2.3 Recent Advancements in Convective-Scale Storm Prediction with the Coupled Rapid Refresh (RAP) / High-Resolution Rapid Refresh (HRRR) Forecast System

Alexander, Curtis, Tanya Smirnova, Ming Hu, Steve Weygandt, David Dowell, Joe Olson, Stan Benjamin, Eric James, John Brown, Patrick Hofmann, Haidao Lin, Kevin Brundage, and Brian Jamison, National Oceanic and Atmospheric Administration

The Rapid Refresh (or RAP) is an hourly assimilation / forecast system that replaced the old Rapid Update Cycle at the National Centers for Environmental Prediction of the US National Weather Service in May 2012. The High-Resolution Rapid Refresh (HRRR) is a CONUS 3-km convection permitting atmospheric prediction system coupled to the HRRR through partial cycling and run hourly out to fifteen hours in real-time at the NOAA Earth System Research Laboratory (ESRL). The purpose of this talk is to report on an upgraded version of the RAP and HRRR. With the availability of the new NCEP WCOSS computer, there are prospects that both RAP and HRRR may go into NCEP operations prior to the 2014 WRF Workshop.

Both RAP and HRRR use version 3.4.1 of the Advanced Research WRF (ARW) model. Identical physics is used with the exception that the HRRR is run without parameterization of convection. Major physics changes since last year include a 9-layer version of the RUC LSM together with modification to roughness length for certain land-use categories (both to be discussed more fully in a poster by Tanya Smirnova at this workshop), and replacement of the Mellor-Yamada Janjic (MYJ) PBL and surface layer schemes by the corresponding Mellor-Yamada-Nakanishi-Niino (MYNN) schemes as modified by Joe Olson. Both RAP and HRRR now use the Goddard short-wave radiation.

The RAP uses the new hybrid-variational option within the Gridpoint Statistical Interpolation (GSI) analysis scheme together with twice-per-day partial cycling from the NCEP Global Forecast System, which also provides lateral boundary conditions. The RAP in turn provides the lateral boundary conditions for the HRRR.

Other important changes to the HRRR configuration include the establishment of 3-km data assimilation to incorporate storm-scale information using Gridpoint Statistical Interpolation (GSI) that includes sub-hourly 3-km radar data assimilation, 3-D variational assimilation of conventional observations and a non-variational cloud and precipitating hydrometeor analysis similar to that used in the RAP, but at 3-km grid spacing instead of 13km.

We will document RAP and HRRR analysis and forecast improvements with retrospective and real-time verification statistics and case studies from 2012-2013.