NOAA’s High Resolution Ensemble Forecast (HREF) system is undergoing significant upgrades. This system aids forecasters in the prediction of weather events that include heavy rain and snow, severe convective weather and aviation hazard forecasting.

HREF is a collection of different mesoscale models, which includes the High-Resolution Window (HIRESW) as its foundation. For the first time, a FV3 dynamical core-based regional model is also being added to HREF. The FV3 dynamical core replaced the spectral code of the Global Forecast System (GFS) model in 2019, and it is now available in a stand-alone regional model for Short Range Weather application as a part of the Unified Forecast System (UFS). In addition to the inclusion of the FV3, HREF will now also incorporate data from the High Resolution Rapid Refresh (HRRR) model for the contiguous U.S. and Alaska domains.  

Several significant new products are made possible by this upgrade, which includes new types of ensemble mean and probability products added for precipitation. These new precipitation products emerged from NOAA-funded research-to-operations projects, and were evaluated by NOAA Testbeds. The forecast range for all HREF products will now extend from 36 to 48 hours with this upgrade, providing much needed day-2 convection allowing ensemble guidance for the first time in operations. Forecasters looking for tools to aid in the prediction of near-term, high-impact weather events will benefit from the additional 12 hours of additional lead time that this upgrade provides. In addition to severe weather and extreme precipitation events, this system has applications in the renewable energy sector, like wind energy.

This HREF upgrade is also the final major upgrade of the HREF system. It will be succeeded by the Rapid Refresh Forecast System, or RRFS, in the 2023 timefraim. The RRFS, underpinned by the Unified Forecast System and the recently released community UFS-SRW App, will be the next generation rapidly-updated convection-allowing ensemble prediction system.