Extreme weather events (e.g. tropical cyclones, thunderstorms, atmospheric rivers, winter storms) produce multiple hazards including freshwater flooding, storm surges, tornados, hail, and landslides. The scale of the problem is staggering, with nearly 100 “billion dollar disasters” due to such events in the United States in the past decade alone. Reducing these impacts requires better prediction and forecasting on timescales ranging from subdaily to decadal, and expertise across a range of disciplines, including meteorologists, climate scientists, hydrologists, and risk experts. This session seeks to bring together early, mid, and late-career experts from across the AGU Natural Hazards community to share approaches and build awareness along the entire chain of extreme weather prediction using theory, models, and observations—from their origins as atmospheric phenomena to their wide-ranging impacts on human and natural systems. In recognition of AGU’s centennial celebration, we also invite submissions that review advances in prediction over the past century.