???ViewItemPage???

???ViewItemOverview_lblAbstractSection???

Climate change impacts are already evident and projected to worsen throughout the 21st century, even with mitigation efforts. Systematic mapping is key to organizing scientific evidence and identifying gaps, but current methods lack geographical context in relation to climate impact risk. In this study, we leverage machine learning to scale up systematic mapping and use automatic geolocation to track place-based research. We then enhance conventional systematic mapping by integrating location-based climate risk components—hazard, exposure, and vulnerability—to create an evidence gap index. This identifies high-risk regions that lack sufficient scientific study. We demonstrate this method using fluvial floods, combining research distribution with a flood-risk indicator (hazard), population density (exposure), and the Human Development Index (vulnerability). Our novel approach refines evidence mapping, supporting data-driven policymaking and directing research resources to the most urgent areas.