Weather, climate, war

Print Friendly
Cycles of war = cycles of weather?

El Niños, the global cycles of weather that are driven by a hot spot in the tropical Pacific Ocean, have been linked to drought, storms and famine in many parts of the tropics.


Dozens of people standing in rain outside long wooden buildings, child in oversized coat standing in foreground

Photo: 2007, Julien Harneis
The Democratic Republic of Congo, in the el Niño “hot zone,” has been battered by years of conflict. Hundreds of people who fled their village to escape attacks by militia and government forces found a haven in this school.

Today, a study in Nature finds that deadly conflicts have started twice as often during the el Niño years – but only in the many countries affected by el Niño.

Scientific interest in el Niño mushroomed during the 1980s, when climate experts began to correlate historic cycles of anchovy harvests along the west coast of South America with changes in weather thousands of kilometers distant, and eventually unraveled a planetary cycle driven by the appearance of huge pools of warm water in the western Pacific.

Because the warming seemed to coincide with Christmas, it was called el Niño, for the Christ Child.

El Niño is now recognized as the warm-water segment of the el Niño Southern Oscillation (ENSO), which also includes a cold-water counterpart called la Nina. Now acknowledged as an engine of global climate, el Niño is linked to prolonged droughts, heat waves and crop failures.

Previous efforts to study whether weather and global warming could affect war have related past environmental changes with conflict and the decline of civilizations, says Solomon Hsiang, who completed the new study as a graduate student at the Lamont Doherty Earth Observatory at Columbia University. But the studies tended to be case-by-case, he notes, and “even if every conflict or collapse happened at random, some would occur during a period of environmental change, so this isn’t compelling evidence.”


Central America, northern half of South America, most of Africa, Southeast Asia, and Australia in red

Courtesy Hsiang et al, 2011
Countries where the weather is strongly affected by el Niño are red.

Looking carefully

To study the issue more systematically, Hsiang and collaborators Mark Cane and Kyle Meng:

Classified nations according to whether their weather responds to el Niño

Culled records from the Peace Research Institute (Oslo, Norway) on the start of 234 civil or intrastate conflicts that killed at least 25 people between 1950 and 2004

Compared the incidence of conflict among the two groups of countries when el Niño was active or inactive

The data showed that conflicts are twice as likely to start during an el Niño, says Hsiang, and that 21 percent of overall conflicts can be attributed to el Niño. The increase was only seen in countries strongly affected by el Niño.

Surprisingly, the average changes wrought by an el Niño are quite minor, Hsiang admits – about 0.05°C rise in temperature, and about 0.1 millimeter reduction in daily rainfall.

Small is … powerful?

How could such minor changes affect warfare?

A study that correlates data does not show why they are related, but there are many ways that seemingly small effects could change human behavior, says Hsiang, who is now at Princeton University, especially considering that averages can conceal major alterations in different locations:

Laboratory studies show that people become more aggressive in hotter conditions.

Economics matters: Staging a rebellion requires a rebel army, which could be too expensive when times are lean. Alternatively, as Hsiang notes, “when it’s harder to find a job, it’s more attractive to work in the local militia.” and

Small weather changes may boost global food prices, causing starvation and increasing dissatisfaction in poor countries. “El Niño may not induce conflict by influencing the local situation,” says Hsiang, but rather by an indirect effects on climate, food supply, refugee flows or politics.


Two men carrying large pole on their shoulders, hammock with wrapped body of victim tied to pole

A victim of El Salvador’s long civil war (1980 – 1992) is returned to his village for burial in 1982

However, Marshall Burke, who published an influential 2009 paper 1 that found a significant increase in warfare during hot weather in sub-Saharan Africa, noted by email that the increase in conflict was seen only inside the el Niño region, and thus, “We might conclude that these global market mechanisms are not at work.”

Still, the new study adds something to the discussion, Burke says. “The [Hsiang] paper’s main innovation is in linking historical changes in the global climate to conflict risk, whereas past studies (including ours in PNAS) looked only at the effect of local weather variations on conflict.”

Burke, a Ph.D. candidate at UC Berkeley department of agricultural and resources economics, added, “They provide very convincing evidence that ENSO-related changes in the global climate are strong drivers of conflict risk in the regions whose weather is affected by ENSO.”

Looking at limits

Mark Cane, a climate scientist at Columbia’s Lamont-Doherty Earth Observatory and a co-author of the new study, said weather does not equal destiny. “No one should take this to say that climate is our fate. Rather, this is compelling evidence that it has a measurable influence on how much people fight overall.”

Strength of el Niño and la Niña

 el Niño has highest peaks at 1983 and 1997, longest period between 1990 and 1995

Graph: NOAA
This graph summarizes the el Niño Southern Oscillation, according to air pressure and temperature, wind, sea surface temperature, and cloudiness.

Ultimately, the motivation for the new study was to peer through the keyhole of time and anticipate a warmed world, Hsiang says, but he admits that the predictive power is limited. “In relationship to global warming, we want to be careful. El Niño is very different … in terms of its spatial pattern, the changes on the ground, and the rate of change. Until we have a much better grasp of these, it’s very hard to take these results and produce any kind of projection for future climate change.”

Still, he adds, “The debate until now has been whether there is any reason to believe that a shift in climate can produce conflict.” Now, “The question is not whether it’s possible, but how much global climate will influence conflict.”

– David J. Tenenbaum


Terry Devitt, editor; S.V. Medaris, designer/illustrator; Jenny Seifert, project assistant; David J. Tenenbaum, feature writer; Amy Toburen, content development executive


  1. Burke, M., Miguel, E., Satyanath, S., Dykema, J. & Lobell, D. Warming increases risk of civil war in Africa. Proc. Natl Acad. Sci. USA 106, 20670–20674 (2009).
  2. Civil conflicts are associated with the global climate, Solomon M. Hsiang et al, Nature, 25 August 2011.
  3. Radio: study’s author speaks.
  4. Weather and war: Scientific American.
  5. El Niño at NOAA.
  6. El Niño effects in 1997-1998.
  7. Peace Research Institute.
  8. Climate change and conflict.
  9. More climate change and conflict.
  10. Darfur conflict and climate.