Rapid Refresh (RAP)

Description

RAP (Rapid Refresh) is NOAA/NCEP’s hourly-updated, continental-scale mesoscale modeling system covering North America. It combines a short-range forecast model with a data-assimilation cycle to produce near-current gridded analyses and forecasts. RAP uses a version of the Weather Research and Forecasting (WRF-ARW) core, initialized hourly via a Rapid Data Assimilation (RAP-DAS) scheme that ingests in situ observations, satellite radiances, radar-derived precipitation, and conventional upper-air data.

Its emphasis is on capturing mesoscale and boundary-layer phenomena at high frequency—critical for thunderstorm development, frontal passages, and low-level wind patterns. The assimilation system uses a hybrid ensemble-3DVar approach to blend the short-term forecast background with observations, enhancing atmospheric structures like temperature, moisture, wind, and vertical motion.

RAP operates at approximately 13 km grid spacing, outperforming global models in resolving small-scale variability. It runs hourly with forecasts extending out to roughly 54 h, offering timely guidance for several downstream systems. In fact, RAP forecasts are a key input into the National Blend of Models (NBM), enhancing short-term accuracy by feeding into its multi-model ensemble and blending routines.

Due to its high-frequency output and fine resolution, RAP is widely used in operational contexts—such as aviation planning, lightning-threat monitoring, convective initiation forecasting, emergency response to rapidly evolving weather (e.g. snowbands, flash flooding), and renewable energy ramp-forecasting. It also serves as the short-term “first guess” in assimilation and as a valuable feature for machine learning models aiming at nowcasting or bias correction.

Detail

• Product page: https://www.nco.ncep.noaa.gov/pmb/products/rap/
• Key Features:
  • Continental-scale, high-frequency (hourly) mesoscale analysis and forecast system.
  • Hybrid ensemble-3DVar (RAP-DAS) assimilation integrating in situ, satellite, radar, and upper-air data.
  • WRF-based forecast core capturing boundary layer and convective-scale features.
  • Feeds into downstream products like NBM and serves as operational first-guess analyses.
• Resolution: ~13 km grid spacing.
• Temporal Coverage: Hourly analyses and 54-h forecasts.
• Primary Use Cases: Aviation decision support, convective initiation and severe weather monitoring, emergency management, renewable energy forecasting, and short-range verification and bias-adjustment pipelines.

Parameters