Coastal communities face growing flood risk, and decisions about adaptation, public finance, and insurance increasingly depend on translating large-scale climate and weather information into local, building-level impacts. This talk explores how the rapid rise of AI-based weather and climate models is reshaping that translation, and how these models might complement, rather than replace, established physics-based and engineering approaches. Using coastal Louisiana as an illustrative setting, I walk through a workflow that links atmospheric and flood-hazard information to property-level exposure, damage, and the downstream questions of fiscal impact and insurance affordability. Along the way, I reflect on practical lessons about uncertainty, data quality, and the modeling choices that most shape the results, and I consider how geospatial and "Geo-AI" tools can make such analyses more transparent and useful for decision-makers. The aim is to open a conversation about the opportunities, limitations, and responsible use of these methods, and to sketch promising directions for coupling AI forecasting with regional flood-risk and resilience research.

EdEC logo color

The visiting faculty is sponsored by the EdEC Faculty Fellowship Program.