MELISSA BLOCK, host: We're going to take a closer look now at a possible relationship between the climate and conflict. NPR's Jon Hamilton reports that scientists have found statistical evidence of a link between climate and political violence.
JON HAMILTON: The researchers studied civil conflicts and the climate pattern known as El Nino. Solomon Hsiang, of Princeton University, says El Nino starts with unusually warm water in the Pacific Ocean near the equator. But pretty soon, he says, it affects weather on land thousands of miles away.
SOLOMON HSIANG: It's not just that an individual country has slightly different weather. It's that all countries around the entire tropical belt, okay - half the world's population - experiences a completely different climate regime.
HAMILTON: In most places, El Nino means the weather gets warmer and drier, often for years. The opposite occurs with La Nina conditions, which occur when waters in the Pacific become unusually cool.
Hsiang and his colleagues did a statistical analysis to see whether El Nino or La Nina conditions were associated with civil conflict. They define civil conflict as a new dispute between the government and an organized group that results in at least 25 battle-related deaths. And Hsiang says the analysis produced a striking result.
HSIANG: When the global climate is in its relatively cooler and wetter, a La Nina state, the rate of conflict is only three percent. So out of 100 countries in any given year, only three countries would be expected to begin a new civil war. However, when the global climate shifts into its relatively hotter and drier El Nino state, the rate of conflict jumps - it actually doubles all the way up to six percent.
HAMILTON: But Hsiang says not all countries are affected.
HSIANG: The countries that are most sensitive to El Nino are the poorer countries. Wealthier countries that are in the tropical belt - so for example, Brazil or Australia, don't exhibit civil conflicts during El Nino events.
HAMILTON: There's also no effect in countries such as Libya, which are outside the tropical belt.
Andy Solow, a statistician at the Woods Hole Oceanographic Institution in Massachusetts, says Hsiang and his team have made the strongest case yet that there is a link between climate and conflict. And Solow can think of some possible reasons that the two are related.
Dr. ANDY SOLOW: People don't just go to war because the weather changes. The effect of weather on human behavior is presumably operating through resource scarcity or food scarcity or something like that.
HAMILTON: That could have been the case before both the French Revolution and the Rwandan genocide. In both cases, hot, dry weather had caused food shortages.
And Solow says El Nino is well-known to reduce crop yields in tropical countries.
SOLOW: A lot of these countries are poor and mainly agricultural. And as climate conditions change, that can put stress on the agricultural system in those countries; also possibly on water resources and other resources like that and that that may lead to conflict.
HAMILTON: Solow says researchers still need to figure out if that's really what's going on with El Nino. But Solomon Hsiang says that because the El Nino conditions can be forecast, it may already be possible to use his study to help anticipate conflicts. He says that doesn't mean other countries should necessarily intervene, though.
HSIANG: Some conflicts are important. People engage in conflicts because they're things they really care about. So it's not up to us to say that the conflicts per se should be stopped. But there are a lot of very innocent bystanders who often suffer as a result of conflicts. And so, it's those people we're really hoping to support and help out.
HAMILTON: For instance, he says, aid groups might have been able to anticipate the current problems in Somalia if they had factored in El Nino. Hsiang's paper appears in the journal Nature.
Jon Hamilton, NPR News.