Convective weather systems represent highly non-linear, rapidly evolving systems involving small-scale processes and fine-scaled features that are challenging to simulate in numerical weather prediction (NWP) models. Here, we present the results of 30-minute precipitation forecasts generated from an experimental real-time NWP modeling system that updates simulations every 30-seconds with observations from a multi-parameter phased array weather radar (MP-PAWR). The forecasts are compared to nowcasts from a spatiotemporal extrapolation-based precipitation nowcasting system that uses MP-PAWR observations with a 30-second update interval to provide 30-minute forecasts. The NWP model successfully predicts rapid changes in the storm’s structure and intensity, resulting in it outperforming the nowcasts at up to 30-minute lead times, demonstrating the advantage of the NWP system over the nowcasting system for very-short range rain forecasts. The 30-second updating was demonstrated to improve rain forecasts by promoting convective growth through increasing moistening and upward motion in the storm environment.