Development of HMON (Hurricanes in a Multi-scale Ocean coupled Non-hydrostatic model) is an important step towards implementing a long-term strategy at NCEP/EMC for multiple static and moving nests globally, with one- and two-way interaction and coupled to other (ocean, wave, sea ice, surge, inundation, etc.) models using NEMS (NOAA’s Environmental Modeling Environment) infrastructure. Developed initially as a collaborative effort between NOAA’s Environmental Modeling Center (EMC) and Hurricane Research Division (HRD) as a component of High Impact Weather Prediction Project (HIWPP), HMON was referred to as H-NMMB (Hurricanes in a Non-hydrostatic Multiscale Model on a B grid). HMON’s development has also been supported by HFIP and NGGPS (Next Generation Global Prediction System) programs.

HMON is based on the NMMB dynamic core which is currently being used in other NCEP’s operational NAM (North American Mesoscale Model) and SREF (Short Range Ensemble Forecast) systems. It has been built using shared infrastructure with unified model development within NEMS. The NMMB dynamic core is much faster and more scalable than other contemporary dynamic cores deployed for modeling Hurricanes at NCEP. Use of NEMS also paves the way for the future use of CCPP (Common Community Physics Package) style physics packages.

The objective for HMON/HNMMB development is to provide high-resolution intensity forecast guidance to NHC along with HWRF (and to replace the legacy GFDL hurricane model). Robust retrospective testing and evaluation in the past year has shown that HMON has superior results over GFDL and different characteristics than HWRF (Figure 1) for both North Atlantic and North East Pacific basins.

alternate text alternate text alternate text alternate text

Figure 1: Skill plots for 2014-2016 seasons relative to GFDL Hurricane Model (cyan) for HMON (blue), HWRF (red) and official results (black) for North Atlantic Tracks and Intensity (top panels) and North East Pacific Tracks and Intensity (bottom panels).