually showed the hypothesis was incorrect and seeding the clouds didn’t work to weaken the structure of hurricanes.

But the mechanism’s ineffectiveness didn’t stop Fidel Castro of accusing the United States of weaponizing hurricanes. And it didn’t stop some communities in Florida and the Gulf Coast who were hit by Hurricane Betsy in 1965 from blaming Stormfury for the damage. (Scientists originally had thought of seeding the storm but decided against it. This caused the media to incorrectly report the facts, leading to confusion.)

Although both Castro and the communities affected by Betsy were not correct in those instances, their concerns were valid. If the flap of a butterfly’s wings could theoretically snowball into a hurricane weeks later, as attributed to mathematician and meteorologist Edward Norton Lorenz, what effects could altering an atmospheric component have on communities near and far?

Stormfury ended in 1983. Since then, average global temperatures have risen 0.32° F (0.18° C) per decade. And with years of messy politics holding back effective climate action, scientists are once again revisiting ways to alter Earth systems. One of these is stratospheric aerosol injection, a type of solar radiation management that would mimic the cooling that occurs after volcanic eruptions. The idea is to inject materials such as sulfur into the stratosphere, to make particles that reflect sunlight back into space and cool the Earth. And since aerosols in the stratosphere have a limited lifespan, such aerosol injections would have to be renewed on an approximately annual basis for a long time.

But what Stormfury and other scientific endeavours have repeatedly shown is that idealized science does not easily overcome messy realpolitik.

Stratospheric aerosol injection would be an extraordinarily dangerous way to attempt to lessen the impacts of global warming—one that most climate scientists roundly warn against. But if leaders around the world ignore those warnings and proceed with a stratospheric injection program or other solar radiation management initiative, that program or initiative will need to be managed. And history shows that environmental policy regulation is insufficient and unreliable as a means of controlling any technology with the catastrophically high-risk profile that stratospheric aerosol injection carries.

What would deployment governance look like? Even an initial and very incomplete canvas of governance for a stratospheric injection program paints a demanding picture, raising questions like who gets to control the world’s climate and how they would go about it.

At a fundamental level, decision-makers would have to determine how to best ensure broad international representation and democratic processes. They would need to agree on who gets to make deployment decisions, and how. This would pose the difficult question of how to govern not only countries, but also decentralized private-sector actors. Such actors do not have a mandate to act in the public interest and instead could be driven by profit. One early example is Make Sunsets, a start-up that is controversially selling “cooling credits” to release particles into the atmosphere.

Decision-makers would then have to determine deployment objectives and configurations. How much warming should stratospheric aerosol injection moderate, and what ecological side effects can be mitigated? Where should injections therefore happen, what substances should be used, and how much of these particles should be released?

To prevent global catastrophe, governments must first admit there’s a problem

A critical related issue is making sure deployment is not used as a perpetual offset or in attempts to make “designer” climates. Some might argue that stratospheric aerosol injection is too uncertain to deliberately make optimal regional climates, but decisions (and political promises) are often made based on incomplete information, poor knowledge, and misguided intentions. Such misguided intentions are especially dangerous when deliberately modifying the entire climate.

Furthermore, a robust international system for monitoring and evaluation of global and regional ecosystems is necessary to identify and address side effects. This optimistically assumes that ecological consequences will always be clearly measurable. And given solar radiation management’s potentially enormous negative side effects, decision-makers may be called to develop accountability and liability mechanisms (again assuming that clear attribution of ecological impacts of stratospheric injection is even possible).

Most important, absent a quick and enormous breakthrough in carbon dioxide removal technology, all of these policy elements would have to be maintained for an exceedingly long time—quite possibly a great deal longer than a century. Continuous stratospheric aerosol injections would last for however long it takes to reduce greenhouse gas emissions to zero and to remove excess greenhouse gases from the atmosphere. Its associated policy and regulation regimes would need a similarly long life.

The fairy tale ends. Such policy consistency is a pipe dream, as evidenced by human history. Societies, politics, and governments change. An election won or lost can be the birth, death, and resurrection of disparate policy approaches. Donald Trump won the American presidency in January 2017 and announced withdrawal from the Paris Agreement five months later. Joe Biden rejoined the treaty on January 20, 2021, his first day in office.

Inconsistency is unfortunately a part of the core nature of policy. This is especially the case for intensely political issues, which stratospheric aerosol injection would certainly be. A reasonably well-governed initial deployment could be undone by political shifts. Solar radiation management with whiplash-inducing policy shifts does not bode well for effective implementation—or Earth’s climate.

The needed policy depth, while not strictly impossible, seems improbable given the difficulties of international governance. For example, one option is to house stratospheric aerosol injection governance within the UN climate system. An amendment to the Paris Agreement or a new post-Paris treaty could include such governance provisions. However, the UN climate system is notoriously ineffective. It works by consensus—each country essentially has a veto vote. Any policy that gets passed is heavily watered down. This does not bode well for robust and stringent governance required for technologies as potentially catastrophic as solar radiation management.

Some may suggest alternative UN bodies or organizations for a governance home, like the UN Environment Programme (UNEP). But such bodies would also run into similar consensus related issues. Where broad consensus is present, robust policy cannot be.

An alternative would be to work in a treaty system with exclusive participation. Instead of being broad and shallow, stratospheric aerosol injection governance could be deep and narrow. The downside is that this could mean a lack of adequate representation. A small number of countries taking it upon themselves to engineer the world’s climate would run counter to principles of democracy and create an almost certain backlash among countries that have no say in the process.

Power and politics define the rhyme and rhythm of international governance. Ultimately, the specific governance approach and structure for a stratospheric aerosol injection program would not be the remit of a group of well-informed and well-intentioned academics. Politics define policy windows, and thus policy discussions.

$600 million battery storage project meets resistance in California

For instance, housing the governance of aerosol injections would not be solely based on rational calculations of policy effectiveness. Countries will “forum shop,” intentionally choosing which structure best fits their interests. For example, initial discussions of governance at the UN Environment Assembly in 2019 met immediate political roadblocks, with some countries, including the United States, preferring the Intergovernmental Panel on Climate Change (IPCC). Even the creation of the IPCC itself, jointly by the UNEP and World Meteorological Organization, was driven by the United States wanting to limit UNEP’s influence over the emerging international climate policy landscape.

Similarly intense political drivers can be seen in specific policy design as well. For example, The Paris Agreement uses a “pledge and review” approach; its main commitment for countries is putting forward, not meeting, a new climate target every five years.

Paris’ pledge and review wasn’t chosen because countries thought it the most effective approach to reduce emissions. It was used because the lack of substance allowed the United States to join the Paris Agreement via presidential executive order. This intentionally bypassed a heavily deadlocked United States Senate that would never ratify a climate treaty. The world essentially traded policy substance for American participation.

Good science and good policy proposals do not fix ugly politics. Rather, ugly politics taint what they touch.[i]

Imperfect governance. Because of power and politics, most international treaties fail to meet their stated goals (as found by a recent synthesis of more than 200 studies). Flawed governance is an acceptable and manageable risk for some policy areas. Carbon markets, for instance, tolerate a level of imperfection, with up to 20 percent of carbon credits usually accepted as flawed, in order to incentivize additional climate action overall. Furthermore, just having a treaty in place can affect social and political norms, even if its substantive core is lacking.

But soft governance is simply not acceptable for high-risk technologies like solar radiation management initiatives. Stratospheric aerosol injection with deeply flawed governance would be like hoisting a climatic Sword of Damocles—it may allay climate risks in the short term, but when disaster strikes (and over several decades, one will) it would be catastrophic.

As technological capabilities grow, the margin of error for effective governance shrinks. In the mid 1900s, humans were attempting to affect a specific weather event—cyclones. With stratospheric aerosol injection, they’d be affecting the entire world’s climate. The risks are high, and that makes the required bar for effective governance even higher.

Given humanity’s political track record, that bar seems out of reach.


[i] As a counterpoint, some may point to our global aviation system as an indication that complex, international, and rather technocratic governance is possible. However, one must also consider the political context that enabled and drove such efforts. By the end of World War II, the United States had developed substantial advantages in its aviation infrastructure. They had an economic interest in aggressively expanding the global aviation market, against the wishes of less wealthy countries who wanted to protect their own aviation industries. It was also a time of historical power imbalances. America’s only rival was the Soviet Union, and the United States still had an economic and technological advantage. The world today, while still deeply unequal, is far more multi-polar. At the time, a complex international governance system was only possible because a powerful country was willing, and able, to force it into existence.

Together, we make the world safer.

The Bulletin elevates expert voices above the noise. But as an independent nonprofit organization, our operations depend on the support of readers like you. Help us continue to deliver quality journalism that holds leaders accountable. Your support of our work at any level is important. In return, we promise our coverage will be understandable, influential, vigilant, solution-oriented, and fair-minded. Together we can make a difference.

Get alerts about this thread
Notify of
Newest Most Voted
Inline Feedbacks
View all comments
Matthias Honegger
Matthias Honegger
1 year ago

I share your characterization of the challenge. But I would add that inaction or passivity is not an option, precisely because the default is messy politics and ineffectual policy. Rather, I seek to adopt a stubborn optimism that is instrumental to trying when it looks impossible – to strengthen the governance basis for SRM. It is that stubborn optimism that allowed Christiana Figueres to shepherd the international community to the first-ever global-in-scope agreement on climate.

Charles Forsberg
Charles Forsberg
1 year ago

Wrong question. If the Chinese get too many crop failures, they will act–probably with India. The West is incapable of making major decisions. After they act, there will then be discussions on the optimum temperature of the earth. If a specially designed aircraft for efficient aerosol delivery is spotted rolling out of a Chinese aircraft manufacturing facility, I will not be surprised. Any competent government would develop the technology to have the option to deploy. It is no different than buying weapons in peacetime. Only a fully disfunctional government would not prepare for such a contingency.

10 months ago

Some valid points but started with a weak analogy and unscientific fear mongering. There is no do-nothing option and we waited too long to rely solely on co2 and methane reduction. Climate change is aggressively and permanently destroying assets needed to sustain life. 20 years ago, refraining from geoengineering may have been valid. That ship has passed, unfortunately. Now we will have to try to intervene in more artificial and risky ways. The effects of solar aerosols happen to be fairly well known as a naturally occurring phenomenon via volcanic activity and there is the possibility of using more inert… Read more »

Last edited 10 months ago by Wills
Stephen Schiff
1 month ago

Even now discussions concening SAI are so shallow as to miss a number of crucial points. For instance Modeling using current global circulation models suggests the need to inject ca 10 – 20 million metric tons of sulfates at stratospheric altitudes of 20 km or above annually. The only currently technologically feasible way of doing so is via aircraft, and no aircraft capable of carrying as little as a few tons of material per flight to such altitudes exists. Moreover, the amount of sulfur required is a significant fraction of that produced, promising a disruption in the global sulfur market.… Read more »


Receive Email