[Met_help] [rt.rap.ucar.edu #55008] History for Forecast valid times and point_stat tool

John Halley Gotway via RT met_help at ucar.edu
Mon Mar 12 12:19:38 MDT 2012


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

Dear MetHelp Staff,

I am doing verification of APCP_24 for a lead time of 96h using MET v3.1 
tool. As previously suggested I included D+0, D+1, D+2 and D+3 APCP_24 
fields inside the same NetCDF file, aiming to reduce number of system 
calls to point_stat tool. Apparently it runs ok since it does not report 
any warning or error. However output files don't seem as expected 
because of valid times in each row.

Command line I run is:
$ point_stat wrfprs_d02_m00_2011102600_APCP24_00-96hr_acc.nc 
precip_24h.nc PointStatConfig_APCP_24h_m00 -v 3

Where ncdump of forecast file shows:
$ ncdump -h wrfprs_d02_m00_2011102600_APCP24_00-96hr_acc.nc
netcdf wrfprs_d02_m00_2011102600_APCP24_00-96hr_acc {
dimensions:
    lat = 138 ;
    lon = 171 ;
variables:
    float lon(lat, lon) ;
        lon:units = "degrees_east" ;
        lon:long_name = "longitude" ;
        lon:standard_name = "longitude" ;
    float lat(lat, lon) ;
        lat:units = "degrees_north" ;
        lat:long_name = "latitude" ;
        lat:standard_name = "latitude" ;
    float APCP_24_f003(lat, lon) ;
        APCP_24_f003:_FillValue = -9999.f ;
        APCP_24_f003:init_time_ut = 1319587200 ;
        APCP_24_f003:valid_time_ut = 1319932800 ;
        APCP_24_f003:name = "APCP_24" ;
        APCP_24_f003:level = "A24" ;
        APCP_24_f003:valid_time = "20111030_000000" ;
        APCP_24_f003:init_time = "20111026_000000" ;
        APCP_24_f003:long_name = "Total precipitation" ;
        APCP_24_f003:accum_time_sec = 86400 ;
        APCP_24_f003:accum_time = "240000" ;
        APCP_24_f003:units = "kg/m^2" ;
        APCP_24_f003:grib_code = 61 ;
    float APCP_24_f002(lat, lon) ;
        APCP_24_f002:_FillValue = -9999.f ;
        APCP_24_f002:init_time_ut = 1319587200 ;
        APCP_24_f002:valid_time_ut = 1319846400 ;
        APCP_24_f002:name = "APCP_24" ;
        APCP_24_f002:level = "A24" ;
        APCP_24_f002:valid_time = "20111029_000000" ;
        APCP_24_f002:init_time = "20111026_000000" ;
        APCP_24_f002:long_name = "Total precipitation" ;
        APCP_24_f002:accum_time_sec = 86400 ;
        APCP_24_f002:accum_time = "240000" ;
        APCP_24_f002:units = "kg/m^2" ;
        APCP_24_f002:grib_code = 61 ;
    float APCP_24_f001(lat, lon) ;
        APCP_24_f001:_FillValue = -9999.f ;
        APCP_24_f001:init_time_ut = 1319587200 ;
        APCP_24_f001:valid_time_ut = 1319760000 ;
        APCP_24_f001:name = "APCP_24" ;
        APCP_24_f001:level = "A24" ;
        APCP_24_f001:valid_time = "20111028_000000" ;
        APCP_24_f001:init_time = "20111026_000000" ;
        APCP_24_f001:long_name = "Total precipitation" ;
        APCP_24_f001:accum_time_sec = 86400 ;
        APCP_24_f001:accum_time = "240000" ;
        APCP_24_f001:units = "kg/m^2" ;
        APCP_24_f001:grib_code = 61 ;
    float APCP_24_f000(lat, lon) ;
        APCP_24_f000:_FillValue = -9999.f ;
        APCP_24_f000:init_time_ut = 1319587200 ;
        APCP_24_f000:valid_time_ut = 1319673600 ;
        APCP_24_f000:name = "APCP_24" ;
        APCP_24_f000:level = "A24" ;
        APCP_24_f000:valid_time = "20111027_000000" ;
        APCP_24_f000:init_time = "20111026_000000" ;
        APCP_24_f000:long_name = "Total precipitation" ;
        APCP_24_f000:accum_time_sec = 86400 ;
        APCP_24_f000:accum_time = "240000" ;
        APCP_24_f000:units = "kg/m^2" ;
        APCP_24_f000:grib_code = 61 ;

// global attributes:
        :nx = "171" ;
        :Projection = "Lambert Conformal" ;
        :FileOrigins = "File 
wrfprs_d02_m00_2011102600_APCP24_72-96hr_acc.nc generated 
20120308_092036 UTC on host sor-num02 by the MET pcp_combine tool" ;
        :x_pin = "0.000000" ;
        :lon_orient = "-14.100000" ;
        :MET_version = "V3.1" ;
        :scale_lat_2 = "43.000000" ;
        :ny = "138 grid_points" ;
        :lat_pin = "34.823000" ;
        :scale_lat_1 = "43.000000" ;
        :d_km = "12.000000" ;
        :r_km = "6371.200000" ;
        :lon_pin = "-19.911000" ;
        :y_pin = "0.000000" ;
        :MET_tool = "pcp_combine" ;
        :RunCommand = "Subtraction: 
wrfprs_d02_m00_2011102600_APCP_0-96hr_acc.grib with accumulation of 
960000 minus wrfprs_d02_m00_2011102600_APCP_0-72hr_acc.grib with 
accumulation of 720000." ;
        :history = "08/03/2012 as 09:20:42 -- Creado a partir de 
wrfprs_d02_m00_2011102600_APCP24_72-96hr_acc.nc\n",
"08/03/2012 as 09:20:42 -- Engadidas as variables APCP_24 aos arquivos: \n",
            "\twrfprs_d02_m00_2011102600_APCP24_00-24hr_acc.nc\n",
            "\twrfprs_d02_m00_2011102600_APCP24_24-48hr_acc.nc\n",
            "\twrfprs_d02_m00_2011102600_APCP24_48-72hr_acc.nc\n",
            "\twrfprs_d02_m00_2011102600_APCP24_72-96hr_acc.nc" ;
}

and observation file has been obtained from ascii2nc tool, containing 
data from 10/26/2011 to 10/30/2011. Configuration file has rows:
fcst_field[] = [ "APCP_24_f000/A24", "APCP_24_f001/A24", 
"APCP_24_f002/A24", "APCP_24_f003/A24" ];
obs_field[] = [ "APCP_24/A24", "APCP_24/A24", "APCP_24/A24", 
"APCP_24/A24" ];

either beg_ds and end_ds are set to 54000 (minus when needed).

Output files has only FCST_VALID 20111027. On CNT (as exemple) edited 
lines are:
VERSION MODEL       FCST_LEAD FCST_VALID_BEG  FCST_VALID_END  OBS_LEAD 
OBS_VALID_BEG   OBS_VALID_END   FCST_VAR FCST_LEV OBS_VAR OBS_LEV
V3.1    WRF_ens_m00 240000    20111027_000000 20111027_000000 000000   
20111026_223000 20111027_013000 APCP_24  A24      APCP_24 A24    
V3.1    WRF_ens_m00 240000    20111027_000000 20111027_000000 000000   
20111026_223000 20111027_013000 APCP_24  A24      APCP_24 A24    
V3.1    WRF_ens_m00 240000    20111027_000000 20111027_000000 000000   
20111026_223000 20111027_013000 APCP_24  A24      APCP_24 A24    
V3.1    WRF_ens_m00 240000    20111027_000000 20111027_000000 000000   
20111026_223000 20111027_013000 APCP_24  A24      APCP_24 A24    

It seems as if point_stat first read APCP_24_f000 forecast valid time 
and maintain it through _f001, _f002 and _f003 calls. All four rows are 
different each other in some fields indeed.

For all four fields as separated files using: fcst_field[] = [ 
"APCP_24_f00n/A24" ] with 0<=n<=3, FCST_VALID maintain to 20111027.

Is there something I'm doing wrong? I checked it several times and don't 
identify any error. Any suggestion?

Thanks in advance,

Víctor M. Álvarez
Numerical Department - MeteoGalicia
Consellería de Medio Ambiente, Territorio e Infraestruturas - Xunta de 
Galicia
Santiago de Compostela - Spain




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

Subject: Re: [rt.rap.ucar.edu #55008] Forecast valid times and point_stat tool
From: John Halley Gotway
Time: Thu Mar 08 08:50:46 2012

Victor,

The easiest way to debug this issue would be for you to send me sample
files to work with.  Please follow the instructions posted here:
    http://www.dtcenter.org/met/users/support/met_help.php#ftp

Please include:
  - your NetCDF forecast file named
"wrfprs_d02_m00_2011102600_APCP24_00-96hr_acc.nc"
  - your NetCDF observation file named "precip_24h.nc"
  - your Point-Stat config file named "PointStatConfig_APCP_24h_m00"

I'll take a look and see if I can figure it out.

Thanks,
John

On 03/08/2012 06:42 AM, Victor M. Alvarez via RT wrote:
>
> Thu Mar 08 06:42:08 2012: Request 55008 was acted upon.
> Transaction: Ticket created by victor.alvarez at meteogalicia.es
>         Queue: met_help
>       Subject: Forecast valid times and point_stat tool
>         Owner: Nobody
>    Requestors: victor.alvarez at meteogalicia.es
>        Status: new
>   Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55008>
>
>
> Dear MetHelp Staff,
>
> I am doing verification of APCP_24 for a lead time of 96h using MET
v3.1
> tool. As previously suggested I included D+0, D+1, D+2 and D+3
APCP_24
> fields inside the same NetCDF file, aiming to reduce number of
system
> calls to point_stat tool. Apparently it runs ok since it does not
report
> any warning or error. However output files don't seem as expected
> because of valid times in each row.
>
> Command line I run is:
> $ point_stat wrfprs_d02_m00_2011102600_APCP24_00-96hr_acc.nc
> precip_24h.nc PointStatConfig_APCP_24h_m00 -v 3
>
> Where ncdump of forecast file shows:
> $ ncdump -h wrfprs_d02_m00_2011102600_APCP24_00-96hr_acc.nc
> netcdf wrfprs_d02_m00_2011102600_APCP24_00-96hr_acc {
> dimensions:
>      lat = 138 ;
>      lon = 171 ;
> variables:
>      float lon(lat, lon) ;
>          lon:units = "degrees_east" ;
>          lon:long_name = "longitude" ;
>          lon:standard_name = "longitude" ;
>      float lat(lat, lon) ;
>          lat:units = "degrees_north" ;
>          lat:long_name = "latitude" ;
>          lat:standard_name = "latitude" ;
>      float APCP_24_f003(lat, lon) ;
>          APCP_24_f003:_FillValue = -9999.f ;
>          APCP_24_f003:init_time_ut = 1319587200 ;
>          APCP_24_f003:valid_time_ut = 1319932800 ;
>          APCP_24_f003:name = "APCP_24" ;
>          APCP_24_f003:level = "A24" ;
>          APCP_24_f003:valid_time = "20111030_000000" ;
>          APCP_24_f003:init_time = "20111026_000000" ;
>          APCP_24_f003:long_name = "Total precipitation" ;
>          APCP_24_f003:accum_time_sec = 86400 ;
>          APCP_24_f003:accum_time = "240000" ;
>          APCP_24_f003:units = "kg/m^2" ;
>          APCP_24_f003:grib_code = 61 ;
>      float APCP_24_f002(lat, lon) ;
>          APCP_24_f002:_FillValue = -9999.f ;
>          APCP_24_f002:init_time_ut = 1319587200 ;
>          APCP_24_f002:valid_time_ut = 1319846400 ;
>          APCP_24_f002:name = "APCP_24" ;
>          APCP_24_f002:level = "A24" ;
>          APCP_24_f002:valid_time = "20111029_000000" ;
>          APCP_24_f002:init_time = "20111026_000000" ;
>          APCP_24_f002:long_name = "Total precipitation" ;
>          APCP_24_f002:accum_time_sec = 86400 ;
>          APCP_24_f002:accum_time = "240000" ;
>          APCP_24_f002:units = "kg/m^2" ;
>          APCP_24_f002:grib_code = 61 ;
>      float APCP_24_f001(lat, lon) ;
>          APCP_24_f001:_FillValue = -9999.f ;
>          APCP_24_f001:init_time_ut = 1319587200 ;
>          APCP_24_f001:valid_time_ut = 1319760000 ;
>          APCP_24_f001:name = "APCP_24" ;
>          APCP_24_f001:level = "A24" ;
>          APCP_24_f001:valid_time = "20111028_000000" ;
>          APCP_24_f001:init_time = "20111026_000000" ;
>          APCP_24_f001:long_name = "Total precipitation" ;
>          APCP_24_f001:accum_time_sec = 86400 ;
>          APCP_24_f001:accum_time = "240000" ;
>          APCP_24_f001:units = "kg/m^2" ;
>          APCP_24_f001:grib_code = 61 ;
>      float APCP_24_f000(lat, lon) ;
>          APCP_24_f000:_FillValue = -9999.f ;
>          APCP_24_f000:init_time_ut = 1319587200 ;
>          APCP_24_f000:valid_time_ut = 1319673600 ;
>          APCP_24_f000:name = "APCP_24" ;
>          APCP_24_f000:level = "A24" ;
>          APCP_24_f000:valid_time = "20111027_000000" ;
>          APCP_24_f000:init_time = "20111026_000000" ;
>          APCP_24_f000:long_name = "Total precipitation" ;
>          APCP_24_f000:accum_time_sec = 86400 ;
>          APCP_24_f000:accum_time = "240000" ;
>          APCP_24_f000:units = "kg/m^2" ;
>          APCP_24_f000:grib_code = 61 ;
>
> // global attributes:
>          :nx = "171" ;
>          :Projection = "Lambert Conformal" ;
>          :FileOrigins = "File
> wrfprs_d02_m00_2011102600_APCP24_72-96hr_acc.nc generated
> 20120308_092036 UTC on host sor-num02 by the MET pcp_combine tool" ;
>          :x_pin = "0.000000" ;
>          :lon_orient = "-14.100000" ;
>          :MET_version = "V3.1" ;
>          :scale_lat_2 = "43.000000" ;
>          :ny = "138 grid_points" ;
>          :lat_pin = "34.823000" ;
>          :scale_lat_1 = "43.000000" ;
>          :d_km = "12.000000" ;
>          :r_km = "6371.200000" ;
>          :lon_pin = "-19.911000" ;
>          :y_pin = "0.000000" ;
>          :MET_tool = "pcp_combine" ;
>          :RunCommand = "Subtraction:
> wrfprs_d02_m00_2011102600_APCP_0-96hr_acc.grib with accumulation of
> 960000 minus wrfprs_d02_m00_2011102600_APCP_0-72hr_acc.grib with
> accumulation of 720000." ;
>          :history = "08/03/2012 as 09:20:42 -- Creado a partir de
> wrfprs_d02_m00_2011102600_APCP24_72-96hr_acc.nc\n",
> "08/03/2012 as 09:20:42 -- Engadidas as variables APCP_24 aos
arquivos: \n",
>              "\twrfprs_d02_m00_2011102600_APCP24_00-24hr_acc.nc\n",
>              "\twrfprs_d02_m00_2011102600_APCP24_24-48hr_acc.nc\n",
>              "\twrfprs_d02_m00_2011102600_APCP24_48-72hr_acc.nc\n",
>              "\twrfprs_d02_m00_2011102600_APCP24_72-96hr_acc.nc" ;
> }
>
> and observation file has been obtained from ascii2nc tool,
containing
> data from 10/26/2011 to 10/30/2011. Configuration file has rows:
> fcst_field[] = [ "APCP_24_f000/A24", "APCP_24_f001/A24",
> "APCP_24_f002/A24", "APCP_24_f003/A24" ];
> obs_field[] = [ "APCP_24/A24", "APCP_24/A24", "APCP_24/A24",
> "APCP_24/A24" ];
>
> either beg_ds and end_ds are set to 54000 (minus when needed).
>
> Output files has only FCST_VALID 20111027. On CNT (as exemple)
edited
> lines are:
> VERSION MODEL       FCST_LEAD FCST_VALID_BEG  FCST_VALID_END
OBS_LEAD
> OBS_VALID_BEG   OBS_VALID_END   FCST_VAR FCST_LEV OBS_VAR OBS_LEV
> V3.1    WRF_ens_m00 240000    20111027_000000 20111027_000000 000000
> 20111026_223000 20111027_013000 APCP_24  A24      APCP_24 A24
> V3.1    WRF_ens_m00 240000    20111027_000000 20111027_000000 000000
> 20111026_223000 20111027_013000 APCP_24  A24      APCP_24 A24
> V3.1    WRF_ens_m00 240000    20111027_000000 20111027_000000 000000
> 20111026_223000 20111027_013000 APCP_24  A24      APCP_24 A24
> V3.1    WRF_ens_m00 240000    20111027_000000 20111027_000000 000000
> 20111026_223000 20111027_013000 APCP_24  A24      APCP_24 A24
>
> It seems as if point_stat first read APCP_24_f000 forecast valid
time
> and maintain it through _f001, _f002 and _f003 calls. All four rows
are
> different each other in some fields indeed.
>
> For all four fields as separated files using: fcst_field[] = [
> "APCP_24_f00n/A24" ] with 0<=n<=3, FCST_VALID maintain to 20111027.
>
> Is there something I'm doing wrong? I checked it several times and
don't
> identify any error. Any suggestion?
>
> Thanks in advance,
>
> Víctor M. Álvarez
> Numerical Department - MeteoGalicia
> Consellería de Medio Ambiente, Territorio e Infraestruturas - Xunta
de
> Galicia
> Santiago de Compostela - Spain
>
>

------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #55008] Forecast valid times and point_stat tool
From: Victor M. Alvarez
Time: Fri Mar 09 05:18:19 2012

John,

it's done. Path to files is: 'incoming/irap/met_help/alvarez.victor'.
'met_help' directory did not exist so I make it.

I also uploaded ascii file 'precip_24h.obs'.

Thanks,
Victor.


John Halley Gotway via RT escribió:
> Victor,
>
> The easiest way to debug this issue would be for you to send me
sample files to work with.  Please follow the instructions posted
here:
>     http://www.dtcenter.org/met/users/support/met_help.php#ftp
>
> Please include:
>   - your NetCDF forecast file named
"wrfprs_d02_m00_2011102600_APCP24_00-96hr_acc.nc"
>   - your NetCDF observation file named "precip_24h.nc"
>   - your Point-Stat config file named "PointStatConfig_APCP_24h_m00"
>
> I'll take a look and see if I can figure it out.
>
> Thanks,
> John
>
> On 03/08/2012 06:42 AM, Victor M. Alvarez via RT wrote:
>
>> Thu Mar 08 06:42:08 2012: Request 55008 was acted upon.
>> Transaction: Ticket created by victor.alvarez at meteogalicia.es
>>         Queue: met_help
>>       Subject: Forecast valid times and point_stat tool
>>         Owner: Nobody
>>    Requestors: victor.alvarez at meteogalicia.es
>>        Status: new
>>   Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55008>
>>
>>
>> Dear MetHelp Staff,
>>
>> I am doing verification of APCP_24 for a lead time of 96h using MET
v3.1
>> tool. As previously suggested I included D+0, D+1, D+2 and D+3
APCP_24
>> fields inside the same NetCDF file, aiming to reduce number of
system
>> calls to point_stat tool. Apparently it runs ok since it does not
report
>> any warning or error. However output files don't seem as expected
>> because of valid times in each row.
>>
>> Command line I run is:
>> $ point_stat wrfprs_d02_m00_2011102600_APCP24_00-96hr_acc.nc
>> precip_24h.nc PointStatConfig_APCP_24h_m00 -v 3
>>
>> Where ncdump of forecast file shows:
>> $ ncdump -h wrfprs_d02_m00_2011102600_APCP24_00-96hr_acc.nc
>> netcdf wrfprs_d02_m00_2011102600_APCP24_00-96hr_acc {
>> dimensions:
>>      lat = 138 ;
>>      lon = 171 ;
>> variables:
>>      float lon(lat, lon) ;
>>          lon:units = "degrees_east" ;
>>          lon:long_name = "longitude" ;
>>          lon:standard_name = "longitude" ;
>>      float lat(lat, lon) ;
>>          lat:units = "degrees_north" ;
>>          lat:long_name = "latitude" ;
>>          lat:standard_name = "latitude" ;
>>      float APCP_24_f003(lat, lon) ;
>>          APCP_24_f003:_FillValue = -9999.f ;
>>          APCP_24_f003:init_time_ut = 1319587200 ;
>>          APCP_24_f003:valid_time_ut = 1319932800 ;
>>          APCP_24_f003:name = "APCP_24" ;
>>          APCP_24_f003:level = "A24" ;
>>          APCP_24_f003:valid_time = "20111030_000000" ;
>>          APCP_24_f003:init_time = "20111026_000000" ;
>>          APCP_24_f003:long_name = "Total precipitation" ;
>>          APCP_24_f003:accum_time_sec = 86400 ;
>>          APCP_24_f003:accum_time = "240000" ;
>>          APCP_24_f003:units = "kg/m^2" ;
>>          APCP_24_f003:grib_code = 61 ;
>>      float APCP_24_f002(lat, lon) ;
>>          APCP_24_f002:_FillValue = -9999.f ;
>>          APCP_24_f002:init_time_ut = 1319587200 ;
>>          APCP_24_f002:valid_time_ut = 1319846400 ;
>>          APCP_24_f002:name = "APCP_24" ;
>>          APCP_24_f002:level = "A24" ;
>>          APCP_24_f002:valid_time = "20111029_000000" ;
>>          APCP_24_f002:init_time = "20111026_000000" ;
>>          APCP_24_f002:long_name = "Total precipitation" ;
>>          APCP_24_f002:accum_time_sec = 86400 ;
>>          APCP_24_f002:accum_time = "240000" ;
>>          APCP_24_f002:units = "kg/m^2" ;
>>          APCP_24_f002:grib_code = 61 ;
>>      float APCP_24_f001(lat, lon) ;
>>          APCP_24_f001:_FillValue = -9999.f ;
>>          APCP_24_f001:init_time_ut = 1319587200 ;
>>          APCP_24_f001:valid_time_ut = 1319760000 ;
>>          APCP_24_f001:name = "APCP_24" ;
>>          APCP_24_f001:level = "A24" ;
>>          APCP_24_f001:valid_time = "20111028_000000" ;
>>          APCP_24_f001:init_time = "20111026_000000" ;
>>          APCP_24_f001:long_name = "Total precipitation" ;
>>          APCP_24_f001:accum_time_sec = 86400 ;
>>          APCP_24_f001:accum_time = "240000" ;
>>          APCP_24_f001:units = "kg/m^2" ;
>>          APCP_24_f001:grib_code = 61 ;
>>      float APCP_24_f000(lat, lon) ;
>>          APCP_24_f000:_FillValue = -9999.f ;
>>          APCP_24_f000:init_time_ut = 1319587200 ;
>>          APCP_24_f000:valid_time_ut = 1319673600 ;
>>          APCP_24_f000:name = "APCP_24" ;
>>          APCP_24_f000:level = "A24" ;
>>          APCP_24_f000:valid_time = "20111027_000000" ;
>>          APCP_24_f000:init_time = "20111026_000000" ;
>>          APCP_24_f000:long_name = "Total precipitation" ;
>>          APCP_24_f000:accum_time_sec = 86400 ;
>>          APCP_24_f000:accum_time = "240000" ;
>>          APCP_24_f000:units = "kg/m^2" ;
>>          APCP_24_f000:grib_code = 61 ;
>>
>> // global attributes:
>>          :nx = "171" ;
>>          :Projection = "Lambert Conformal" ;
>>          :FileOrigins = "File
>> wrfprs_d02_m00_2011102600_APCP24_72-96hr_acc.nc generated
>> 20120308_092036 UTC on host sor-num02 by the MET pcp_combine tool"
;
>>          :x_pin = "0.000000" ;
>>          :lon_orient = "-14.100000" ;
>>          :MET_version = "V3.1" ;
>>          :scale_lat_2 = "43.000000" ;
>>          :ny = "138 grid_points" ;
>>          :lat_pin = "34.823000" ;
>>          :scale_lat_1 = "43.000000" ;
>>          :d_km = "12.000000" ;
>>          :r_km = "6371.200000" ;
>>          :lon_pin = "-19.911000" ;
>>          :y_pin = "0.000000" ;
>>          :MET_tool = "pcp_combine" ;
>>          :RunCommand = "Subtraction:
>> wrfprs_d02_m00_2011102600_APCP_0-96hr_acc.grib with accumulation of
>> 960000 minus wrfprs_d02_m00_2011102600_APCP_0-72hr_acc.grib with
>> accumulation of 720000." ;
>>          :history = "08/03/2012 as 09:20:42 -- Creado a partir de
>> wrfprs_d02_m00_2011102600_APCP24_72-96hr_acc.nc\n",
>> "08/03/2012 as 09:20:42 -- Engadidas as variables APCP_24 aos
arquivos: \n",
>>              "\twrfprs_d02_m00_2011102600_APCP24_00-24hr_acc.nc\n",
>>              "\twrfprs_d02_m00_2011102600_APCP24_24-48hr_acc.nc\n",
>>              "\twrfprs_d02_m00_2011102600_APCP24_48-72hr_acc.nc\n",
>>              "\twrfprs_d02_m00_2011102600_APCP24_72-96hr_acc.nc" ;
>> }
>>
>> and observation file has been obtained from ascii2nc tool,
containing
>> data from 10/26/2011 to 10/30/2011. Configuration file has rows:
>> fcst_field[] = [ "APCP_24_f000/A24", "APCP_24_f001/A24",
>> "APCP_24_f002/A24", "APCP_24_f003/A24" ];
>> obs_field[] = [ "APCP_24/A24", "APCP_24/A24", "APCP_24/A24",
>> "APCP_24/A24" ];
>>
>> either beg_ds and end_ds are set to 54000 (minus when needed).
>>
>> Output files has only FCST_VALID 20111027. On CNT (as exemple)
edited
>> lines are:
>> VERSION MODEL       FCST_LEAD FCST_VALID_BEG  FCST_VALID_END
OBS_LEAD
>> OBS_VALID_BEG   OBS_VALID_END   FCST_VAR FCST_LEV OBS_VAR OBS_LEV
>> V3.1    WRF_ens_m00 240000    20111027_000000 20111027_000000
000000
>> 20111026_223000 20111027_013000 APCP_24  A24      APCP_24 A24
>> V3.1    WRF_ens_m00 240000    20111027_000000 20111027_000000
000000
>> 20111026_223000 20111027_013000 APCP_24  A24      APCP_24 A24
>> V3.1    WRF_ens_m00 240000    20111027_000000 20111027_000000
000000
>> 20111026_223000 20111027_013000 APCP_24  A24      APCP_24 A24
>> V3.1    WRF_ens_m00 240000    20111027_000000 20111027_000000
000000
>> 20111026_223000 20111027_013000 APCP_24  A24      APCP_24 A24
>>
>> It seems as if point_stat first read APCP_24_f000 forecast valid
time
>> and maintain it through _f001, _f002 and _f003 calls. All four rows
are
>> different each other in some fields indeed.
>>
>> For all four fields as separated files using: fcst_field[] = [
>> "APCP_24_f00n/A24" ] with 0<=n<=3, FCST_VALID maintain to 20111027.
>>
>> Is there something I'm doing wrong? I checked it several times and
don't
>> identify any error. Any suggestion?
>>
>> Thanks in advance,
>>
>> Víctor M. Álvarez
>> Numerical Department - MeteoGalicia
>> Consellería de Medio Ambiente, Territorio e Infraestruturas - Xunta
de
>> Galicia
>> Santiago de Compostela - Spain
>>
>>
>>
>
>


------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #55008] Forecast valid times and point_stat tool
From: John Halley Gotway
Time: Fri Mar 09 11:56:52 2012

Victor,

Thanks for sending your data.

Unfortunately, I have bad news.  Upon further investigation into the
code, I found the following assumption:
   Our code for reading MET NetCDF files assumes that all variables in
the file contain the same timing information.

That's why you're seeing identical timing information in all of the
output files.  I apologize for our suggestion of going this route in
the first place!  But in order to use the MET tools, you'll
need to organize your forecast NetCDF files with one NetCDF file for
each output time.  I understand that that may lead to an unmanageable
number of NetCDF files for you, but unfortunately, there's no
way around it in METv3.1.

Sorry for the bad news.

Thanks,
John

On 03/09/2012 05:18 AM, Victor M. Alvarez via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55008>
>
> John,
>
> it's done. Path to files is:
'incoming/irap/met_help/alvarez.victor'.
> 'met_help' directory did not exist so I make it.
>
> I also uploaded ascii file 'precip_24h.obs'.
>
> Thanks,
> Victor.
>
>
> John Halley Gotway via RT escribió:
>> Victor,
>>
>> The easiest way to debug this issue would be for you to send me
sample files to work with.  Please follow the instructions posted
here:
>>      http://www.dtcenter.org/met/users/support/met_help.php#ftp
>>
>> Please include:
>>    - your NetCDF forecast file named
"wrfprs_d02_m00_2011102600_APCP24_00-96hr_acc.nc"
>>    - your NetCDF observation file named "precip_24h.nc"
>>    - your Point-Stat config file named
"PointStatConfig_APCP_24h_m00"
>>
>> I'll take a look and see if I can figure it out.
>>
>> Thanks,
>> John
>>
>> On 03/08/2012 06:42 AM, Victor M. Alvarez via RT wrote:
>>
>>> Thu Mar 08 06:42:08 2012: Request 55008 was acted upon.
>>> Transaction: Ticket created by victor.alvarez at meteogalicia.es
>>>          Queue: met_help
>>>        Subject: Forecast valid times and point_stat tool
>>>          Owner: Nobody
>>>     Requestors: victor.alvarez at meteogalicia.es
>>>         Status: new
>>>    Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55008>
>>>
>>>
>>> Dear MetHelp Staff,
>>>
>>> I am doing verification of APCP_24 for a lead time of 96h using
MET v3.1
>>> tool. As previously suggested I included D+0, D+1, D+2 and D+3
APCP_24
>>> fields inside the same NetCDF file, aiming to reduce number of
system
>>> calls to point_stat tool. Apparently it runs ok since it does not
report
>>> any warning or error. However output files don't seem as expected
>>> because of valid times in each row.
>>>
>>> Command line I run is:
>>> $ point_stat wrfprs_d02_m00_2011102600_APCP24_00-96hr_acc.nc
>>> precip_24h.nc PointStatConfig_APCP_24h_m00 -v 3
>>>
>>> Where ncdump of forecast file shows:
>>> $ ncdump -h wrfprs_d02_m00_2011102600_APCP24_00-96hr_acc.nc
>>> netcdf wrfprs_d02_m00_2011102600_APCP24_00-96hr_acc {
>>> dimensions:
>>>       lat = 138 ;
>>>       lon = 171 ;
>>> variables:
>>>       float lon(lat, lon) ;
>>>           lon:units = "degrees_east" ;
>>>           lon:long_name = "longitude" ;
>>>           lon:standard_name = "longitude" ;
>>>       float lat(lat, lon) ;
>>>           lat:units = "degrees_north" ;
>>>           lat:long_name = "latitude" ;
>>>           lat:standard_name = "latitude" ;
>>>       float APCP_24_f003(lat, lon) ;
>>>           APCP_24_f003:_FillValue = -9999.f ;
>>>           APCP_24_f003:init_time_ut = 1319587200 ;
>>>           APCP_24_f003:valid_time_ut = 1319932800 ;
>>>           APCP_24_f003:name = "APCP_24" ;
>>>           APCP_24_f003:level = "A24" ;
>>>           APCP_24_f003:valid_time = "20111030_000000" ;
>>>           APCP_24_f003:init_time = "20111026_000000" ;
>>>           APCP_24_f003:long_name = "Total precipitation" ;
>>>           APCP_24_f003:accum_time_sec = 86400 ;
>>>           APCP_24_f003:accum_time = "240000" ;
>>>           APCP_24_f003:units = "kg/m^2" ;
>>>           APCP_24_f003:grib_code = 61 ;
>>>       float APCP_24_f002(lat, lon) ;
>>>           APCP_24_f002:_FillValue = -9999.f ;
>>>           APCP_24_f002:init_time_ut = 1319587200 ;
>>>           APCP_24_f002:valid_time_ut = 1319846400 ;
>>>           APCP_24_f002:name = "APCP_24" ;
>>>           APCP_24_f002:level = "A24" ;
>>>           APCP_24_f002:valid_time = "20111029_000000" ;
>>>           APCP_24_f002:init_time = "20111026_000000" ;
>>>           APCP_24_f002:long_name = "Total precipitation" ;
>>>           APCP_24_f002:accum_time_sec = 86400 ;
>>>           APCP_24_f002:accum_time = "240000" ;
>>>           APCP_24_f002:units = "kg/m^2" ;
>>>           APCP_24_f002:grib_code = 61 ;
>>>       float APCP_24_f001(lat, lon) ;
>>>           APCP_24_f001:_FillValue = -9999.f ;
>>>           APCP_24_f001:init_time_ut = 1319587200 ;
>>>           APCP_24_f001:valid_time_ut = 1319760000 ;
>>>           APCP_24_f001:name = "APCP_24" ;
>>>           APCP_24_f001:level = "A24" ;
>>>           APCP_24_f001:valid_time = "20111028_000000" ;
>>>           APCP_24_f001:init_time = "20111026_000000" ;
>>>           APCP_24_f001:long_name = "Total precipitation" ;
>>>           APCP_24_f001:accum_time_sec = 86400 ;
>>>           APCP_24_f001:accum_time = "240000" ;
>>>           APCP_24_f001:units = "kg/m^2" ;
>>>           APCP_24_f001:grib_code = 61 ;
>>>       float APCP_24_f000(lat, lon) ;
>>>           APCP_24_f000:_FillValue = -9999.f ;
>>>           APCP_24_f000:init_time_ut = 1319587200 ;
>>>           APCP_24_f000:valid_time_ut = 1319673600 ;
>>>           APCP_24_f000:name = "APCP_24" ;
>>>           APCP_24_f000:level = "A24" ;
>>>           APCP_24_f000:valid_time = "20111027_000000" ;
>>>           APCP_24_f000:init_time = "20111026_000000" ;
>>>           APCP_24_f000:long_name = "Total precipitation" ;
>>>           APCP_24_f000:accum_time_sec = 86400 ;
>>>           APCP_24_f000:accum_time = "240000" ;
>>>           APCP_24_f000:units = "kg/m^2" ;
>>>           APCP_24_f000:grib_code = 61 ;
>>>
>>> // global attributes:
>>>           :nx = "171" ;
>>>           :Projection = "Lambert Conformal" ;
>>>           :FileOrigins = "File
>>> wrfprs_d02_m00_2011102600_APCP24_72-96hr_acc.nc generated
>>> 20120308_092036 UTC on host sor-num02 by the MET pcp_combine tool"
;
>>>           :x_pin = "0.000000" ;
>>>           :lon_orient = "-14.100000" ;
>>>           :MET_version = "V3.1" ;
>>>           :scale_lat_2 = "43.000000" ;
>>>           :ny = "138 grid_points" ;
>>>           :lat_pin = "34.823000" ;
>>>           :scale_lat_1 = "43.000000" ;
>>>           :d_km = "12.000000" ;
>>>           :r_km = "6371.200000" ;
>>>           :lon_pin = "-19.911000" ;
>>>           :y_pin = "0.000000" ;
>>>           :MET_tool = "pcp_combine" ;
>>>           :RunCommand = "Subtraction:
>>> wrfprs_d02_m00_2011102600_APCP_0-96hr_acc.grib with accumulation
of
>>> 960000 minus wrfprs_d02_m00_2011102600_APCP_0-72hr_acc.grib with
>>> accumulation of 720000." ;
>>>           :history = "08/03/2012 as 09:20:42 -- Creado a partir de
>>> wrfprs_d02_m00_2011102600_APCP24_72-96hr_acc.nc\n",
>>> "08/03/2012 as 09:20:42 -- Engadidas as variables APCP_24 aos
arquivos: \n",
>>>               "\twrfprs_d02_m00_2011102600_APCP24_00-
24hr_acc.nc\n",
>>>               "\twrfprs_d02_m00_2011102600_APCP24_24-
48hr_acc.nc\n",
>>>               "\twrfprs_d02_m00_2011102600_APCP24_48-
72hr_acc.nc\n",
>>>               "\twrfprs_d02_m00_2011102600_APCP24_72-96hr_acc.nc"
;
>>> }
>>>
>>> and observation file has been obtained from ascii2nc tool,
containing
>>> data from 10/26/2011 to 10/30/2011. Configuration file has rows:
>>> fcst_field[] = [ "APCP_24_f000/A24", "APCP_24_f001/A24",
>>> "APCP_24_f002/A24", "APCP_24_f003/A24" ];
>>> obs_field[] = [ "APCP_24/A24", "APCP_24/A24", "APCP_24/A24",
>>> "APCP_24/A24" ];
>>>
>>> either beg_ds and end_ds are set to 54000 (minus when needed).
>>>
>>> Output files has only FCST_VALID 20111027. On CNT (as exemple)
edited
>>> lines are:
>>> VERSION MODEL       FCST_LEAD FCST_VALID_BEG  FCST_VALID_END
OBS_LEAD
>>> OBS_VALID_BEG   OBS_VALID_END   FCST_VAR FCST_LEV OBS_VAR OBS_LEV
>>> V3.1    WRF_ens_m00 240000    20111027_000000 20111027_000000
000000
>>> 20111026_223000 20111027_013000 APCP_24  A24      APCP_24 A24
>>> V3.1    WRF_ens_m00 240000    20111027_000000 20111027_000000
000000
>>> 20111026_223000 20111027_013000 APCP_24  A24      APCP_24 A24
>>> V3.1    WRF_ens_m00 240000    20111027_000000 20111027_000000
000000
>>> 20111026_223000 20111027_013000 APCP_24  A24      APCP_24 A24
>>> V3.1    WRF_ens_m00 240000    20111027_000000 20111027_000000
000000
>>> 20111026_223000 20111027_013000 APCP_24  A24      APCP_24 A24
>>>
>>> It seems as if point_stat first read APCP_24_f000 forecast valid
time
>>> and maintain it through _f001, _f002 and _f003 calls. All four
rows are
>>> different each other in some fields indeed.
>>>
>>> For all four fields as separated files using: fcst_field[] = [
>>> "APCP_24_f00n/A24" ] with 0<=n<=3, FCST_VALID maintain to
20111027.
>>>
>>> Is there something I'm doing wrong? I checked it several times and
don't
>>> identify any error. Any suggestion?
>>>
>>> Thanks in advance,
>>>
>>> Víctor M. Álvarez
>>> Numerical Department - MeteoGalicia
>>> Consellería de Medio Ambiente, Territorio e Infraestruturas -
Xunta de
>>> Galicia
>>> Santiago de Compostela - Spain
>>>
>>>
>>>
>>
>>
>

------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #55008] Forecast valid times and point_stat tool
From: Victor M. Alvarez
Time: Mon Mar 12 07:01:32 2012

John,

thanks for news, though bad news :D

I'm going to close this branch on my development and continue where I
left before.

Another route I should explore is changing accumulation times on GRIB
files and use it as input files to point_stat. However I'm a little
hurried up and have to skip this path by now. I'll have in it mind.

Excuse my boldness and forget it if I'm proposing a nonsense but I'd
like to suggest to implement support for a temporal dimension on MET
tools NetCDF input files. I consider it useful in order to develop an
operational validation task based on MET tools. Though there would be
other points of view I'm not considering.

Thanks for your valuable help,
Víctor.


John Halley Gotway via RT escribió:
> Victor,
>
> Thanks for sending your data.
>
> Unfortunately, I have bad news.  Upon further investigation into the
code, I found the following assumption:
>    Our code for reading MET NetCDF files assumes that all variables
in the file contain the same timing information.
>
> That's why you're seeing identical timing information in all of the
output files.  I apologize for our suggestion of going this route in
the first place!  But in order to use the MET tools, you'll
> need to organize your forecast NetCDF files with one NetCDF file for
each output time.  I understand that that may lead to an unmanageable
number of NetCDF files for you, but unfortunately, there's no
> way around it in METv3.1.
>
> Sorry for the bad news.
>
> Thanks,
> John
>

------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #55008] Forecast valid times and point_stat tool
From: John Halley Gotway
Time: Mon Mar 12 12:19:22 2012

Victor,

Thanks for the feedback.  I will forward your suggestion to the rest
of the MET team for consideration for future releases.

Thanks,
John

On 03/12/2012 07:01 AM, Victor M. Alvarez via RT wrote:
> John,
>
> thanks for news, though bad news :D
>
> I'm going to close this branch on my development and continue where
I
> left before.
>
> Another route I should explore is changing accumulation times on
GRIB
> files and use it as input files to point_stat. However I'm a little
> hurried up and have to skip this path by now. I'll have in it mind.
>
> Excuse my boldness and forget it if I'm proposing a nonsense but I'd
> like to suggest to implement support for a temporal dimension on MET
> tools NetCDF input files. I consider it useful in order to develop
an
> operational validation task based on MET tools. Though there would
be
> other points of view I'm not considering.
>
> Thanks for your valuable help,
> Víctor.

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


More information about the Met_help mailing list