Climate change is set to disrupt weather patterns in many parts of the world. Many places will become hotter on average, some will see more rainfall than they used to, others less. But changes to average temperatures and precipitation are not the only thing we should worry about. Recent evidence shows that, over the past century, wet seasons have become progressively wetter and dry seasons drier. Atmospheric models suggest that this shift in seasonal rainfall patterns will intensify in the future.
This shift will create enormous challenges for agriculture. Many tropical regions, for instance, have two growing seasons per year: a dry season and a wet season. In the dry season, crop yields are often limited by insufficient rainfall; in the wet season, yields are limited by an excess of rainfall that leads to waterlogging and flooding. Drier dry seasons and wetter wet seasons – what climate scientists call an increase in the “annual range of precipitation” – can have devastating effects on yields even if annual rainfall totals stay the same.
In a recent working paper, my co-authors and I show evidence for negative effects of an increased annual range of precipitation in the Philippines, a country with two distinct growing seasons: a dry season from November to April and a wet season from May to October. We show that below-average rainfall in the dry season leads to reduced production of rice and corn, the country’s most important staple crops. In the wet season, however, the effect of rainfall points in the opposite direction and above-average rainfall leads to poor harvests.
We then examine the effect of seasonal rainfall on civil conflict between the government of the Philippines and several armed groups operating in the country, including the communist New People’s Army and the Muslim-separatist Moro-Islamic Liberation Front. Conflict with both of these groups has been ongoing since the 1960s and has claimed hundreds of thousands of lives and displaced millions from their homes.
We find that the effect of rainfall on conflict also has a pronounced seasonal pattern, consistent with its effect on agriculture: below-average dry season rainfall leads to increased conflict in the following year, as does above-average wet season rainfall. This increase in violence is almost exclusively due to insurgent-initiated violence against civilians and government forces, perhaps because bad harvests make it easier for insurgents to recruit fighters to carry out attacks.
Our results are important for two reasons. First, they show that a shift toward drier dry seasons and wetter wet seasons can harm agriculture and exacerbate conflict even if annual rainfall totals remain stable. Analyses that do not account for changes in the seasonal distribution of rainfall are therefore likely to underestimate the true effect of climate change on conflict. Second, our results indicate that changing rainfall patterns affect conflict at least partly through their effect on agriculture.
This has potentially important implications for adaptive policies. Even under the most optimistic scenarios, substantial changes to the world’s climate appear to be unavoidable. Designing policies that can increase resilience to these changes will become increasingly important. Our results suggest that policies that mitigate the effect of climate change on agriculture could have the added benefit of reducing civil conflict.