EMC/HPC/MPC/NCO/CPC synergy meeting highlights 03/25/02

This meeting was led by Peter Manousos and followed a generic agenda denoted by the subtitles below. Attendees included Hua-Lu Pan, Zoltan Toth, Steve Tracton, Jun Du, Keith Brill, Larry Burroughs, Steve Jascourt, and Bill Bua.


NCO members were tied up at a meeting dealing with the purchase of the next super computer. A status was unable to be provided.

2. Notes from EMC

a. Global Modeling Group: Hua-Lu reported no changes from last meeting.

b. Mesoscale Modeling Group: No representative from the Mesoscale Modeling Group was available at this time to provide input on the mesoscale modeling effort.

c. Global Ensembles: Zoltan Toth reported that NCEP now has access to Canadian EPS output grids. Preliminary verification results indicate the Canadian EPS performs similarly to the NCEP EPS. Interestingly the NCEP EPS performs slightly better during the first 6 days with the Canadian performing better afterward, which is the opposite of how their opeational counterparts seem to perform. The next step is to perform the same tests with the ECMWF EPS. Ultimately each EPS would be bias corrected and then combined to create a “super ensemble”. Some differential weighting between models will likely be needed as the ECMWF is intiated at the 12Z cycle where as the Canadian and NCEP EPS are 00Z and 12Z. The Winter Storm Recon. Program ended about a week ago with over 20 flights and 15 cases in a 2 month period. This is the first year however real time evaluation was not performed. Results can be found at the site http://sgi62.wwb.noaa.gov:8080/ens/target/wsr2002.html .


d. Short Range Ensembles: Steve Tracton and Jun-Du reported that test runs using a upgraded rescaling on the perturbations are being verified in order to ascertain improvement. These results will be available shortly. Work continues toward generating Eta members using the Kain-Fritsch (KF) convective parameterization, though depending on the amount of computer resources available for the SREF package, the total number of Eta members might not be able to be increased. Instead, a few KF members would have to substitute for a few that now use the operational BMJ parameterization. Progress on adding RUC members awaits action by FSL.

e. Marine Modeling and Analysis Branch (MMAB) - Larry Borroughs reported COFS model was implemented March 12th. They are going to CAFTI April 8th for NWW3 model upgrades to physics packages and computational improvements and the Eastern North Pacific Regional Wave Model (ENP) (which is being added to the suite based on the NOAA WAVEWATCH III). The North Atlantic Hurricane model will be ugraded to accommodate 1 hourly output from the GFDL.


3. Input to EMC from Operational Centers

Concerns have been raised by some NCEP operational centers and forecasters from the WFOs as to the validity of running the Global model out to 384 hours 4x a day. Hua Lu had addressed these issues in an email to the Global Model users - “Many people would ask if we have any skill credential to issue two week forecasts every six hours. Our strategy is to make these forecasts part of our ensemble strategy. Given that week-two forecasts can only be used as outlooks and the skill of any individual forecast is very low, we can only hope to use the ensemble strategy to provide information about the atmosphere regimes. Running all the ensemble at 00z or running them at 4 different times are just different strategies. Later this year when we get more computer resource, we plan to run the global ensemble four times daily with ten members per run (a doubling of our ensemble members).” Further, CPC operational forecasters have stated that although any one member of the Global Ensemble does not show much skill at the 6-10 day and 8-14 day time frame, the ensemble mean does exhibit usable skill at these lead times. Since CPC has a requirement to issue these operational forecasts routinely to the public, CPC has a requirement for EMC to supply model data at that time range to support their operations.


4. Next Meeting Proposed Monday April 22nd at noon in room 209.