EMC Synergy Meeting Highlights April 23, 2005


This meeting was led by Peter Manousos and attended by Larry Burroughs, Eric Rogers, David Plummer, Zoltan Toth, John Ward, Jim Hoke, Geoff Dimego, Mary Hart, Keith Brill, Stephen Jascourt, Ed Danaher, Stephen Lord, Joe Sienkiewicz, and by phone Stephen Weiss, David Bright, and Greg Grosshans from SPC.


1. CCS

John Ward reported that the only major issue with the CCS is overheating of White, which became an issue in warm weather when opening the outside door no longer solved the problem. Therefore, the day after the meeting, White was taken down for roughly 30 hours while operations were transferred to Blue (normally the development machine), creating a period when no EMC development/testing could be done.  Disk space increase proposal is pending a “needs assessment” by NOAA COTAR (money is available to pay for the increase).  TOC will have a back up in VA up and running by end of calendar year.


2. Notes from EMC

a.     Global Modeling Group: Stephen Lord reported the parallel GFS has been checked out (except for summer QPF).  Summer reruns are being conducted and evaluations are pending.  The GFS parallel is targeted for operational implementation by the end of May.

b.     Mesoscale Modeling Group: Eric Rogers reported the parallel NAM evaluation ended April 15, 2005.  A briefing to the NCEP Director is scheduled for April 28th with an operational implementation May 3rd.

c.     Global Ensemble Prediction System: Zoltan Toth reported work on the pre-hurricane season implementation of the ensemble system (6-hrly breeding, vortex relocation algorithm – control of perturbation size and location, and extending T126 resolution out to 16 days) has been delayed, partly due to chronic disk space shortage on the CCS.  Relocation tests have just started (too early for results).  There is still the possibility to implement by June, but code has not yet been offered to NCO.   After implementation the next parallel (fall package) will feature 4 cycles per day.  Additionally, membership will increase from 10-20 per cycle and the Ensemble Transform Kalman Filter will be tested for generating initial conditions.  The NAEFS 1st operational implementation (products from combined CMC/NCEP output) is targeted for March 2006.  This will include output covering South America and WMO regions 3 and 4.  Finally, the Winter Storm Recon program ended in March (results look good – out of 30 cases run at T126 resolution, 73% showed positive impact from including the dropsonde data).

d.     Short Range Ensemble Prediction System: Geoff Dimego reported a minor June or July upgrade which includes a faster version of the RSM, switching to the WRF postprocessor (which avoids the CAPE problem in the RSM postprocessor), and BUFR output will be added for the RSM members.  After additional disk space is installed, SREF lead time will be extended from f63 to f87 and output grids covering Alaska and perhaps Hawaii-east Pacific will be added including domain covering AK. 

e.     Marine Modeling and Analysis Branch (MMAB): Larry Burroughs reported the Tracker System showing cyclone tracks for many different models is online. A “Mean Tracker” is being worked on for ensembles (includes indication of an envelope of uncertainty and forecast duration is being extended to 384 h).  OPC evaluated it for a snowstorm case, and NCO is now putting the tracker in NAWIPS. In anticipation of NAEFS, Ensembles Group has requested MMAB to include a tracker region surrounding South America.  The Real-Time Global SST (RTG_SST) analysis is testing a 1/12th-degree resolution version to replace the presently operational ˝-degree version, including lake temperatures. It will utilize data from a second DMSP satellite. Hi-Res SST is in parallel (1/12th degree resolution by fall including the Great Lakes).  HYCOM is about to get into parallel production.


3. Input to EMC and NCO

a.     OPC – stated to MMAB it has not yet evaluated HYCOM output.

b.     SPC – expressed concerns about jobs running on the development machine getting terminated due to disk space shortage.  NCO responded that critical jobs should be transitioned to operational status and they would be happy to help with the transition.

c.     HPC – stated it completed its evaluation of the parallel NAM, is currently evaluating the parallel GFS (and in a preliminary sense has a favorable impression of the global parallel output), and requested a rerun of a single SREF run to be used for experimental product development for next year’s Winter Weather Desk product suite.


4. The next meeting will be held Monday May 23rd, 2005 at noon in room 209 with remote conference capability. The last Monday of the month, May 30, would have been the regularly scheduled date but is the Memorial Day holiday.