One in five major civil conflicts since 1950 may be linked to climate extremes associated with El Ni±os — periods of warming lasting a year or longer in surface waters of the central equatorial Pacific, a new study finds.
Solomon Hsiang of Princeton University and his coauthors at Columbia University emphasize that they don’t know what mechanism might link El Niños to eruptions of major civil unrest, which they define as disputes between governments and other organized parties that claim at least 25 lives. But the researchers point out that droughts, torrential rains and other weather extremes that tend to develop during El Niño years can devastate crop yields, leading to higher food prices and unemployment in affected nations — home to half of the world’s population.
“I’m one of those people who would be generally skeptical about correlating things to climate,” says statistician Andrew Solow of the Woods Hole Oceanographic Institution in Massachusetts. But the new report’s finding makes sense, he says: In poor countries where the economy is closely linked to agriculture — and therefore the weather — poor harvests and diminishing food supplies can leave large numbers of people available to engage in civil uprisings.
Hsiang and his colleagues analyzed 234 civil conflicts that broke out within 175 nations between 1950 and 2004. In any given year, the probability that a new conflict would erupt among the 90 or so nations whose climates can be heavily affected by El Ni±o events was 4.1 percent. That’s twice the conflict rate in countries largely immune to El Niño effects, Hsiang’s group reports in the Aug. 25 Nature.
About every seven years, climates in tropical regions swing between conditions dominated by an El Niño and those moderated by a La Niña (cooling in equatorial Pacific waters). During El Niño years, the likelihood that a new civil conflict would erupt in equatorial nations was roughly 6 percent, or twice that for La Niña periods.
This disparity in violence rates is huge, Hsiang says: It’s comparable to differences that others have reported between populations where annual incomes were $1,000 per capita versus $10,000.
In Afghanistan, where Neil Johnson of the University of Miami conducts studies, “we all know there’s a fighting season, where weather affects conflict.” But he worries that the authors’ aggregation of El Niño–driven changes across the globe “could be masking all sorts of subtleties [in local environmental conditions] that would actually change their conclusions.”
Yaneer Bar-Yam of the New England Complex Systems Institute in Cambridge, Mass., argues that the new study is a good example “of the progress that has been made in understanding how we can predict social behavior writ large.” The correlation doesn’t prove that El Niño weather disruptions cause civil unrest, he acknowledges. “But a properly used and understood correlation is an important tool for understanding interdependencies in complex systems — and the world around us.”