On Fri, 2007-09-28 at 10:01 -0700, Harry Edmon wrote:
I have had a couple of users complain about changes in the GFS data
starting with the 1200 UTC run on 25 September. I notice there is a
jump up in the size of the gempak files starting with that run. Does
anyone know of any changes? If you are curious about what I am talking
about, look at this page, which uses the 0.5 degree data:
http://www.atmos.washington.edu/~hakim/loop.psp?url=~hakim/tropo/theta/
Start the loop and look at the areas in white go from smooth to noisy
with the 1200 UTC 25th run. Another user is now having problems with
his WRF initialization that he did not have before this.
I just received the following:
All,
You are receiving this message because you have been identified as a> user of
NCEP model data.
Users reported some problems with the GFS model changes which were implemented
on Sep 25 2007. NCEP's EMC and NCO have coordinated and tested a package
which addresses these problems. The results have been verified by parties in
the public and private sectors.
This improved GFS package will be implemented on Oct 10 2007 at 12Z.
The specific problems addressed are:
Grid Interpolation: A change will be made to the interpolation routine for
AWIPS and WAFS products, as well as GFS grids sets sent to public servers. The
change will reflect a more consistent bi-linear interpolation. This will
remove much of the noise observed in contours, as well as soil/land
inconsistencies.
Spurious High Pressure over Mountainous Coasts: A change to the GFS post codes
will remove a persistent, spurious high pressure system observed over Hawaii
and the Alaskan coast.
If you encounter any problems with this GFS implementation, please contact the
following PMB Data Flow helpdesk email address with your concern.
NCEP.PMB.Dataflow@xxxxxxxx
Thank you and we apologize for any inconvenience this model change may have
caused.
NCEP/NCO/PMB
End of Message.
All,
You are receiving this message because you have been identified as a
user of NCEP model data on the NCEP and/or NWS ftp servers.
The GFS model will be upgraded on Sep 25 2007. The most significant
impact of this change will result in grib1 and grib2 files which are
10-30% larger than with the current GFS. A few other changes are
occurring in the modeling system. Please see the NWS TIN which is attached.
All other changes to NCEP modeling systems can be found at the following
website:
http://www.nco.ncep.noaa.gov/pmb/changes/
Thanks,
Joey Carr
SUBJECT: UPGRADE TO NCEP GLOBAL FORECAST SYSTEM /GFS/
EFFECTIVE TUESDAY SEPTEMBER 25 2007...WITH THE 1200 COORDINATED
UNIVERSAL TIME /UTC/ RUN...THE NATIONAL CENTERS FOR ENVIRONMENTAL
PREDICTION /NCEP/ WILL UPGRADE THE GFS POST PROCESSOR.
THE PRIMARY EFFORT BEHIND THIS UPGRADE WILL BE TO UNIFY THE POST
PROCESSING CODE FOR THE NORTH AMERICAN MESO SCALE /NAM/ MODEL AND THE
GFS INTO THE NEW NCEP POST. AFTER THE UPGRADE...BOTH THE NAM AND GFS
MODELS WILL BE ABLE TO GENERATE FIELDS ORIGINALLY CREATED FOR EACH
UNIQUE MODEL.
THE PRIMARY IMPACT TO CUSTOMERS RECEIVING THE GFS WILL BE THAT THE
DEFAULT PRECIPITATION TYPE ALGORITHM WILL CHANGE FROM THE BALDWIN METHOD
TO THE DOMINANT PRECIPITATION TYPE. THE DOMINANT PRECIPITATION TYPE IS
DETERMINED BY TAKING THE PREDICTED TEMPERATURE AND MOISTURE PROFILES AND
RUNNING THEM THROUGH FIVE ALGORITHMS: BALDWIN (NCEP)...REVISED
NCEP...RAMER...BOURGOUIN
...AND AN EXPLICIT MICROPHYSICS ALGORITHM. AT EACH POINT...THE CODE
DETERMINES WHICH PRECIPITATION TYPE WAS PREDICTED BY THE MAJORITY OF THE
ALGORITHMS AND USES THAT AS THE DOMINANT TYPE. TIES ARE BROKEN IN FAVOR
OF THE MORE DANGEROUS WEATHER (ZR>SN>IP>RA).
IN ADDITION...THE SAME ALGORITHMS WILL BE USED TO DERIVE MOST FIELDS FOR
THE TWO MODELS.
THE PRODUCTS/GRIDS WHICH ARE DISSEMINATED TO NOAAPORT AND/OR THE
SATELLITE BROADCAST NETWORK /SBN/ WILL NOT BE CHANGED IN VOLUME OR CONTENT.
NCEP CENTRAL OPERATIONS IS CURRENTLY RUNNING A REAL-TIME PARALLEL.
RESULTS FROM THESE PARALLEL RUNS CAN BE VIEWED AT /USE LOWER CASE LETTERS/:
HTTP://WWW.NCO.NCEP.NOAA.GOV/PMB/NWPARA/ANALYSIS/
IF YOU HAVE ANY QUESTIONS CONCERNING THESE CHANGES...PLEASE CONTACT:
DR MARK IREDELL
NCEP/EMC...CHIEF GLOBAL MODELING BRANCH
CAMP SPRINGS MARYLAND
PHONE: 301 763 8000 X7231
EMAIL: MARK.IREDELL@xxxxxxxx
OR
KEN CAMPANA
NCEP/EMC...GLOBAL MODELING BRANCH
CAMP SPRINGS MARYLAND
PHONE: 301 763 8000 X7228
EMAIL: KENNETH.CAMPANA@xxxxxxxx
Kevin W. Thomas
Center for Analysis and Prediction of Storms
University of Oklahoma
Norman, Oklahoma
Email: kwthomas@xxxxxx
