loading page

The High Resolution Ensemble Forecast (HREF) system: Applications and Performance for Forecasting Convective Storms
  • +1
  • Brett Roberts,
  • Burkely T. Gallo,
  • Israel L. Jirak,
  • Adam J. Clark
Brett Roberts
Cooperative Institute for Mesoscale Meteorological Studies, NOAA/NCEP/Storm Prediction Center, NOAA/OAR/National Severe Storms Laboratory

Corresponding Author:[email protected]

Author Profile
Burkely T. Gallo
Cooperative Institute for Mesoscale Meteorological Studies, NOAA/NCEP/Storm Prediction Center
Author Profile
Israel L. Jirak
NOAA/NCEP/Storm Prediction Center
Author Profile
Adam J. Clark
NOAA/OAR/National Severe Storms Laboratory
Author Profile


In November 2017, the High Resolution Ensemble Forecast version 2 (HREFv2) system was implemented by the U.S. National Oceanic and Atmospheric Administration (NOAA). The HREFv2 is NCEP’s first operational convection-allowing model (CAM) ensemble and runs twice daily at ~3-km horizontal grid spacing. Coinciding with the operationalization of the HREFv2, the NOAA National Weather Service’s Storm Prediction Center (SPC) launched its HREF Ensemble Viewer (https://www.spc.noaa.gov/exper/href), a public web display for real-time HREF data. In addition to products traditionally used in operational forecasting, the HREF Ensemble Viewer also employs post-processing and visualization techniques which leverage the convective structures explicitly modeled in CAM ensembles. This presentation describes and illustrates products from the SPC viewer unique to CAM ensembles which are targeted at forecasting convective storm coverage, intensity, and evolution. A key aspect of HREF’s ensemble design is membership diversity with respect to model cores and parameterization schemes, in conjunction with a time-lagging approach. Verification is performed on ensemble fields related to explicitly modeled convection for various modified configurations of the HREF membership, illustrating how each dimension of membership diversity contributes to the system’s skill in forecasting convective evolution. These results help to illuminate potential tradeoffs entailed in configuring future HREF iterations beyond version 2 as CAMs using new dynamical cores and parameterization schemes become available.