The High-Resolution Rapid Refresh (HRRR)


image not available
Composite Reflectivity

Due to excessive resource usage, automated scrapers will be blocked in the near future. If you regularly use these kinds of scripts to download GSL graphics, please contact us at [email protected] to arrange an alternative method for obtaining our model images. Thank you.


The HRRR is a NOAA real-time 3-km resolution, hourly updated, cloud-resolving, convection-allowing atmospheric model, initialized by 3km grids with 3km radar assimilation. Radar data is assimilated in the HRRR every 15 min over a 1-h period adding further detail to that provided by the hourly data assimilation from the 13km radar-enhanced Rapid Refresh .
    HRRR implementations at NCEP
  • HRRRv1 - 30 Sept 2014
  • HRRRv2 - 23 Aug 2016
  • HRRRv3 - 12 July 2018
  • HRRRv4 - 2 Dec 2020
    Information here in HRRRv4/RAPv5 summary - Jan 2020

HRRR Colorado Labs Award video - a 2-minute layperson-level description on the HRRR from late 2015 and why it is important. (October 2015)
HRRRv4/RAPv5 .

    Key features for HRRRv4:
  • improved cloud representation for boundary-top clouds, especially for shallow cold-air layers with cold-air retention
  • better cloud bands (snow squalls, hurricane bands, lake-effect bands)
  • 3km ensemble data assimilation for improved storm prediction for 1-12h
  • inline smoke prediction
  • improved lake temperatures
  • Extension to 48h forecast every 6h.

HRRR Ensemble (HRRRE) prediction - documentation as of March 2020
2022 HRRR description - Dowell et al 2022
HRRRv2 physics description in Benjamin et al. 2016, A North American Hourly Assimilation and Model Forecast Cycle: The Rapid Refresh. Mon. Wea. Rev., 144, 1669-1694. 2020 RAPv5/HRRRv4 Changes RAPv4/HRRRv3 Change Highlights

News items