Ignoring all of the likely reasons why Africa may see more civil wars in the future, a study published this week in the Proceedings of the National Academy of Sciences suggests that global warming could increase the likelihood of civil war in sub-Saharan Africa by over 50 percent within the next two decades.

The study, conducted by researchers at UC Berkeley as well as at Stanford University, New York University and Harvard University, provides the first "quantitative" evidence linking climate change and the risk of civil conflict, the authors claim. They conclude by urging accelerated support by African governments and foreign aid donors for new and/or expanded policies to assist with African adaptation to climate change.

In the study, the researchers first combined historical data on civil wars in sub-Saharan Africa with rainfall and temperature records across the continent. They found that between 1980 and 2002, civil wars were significantly more likely in warmer-than-average years, with a 1 degree Celsius increase in temperature in a given year raising the incidence of conflict across the continent by nearly 50 percent.

Building on this historical relationship between temperature and conflict, the researchers then used projections of future temperature and precipitation change to quantify future changes in the likelihood of African civil war. Based on climate projections from 20 global climate models, the researchers found that the incidence of African civil war could increase 55 percent by 2030, resulting in an additional 390,000 battle deaths if future wars are as deadly as recent ones.

All climate models project rising temperatures in coming decades, said David Lobell, study co-author and an assistant professor of environmental earth systems science at Stanford. "On average, the models suggest that temperatures over the African continent will increase by a little over 1 degree Celsius by 2030," he added. "Given the strong historical relationship between temperature rise and conflict, this expected future rise in temperature is enough to cause big increases in the likelihood of conflict."

To confirm that this projection was not the result of large effects in just a few countries or due to overreliance on a particular climate model, the researchers recalculated future conflict projections using alternate data. "No matter what we tried - different historical climate data, different climate model projections, different subsets of the conflict data - we still found the same basic result," said Lobell.

Although, little thought was apparently given to the idea that correlating temperature records and the incidence of civil war is hardly enough to prove that one causes the other. The possibility that their chosen climate models are seriously flawed also seems to have eluded the authors.

Applying findings from this study could prove useful to policy makers at the upcoming Copenhagen negotiations in December in determining both the speed and magnitude of response to climate change, the authors said.

Citation: This study will appear later this week in the Proceedings of the National Academy of Sciences