Authors: Erin Bunting, Michigan State University, James Herdon, University of Alabama, Elizabeth Mack, Michigan State Universiy, Amanda Ross*, University of Alabama
Topics: Africa, Water Resources and Hydrology, Economic Geography
Keywords: conflict, rainfall, climate change, Markov analysis, logit model
Session Type: Paper
Presentation File: No File Uploaded
Climate change and its impact on people is a growing area of research in light of projected changes in climate and the environment. One important impact that may occur as temperatures and rainfall change is that there may be changes in the frequency and likelihood of conflict, both violent and non-violent. Overall, the research is mixed with regards to the impact of changes in temperature and rainfall on conflict. To examine the effects of changes in climate on conflict, this paper uses Markov chain analysis to estimate the probability of regions in Africa converting between non-conflict and conflict status. In particular, we are interested in whether regions move from a no-conflict state to a conflict state in the subsequent year or whether regions transition out of a conflict state to a no- conflict state. We examine both violent and non-violent conflict, as it is likely that the drivers of these two different types of conflict are different. These results are then incorporated into panel logit models to assess whether climate change operationalized as reduced precipitation and increased surface temperature are associated with changes in conflict status controlling for a variety of regional factors. Our results will be important to policy makers in the future, as they will provide insights into not just what may drive conflict, but also what can help reduce conflict.
To access contact information login