NCEP Synergy Meeting Highlights: December 1, 2008


This meeting was led by Bill Bua and attended by Chris Magee, Keith Brill, James Hoke, Joe Sienkiewicz, Geoff DiMego, John Derber, Bill Lapenta, Joe Carr, and Allan Darling. Steve Silberberg of AWC and Steve Weiss of SPC attended by remote teleconference.


Checkers:  Please pay particular attention to highlighted items.


1. NCO


Chris McGee of NCO reported that the new RTOFS (Atlantic basin ocean forecast model) implementation (based on HYCOM) will occur on 16 December 2008. Scheduled for 9 December implementations are the HYSPLIT (dispersion) model upgrade and the Unified Real-Time Mesoscale Analysis (RTMA). Operational center evaluations of the new NAM-WRF parallel need to be turned in to NCO by Friday, 5 December 2008 to be included in the briefing to Dr. Uccellini on 12 December. Tentative implementation date is 12 UTC 16 December 2008.


A Global Statistical Interpolation (GSI) and post-processing 30-day parallel will be started as soon as possible, for an implementation of hopefully early February 2009, if successful. This implementation would be an exception to the system moratorium being held from Christmas 2008 into April 2009, during IBM System 6 testing.


A Short-Range Ensemble Forecast (SREF) System implementation will occur after the system moratorium ends.  Steve W. from SPC requested a parallel SREF be started during the moratorium; the data would not be needed real-time necessarily since it will be used to calibrate severe weather, lightning, and other probabilities based on the new SREF.  Allan Darling of NCO said he would work with SPC to come up with a way to do this (which would just involve making sure data is available to SPC).


RUC implementation omitted the TAMDAR data.  A fix will be decided on by the principals at NCO, the Mesoscale Branch, and the Earth Systems Lab in Boulder CO in a future meeting.


As an aside, we briefly discussed reorganizing the Synergy meeting agenda so that systems, rather than branches are discussed.  A possibility would be to rotate discussion for different systems every other or every third month.  Bill Bua said he’d add that to next agenda for discussion.




2a. Global Climate and Weather Modeling Branch


No report this month.


2b. Mesoscale Modeling Branch



Geoff DiMego gave further details on the NAM-WRF implementation.  The big change is partial cycling of data assimilation on GFS (using GDAS as the first guess at t-12 hours) and then ingesting and assimilating subsequently at the mesoscale. The large and consistent forecast improvements through 84 hours in the NAM-WRF that have resulted indicate an improved synoptic scale analysis.  This complements the improvements resulting from the addition of gravity wave drag parameterization in the previous implementation. 


No further implementations will be done in the WRF framework; future changes to the NAM will wait until 2010 when the Earth System Modeling Framework (ESMF) has been successfully implemented.  At that time, 4-km nests will be looked into and implemented upon running successfully in parallel.  Fire weather support runs would be considered for reinstatement subsequently.



The next implementation will be done in Q4 2009, increasing resolution from 5- to 2.5-km (3-km in AK).  For Guam, smart-init will be applied to the GFS at its native grid resolution to provide a “first guess” for an RTMA as part of a future implementation.  This same smart-init will provide a downscaled GFS to all WFOs from the same GFS native grid, a better product that the WFO produced smart-init from the interpolated GFS grids that the WFOs get through AWIPS.



The NGM will be terminated, as will the Eta-32, since the NAM-MOS will be replacing the NGM and Eta MOS products.


2c. Global Ensemble Prediction System


No report.


2d. Gridpoint Statistical Interpolation (Data Assimilation) System


John Derber reported further on the implementation brought up in the NCO section.  The next implementation will include more post-processed products and the inclusion of new satellite observations from IASI (an NPOESS source similar to AIRS data), plus improved complex QC inside the analysis.  As previously stated, there will be a 30-day evaluation period of a parallel run done by the operational centers starting in early February 2009.


Future work:  In Q4 of 2009, a change to how NPOESS sounder data is assimilated is to be tested, that will get the starting GFS forecast analysis closer to the final Global Data Assimilation System (GDAS) analysis.  Preliminarily, this change seems to affect the number and size of so-called forecast “drop-outs” in the medium range period.  Additional work will be done on ozone assimilation, 4-D Variational analysis, and situational background error covariances in the next couple of years.


2e. Short Range Ensemble Forecast System


See NCO above.  No report otherwise.


2f. Marine Modeling and Analysis Branch


No report.






No feedback at this time.


4. The next Synergy Meeting will be held Monday, January 5, 2008, at 12:00 pm EST in Room 209 at EMC, with remote conference capability.