[Met_help] [rt.rap.ucar.edu #45262] History for question about MET

[RAL HelpDesk {for Tressa Fowler}]

----------------------------------------------------------------
  Initial Request
----------------------------------------------------------------

Dear MET help desk,

I am using MET to evaluate WRF results. I use 4 types of GDAS prebufr 
observations ( ADPUPA, ADPSFC, SFCSHP, PROFLR).

I found that there existed great differences between results of ADPUPA 
and PROFLR. For example, at around 500mb, mean bias of geopotential 
height between WRF and ADPUPA is  -3m, but it is 228.5m between WRF and 
PROFLR.

I wonder whether this is caused by great difference between ADPUPA and 
PROFLR, or it is possibly a bug in MET?

Thanks a lot !

Ming





----------------------------------------------------------------
  Complete Ticket History
----------------------------------------------------------------

Subject: question about MET
From: Tressa Fowler
Time: Fri Mar 18 11:20:09 2011


Hi Ming,

There are many possible reasons that your results could be very
different. The spatial and temporal resolution of the profiler data
and sounding data should be quite different. Also, perhaps the
sounding data was used to initialize the model? Then the errors should
be much less when comparing the model to soundings rather than
profilers.

You might want to dump out some matched pair data to eyeball (set the
last output flag to 2 in the config file), just to be sure that your
observations look like you think they should. Then you will be able to
see the observation data so you will know why your statistics look
like they do.

Please let me know if you have any further questions.

Tressa



On Thu Mar 10 16:34:07 2011, chenming at ucar.edu wrote:
> Dear MET help desk,
>
> I am using MET to evaluate WRF results. I use 4 types of GDAS
prebufr
> observations ( ADPUPA, ADPSFC, SFCSHP, PROFLR).
>
> I found that there existed great differences between results of
ADPUPA
> and PROFLR. For example, at around 500mb, mean bias of geopotential
> height between WRF and ADPUPA is  -3m, but it is 228.5m between WRF
and
> PROFLR.
>
> I wonder whether this is caused by great difference between ADPUPA
and
> PROFLR, or it is possibly a bug in MET?
>
> Thanks a lot !
>
> Ming
>
>
>



------------------------------------------------