NOTE: The nws-changes mailing list is no longer active. The list archives are made available for historical reasons.
-------- Original Message --------815 NOUS42 KWNO 041102
ADMNFD SENIOR DUTY METEOROLOGIST NWS ADMINISTRATIVE MESSAGE NWS NCEP CENTRAL OPERATIONS CAMP SPRINGS MD 1103Z TUE AUG 04 2009 *********************************************************** NCEP WILL BE CONDUCTING A LIVE PRODUCTION RUN FROM THE CIRRUS SUPERCOMPUTER FOR THIS UPCOMING /4 AUG/ PROD CYCLE PER SCHEDULE BELOW. PRODUCTION WILL RETURN TO DEW FOR TODAYS 18Z CYCLE. *********************************************************** THE NEW NOAA NWS NCEP SUPERCOMPUTERS...STRATUS/CIRRUS...ARE SCHEDULED TO BECOME OPERATIONAL IN AUGUST. NCEP/NCO WILL BE PERFORMING TESTS DURING THE NEXT SEVERAL WEEKS IN ORDER TO ENSURE THAT PRODUCT DELIVERY IS NOT IMPACTED ON THE IMPLEMENTATION DATE. NCEP/NCO IS PLANNING
TO DISSEMINATE PRODUCTS OPERATIONALLY FROM CIRRUS AND STRATUS ON THE DATES LISTED BELOW: CIRRUS - 12Z CYCLE TUE AUG 4TH...ONE CYCLE ONLY STRATUS - 12Z CYCLE MON AUG 10TH...ONE CYCLE ONLY IMPLEMENTATION DATE: 12Z CYCLE TUE AUG 18THAS WITH PREVIOUS UPGRADES...THE MODEL FORECASTS FROM THE NEW COMPUTER MAY BE DISCONTINUOUS FROM THOSE PRODUCED ON THE OLD COMPUTER/PREVIOUS CYCLE ESPECIALLY AFTER DAY 5. WHY? THE HARDWARE/COMPILERS USED BY THE TWO COMPUTERS DIFFER AND THE DATA DECODERS/ASSIMILATION SYSTEMS HAVE BEEN RUNNING INDEPENDENTLY ON EACH SYSTEM. THESE TWO FACTORS WILL CREATE SMALL DIFFERENCES IN MODEL INITIAL CONDITIONS CAUSED BY ASSIMILATING DIFFERENT AMOUNTS OF DATA AND USING SLIGHTLY DIFFERENT FIRST GUESS FIELDS. SMALL NON-METEOROLOGICAL DIFFERENCES EVOLVE QUICKLY INTO RANDOM NOISE - TRIGGERING CONVECTION IN DIFFERENT LOCATIONS AT DIFFERENT TIMES. THESE TYPES OF DIFFERENCES GROW QUICKLY AND TEND TO SATURATE AFTER A FEW HOURS OF MODEL TIME. AVERAGED OVER A LARGE NUMBER OF CASES THERE IS NO EVIDENCE THAT ONE MACHINE IS BETTER THAN THE OTHER...OR DIFFERENT IN ANY OTHER SYSTEMATIC WAY.
EMC HAS BEEN MONITORING THE MODEL SUITE RUNNING ON THE NEW COMPUTER AND RESULTS HAVE SHOWN THAT THERE IS NO APPRECIABLE
DIFFERENCE IN SKILL SCORE BETWEEN THE SYSTEMS. THIS IS TO BEEXPECTED SINCE THE MODELING SUITE HAS NOT CHANGED.
WE APOLOGIZE FOR ANY INCONVENIENCES THIS MAY CAUSE. ALSO...THE SCHEDULE IS HIGHLY SUBJECT TO CHANGE IN THE EVENT OF A CRITICAL WEATHER DAY DURING THE TEST PERIOD. WE WILL
STRIVE TO INFORM USERS AS SOON AS POSSIBLE IN THE EVENT OF ANY SCHEDULE CHANGES. $$ STOUDT/SDM/NCO/NCEP
nws-changes archives: