Nuclear War Project

PROJECT LEADS: Tony Barrett and Seth Baum

GCRI’s work on nuclear war covers the full range of topics for nuclear war as a GCR. We conduct research on the probability of various nuclear war scenarios, the severity of the consequences if nuclear war occurs, and the interventions that can be performed to reduce the risk. We also do what we call “stakeholder engagement” by participating in debates about nuclear weapons issues with leading experts, activists, and policy makers. Our nuclear war research and stakeholder engagement are an excellent example of the range of work we have set up in our flagship Integrated Assessment project.

Nuclear weapons were invented in 1945, at the height of World War II. They were used twice in the war, killing hundreds of thousands of people. They have been a major international issue ever since. The risk of nuclear war has received less attention since the end of the Cold War. However, there are still thousands of nuclear weapons. More than 90% of these weapons are still held by Russia and the United States [1]. New research on nuclear winter shows that it could have catastrophic global consequences even from war with the much smaller nuclear arsenals of other countries [2]. And there are still tensions between nuclear-armed states, for example in disputes over Kashmir, Taiwan, and Ukraine. Nuclear war thus remains a major GCR.

The Probability Of Nuclear War

Although people have been concerned about nuclear war risk for a long time, for years very little research has considered how likely a nuclear war actually is. That began to change in recent years with the work of our colleague Martin Hellman [3]. GCRI researchers Tony Barrett, Seth Baum, and Kelly Hostetler contributed to this topic with a detailed study of the probability of an inadvertent nuclear war between Russia and the United States. Inadvertent nuclear war occurs when one side misinterprets a false alarm as a real attack and launches nuclear weapons in what it believes is a counterattack, but is in fact the first strike. Barrett and colleagues developed a detailed fault tree model of inadvertent nuclear war scenarios. They used a detailed survey of prior literature to inform the selection of model parameter values. They also used probability distributions to represent uncertainty about the parameter values. They found that the annual probability could be anywhere within a wide range. The range is wide because of the uncertainty in parameter values. For about 90% of the combinations of parameter values, the annual probability was between 0.07 and 0.00001, or on average roughly once per 14 years to once per 100,000 years [4]. This study exemplifies the detailed and rigorous probability analysis that GCRI does.

The Consequences Of Nuclear War

For decades, people believed that nuclear war would primarily affect just the countries directly hit by nuclear weapons. That belief proved false in the 1980s, when new research showed that a nuclear war could cause catastrophic global environmental changes, popularly known as nuclear winter. Indeed, some key early work on GCR came out of the 1980s nuclear winter research [5]. More recent work on nuclear winter uses state-of-the-art models to give unprecedented insight into the environmental [2] and agricultural [6] consequences of nuclear war. GCRI contributes to this research by looking more carefully at the full range of consequences for humanity and the uncertainties about these consequences. A key consequence we look at is the possibility of permanent collapse of human civilization [7]. We study the risk of permanent collapse for a range of nuclear war scenarios [8]. Our work on the consequences of nuclear war is closely related to our project on the Aftermath of Global Catastrophes.

Risk Reduction Interventions

Ultimately, what is most important is how risks can be reduced. This holds for nuclear war as much as it holds for any other risk. GCRI nuclear war research catalogs existing interventions, develops new ones, and evaluates their effectiveness [9]. For example, our study on inadvertent nuclear war looked at two specific risk-reducing interventions: (1) moving submarines further from adversary coasts to give adversaries more time to correctly identify alarms as false alarms; and (2) part-time lowering of weapon system alert levels to make it less likely that nuclear weapons are launched in response to a false alarm. Our study found that each intervention could cut the risk approximately in half [4]. Another example is alternative food. This new form of nuclear war intervention was developed by GCRI Associate Dave Denkenberger and Joshua Pearce of Michigan Technological University. If nuclear winter destroys agriculture, then alternative foods might help people survive. The alternative foods would be produced using energy from stored biomass (such as trees) or fossil fuels, instead from sunlight. Initial research by Denkenberger and Pearce suggests that alternative foods might be able to feed all nuclear war survivors [10]. A third example is alternative deterrents. A study by Seth Baum considered what other types of weapons besides nuclear weapons could help countries achieve their deterrence goals without risking global catastrophe [8].

Stakeholder Engagement

GCRI presents its nuclear war work to key nuclear war stakeholders, building relationships with them, and getting their input for future work. This ensures that our work is both relevant to and used by nuclear war decision makers. In 2012-2013, Tony Barrett held a Stanton Foundation Nuclear Security Fellowship at the RAND Corporation in Washington, DC. Barrett spent time with leading experts and policy makers. Barrett presented his work at the Center for Strategic and International Studies, a leading Washington think tank on nuclear war and other security issues. More recently, Seth Baum has been active in international discussions of nuclear war. He has presented at international events hosted by the United Nations, the U.S. State Department, the Foreign Ministry of Austria, and others. He also has a monthly column in the Bulletin of Atomic Scientists, in which he often discusses nuclear war. These and other activities offer important platforms for GCRI to engage stakeholders on nuclear war.


[1] Status of World Nuclear Forces 2014. Federation of American Scientists.

[2] Alan Robock, 2010. Nuclear winter. Wiley Interdisciplinary Reviews: Climate Change 1, 418-427. See also Robock’s nuclear winter page and nuclear winter publications page.

[3] Martin Hellman, 2008. Risk analysis of nuclear deterrence. The Bent of Tau Beta Pi, Spring, 14-22.

[4] Anthony M. Barrett, Seth D. Baum, and Kelly R. Hostetler, 2013. Analyzing and reducing the risks of inadvertent nuclear war between the United States and Russia. Science and Global Security 21(2), 106-133.

[5] Carl Sagan, 1983. Nuclear war and climatic catastrophe: Some policy implications. Foreign Affairs 62, 257-292.

[6] Lili Xia and Alan Robock, 2013. Impacts of a nuclear war in South Asia on rice production in mainland China. Climatic Change 116, 357-372. Mutlu Özdoğan, Alan Robock, and Christopher Kucharik, 2013. Impacts of a nuclear war in South Asia on soybean and maize production in the Midwest United States. Climatic Change 116, 373-387.

[7] Timothy M. Maher, Jr. and Seth D. Baum, 2013. Adaptation to and recovery from global catastrophe. Sustainability 5(4), 1461-1479.

[8] Seth D. Baum, 2015. Winter-safe deterrence: The risk of nuclear winter and its challenge to deterrence. Contemporary Security Policy 36(1), 123-148.

[9] Seth D. Baum. Confronting the threat of nuclear winter. Futures, forthcoming, DOI 10.1016/j.futures.2015.03.004.

[10] David Denkenberger and Joshua Pearce, 2014. Feeding Everyone No Matter What: Managing Food Security After Global Catastrophe. Waltham, MA: Academic Press.