"Our paper provides the first strong evidence that climate change could greatly increase the risk of civil conflict in Africa," Marshall Burke of the University of California Berkeley and Stanford University told environmentalresearchweb. "There had been lots of speculation…but to our mind little hard evidence. Given the humanitarian costs of past conflict, resolving this more quantitatively seemed important."
Burke and colleagues at the University of California Berkeley, New York University, Harvard University and Stanford University examined the link between temperature and incidence of conflict for the period from 1981 to 2002. This revealed a strong correlation between the two, with a 1 °C increase in temperature corresponding to a 4.5% increase in civil war in the same year or a 0.9% increase in conflict incidence the following year. This is equivalent to a 49% relative increase in the incidence of civil war.
The team then combined this result with temperature projections for Africa to 2030 from 20 general circulation models from the World Climate Research Program's Coupled Model Intercomparison Project phase 3 (WCRP CMIP3). The average temperature increase projected by the models was around one degree, with a range of 0.7 to 1.6 degrees.
For the continent as a whole this study indicates a 54% rise in the average likelihood of conflict by 2030. Even if economic conditions improve and democracy becomes more widespread, this would only reduce the relative risk of conflict by around 20% – the climate-change effect would dominate.
Higher temperatures affect crops both by raising evapotranspiration, which increases water stress, and by accelerating crop development. The result can be a drop in yield of as much as 30% per degree of warming. The poorest rural African households typically rely on agriculture for 60–100% of their income. That means climate change is likely to have a serious effect on their economic welfare – a factor linked to conflict.
"Our study suggests that, in the absence of any intervention, the humanitarian costs of conflict could be much higher than previously thought, and that these impacts could be realised in as few as two decades," said Burke. "Because climate appears linked to conflict mainly through agricultural productivity, our results provide strong impetus for investment in agricultural adaptation to climate change."
Burke says that this could include increased investment in the development of crop varieties better adapted to hot temperatures, or in crop-insurance programmes to help farmers to buffer their income when the climate takes a turn for the worse.
That said, issues such as temperature-related increases in violent crime and drops in productivity of non-farm workers could also be playing a role in the link between temperature rise and conflict. The researchers say that examining the relative contributions of these factors is a critical area for future research.
Because of Africa's dependence on rain-fed agriculture, previous studies had focused on the link between conflict in the region and precipitation, rather than temperature. Sure enough, drier years were correlated with past conflict. But climate projections for African precipitation contain a lot of uncertainty; it's not even clear whether rainfall will increase or decrease in some regions, which in turn, made predicting future effects on conflict difficult.
The researchers reported their work in PNAS.