NSA/X1 - ECMWF Error in Derived Physical Tendencies for Temperature and Specific Humidity

Christian Jakob has found an error in the post-processing
code that creates ECMWF diagnostic files for ARM.  The sign
of the cloud water change term was wrong, and this resulted
in erroneous physical tendency terms for temperature and
specific humidity.  As a result, the advective forcing
(which is the difference between the total tendency and
physical tendency) cannot be determined.

Reprocessing is underway, and the XDC web page will be updated
to reflect the availability of corrected tendency files.

The additional information on D000726.4 received from Joyce Tichler
on 09/06/2000.

Christian Jacob has provided us with corrected tendency data for
June and July 1999 for all ARM areas. These files has been
transferred to the Archive.

Christian further reports that he is working his way through the periods 
that Ric Cederwall specified as most urgent to be corrected and will then 
try and reprocess all data since October 1999.

• Humidity, specific, tendency from phys. procs, at altitude, ECMWF model-derived(q_physics)
• Temperature - physical tendencies(T_physics)

NSA/ECMWF/X1 - Processing error in ECMWF diagnostics

A change in the production processing system at ECMWF caused these data to be corrupted.

• Wind spd, S-to-N component, tend. from phys procs & advect, @alt, ECMWF-derived(v_total)
• Temperature - total tendencies(T_total)
• Humidity, specific, tendency from phys. procs, at altitude, ECMWF model-derived(q_physics)
• Temperature - physical tendencies(T_physics)
• Wind speed, S-to-N component, tendency from phys. procs, @alt, ECMWF-derived(v_physics)
• zonal wind component - physical tendencies(u_physics)
• specific humidity - total tendencies(q_total)
• Wind spd, W-to-E component, tend. from phys procs & advect, @alt, ECMWF-derived(u_total)