NCEP Synergy Meeting Highlights: March 30, 2009


This meeting was led by Bill Bua and attended by John Ward; Geoff DiMego; Stephen Jascourt; Mike Brennan and Richard Pasch, NHC; Dave Novak and Keith Brill, HPC; Steve Weiss and David Bright, SPC; Bernard Meisner, Southern Region; David Myrick, Western Region; Bill Ward, Pacific Region; and Vera Gerald, MMAB.


1. NCO


Geoff DiMego reported on behalf of NCO. 95% of production is running on the new Power-6 cirrus machine in Fairmont WV. There are still some problems with consistent running times for the Gridpoint Statistical Interpolation (GSI) system. The Marine Modeling and Analysis Branch and Hurricane WRF models are still having difficulties with timing issues as well. The Power-6 development (stratus) machine is being installed in Gaithersburg MD, with preparatory infrastructure work proceeding on schedule.


A minor change to the RUC was being implemented on Tuesday, 31 March, including a fix to bring in observations (daily NESDIS analysis) to correct snow cover errors in the self-cycling, correct an error in calculation of forecast total precipitation, and a change moving cloud analysis to the end of the analysis process to make the forecast of cloud more consistent from run to run.


John Ward noted that the computer moratorium is scheduled, per the IBM contract, to end no later than October of 2009, and most likely will end in July of 2009.




2a. Global Climate and Weather Modeling Branch (GCWMB)


John Ward reported that upgrades are still waiting in the queue for more computer resources.  Post-processing changes to the GDAS need to be made to match the GDAS to the GFS. An eye will need to be kept on the vortex relocation for hurricanes prior to implementation. For the GFS, new post-processed fields are also to be added along with some bug fixes, to the next minor implementation. 


The next GFS implementation contains a few bug fixes. Currently, a “null” test to produce identical results to operations cannot be run with the current parallel GFS. GCWMB wants to be able to get identical results as a “sanity check” prior to running parallel tests.  The problem has been isolated to the aerosol routine in the code, and the global branch continues to work on the issue.


Hurricane WRF (HWRF) is now running in parallel; it will not be implemented on the Power-5 (mist), but will be on the Power-6 (cirrus) for the entire season. Changes to the HWRF include the effect of sea spray and gravity wave drag on the forecast, and a change in the initialization procedure. TPC and Central Pacific Hurricane Center will want to talk to EMC further about this before implementation. Additionally, information was requested on having 5 runs available for the HWRF and the GFDL hurricane models; this will be followed up on with NCO (contact: Dan Sterosta).


Testing is being done on a new semi-Lagrangian model which will be run at T878. The goal is to implement this model sometime between November 2009 and February 2010. In the meantime, further changes to the GSI are scheduled for 1st quarter of FY 2010.


Climate Forecast System (CFS):  reanalysis and reforecasting has been continuing on Power-5 (mist).  Runs will be suspended if there is a need to run production on mist.  Next major implementation for the CFS isn’t scheduled until 1st quarter of FY2011. 


Note was made that global data will be put into GRIB2 format in parallel in a few months to save disk space. This however will not be implemented operationally until FY 2010.


2b. Mesoscale Modeling Branch (MMB)


Geoff DiMego reported on a minor implementation for the RUC (see NCO section).


WRF-NMM has a parallel on the Power-6 cirrus computer that have 3 bug fixes. So far the aggregate impact is nil, though specific forecasts may show small differences.  This will be made operational once the computer moratorium is lifted. Additional work will be done to work toward having the NMM Data Assimilation System (NDAS) use the same data as the GDAS; this will initially involve instituting monitoring of data NOT used in NDAS but used in GDAS. 


The Real-Time Mesoscale Analysis (RTMA) needs more disk space before it can be increased in resolution from 5 km to 2.5 km. This problem will not be solved when the Power-6 becomes operational.


4-km NMM and NMMb runs have been placed on the development computer by Matt Pyle.  MMB expects to make these operational in the hi-res window slots later this year.


Recasting MMB models into the NEMS framework is coming in FY2010.


2c. Global Ensemble Prediction System


Yuejian Zhu is testing a parallel GEFS run which increases resolution by about 50%, which would be implemented once the computer moratorium is over.


2d. Gridpoint Statistical Interpolation (GSI) Data Assimilation System


See above on GFS/GSI implementation.


2e. Short Range Ensemble Forecast System (SREF)


See last month’s report; SREF still running in parallel on Power-6 cirrus computer and MMB is hoping that implementation can take place before the moratorium is lifted. SPC,HPC, and AWC are strongly in favor of this effort.


2f. Marine Modeling and Analysis Branch


No report.




NCEP Centers and NWS Regions


Migration of data from Power-5 to Power-6 has not resulted in any problems thus far. SPC and HPC have been actively getting data from the new machine through dbnet.


The NCEP office move to the new building will not result in a CCS moratorium, though there may be a bit of disruption of staff at the in-house operational centers.  As of last week, there was still no further action on completion of the new building.  At present, the delay is at least 7 months from the original planned move dates for each in-house center.


No issues were brought up by the regions participating in the call.


4. The next Synergy Meeting will be held at 2:30 pm EDT on Monday, 27 April 2009 in room 209, with remote teleconferencing capability.