NCEP Modeling Synergy Meeting Highlights: August 27, 2007


This meeting was led my Michael Brennan and attended by Keith Brill, Joey Carr, Joe Sienkiewicz, Mark Iredell, Mary Hart, and Geoff DiMego. Bill Bua and Stephen Jascourt of UCAR/COMET, David Bright of SPC, and Steve Silberberg of AWC attended by remote conference.


1. CCS


Joey Carr of NCO reported that four implementations are planned by NCO during the remainder of FY07:


        Upgrade to the high resolution window (HRW) WRF runs scheduled for 11 September

        Implementation of new multi-grid wave model for the NOAA Wave Watch III scheduled for 18 September

        Expansion of the air quality model domain scheduled for 18 September

        Implementation of the unified GFS post scheduled for 25 September


A detailed list of upgrades and implementations can be found here, and further details on several of these implementations were provided in last months meeting summary.


The implementation of the real time mesoscale analysis (RTMA) over Alaska will likely be delayed until the 1st quarter of FY08.



2a. Global Climate and Weather Modeling Branch


Mark Iredell discussed several projects that will be worked on in the 1st quarter of FY08:


        A minor data assimilation upgrade for the GSI is scheduled for the 1st quarter of FY08 to accommodate some data format changes. Also, this upgrade will result in a fully-reproducible GSI.

        A CFS upgrade that will use 1-day old initial conditions instead of 7-day old initial conditions for the model run. The upgrade will also add two T126 members to the CFS running out to 60 days to provide more detailed guidance to CPC for their one-month outlooks.

        Testing is underway on a major GFS upgrade planned beyond the 1st quarter of FY08, including potential physics upgrades to the GFS.

        Merged GFS and ensemble calibration. This will result in more weight being placed on the operational GFS during the first three days of the forecast period when it performs best. After three days, the ensemble will be weighted more heavily.

        New products for the NAEFS to include 10% and 90% probabilities of the forecast distribution.


2b. Mesoscale Modeling Branch


Geoff DiMego reported that NAM upgrades will be tested during the 1st quarter of FY08. Additionally, the unified post code for the NAM may be implemented in January or February 2008.


2c. Global Ensemble Prediction System


No representative present.


2d. Short Range Ensemble Forecast System


The upcoming SREF upgrade will be split into two packages. The first package is scheduled for implementation on 9 November 2007 and will include bias correction, expansion of the regional spectral model (RSM) domain to cover more of Alaska and the inclusion of additional aviation output products. The bias correction scheme implemented will be the same as that used in the global NCEP ensemble while development on other methods continues. Also, bias correction of precipitation will not be included in this upgrade.


The second SREF upgrade package, tentatively scheduled for the 3rd quarter of FY08, will upgrade the WRF members to the latest version and increase the resolution of all the SREF members to as close as 32 km as possible with current computational resources (Eta SREF members are already running at 32 km).


2e. Marine Modeling and Analysis Branch


No representative present.


3. Feedback from operational centers


David Bright of SPC reminded EMC of a recent request to have hourly output from the HRW runs made available.


Joe Sienkiewicz of OPC asked about the status of verification for model track forecasts of extratropical cyclones. Geoff DiMego reported that work was ongoing, and focusing on using model analyses as verification for the cyclone tracks.


Keith Brill of HPC asked if all the precipitation type values were instantaneous in the unified post for the GFS. Mark Iredell reported they were not yet, but would eventually be changed.


Steve Silberberg of AWC stated they were examining HRW output and would be providing feedback shortly.


Stephen Jascourt noted that while investigating the large track differences for Hurricane Dean between the HWRF and GFDL models, he noticed major differences in the distribution of upper-level outflow between the models. The GFDL had outflow through a deep layer between 300 mb and 150 mb, while the HWRF outflow was restricted to 150 mb. These differences greatly impacted the upper-tropospheric flow in the environment around Dean.


4. The next synergy meeting will be held Monday, 24 September 2007, at noon EDT in Room 209 at EMC, with remote conference capability.