GFS QPF bug found at hour 192

[Cory]

SENIOR DUTY METEOROLOGIST NWS ADMINISTRATIVE MESSAGE

NWS NCEP CENTRAL OPERATIONS CAMP SPRINGS MD

1526Z THU DEC 23 2010

ATTN...GFS MODEL INFORMATION..

AN INCONSISTENCY HAS BEEN UNCOVERED IN THE GFS 12 HR ACCUMULATED

PRECIPITATION AT FORECAST HOUR 192. THE 12 HR ACCUMULATION DOES

NOT REPRODUCE THE SUM OF THE PREVIOUS 2 SIX HOUR ACCUMULATIONS.

THIS ERROR WAS INTRODUCED WITH THE 27 JULY RESOLUTION INCREASE.

WE HOPE TO IMPLEMENT A CORRECTION BY THE END OF JANUARY, 2011.

AN ANNOUNCEMENT WITH THE ACTUAL IMPLEMENTATION DATE WILL BE MADE

AS SOON AS IT IS DETERMINED. QUESTIONS SHOULD BE ADDRESSED TO

[email protected]

[nevildev]

interesting... I've noticed some inconsistencies in other fields around the 192hr change (in particular RH). I'll have to check it out when I get back to work on Monday. That was an awesome upgrade but a PAIN to update all of our legacy GFS code.

[Yeoman]

Definitely a non-priority item

[dtk]

interesting... I've noticed some inconsistencies in other fields around the 192hr change (in particular RH). I'll have to check it out when I get back to work on Monday. That was an awesome upgrade but a PAIN to update all of our legacy GFS code.

The QPF issue is a post-processing thing (the model itself doesn't keep track of accumulations at multiple time frequencies).

Any other inconsistencies at that time are certainly due to the model truncating resolution (which would especially show up in fields with sharp gradients...moisture for example).

[Guest someguy]

The QPF issue is a post-processing thing (the model itself doesn't keep track of accumulations at multiple time frequencies).

Any other inconsistencies at that time are certainly due to the model truncation resolution (which would especially show up in fields with sharp gradients...moisture for example).

first thing I thought of when I saw the time frame