Codes Employed

The software packages used for the HWRF pre-implementations tests were developed at NCEP EMC. Code innovations were brought into the HWRF system for testing by the DTC in close collaboration with physics scheme developers at Atmospheric and Environmental Research (AER), NOAA Earth Systems Research Labortory (ESRL), the National Center for Atmospheric Research (NCAR) and DTC.

The code repositories are:
• HWRF (scripts and namelists): https://svn-dtc-hwrf.cgd.ucar.edu
• WRF: https://github.com/wrf-model/WRF
• WPS: https://github.com/wrf-model/WPS
• UPP: https://svn-dtc-unifiedpostproc.cgd.ucar.edu
• GSI: https://vlab.ncep.noaa.gov/svn/comgsi
• HWRF-utilities: https://svn-dtc-hwrf-utilities.cgd.ucar.edu
• MPIPOM-TC: https://svn-dtc-pomtc.cgd.ucar.edu
• NCEP Coupler: https://svn-dtc-ncep-coupler.cgd.ucar.edu
• GFDL Vortex Tracker: https://svn-dtc-gfdl-vortextracker.cgd.ucar.edu

The branches for the H6CL configuration are:
• HWRF: branches/HWRF_2016_phystest_cntrl
• WRF: branches/HWRF_2016_phystest_cntrl
• WPS: branches/HWRF_2016_phystest_cntrl
• UPP: branches/HWRF_2016_phystest_cntrl
• GSI: branches/HWRF_2016_phystest_cntrl
• HWRF-utilities: branches/HWRF_2016_phystest_cntrl
• Princeton Ocean Model for TCs: branches/HWRF_2016_phystest_cntrl
• NCEP Coupler: branches/HWRF_2016_phystest_cntrl
• GFDL Vortex Tracker: branches/HWRF_2016_phystest_cntrl

Additional branches for the H6CO configuration are:
• HWRF: branches/HWRF_2016_phystest_CldOvlp
• WRF: branches/HWRF_2016_phystest_CldOvlp

Additional branches for the H6PC configuration are:
• HWRF: branches/HWRF_2016_phystest_PartCld
• WRF: branches/HWRF_2016_phystest_PartCld

Additional branches for the H6GF configuration are:
• HWRF: branches/HWRF_2016_phystest_GF
• WRF: branches/HWRF_2016_phystest_GF

HWRF Control: H6CL Configuration

This code is very similar to the 2016 operational model, with slight differences from bug fixes. The operational 1-way wave coupling (WAVEWATCH III) is not employed for this version.

• Cumulus: SA SAS
• Microphysics: Ferrier-Aligo
• Radiation LW: RRTMG, MR cloud overlap
• Radiation SW: RRTMG, MR cloud overlap
• Partial Cloudiness: 2016 operational (icloud=3)
• Planetary Boundry Layer: GFS-EDMF
• Surface Layer: GFDL
• Model Time step 30 s

HWRF Experimental Model: H6CO Configuration

Configuration is similar to the H6CL. Modifications include exponential random (ER) cloud overlap methodology within RRTMG radiation parameterization.

HWRF Experimental Model: H6PC Configuration

Configuration is similar to the H6CL. Modifications include updates to operational partial cloudiness scheme within RRTMG radiation parameterization.

HWRF Experimental Model: H6GF Configuration

Configuration is similar to the H6CL, but with Grell-Freitas cumulus parameterization.

Domain Configuration

The HWRF domain configuration for 2016 physics experiments consists of a parent with one set of storm-following nests. The parent grid covered an 80 x80 degree area with approximately 18 km horizontal grid spacing. The d02 domain covers an area of approximately 25 x 25 degrees with 6 km grid spacing centered on the storm, while d03 domain covers 8.3 x 8.3 degree area with 2 km grid spacing.

The location of the parent and nests as well as the pole of the projection varied from run to run and were dictated by the location of the storm at the time of initialization in all experiments.

As with the operational configuration, HWRF was run coupled to the three-dimensional MPIPOM-TC model, which used approximately 9 km grid spacing. There are two domains to cover the North Atlantic (also called Transatlantic) and Eastern North Pacific domains. The North Atlantic domain was initialized bu the GDEM climatology and the Eastern North Pacific domain was initialized with RTOFS data.

Track and Intensity Verification

Track, Intensity, and Absolute intensity errors were computed relative to the NHC Best Track out to 120 hours using MET-TC software.

Satellite Verification

Verification of HWRF-simulated brightness temperatures was performed on the d01 and d03 domains for 15 model cycles of Hurricane Matthew (2016). Satellite-observed and model-similated brightness temperatures from channel 4 of the GOES-13 platform were selected.