[Met_help] [rt.rap.ucar.edu #55737] History for Problem with point stat tool

[John Halley Gotway via RT]

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

***Dear all,

I received the following error message while running the point_stat tool:-*

[term at term METv3.1]# bin/point_stat ../input/eth_for.nc
../input/eth_obs.nc../input/PointStatConfig_ethiopia -outdir ../output
-v 2
DEBUG 1: Default Config File:
/home/term/MET/METv3.1/data/config/PointStatConfig_default
DEBUG 1: User Config File: ../input/PointStatConfig_ethiopia
ERROR  :
ERROR  : PointStatConfInfo::process_config() -> When the forecast file is
not GRIB1, "obs_field" cannot be blank.  You must specify the verifying
observations following the GRIB1 convention.
ERROR  :

*The observation file (eth_obs.nc) was generated using the ASCII2NC tool,
while the forecast file (eth_for.nc) was generated using the PCP-combine
tool.


Attached here with, please find the pointstat config file.

Kindly advise.

Thanks in advance,
Endalk*




-- 
-----------------------------------------------------------------------------------
Endalkachew Bekele
NOAA Climate Prediction Center / African Desk
5200 Auth Road, Rm 811, Camp Springs, MD 20746, USA
Tel: (301) 763 8000 Ext. 7014
e-mail: endalkachew.bekele at noaa.gov


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

Subject: Re: [rt.rap.ucar.edu #55737] Problem with point stat tool
From: John Halley Gotway
Time: Fri Apr 06 13:03:31 2012

Endalk,

I see the following in the Point-Stat config file you sent:
    fcst_field[] = ["APCP_24(89,99)"];
    obs_field[]  = [];

You are trying to use Point-Stat to compare the NetCDF output of PCP-
Combine to point observations generated using the ASCII2NC tool.

The short answer is... try using the following settings:

    fcst_field[] = ["APCP_24(*,*)"];
    obs_field[]  = ["APCP/A24"];

The long answer is...

When you're verifying data in a NetCDF file, you specify in the config
file the name of the NetCDF variable to use.  In your case, the NetCDF
variable is named "APCP_24".  The "(*,*)" tells MET that
those are the two dimensions for the grid.  If you'd like, you can
leave the stars off and MET will assume them.  Meaning, if you simply
put "APCP_24", MET will assume you mean "APCP_24(*,*)".  You're
telling MET the name of a NetCDF variable to use, but the software
doesn't make any assumptions about what that variable contains.  You
need to tell Point-Stat what observations should be compared to
the forecast field.  And you do that using GRIB conventions.  In your
case, that's precipitation accumulated over 24-hours, or "APCP/A24".
We chose to use the existing GRIB conventions instead of
making up a new convention on our own.

In this case, it seems obvious - the forecast NetCDF variable is named
"APCP_24" so MET should know that it's dealing with precipitation.
But in the general case, for a NetCDF variable named "data",
MET has no way of knowing what it contains.  So it's up to the user to
specify what observations should be used.

Please give that a try and let me know if you run into more issues.

Thanks,
John Halley Gotway
met_help at ucar.edu


On 04/06/2012 12:01 PM, Endalkachew Bekele via RT wrote:
>
> Fri Apr 06 12:01:31 2012: Request 55737 was acted upon.
> Transaction: Ticket created by endalkachew.bekele at noaa.gov
>         Queue: met_help
>       Subject: Problem with point stat tool
>         Owner: Nobody
>    Requestors: endalkachew.bekele at noaa.gov
>        Status: new
>   Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55737>
>
>
> ***Dear all,
>
> I received the following error message while running the point_stat
tool:-*
>
> [term at term METv3.1]# bin/point_stat ../input/eth_for.nc
> ../input/eth_obs.nc../input/PointStatConfig_ethiopia -outdir
../output
> -v 2
> DEBUG 1: Default Config File:
> /home/term/MET/METv3.1/data/config/PointStatConfig_default
> DEBUG 1: User Config File: ../input/PointStatConfig_ethiopia
> ERROR  :
> ERROR  : PointStatConfInfo::process_config() ->  When the forecast
file is
> not GRIB1, "obs_field" cannot be blank.  You must specify the
verifying
> observations following the GRIB1 convention.
> ERROR  :
>
> *The observation file (eth_obs.nc) was generated using the ASCII2NC
tool,
> while the forecast file (eth_for.nc) was generated using the PCP-
combine
> tool.
>
>
> Attached here with, please find the pointstat config file.
>
> Kindly advise.
>
> Thanks in advance,
> Endalk*
>
>
>
>

------------------------------------------------
Subject: Problem with point stat tool
From: Endalkachew Bekele
Time: Fri Apr 06 13:20:28 2012

John,
Thank you so much for your clear explanation. The tool works now
perfect.
Thanks again.
Have a great weekend,
Endalk

On Fri, Apr 6, 2012 at 3:03 PM, John Halley Gotway via RT
<met_help at ucar.edu
> wrote:

> Endalk,
>
> I see the following in the Point-Stat config file you sent:
>    fcst_field[] = ["APCP_24(89,99)"];
>    obs_field[]  = [];
>
> You are trying to use Point-Stat to compare the NetCDF output of
> PCP-Combine to point observations generated using the ASCII2NC tool.
>
> The short answer is... try using the following settings:
>
>    fcst_field[] = ["APCP_24(*,*)"];
>    obs_field[]  = ["APCP/A24"];
>
> The long answer is...
>
> When you're verifying data in a NetCDF file, you specify in the
config
> file the name of the NetCDF variable to use.  In your case, the
NetCDF
> variable is named "APCP_24".  The "(*,*)" tells MET that
> those are the two dimensions for the grid.  If you'd like, you can
leave
> the stars off and MET will assume them.  Meaning, if you simply put
> "APCP_24", MET will assume you mean "APCP_24(*,*)".  You're
> telling MET the name of a NetCDF variable to use, but the software
doesn't
> make any assumptions about what that variable contains.  You need to
tell
> Point-Stat what observations should be compared to
> the forecast field.  And you do that using GRIB conventions.  In
your
> case, that's precipitation accumulated over 24-hours, or "APCP/A24".
We
> chose to use the existing GRIB conventions instead of
> making up a new convention on our own.
>
> In this case, it seems obvious - the forecast NetCDF variable is
named
> "APCP_24" so MET should know that it's dealing with precipitation.
But in
> the general case, for a NetCDF variable named "data",
> MET has no way of knowing what it contains.  So it's up to the user
to
> specify what observations should be used.
>
> Please give that a try and let me know if you run into more issues.
>
> Thanks,
> John Halley Gotway
> met_help at ucar.edu
>
>
> On 04/06/2012 12:01 PM, Endalkachew Bekele via RT wrote:
> >
> > Fri Apr 06 12:01:31 2012: Request 55737 was acted upon.
> > Transaction: Ticket created by endalkachew.bekele at noaa.gov
> >         Queue: met_help
> >       Subject: Problem with point stat tool
> >         Owner: Nobody
> >    Requestors: endalkachew.bekele at noaa.gov
> >        Status: new
> >   Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55737>
> >
> >
> > ***Dear all,
> >
> > I received the following error message while running the
point_stat
> tool:-*
> >
> > [term at term METv3.1]# bin/point_stat ../input/eth_for.nc
> > ../input/eth_obs.nc../input/PointStatConfig_ethiopia -outdir
../output
> > -v 2
> > DEBUG 1: Default Config File:
> > /home/term/MET/METv3.1/data/config/PointStatConfig_default
> > DEBUG 1: User Config File: ../input/PointStatConfig_ethiopia
> > ERROR  :
> > ERROR  : PointStatConfInfo::process_config() ->  When the forecast
file
> is
> > not GRIB1, "obs_field" cannot be blank.  You must specify the
verifying
> > observations following the GRIB1 convention.
> > ERROR  :
> >
> > *The observation file (eth_obs.nc) was generated using the
ASCII2NC
> tool,
> > while the forecast file (eth_for.nc) was generated using the PCP-
combine
> > tool.
> >
> >
> > Attached here with, please find the pointstat config file.
> >
> > Kindly advise.
> >
> > Thanks in advance,
> > Endalk*
> >
> >
> >
> >
>
>


--
-----------------------------------------------------------------------------------
Endalkachew Bekele
NOAA Climate Prediction Center / African Desk
5200 Auth Road, Rm 811, Camp Springs, MD 20746, USA
Tel: (301) 763 8000 Ext. 7014
e-mail: endalkachew.bekele at noaa.gov

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