What climate? The different meaning of climate indicators in violent conflict studies

Abstract This paper explores the operationalization of climate-related indicators in violent conflict research. The climate-conflict narrative gained traction in recent decades and climate change is often referred to as a ‘threat multiplier’ by both policy makers and scholars. Yet, the relationships between climate-related phenomena and violent conflict are complex and context-specific. However, limited attention has been given to the climatic indicators applied in climate-conflict research. This paper addresses that gap by analyzing 32 studies published from 2004 to 2020 on the operationalization of climatic indicators and their relationship with violent conflict. It first categorizes climate indicators operationalization into five clusters: natural disasters, basic climate variability, advanced climate variability, freshwater availability, and the ENSO. The study evaluates the climate indicators for each cluster and shows that at an aggregate level these clusters examine 68 different climate representations. When paired with their respective conflict types, it finds a total of 113 climate-conflict combinations. Most operationalizations represent various forms of climate-related phenomena and variability rather than climate change. Some indicators are advancements over time, for example moving from changes in average rainfall to standardized precipitation indices. However, other indicators refer to various natural processes, making it challenging to determine whether climatic variability impacts conflict. The paper then demonstrates a discrepancy between the pathways through which climate may affect violent conflict and the representation of these pathways in the selected climate indicators. It discusses how the selection and operationalization of climate indicators requires careful consideration, and the phenomena researched should be well-specified in research findings.