This post is part of a series of rough posts on nuclear risk research ideas. I strongly recommend that, before you read this post, you read the series’ summary & introduction post for context, caveats, and to see the list of other ideas. One caveat that’s especially worth flagging here is that I drafted this in late 2021 and haven’t updated it much since. I’m grateful to Will Aldred for help with this series.
One reason I'm publishing this now is to serve as one menu of research project ideas for upcoming summer research fellowships. I'm publishing a few of the posts to the frontpage to raise awareness that this series exists, but I'll keep the rest as personal blogposts to avoid spamming the frontpage.
Some tentative bottom-line views about this project idea
What is this idea? How could it be tackled?
This topic could be broken down into the following subtopics:
- What climate effects would occur, given various possible nuclear conflicts?
- One could try to cover a range of specific possible conflicts, or to cover particularly extreme possible conflicts (e.g., >1000 warheads used against cities), or to make broader points about the debate over nuclear winter and relevant key factors and uncertainties.
- How much would agricultural production decline, given various possible effects of nuclear conflict?
- This should consider changes over time, since the declines could grow or shrink over the months and years following the conflict.
- This would ideally consider possible adaptive measures people could engage in, such as changing what crops they grow, where they grow them, and how much they rely on crops vs other types of food (including “resilient food”).
- This could consider how the declines vary between countries and between types of crops or other agricultural goods.
- This could account for not just the climate effects of nuclear conflict but also the effects on trade (e.g., of fuel or fertiliser), immigration, labour forces, etc.
- How many people would die due to famine, given various possible agricultural production declines following nuclear conflict?
- This could consider not just deaths “directly” caused by famine but also those caused by disease outbreaks, further conflicts, etc. that are in turn partly caused by famine.
I’d intended to research those three subtopics myself, so I made rough notes on what the topics mean, why they matter, how to tackle them, what the relevant variables might be, what previous work says, etc. I could share the first batch of notes on request. The second and third batches can be found here and here, respectively.
It seems probably ideal for a single person or team to tackle all three subtopics, and in roughly that order. But I think it would also be possible and useful for a person to tackle any one of the subtopics - or one of its own many sub-subtopics - by itself.
I think projects on those subtopics could span anywhere from a single person working for a single week to a team working for (say) six months. Brief projects could look like:
- Simply summarising bodies of existing work that are relevant to those subtopics
- Red-teaming existing work relevant to those subtopics
- Working out which variables seem relevant (e.g., thinking about various ways people could adapt in the face of climate effects of agricultural production declines and what variables would affect their ability to do so)
- Making very rough estimates for some of the relevant variables
- Constructing very simple models
- Brief expert elicitation, such as having several calls or email exchanges regarding those subtopics and summarising takeaways
More extensive projects could look like:
- Constructing and running proper fire models, climate models, models of agricultural production, or economic models of how prices and consumption might change
- Working out good estimates for the relevant variables and plugging them into models
- Doing extensive expert elicitation, such as running workshops or sending out carefully designed surveys to a wide variety of relevant experts
Why might this research be useful?
The harms from climate and famine effects following nuclear war could dwarf the more immediate harms, and also seem the most likely way for nuclear war to cause an existential catastrophe. But there’s very large uncertainty around this; each of the above questions has received relatively little attention, the research that has been produced has been contested, and it has various flaws in my view. Unfortunately, many people - including people within the EA community or non-EAs focused on nuclear risk reduction - either overlook such climate and famine effects as a key consideration or overstate the likelihood and likely severity of such effects. For further discussion, see my docs of rough notes (here and here) and Aird (2022). As such, I think good research on these questions could substantially advance the frontiers of EA knowledge on these topics and/or bring more people up to those frontiers.
What sort of person might be a good fit for this?
This depends on the sub-question and the approach taken to it. I expect any good generalist researcher could usefully do some version of this project. But some projects on this topic would benefit substantially from knowledge and skills in the domains of climate science, agricultural science, economics, modelling, statistics, and maths.
Some relevant previous work
- Aird, 2020, “Why I'm less optimistic than Toby Ord about New Zealand in nuclear winter, and maybe about collapse more generally”
- Denkenberger & Pearce, 2018, “Cost-effectiveness of interventions for alternate food in the United States to address agricultural catastrophes”
- Ladish, 2020, “Nuclear war is unlikely to cause human extinction”, in particular the section on “Climate alteration”
- Ord, 2020, The Precipice, Chapter 4, the section on “Nuclear Weapons”
- Toon et al., 2007, “”Atmospheric effects and societal consequences of regional scale nuclear conflicts and acts of individual nuclear terrorism”
- Reisner et al., 2018, “Climate Impact of a Regional Nuclear Weapons Exchange: An Improved Assessment Based On Detailed Source Calculations”
- Robock et al., 2019, “Comment on [...] Reisner et al.”
- Reisner et al., 2019, “Reply to Comment by Robock et al. […]”
- Rodriguez, 2019, “How bad would nuclear winter caused by a US-Russia nuclear exchange be?”
- Shulman, 2012, “Nuclear winter and human extinction: Q&A with Luke Oman”
One could perhaps also add “What would be the long-term effects of various possible levels of famine effects (e.g., of 2 billion vs 6 billion vs 7.5 billion deaths)?” I think that’s a key question, but it feels to me better treated as a separate project. I briefly discuss this sort of topic here and in some of the other sections of that appendix.
Note that the second and third of these three questions could also be relevant for risks other than nuclear conflict that could also cause similar climate effects and/or crop yield declines, such as asteroid impacts, supervolcano eruptions, or anti-plant bioengineering attacks or accidents (see Aird, 2021; Dando, 2006; Denkenberger & Pearce, 2015).