High Temps, High Tempers? Weather- Related Natural Disasters & Civil Conflict

The issue of climate change and security is hotly debated in many arenas. Most systematic studies to date have examined the association between natural disasters and civil wars. The results are mixed because there is some confusion due to the mixing of non-climate related disasters, such as earthquakes and volcanoes and extreme weather events, and what type of violence these events may spawn. This study will focus exclusively on weather events, the most relevant disasters for judging the impact of climate change, on civil war at a lower threshold of civil war for a long period of time (1950 to present). Using multivariate methods, this study use data on extreme weather events affecting population centers and use several variables gleaned from theory to test how extreme weather events predict violence within these populations. The behavior of people under these conditions in the past and in recent times will allow us to assess how things might turn out given the changes to weather patterns expected from climate change. This study will be the first to address this issue both theoretically and empirically using longitudinal data in systematic ways. We find that the measured effect of climate-related disasters on conflict risk is highly dependent on model specification, but that when using a well established statistical model of civil conflict, the disasters appear to prevent rather than promote conflict