A look at the current and future status of the HRRR model
Ed Szoke, CIRA and NOAA/ESRL/GSD, Boulder, CO
Stan Benjamin, NOAA/ESRL/GSD
Curtis Alexander, NOAA/ESRL/GSD
John Brown, NOAA/ESRL/GSD
Steve Weygandt, NOAA/ESRL/GSD

Abstract
The High Resolution Rapid Refresh (HRRR) model continues to be a widely used operational NCEP model and the only convection-allowing model with a new run every hour.  The HRRR became operational at NCEP on 30 September 2014, after being available to forecasters for several years on an experimental basis from the Global Systems Division (GSD) of ESRL/NOAA in Boulder.  The HRRR is run once per hour at NCEP out to 18 h at a horizontal grid resolution of 3 km, allowing it to operate without convective parameterization.  It is initialized with the RAP (Rapid Refresh model) model, which is run at NCEP at a resolution of 13 km once per hour out to 21 hours, but the HRRR assimilates radar and other data on the 3-km scale.  The RAP model, also developed at NOAA/ESRL/GSD, became operational on 1 May 2012.  Both the HRRR and RAP are on the NCEP schedule for transition to updated versions; the last update came on 23 August 2016 and the next is scheduled for February 2018.  Ongoing development to address known issues and improve both models occurs within experimental versions that are available at http://rapidrefresh.noaa.gov/.  Since the model codes will be frozen for the next upgrade by early summer 2017, this presentation will summarize the changes that will have been made for the next operational version of the HRRR, using examples from various cases to illustrate some of the issues with the current HRRR.  We will also look at future areas of model improvement that will be tested in the experimental version of the HRRR.  Other areas of focus with the HRRR will also be addressed, including experimental ensemble runs and the latest developments with the HRRR time-lagged ensemble (HRRRTLE), as well as work with new domains in Alaska and elsewhere.