An NWP-Free, Observation-Driven Deep Learning Approach to Heavy-Rainfall Nowcasting Beyond the Three-Hour Limit
FlowsNet is presented, an observation-based multisensor fusion model that learns directly from radar/rain gauge-analyzed precipitation, surface variables from ground stations, geostationary satellite imagery, and satellite-derived precipitation context and outperforms a leading extrapolation-based method and state-of-t...