EMC Synergy Meeting Highlights October 25 2004


This meeting was led by Peter Manousos and was attended by Bill Bua, Zoltan Toth, Jim Hoke, Daniel Pawlak, Pete Caplan, John Ward, Joe Sienkiewicz, Mike Bodner, Ed Danaher, Mark Klein, Keith Brill, Mary Hart, Stephen Jascourt, and Larry Burroughs.


1. CCS

The CCS moratorium commenced October 12th and is still anticipated to end when the new system becomes operational in mid to late January 2005.  The new system will be configured so that the current “Frost/Snow” will be the NOAA Research and Development machine, the production machine “White” will be installed at Fairmont, and the development machine “Blue” will be in Gaithersburg.  This configuration will be temporary until the required communications upgrade is implemented June 2005 – after which the final configuration will feature operations on “Blue” and NOAA R & D on Frost & Snow – both in Gaithersburg.. and development on “White” in Fairmont (and therefore the backup system).   Finally, the naming convention of AVN to GFS, Eta to NAM (North American Mesoscale), and TDL to MDL (for directories containing MOS files) will be completed when the new system comes on line in January – see


for details.


2. Notes from EMC

a.     Global Modeling Group: Pete Caplan reported that a partial bundle for the next version of the GFS has been implemented and available in the parallel slot (MRFx) since October 1.  To date, the bundle includes the hybrid grid coordinate system (sigma – pressure) in the forecast mode, while the analysis scheme still uses just the sigma coordiante. Numerous other items are expected to go in eventually, including the Gridpoint Statistical Interpolation analysis and the NOAH land surface model which is also used in the Eta model.

b.     Mesoscale Modeling Group: The final change package for the Eta and DGEX, as well as an upgrade to the Air Quality Forecast system will be implemented by then end of February 2005.  An upgrade to the HiRes Window ensemble will be implemented by the end of June 2005 followed by an SREF upgrade by the end of September 2005.  The North American Mesoscale Model (NAM) WRF, the NMM replacement of Eta, and Gridpoint Statistical Interpolation (GSI) replacement of the Eta 3D-Var are targeted to be implemented by the end of Februrary 2006.

c.     Global Ensemble Prediction System: Zoltan Toth reported the following changes intended to be implemented next spring: the ensemble members will use hurricane relocation as the GFS analysis now does, increase of membership from 10 to 20 per cycle, and the breeding cycle of initial perturbations will be replaced by an Ensemble Transform Kalman Filter (ETKF) technique.  The new computer has capacity to run more than 20 members or to run 20 members at higher resolution. Evaluation of these options is ongoing.  Tests of physics diversity with RAS and SAS convective parameterization schemes (same as that in SREF RSM members) revealed no significant benefit in medium to long range forecast times. This leads to the next series of tests where stochastic forcing will be imparted on higher wave numbers in the forecast system. On a side note, the CMC will abandon use of physics diversity to generate membership in its global ensemble.  In mid-November a workshop will be held at NCEP WWB to officially kick off the North American Forecast System (combined ensemble effort between NCEP, CMC, and Mexico).

d.     Short Range Ensemble Prediction System: No report.

e.     Marine Modeling and Analysis Branch (MMAB): Larry Burroughs reported the Regional Ocean Forecast System (ROFS) was updated just before the moratorium. GOES data will not be included until January 2005.  AWIPS build OB6 (due out in perhaps spring 2005) will include SSM/I winds in BUFR, as well as global visibility and vessel icing fields, and NOAA Wave Watch III output from using the hurricane model.  The extratropical storm tracker, a post-processing code which tracks and plots cyclone centers from all available models including the Eta, GFS, SREF, and global ensembles, will be in parallel testing in November. Finally, the wave assimilation system is in parallel with an intended implementation toward the end of 2005.


3. Next Meeting Monday Nov 29th, 2004 at noon in room 209.