Geoengineering to reduce global catastrophic risk?

Thanks to Luke Kemp and John G. Halstead for commenting on an earlier version of this post. They have different views on some points that I make throughout and I highly recommend reading their research on the topic. 

My main point is: Positively shaping the global governance of geoengineering is a potentially effective way of reducing global catastrophic risk.

Isn't geoengineering dangerous?

Yes. Geoengineering, in the literature most often called climate engineering, is the deliberate manipulation of the planet's climate. It is easy to see how that can be dangerous. Although geoengineering poses risks of its own, the topic should not be simply discarded because it interacts with global catastrophic risk in numerous ways. In order to better understand geoengineering, let's look at how we might do it.

Researchers have identified several different geoengineering methods for mitigating climate change. Most of these can be grouped into two categories, solar geoengineering and carbon dioxide removal.^[1] Solar geoengineering attempts to lower the temperature by reducing the amount of energy that the earth absorbs from the sun. As an example, one idea is to make clouds brighter so that they reflect more sunlight. This can be done by spraying fine particles into the lower atmosphere from ships.^[2] Carbon dioxide removal, on the other hand, is an approach that directly tackles the root cause of climate change by reducing CO2 levels in the atmosphere. This is much less dangerous than tempering with the earth's sunscreen. Think of it this way: Global greenhouse gas emissions are a geoengineering experiment. Carbon dioxide removal is trying to undo it.

If carbon dioxide removal is the dominant strategy, why even bother about solar geoengineering? Solar geoengineering has three major advantages over reducing carbon dioxide in the atmosphere: It is fast, flexible and dirt cheap.^[3] Compared to the costs of climate change it is so cheap that costs do not matter for deployment decisions. To clarify, India alone can expect to gain in the magnitude of around 1 trillion dollars in wealth through artificially lowered temperatures.^[4] A large-scale solar geoengineering program would cost only around 1-10 billion dollars per year.^[5] Such a program could offset many degrees of warming. 

 The idea of solar geoengineering has such incredible economics that it isn't going to go away soon.^[6]

How does geoengineering affect global catastrophic risk?

Despite the fact that geoengineering is dangerous, it is going to be considered by many nations as a way of mitigating climate change.^[7] It is therefore important to think about how this will affect our future.

Although there are concerns regarding the effects of geoengineering on the ecosystem, these are not what researchers are most concerned about. The intricate relationship of geoengineering with nations sovereignty is what makes geoengineering a huge problem. Most likely, the weather in your place is currently not determined by other humans. Geoengineering would change that. Furthermore, climate is by definition global. The effects of geoengineering are hard to regionalize and all countries will be impacted by geoengineering in some way.^[8] Therefore, just the bare possibility of geoengineering is going to stress international relations.^[9]

India might gain a lot from a geoengineering program, but countries like Canada, Russia, Germany and the UK would lose hundreds of billions of dollars (e.g. through decreased agricultural productivity).^[10] Furthermore, geoengineering might politicize the weather, causing governments and media to blame others for natural disasters or crop loss.^[11] A worst-case scenario would be to have competing geoengineering efforts.^[12] In other words:

[If governed poorly,] climate engineering could complicate ongoing efforts to mitigate and adapt, not least by promoting adversarial friend–enemy logics that sour international relations and make collective action on mitigation [and other pressing problems] more difficult, as well as making climate engineering solutions themselves more costly and difficult to implement.This points to a wider set of neglected issues around how climate engineering might interact politically with other forms of climate policy.^[13]

A causal diagram makes clear the ways in which geoengineering affects global catastrophic risk. In the picture below an arrow  corresponds to an effect. That is,$X\to Y$ means X has an effect on Y.

International relations are a major factor of global catastrophic risk. Peace and cooperation among countries is needed to reduce the chance of future conflicts and to solve international problems, like combating pandemics, together. The main contribution of geoengineering to global catastrophic risk is likely its indirect effect on international relations. That is, the path 

$\text{Geoengineering}\to \text{Global Climate}\to \text{International Relations}\to \text{Global Catastrophic Risk}$

has likely the greatest negative impact. 

Notice that I have drawn no direct arrow from geoengineering to global catastrophic risk. The chance of catastrophe through geoengineering directly, whilst non-zero, is small compared to the other effects.^[14]

Also, there is a chance of extreme warming.^[15] A solar geoengineering program would be an effective protection against that scenario.^[16] Furthermore, should the warming become extreme, other catastrophic environmental damage might result that could be averted by geoengineering. The path 

$\text{Geoengineering}\to \text{Global Climate}\to \text{Global Catastrophic Risk}$

could have a positive impact (total risk reduced). However, a poorly-governed program could also exacerbate current climate risks through the risk of termination shock.^[17] If a geoengineering program were to be used to mask warming and then discontinued, the planet would warm very rapidly.

Geoengineering also interacts with risks from nuclear winter, large volcanic eruption and small asteroid impact.^[18] All of them are of global scale because of their climatic effects (reduced solar radiation and excessive cooling). Once a geoengineering infrastructure is available it seems possible that this infrastructure can be equipped to be able to cause rapid warming, potentially reducing the risks.^[19]

Overall, the complex interdependencies of geoengineering with international relations and the global climate make it difficult to assess whether it will be beneficial or not.

It is clear that irresponsible and uncoordinated use of geoengineering would be terrible. We need a governance mechanism that ensures that nations share risks and benefits of geoengineering and that all activities have international approval. If geoengineering is ever to be used, we should make sure that it is a robust system.

Why work on it?

Geoengineering is uniquely influencing many important factors and thus global catastrophic risk for better or worse. The goal is to get all major powers to sign a geoengineering treaty that protects the future of the planet. This would relieve the strain geoengineering will put on international relations. Furthermore, a properly governed program could largely eliminate the risk from extreme warming. 

Moreover, it is not only about risk reduction. Geoengineering is a very timely issue and the risk landscape matters. I believe that, if left unresolved, geoengineering is likely to cause international friction much sooner. Thus, global catastrophic risk would increase earlier than it would otherwise.^[20]

So, how do we do it?

For [Climate change], governance is concerned with encouraging action, while for solar geoengineering, governance is primarily concerned with inhibiting action.^[21]

Given that humanity has failed to govern greenhouse gas emissions for three decades straight, one should not expect geoengineering governance to be a walk in the park. There are significant challenges ahead that need to be addressed, but there is an existing body of literature on the topic and the path forward is relatively clear. Contrary to some other opportunities to positively shape the future of humankind, it can and therefore should be seized now. Below are some research ideas of mine.

• Find out how to alter the behavior of states toward a solution as close to the optimum as possible. Provide the foundations of a working treaty. (For social scientists) - A lot of research in this direction has been done already but the most interesting questions remain unanswered
• Find out how to monitor geoengineering deployments. Then build such a system. (For physical scientists, engineers, computer scientists) - I am not aware of any such efforts
• Research and implement safe climate engineering. (Both for engineers and physical scientists) - Already some existing research. Current research is not sufficient to deem interventions safe.
  • Whilst most actions taken to alter the climate are unprecedented and pose therefore a tiny risk of catastrophe (because there simply is no way to do the experiment up ahead), it might be possible to do very low-risk climate engineering now. This is because we are blessed with some good natural experiments. For example, it seems as if we have been doing marine cloud brightening unintended for some time without recognizing any catastrophic results. Cloud brightening from ships exhaust emissions has had a measurable cooling effect. As ships emit less of these particles due to increased emissions regulation, the cooling effect is reduced. Therefore, it seems plausible that if we were to reverse back to old levels of marine cloud brightening we must not fear strong side-effects.^[22]
• Research the effect of geoengineering (research) on climate mitigation and other problem areas. (Trans-disciplinary, especially for social scientists and philosophers) - A lot of existing research is available that has not yet reached a conclusion. In fact, I have read basically every imaginable opinion on that topic. For one see here.

A fantastic short movie on geoengineering by Kurzgesagt - In A Nutshell is available here.

Notes and References

1. ^{^}

   All technologies and methods have different properties and greatly varying costs. I summarize them here for the sake of simplicity. For an overview of popular geoengineering ideas see the EuTRACE assessment in Note 7.

2. ^{^}

   This is similar to how vapor trails form behind planes in the blue sky. See Wikipedia for a solid introduction to the method. 80000Hours also has a podcast on this, although I am very skeptical of the content expressed.

3. ^{^}

   See (p.9) in Geoengineering: the Gamble by Gernot Wagner. I also simplify here. Space-based solar dimming e.g. is not dirt cheap at all.

^{^}

This figure relates to earnings over multiple decades this century. The estimates are imprecise to say the least but the magnitude matters here. See: 

Rickels, Wilfried, et al. Turning the global thermostat-who, when, and how much?. No. 2110. Kiel Working Paper, 2018. | Link

And to a lesser extent: 

Emmerling, Johannes, and Massimo Tavoni. "Quantifying non-cooperative climate engineering." (2017). | Link

^{^}

Cost estimates vary and depend on the technology used. For one detailed estimate of costs related to Stratospheric Aerosol Injection see:

Smith, Wake and Gernot Wagner. “Stratospheric aerosol injection tactics and costs in the first 15 years of deployment.” Environmental Research Letters (2018) | Link

^{^}

See (p. 15): 

Weitzman, Martin L. "A voting architecture for the governance of free‐driver externalities, with application to geoengineering." The Scandinavian Journal of Economics 117.4 (2015): 1049-1068. | Link

Or: 

Schelling, Thomas C. "The economic diplomacy of geoengineering." Climatic Change 33.3 (1996): 303-307. | Link (not secure) 

Or: 

Barrett, S., 2008. The incredible economics of geoengineering. Environmental and resource economics, 39(1), pp.45-54. | Link (not secure)

^{^}

John G. Halstead disagrees with that point, see this Forum Post and this. I think it is correct because many sources point to this (see a sample below). In fact, Germany is already pursuing a medium-scale carbon dioxide removal program (see Koalitionsvertrag, p.31).  I do however agree that unilateral deployment of solar geoengineering seems quite unlikely from a International Relations 101 perspective.

For a policy makers perspective see e.g. the EU's assessment on geoengineering: 

(2015) The European Transdisciplinary Assessment of Climate Engineering (EuTRACE):
Removing Greenhouse Gases from the Atmosphere and Reflecting Sunlight away from Earth. | Link

Also see IPCC 2014 Full Report on Climate Change (p.89):

IPCC, 2014: Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, R.K. Pachauri and L.A. Meyer (eds.)]. IPCC, Geneva, Switzerland, 151 pp.

Also see the 2022 Climate Intervention Policy Statement of the American Meteorological Society

^{^}

See e.g.: 

Ricke, K., Morgan, M. & Allen, M. Regional climate response to solar-radiation management. Nature Geosci 3, 537–541 (2010). | Link

^{^}

Martin Weitzman says in his paper from Note 6:

Geoengineering is an externality accident waiting to happen.

^{^}

See Note 4

^{^}

See:

Halstead, John. "Stratospheric aerosol injection research and existential risk." Futures 102 (2018): 63-77. | Link

^{^}

It is not completely clear to what extent that is possible. However, early research shows that geoengineering could be very versatile, potentially giving states the ability to counteract other climate measures. See: 

Parker, Andy, Joshua B. Horton, and David W. Keith. "Stopping solar geoengineering through technical means: a preliminary assessment of counter‐geoengineering." Earth's Future 6.8 (2018): 1058-1065. | Link

or 

Heyen, Daniel, Joshua Horton, and Juan Moreno-Cruz. "Strategic implications of counter-geoengineering: clash or cooperation?." Journal of Environmental Economics and Management 95 (2019): 153-177. | Link

^{^}

Quoted from: 

Corry, Olaf. "The international politics of geoengineering: The feasibility of Plan B for tackling climate change." Security Dialogue 48.4 (2017): 297-315. | Link

^{^}

See: 

Tang, Aaron, and Luke Kemp. "A Fate Worse Than Warming? Stratospheric Aerosol Injection and Global Catastrophic Risk." (2021). | Link

Also, depending on the technology used to lower temperatures some unpleasant side-effect (like e.g. acid rain) might occur. But these are no catastrophic risks and should be treated as "common" environmental externalities.

^{^}

For a brief overview over research related to extreme warming and global catastrophic risk see (p. 102)

Ord, Toby. The precipice: Existential risk and the future of humanity. Hachette Books, 2020.

^{^}

A carbon dioxide removal program is not an effective protection since it takes time to remove CO2 and even more time for the planet to cool. If carbon feedback loops (e.g. through thawing permafrost) kick in, carbon dioxide removal might just be too slow to cool the planet. 

^{^}

See: 

Tang, Aaron, and Luke Kemp. "A Fate Worse Than Warming? Stratospheric Aerosol Injection and Global Catastrophic Risk." (2021). | Link

^{^}

This is very speculative and solely my conjecture based on the sources in Note 12.

^{^}

Luke Kemp has rightfully pointed out that this brings with some problems of its own.

^{^}

I weakly believe this because I have noticed a different writing style in the assessments aimed at policy makers (see Note 1, especially EuTRACE) relative to academic literature. The academic literature is very careful about communicating knowledge around geoengineering, mostly wrapping it in sentences like "first off, this is very dangerous and must be approached with caution". The assessments separately highlight these risks but also present multiple geoengineering technologies as a colorful basket of options in the fashion "we could do a, b, c or d ; would be that much".

^{^}

Quoted from 

Sandler, Todd. "Collective action and geoengineering." The Review of International Organizations 13.1 (2018): 105-125. | Link

^{^}

I am not suggesting that we should lift the emissions regulation. It seems that the cloud brightening effect can be achieved with other harmless particles. 

Show all footnotes