Is Climate Change Likely to Increase Conflict?

Likely not, according to Nils Petter Gledditsch, in the introduction to a special issue of the Journal of Peace Research on this topic. Research to date shows little evidence for systematic relationship between increased global warming, water shortages etc and violent conflict.

Climate change is the world’s first truly global manmade environmental problem and a firm warning that human activities can influence our physical environment on a global scale. The range of possible consequences of climate change is so wide, even for the limited temperature changes foreseen in the IPCC scenarios, that it is difficult to sort out the main priorities. Obviously, if a reversal of the trend towards a more peaceful world was one of these consequences, it should have a prominent place on the policy agenda. Based on the research reported here, such a pessimistic view may not be warranted in the short to medium run.

In other words – there are a large number of unhappy consequences that may flow from continued global warming. But what evidence we have to date (which is admittedly based either on spotty data, or studies over the shorter term) gives us no very strong reason to argue that an increase in violent conflict is among them.

6 Responses to Is Climate Change Likely to Increase Conflict?

  1. Andrew Gelman February 3, 2012 at 2:32 pm #

    We discussed this a couple years ago in the context of claims by Marshall Burke, Edward Miguel, Shanker Satyanath, John Dykema, and David Lobell. As I wrote at the time, they have a bunch of tables, but what I really want to see are the time series and the scatterplot. . . . Burke et al. include several alternative specifications of their model, and I believe they believe their result is robust–but it’s hard for me to know how to think about their results without seeing their data.

  2. Solomon Hsiang February 3, 2012 at 5:14 pm #

    It’s worth noting that Gleditsch writes

    “With a single exception (De Stefano et al.) the articles in this special issue are based on papers or presentations at the international conference on ‘Climate Change and Security’, held in Trondheim, Norway, 21–24 June 2010″

    so his assessment is not entirely up to date.  In 2011, Kyle Meng, Mark Cane and I published this paper in Nature (unpaywalled here; supplement here; decoded in Nature, Science, NPR, Slate and the Economist; video and other materials here) demonstrating a direct association between the global climate and conflict.  Our identification of a causal effect and the robustness of our finding is probably tighter than most, if not all, of these older studies (please read the 50 page supplement if you’re skeptical).  In his introduction to these older papers, Gleditsch suggests that issues have been raised about our findings, but he provides neither a citation nor evidence, so I am not sure what his specific concern is.

    • Solomon Hsiang February 10, 2012 at 3:59 pm #

      Following Nils’ advice, I’m reading the issue. I haven’t gone through the entire thing, but I’ve tried to reconcile the first empirical paper I got to with our 2011 Nature paper here. Our original findings seem to hold up when I merge the two data sets, although the new JPR findings are somewhat more sensitive.

  3. Nils Petter Gleditsch February 9, 2012 at 8:02 am #

    Of course, all the articles in the special issue have been extensively revised or even completely rewritten since the 2010 conference. Moreover, in my introduction to the issue I review several other recent contributions, including that by Hsiang et al. (2011) and acknowledge that CLIMATE may well be related to conflict. I am sure their pioneering article will spawn a secondary literature in due course. But it is not obvious that a possible relationship between El Niño and civil conflict sheds any light on the security implications of CLIMATE CHANGE in the sense discussed by the IPCC. – But read and judge for yourself! All the articles in the special issue is available free until the end of the month from And replication data for all the articles can be found at

    • Mark Cane February 10, 2012 at 12:43 pm #

      First, congratulations on your excellent special issue.

      It is a truism that it is impossible to establish an unimpeachable link between conflict and anthropogenic climate change (ACC; I believe this what Dr. Gleditsch means by “CLIMATE CHANGE in the sense discussed by the IPCC”). Our understanding of conflict is not so advanced as to allow arguments from first principles, so we rely on analysis of past events, either by quantitative analysis or more subjective methods. As discussed, for example, by Hsiang et al, attempts to relate climate and conflict over the long time scales comparable to that for ACC are inevitably confounded by secular social changes. A claim to link conflict and climate on shorter timescales (e.g. the interannual variability of the El Niño-Southern Oscillation (ENSO) cycle) may be dismissed as proving nothing about the longer changes due to ACC. So we either wait until the passage of time and increase in greenhouse gases tells us the answer, or think it through as best we can.

      If “CLIMATE CHANGE in the sense discussed by the IPCC” means a set of gradual (longer than a decade) changes in a set of climate variables, then I doubt that this sort of climate change will have much of an impact on conflict. I think it more likely that we humans will respond much as the proverbial frogs in water brought to a boil; this seems an apt description of our collective response to the ACC already evident. In tacit recognition, much of the recent climate change literature has focused on extreme events such as the 2010 summer heat wave in Russia. While a full account of the impact of ACC on extreme events is complex and has many uncertainties, the leading effect, a shift in the mean, is straightforward. For example, suppose that for people at Port Puerto a storm surge of 1m is an extreme event. If mean sea level rises by 0.5m in a warming world, it now takes only a 0.5m surge to make an extreme event. Since the distribution of storms is unlikely to change greatly, extreme events become far more likely. (See Hansen et al for an accessible account, and Ramstorf and Coumou 2011 for a more comprehensive mathematical treatment.)

      If unusually high temperatures (whatever the cause, including El Niño, as in Hsiang et al) induce more conflict, one would expect anthropogenic warming to worsen the impact. One might object that rainfall is the key climate variable, but one of the more robust IPCC model results, amply supported by theory (Allen and Ingram, Nature 2002; Held and Soden, J. Climate, 2006) is that “the wet places get wetter and the dry places get dryer.” Droughts (e.g. due to ENSO) tend to occur in places where rainfall is marginal, so ACC will tend to make them more frequent.

      There is no sound reason to doubt that the natural climate variations evident in past and present climate will continue as ACC proceeds, and that the climate we will experience will be the combined effect of natural and anthropogenic effects. Hence extremes associated with El Niño are likely to increase, and since Hsiang et al find more conflicts with larger events, it is reasonable to expect the incidence of ENSO related conflicts to increase. Reasonable, but not entirely proven, since it may be that too many extreme events events will provoke a different response.

  4. Idean Salehyan February 14, 2012 at 1:27 pm #

    One important point that emerges from several contributions to the special issue is that better environmental conditions (more rain, foliage, etc) lead to more violence. Hendrix and Salehyan, Theisen, and Adano et al, all show this using different data and units of analysis. Here is an apt quote from Sun Tzu’s Art of War:

    “Make forays in fertile country in order to supply your army with food.”

    Put simply, it is hard to sustain violence when fighters are hungry. This is an argument that is often ignored in the scarcity and conflict literature.