The NCEP North American Mesoscale (NAM) Analysis and Forecast System : Near-term plans and future evolution into a high-resolution ensemble

As part of the NWS mission to provide high-resolution forecast guidance, during the next five years the NCEP North American Mesoscale (NAM) Analysis and Forecast System will be undergo significant near- and long-term upgrades. Near-term model/analysis changes, scheduled for implementation in 2014, include use of the NCEP hybrid-ensemble GSI analysis with the global EnKF, replacing the GFDL radiation scheme with the Rapid Radiative Transfer Model (RRTM), extensive tuning of the Non-hydrostatic Multiscale Model on B-grid (NMMB) gravity wave drag/mountain blocking scheme to make the model forecast more responsive to subgrid-scale terrain variability and improve synoptic performance, a new version of the Betts-Miller-Janjic (BMJ) convection scheme, and modifications to the microphysics. Also part of the 2014 change package will be the increase of both CONUS and Alaska fixed nests from 4 km and 6 km respectively to 3 km, with mostly explicit treatment of convection.  The impact of these changes on the NAM's QPF and synoptic performance will be presented.

Over the longer term (2015-2018) anticipated increases in NCEP's computing capacity as part of the Hurricane Sandy Supplemental Funding Bill will allow for the possible convergence of the NAM, High-Resolution Rapid Refresh (RAP/HRRR) and the NCEP Short-Range Ensemble Forecast System (SREF) into a high-resolution mesoscale ensemble system by 2018. Tentative plans include the establishment of both ARW-based and NMMB-based hourly data assimilation systems, hourly 18-h forecasts for RAP and NMMB parent and nest domains, and up to six (3 ARW, 3 NMMB) 3 km ensemble members over the CONUS and Alaska run every 6 hours. Additional details on this initiative will be presented at the conference.