Climate Change and Conflict: Is There Any Correlation?

Will climate change lead to a future with more war? One popular hypothesis among advocates for action to curb climate change is that, if climate change creates or exacerbates resource scarcity for food and water (e.g., via crop failure and low rainfall), and resource scarcity creates conflict, then climate change could lead to increased conflict. But what is the evidence that past changes in climate have created resource scarcity and that such scarcity contributed to war? And how does the resource scarcity factor compare in importance to political drivers of war?

The question is complicated. For example, you have to specify what kind of war/ conflict you are talking about (e.g., interstate war, civil war, or inter-communal violence) and use consistent criteria for what counts as war (e.g., thresholds for death counts and criteria to exclude one-sided massacres and pogroms). But there is a growing body of literature on this topic, and the surprising answer appears to be that the effect of past climate change on conflict ranges from undetectable to small, and even when present sometimes goes in the opposite direction than you might predict.

Let’s look at some specific examples from the literature. Although some research argued that increasing temperatures are correlated with increased war in Africa (Burke, 2009), this assertion is contradicted by subsequent analysis and by recent events (Buhaug, 2010). For example, although the first decade of this millennium was one of the hottest on record for Africa, average annual battle deaths were down 38-68% from any of the four previous decades (through 2008, the last year for which battle deaths data are available; see Figure 1). Clearly, whatever factors led to declines in violence are much stronger than the effects of increased temperature.

A recent special issue in The Journal of Peace Research provides more detailed case studies of the relationship between climate and conflict. Researchers are turning toward lower-intensity conflict to test for the effects of climate change on violent conflict. As they define it, low-intensity conflict includes riots, protests and inter-communal conflict, such as between farmers and herders. Such conflict is more frequent, allowing a larger sample size and more rigorous statistics. Such conflict is also potentially more easily triggered by climate since, for example, subsistence farmers and herders are among the most exposed to the effects of climate change.

But one recent study of low-intensity conflict in Africa found that it tended to be higher in wetter years (Hendrix and Salehyan, 2012). The causal mechanism is unclear, but these results suggest it is not as simple as climate change -> resource scarcity -> increased conflict.

Of course, lack of past correlation doesn’t mean that there won’t be problems in the future. Rapid climate change could bolt past thresholds of resilience, threatening food security and triggering natural disasters that spur an unsustainable wave of migration from rural to urban areas.

However, political factors are still likely to determine the prevalence of war. The surprising news here is that war deaths (along with pretty much every other form of violence) have been decreasing globally since the end of World War II. But don’t take my word for it. The data are exhaustively compiled in Steven Pinker’s 2011 book The Better Angels of Our Nature: Why Violence Has Declined (Pinker, 2011).

The forces that Pinker has identified as likely causes for the declines in war will still be in effect as the climate changes. Democracies tend to stick around once they are created, and are much less likely to be involved in interstate wars than are other forms of government. Weak democracies are still prone to civil war, but several factors have reduced deaths from civil wars. Peacekeepers have proven to be effective in reducing the probability of civil war. Increasing economic co-dependency and the economic benefits of trade, which are disrupted during civil wars, can provide incentives for citizens to avoid conflict. And one of the biggest predictors of civil war is past civil war. As more countries gain distance from their conflicted pasts, the likelihood that old rivalries will resurface decreases. Even the influx of rural residents to urban areas should ultimately decrease inter-communal conflict, as the diversity and proximity of cultures in the urban milieu promotes increased understanding and tolerance among ethnic groups. In other words, climate change is likely to make a lot people suffer, but not necessarily by increasing war.

So what does all this mean for conservation? The conservation movement recognizes that it is important to get climate change back in the public conversation, following the build-up and crash of media coverage around Copenhagen. Highlighting the linkages between climate change and some of its potential adverse impacts on people – such as food production, natural disasters, and national security – could help renew the conversation. And although the scientific literature to date has not documented a general relationship between climate variability and war deaths, there are still valid national security reasons for concern about climate change. For example, the Department of Defense’s 2010 Quadrennial Defense Review found that “While climate change alone does not cause conflict, it may act as an accelerant of instability or conflict, placing a burden to respond on civilian institutions and militaries around the world.”

The take-home message is that appropriate nuance is required. Based on the existing scientific literature, past climate change is not well correlated with battle deaths. Of course, national security is about more than just the simplistic, if gruesome, metric of death tolls – our military is also asked to respond to natural disasters, refugee situations and political instability – all issues that national security experts have concluded could be made worse by the additional stressors resulting from a rapidly changing climate. Accurate statements about the increased risks that future climate change poses for instability, particularly to already fragile governments, will recognize this distinction.


Burke, M.B., et al. 2009. Warming increases the risk of civil war in Africa. PNAS 106 (49): 20670-20674.

Buhaug, H. 2010. Climate not to blame for African civil wars. PNAS 107(38): 16477-16482.

Hendrix, C.S. and I. Salehyan. 2012. Climate change, rainfall, and social conflict in Africa. Journal of Peace Research 49(1): 35-50.

Pinker, S. 2011. The better angels of our nature: Why violence has declined? New York, New York: Viking.

(Image 1: Battleground. Source: JD Hancock/Flickr through a Creative Commons license. Image 2: Data from Climate Wizard and PRIO Battle Deaths Dataset.)

If you believe in the work we’re doing, please lend a hand.


  1. The article questions the rubric [climate change -> resource scarcity -> conflict] with evidence showing higher temperatures do not correlate with increased battle deaths. Does . . . not . . . compute. Or to be blunter, absolute nonsense.

    Why would anyone expect the actual temperature at a given time to correlate with resource-driven conflict? For example, droughts in the U.S., Australia, and elsewhere drove up food prices severely in 2008 and 2011. Result? Food riots, and bloody revolutions throughout the Middle East.

    Tracking the temperature during the actual riots tells you nothing about their cause. Tracing riots to food prices to droughts a year or more before in other countries – now that makes a bit more sense, doesn’t it? A little more complicated, yes. But also more accurate. And really not that difficult a concept for people to grasp.

    Of course it’s idiotic to tell people that temperature increase by itself will lead to increased conflict. But it’s just as wrong to claim that making the connection between climate change and conflict is “treading on thin ice.” We’ve just seen revolutions sweep across a large portion of the globe spurred in large part by rising food prices caused in large part by drought. But if the temperature in Tahrir Square in early 2011 was lower than the average, I guess we can’t make any connection with climate change.

  2. Joe,
    I love love love this article…I felt refreshed reading your examinations of human security in terms of natural resources and ‘emotional climates’ and our desperate need for sound science.
    One way I am inspired to hope in our collective state of being is through this new platform,
    It seems to me the most democratic and transparent way to navigate toward global scientific solutions. It is still developing, yet I am moved to suggest you circulate this as a viable course of action. It exercises our democratic wisdom to fund and influence the direction of scientific and medical research through peer process
    Please, take ownership of our future unfolding. Review and Recommend; Visit and share; and just in general- pass it on, Please…

    Biologist, parent, mediator

  3. in reply to nnokxs:

    I brought up the issue of whether temperature correlates with violence in response to a paper that claimed to find exactly this correlation [1]. I agree this seems like a poor way to test the hypothesis that climate change will cause conflict. So we agree on the point that it is not nearly that simple.

    I have a hard time viewing the Arab Spring as an example of the negative impacts of climate change, since the ouster of dictators with relatively few fatalities in Tunisia, Egypt, and Yemen seems like a positive outcome. But I see your point that drought and increased food prices appear to have contributed to political instability. Many researchers see this as having more to do with ‘poverty and dysfunctional institutions’ than with resource scarcity, such as poor crop harvests, per se [2].

    I agree that increasing food security is a great reason to fight climate change. But it is not clear which of two alternative hypotheses will hold true: 1) food price volatility will increase the amount of violent conflict in the world, or 2) poverty alleviation and improved governance will continue to decrease the amount of violent conflict in the world, in spite of any increases in food price volatility. If the latter hold true, then using security concerns as justification for fighting climate change is treading on thin ice. Although it seems obvious to test whether the probability of uprisings and revolutions is higher following spikes in food prices (and whether this probability has changed over time), I am not aware of any research that tests this hypothesis. But this is an active area of research, so perhaps we will soon have additional empirical evidence that better allows us to compare these alternative hypotheses.

    1. Burke, M.B., et al. 2009. Warming increases the risk of civil war in Africa. PNAS 106 (49): 20670-20674.

    2. Theisen, O.M. 2008. Blood and Soil? Resource Scarcity and Internal Armed Conflict Revisited. JOURNAL OF PEACE RESEARCH (45): 801-818.

  4. One reason scarcity might lead to less conflict, instead of more, is that armies have to be fed. When people are hungry, they are not such good fighters. I was just reading something about 17th century Irish rebellions against English plantations and was surprised to read how a famine pretty much put a quick end to it, whereas I would have thought a famine would have intensified rebellion against those stealing your land. And of course there was some violence in the 19th century potato blight famines, but remarkably little considering the death and displacement involved. And much of the violence that did occur was in better off areas where better off people attacked the economic refugees. Not to mention the additional pressures on Native American peoples displaced in part by the flood of Irish immigration to America from a natural catastrophe halfway around the world.
    So maybe if you are looking for violence triggered by scarcity, maybe you should look at aggression against the victims at least as much as you look for aggression coming from the victimized regions or classes.
    And also, if you see a true decline in violence, it may just be that modern, more urban societies are better suited to suppressing violence as a viable strategy. In today’s world of interdependent food economies, violence is certainly not going to bring more or cheaper food to your country or region, or gain you the land, water and farming skills most people lack in order to provide food self-sufficiency.

Add a Comment