Skip to main content

New Research on Climate and Conflict Links

Experts discuss new and existing research that shows how climate change could potentially contribute to armed conflict or violent social unrest.

Date & Time

Dec. 19, 2011
9:00am – 11:45am ET


6th Floor, Woodrow Wilson Center
Get Directions


“We know that there will be more conflicts in the future as a result of climate change than there would have been in a hypothetic world without climate change,” said Marc Levy, deputy director of the Center for International Earth Science Information Network (CIESIN) at Columbia University’s Earth Institute, although existing data and methodologies cannot predict how many additional conflicts there will be, or which causal factors will matter most.

Levy spoke at a December 19 panel at the Wilson Center on new research on the linkages between climate change and conflict. He was joined by Joseph Hewitt, technical team leader for USAID’s Office of Conflict Management and Mitigation; Joshua Busby, assistant professor at the Lyndon B. Johnson School of Public Affairs, University of Texas at Austin; and Solomon Hsiang, postdoctoral research associate at Princeton University’s Woodrow Wilson School of Public and International Affairs.

Linking El Niño to Civil Conflict

Princeton University’s Solomon Hsiang recently co-authored a study published in Nature that used statistical analysis to link observable changes in the global climate to conflict outcomes on the ground. The researchers looked at countries strongly impacted by the El Niño/Southern Oscillation and compared the onset of civil conflict in those countries during El Niño, relative to the La Niña state.

“[El Niño] is the single dominant pattern of the entire planet’s climate on annual timescales,” said Hsiang. “So what is convenient here from a statistical standpoint is that the climate is going back and forth very rapidly…so there haven’t been major socio-political changes over that time horizon.”

The study found that conflict risk for a given region doubled during the hotter and drier El Niño state, from an average of around three percent to six percent. “You can make a variety of different assumptions about what kind of statistical model you are using and you generally always get the same estimate,” said Hsiang. “The correlation between the global climate and conflict seems to be very, very robust to a variety of choices…It’s one of the most robust results I have seen in any of the statistical literature.”

Nevertheless, “our study doesn’t say anything about why El Niño might be linked to conflict,” Hsiang clarified. “We are just showing an association. Climate is not the only thing driving conflict in these countries…it exacerbates an existing problem.”

Identifying Chronic Vulnerability in Africa

Working at the University of Texas at Austin, Josh Busby presented the Climate Change and African Political Stability program, a composite index mapping climate security vulnerability in a region with rising strategic significance and low adaptive capacity. The index incorporates not only physical exposure but also demographic, socio-economic, and political indicators.

“We focus on situations where large numbers of people could be exposed to mass death from climate-related hazards,” said Busby. He identified southern Somalia, South Sudan, and much of the Democratic Republic of the Congo as among the most vulnerable regions, relative to the rest of Africa.

These areas might not necessarily appear as the most vulnerable from a strictly climatic point of view, Busby said, but the composite analysis brings them into focus. For instance, many factors, including governance and a strong La Niña year, contributed to the famine Somalia experienced this year. Although the precise role of climate change is unclear, from a chronic vulnerability perspective, southern Somalia remains an area of concern, he said.

Understanding Pre-Existing Conditions in Vulnerable States

The Office of Conflict Management and Mitigation at USAID has commissioned research looking at the relationship between countries that are conflict affected, countries that are fragile, and countries that are highly vulnerable to climate threats, said Hewitt: “We wanted to better understand which countries are likely not to have the capacity, or likely not to have the ability, to manage the stresses and strains of climate threat.”

“[Fragile states] are already characterized by many, many different challenges that contribute to causes of conflict, just aside from climate change itself,” Hewitt pointed out. “Any understanding of the relationship between climate change and conflict needs to understand how climate change is in some sense filtered through all of these existing characteristics.”

On the other hand, many countries identified as highly vulnerable to climate change are not necessarily considered fragile. Despite the predicted changes in climate for these places, they have sources of conflict mitigation and resilience that will likely be able to handle the strains posed by climate change, Hewitt said. “We really want to try and understand what is happening in these countries. How are those countries positioned to confront those stresses, identify coping strategies, and adapt?”

“Any programming that is done to address the consequences of climate threats needs to be attentive to the connections between the program and any pre-existing characteristics that either mitigate conflict or in some sense make the society more vulnerable to conflict” said Hewitt.

Projecting Into the Future

Columbia University’s Marc Levy noted that a strong case for linking climate stress to increased risk of conflict can be made by better explaining the causal chain that leads from environmental change to societal stress. According to the 4th IPCC Assessment, climate change will increase stress on a number of biophysical processes and systems relevant to human societies, such as agriculture, water, ecosystems, and disease. A body of research shows that these natural stresses make societies more vulnerable, consequently increasing their risk of conflict.

Nevertheless, these conclusions are limited by data, according to Levy. Referencing Hsiang et al.’s study, he noted that “there are very few other things that you could measure in a large-end statistical global time series test than inter-annual variability and civil war.” And, importantly, climate change will alter the conditions that the study focused on. “By focusing on variability we know what happens to societies when you get variations around a mean, but we have almost no basis for figuring out what happens when the mean changes,” he said.

“I think we need to firm up our knowledge base by looking more explicitly at how these things operate in high-risk countries. And perhaps start thinking about some customized approaches that might be relevant in high conflict risk countries that wouldn’t necessarily be on the radar outside of those countries,” Levy concluded.

Drafted by Theresa Polk and edited by Schuyler Null and Geoff Dabelko.


Thank you for your interest in this event. Please send any feedback or questions to our Events staff.