[Met_help] [rt.rap.ucar.edu #64437] History for thanks!
John Halley Gotway via RT
met_help at ucar.edu
Tue Feb 18 12:01:22 MST 2014
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
john
hello!
as you send me the information , I have succeeded running my job without the -config file.
the run script like this:
./stat_analysis \
## -config ${inputdir}/statanalysis_aggregate_time${hour_forecast} \
-lookin ${datadir}/20130428/00Z_27km/metout/3km/time${hour_forecast} ${datadir}/20130513/00Z_27km/metout/3km/time${hour_forecast} \
-lookin ${datadir}/20130523/12Z_27km/metout/3km/time${hour_forecast} ${datadir}/20130605/12Z_27km/metout/3km/time${hour_forecast} \
-lookin ${datadir}/20130608/00Z_27km/metout/3km/time${hour_forecast} ${datadir}/20130629/12Z_27km/metout/3km/time${hour_forecast} \
-lookin ${datadir}/20130909/12Z_27km/metout/3km/time${hour_forecast} \
-fcst_lead ${ii}0000 -out ./3km_stat_analysis_${ii}_cts.txt \
-v 2 \
-job aggregate_stat -line_type MPR -out_line_type CTS -out_fcst_thresh \>=10.0 -out_obs_thresh \>=10.0
and I get the output result ,but when I add the -config, there is also error information.
I think this is just way to solve the question.
by the way ask the another question.
I want to verify the radar output of my storm case with the Grid_tat library.
I have wrote the observation(REFO.nc) and the forecast(REFC.nc) file for NETCDF grid files.
and I also repair the my GridStatConfig file(all the file are attached)
when I run ./grid_stat REFC.nc REFO.nc GridStatConfig
the error information:
DEBUG 1: Default Config File: /home/liqiang/wrf33/METv4.0/data/config/GridStatConfig_default
DEBUG 1: User Config File: GridStatConfig
NetCDF: Attribute not found
so I am not sure whether the Met can run radar data verification.
another question I get my NETCDF file using the NCL, but the guider said the NETCDF should output from the Pcp-Combine. I am not sure that is fit enough. the radar just output one file a few minutes and not need pcp-combine processing.
last question about the grid file.
like the guider tell the *,* specifies the two dimensions for the gridded field , and I also set in my config file.
but the write variable name in my NETCDF files according to the on388 table .
I am not sure all the above is available.
and if I want to run radar verification using MET , could you give me some suggestion.
Thank you from soul !
Qiang Li
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #64437] thanks!
From: John Halley Gotway
Time: Mon Dec 02 15:03:45 2013
Qiang,
I tried adding in the config file but was still unable to reproduce
the seg fault you're seeing.
I took a look at the NetCDF files you sent and see why MET is having a
hard time reading them. MET expects to find grid definition
information in the global attributes section and timing information
in the variable attributes. It's erroring out when trying to access
those attributes. In order it use gridded NetCDF files in MET, you'll
need to make them look like the output of the pcp_combine tool.
Listed below is the header from one of the NetCDF files from
pcp_combine created by the METv4.0 test scripts. Here are the
required parts.
(1) dimensions should be named "lat" and "lon"
(2) the "lat" and "lon" variable are **NOT** required
(3) gridded variables (e.g. APCP_12) must use the "lat" and "lon"
dimensions
(4) gridded variables should include the attributes listed in the
example (for timing info, only the init_time_ut, valid_time_ut, and
accum_time_sec are actually used. "ut" stands for unix time,
the number of seconds since Jan 1, 1970).
(5) global attributes should include the grid/projection
information. This example shows how they're specified for a lambert
conformal projection. Let me know if you're using a different
projection and I'll send you an example.
Hope that helps get you going.
Thanks,
John Halley Gotway
met_help at ucar.edu
> ncdump -h
METv4.0/out/pcp_combine/sample_fcst_12L_2005080712V_12A.nc
netcdf sample_fcst_12L_2005080712V_12A {
dimensions:
lat = 129 ;
lon = 185 ;
variables:
float lat(lat, lon) ;
lat:long_name = "latitude" ;
lat:units = "degrees_north" ;
lat:standard_name = "latitude" ;
float lon(lat, lon) ;
lon:long_name = "longitude" ;
lon:units = "degrees_east" ;
lon:standard_name = "longitude" ;
float APCP_12(lat, lon) ;
APCP_12:name = "APCP_12" ;
APCP_12:long_name = "Total precipitation" ;
APCP_12:level = "A12" ;
APCP_12:units = "kg/m^2" ;
APCP_12:_FillValue = -9999.f ;
APCP_12:init_time = "20050807_000000" ;
APCP_12:init_time_ut = 1123372800 ;
APCP_12:valid_time = "20050807_120000" ;
APCP_12:valid_time_ut = 1123416000 ;
APCP_12:accum_time = "120000" ;
APCP_12:accum_time_sec = 43200 ;
// global attributes:
:FileOrigins = "File
../out/pcp_combine/sample_fcst_12L_2005080712V_12A.nc generated
20130723_215844 UTC on host dakota by the MET pcp_combine tool" ;
:MET_version = "V4.0" ;
:MET_tool = "pcp_combine" ;
:RunCommand = "Sum: 4 files with accumulations of
030000." ;
:Projection = "Lambert Conformal" ;
:scale_lat_1 = "25.000000" ;
:scale_lat_2 = "25.000000" ;
:lat_pin = "12.190000" ;
:lon_pin = "-133.459000" ;
:x_pin = "0.000000" ;
:y_pin = "0.000000" ;
:lon_orient = "-95.000000" ;
:d_km = "40.635000" ;
:r_km = "6371.200000" ;
:nx = "185" ;
:ny = "129 grid_points" ;
}
On 11/26/2013 02:22 PM, Li, Qiang via RT wrote:
>
> Tue Nov 26 14:22:58 2013: Request 64437 was acted upon.
> Transaction: Ticket created by liqiang at ou.edu
> Queue: met_help
> Subject: thanks!
> Owner: Nobody
> Requestors: liqiang at ou.edu
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=64437 >
>
>
> john
> hello!
> as you send me the information , I have succeeded running my job
without the -config file.
> the run script like this:
> ./stat_analysis \
> ## -config ${inputdir}/statanalysis_aggregate_time${hour_forecast}
\
> -lookin
${datadir}/20130428/00Z_27km/metout/3km/time${hour_forecast}
${datadir}/20130513/00Z_27km/metout/3km/time${hour_forecast} \
> -lookin
${datadir}/20130523/12Z_27km/metout/3km/time${hour_forecast}
${datadir}/20130605/12Z_27km/metout/3km/time${hour_forecast} \
> -lookin
${datadir}/20130608/00Z_27km/metout/3km/time${hour_forecast}
${datadir}/20130629/12Z_27km/metout/3km/time${hour_forecast} \
> -lookin
${datadir}/20130909/12Z_27km/metout/3km/time${hour_forecast} \
> -fcst_lead ${ii}0000 -out ./3km_stat_analysis_${ii}_cts.txt \
> -v 2 \
> -job aggregate_stat -line_type MPR -out_line_type CTS
-out_fcst_thresh \>=10.0 -out_obs_thresh \>=10.0
>
> and I get the output result ,but when I add the -config, there is
also error information.
>
> I think this is just way to solve the question.
>
> by the way ask the another question.
> I want to verify the radar output of my storm case with the
Grid_tat library.
>
> I have wrote the observation(REFO.nc) and the forecast(REFC.nc) file
for NETCDF grid files.
> and I also repair the my GridStatConfig file(all the file are
attached)
> when I run ./grid_stat REFC.nc REFO.nc GridStatConfig
> the error information:
> DEBUG 1: Default Config File:
/home/liqiang/wrf33/METv4.0/data/config/GridStatConfig_default
> DEBUG 1: User Config File: GridStatConfig
> NetCDF: Attribute not found
>
> so I am not sure whether the Met can run radar data verification.
> another question I get my NETCDF file using the NCL, but the guider
said the NETCDF should output from the Pcp-Combine. I am not sure that
is fit enough. the radar just output one file a few minutes and not
need pcp-combine processing.
>
> last question about the grid file.
> like the guider tell the *,* specifies the two dimensions for the
gridded field , and I also set in my config file.
> but the write variable name in my NETCDF files according to the
on388 table .
> I am not sure all the above is available.
>
> and if I want to run radar verification using MET , could you give
me some suggestion.
>
>
> Thank you from soul !
>
>
> Qiang Li
>
>
>
>
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #64437] thanks!
From: Li, Qiang
Time: Fri Dec 06 16:13:16 2013
thanks your helpful suggestion.
because my radar data do not include the projection information and
your listed information, I think I should give the information for my
data first of all.
I am not sure whether the NCL can do that.
I wan to ask another question about Met
I wan to plot the ROC image, and the guider I must get the hit
rate(POD) and false alarm rate(POFD) for each user-specified
threshold.
I think the POD means the PODY.
but I am not know the threshold meanings. how to set in my config
file.
thanks!
Qiang
________________________________________
From: John Halley Gotway via RT [met_help at ucar.edu]
Sent: Monday, December 02, 2013 4:03 PM
To: Li, Qiang
Subject: Re: [rt.rap.ucar.edu #64437] thanks!
Qiang,
I tried adding in the config file but was still unable to reproduce
the seg fault you're seeing.
I took a look at the NetCDF files you sent and see why MET is having a
hard time reading them. MET expects to find grid definition
information in the global attributes section and timing information
in the variable attributes. It's erroring out when trying to access
those attributes. In order it use gridded NetCDF files in MET, you'll
need to make them look like the output of the pcp_combine tool.
Listed below is the header from one of the NetCDF files from
pcp_combine created by the METv4.0 test scripts. Here are the
required parts.
(1) dimensions should be named "lat" and "lon"
(2) the "lat" and "lon" variable are **NOT** required
(3) gridded variables (e.g. APCP_12) must use the "lat" and "lon"
dimensions
(4) gridded variables should include the attributes listed in the
example (for timing info, only the init_time_ut, valid_time_ut, and
accum_time_sec are actually used. "ut" stands for unix time,
the number of seconds since Jan 1, 1970).
(5) global attributes should include the grid/projection
information. This example shows how they're specified for a lambert
conformal projection. Let me know if you're using a different
projection and I'll send you an example.
Hope that helps get you going.
Thanks,
John Halley Gotway
met_help at ucar.edu
> ncdump -h
METv4.0/out/pcp_combine/sample_fcst_12L_2005080712V_12A.nc
netcdf sample_fcst_12L_2005080712V_12A {
dimensions:
lat = 129 ;
lon = 185 ;
variables:
float lat(lat, lon) ;
lat:long_name = "latitude" ;
lat:units = "degrees_north" ;
lat:standard_name = "latitude" ;
float lon(lat, lon) ;
lon:long_name = "longitude" ;
lon:units = "degrees_east" ;
lon:standard_name = "longitude" ;
float APCP_12(lat, lon) ;
APCP_12:name = "APCP_12" ;
APCP_12:long_name = "Total precipitation" ;
APCP_12:level = "A12" ;
APCP_12:units = "kg/m^2" ;
APCP_12:_FillValue = -9999.f ;
APCP_12:init_time = "20050807_000000" ;
APCP_12:init_time_ut = 1123372800 ;
APCP_12:valid_time = "20050807_120000" ;
APCP_12:valid_time_ut = 1123416000 ;
APCP_12:accum_time = "120000" ;
APCP_12:accum_time_sec = 43200 ;
// global attributes:
:FileOrigins = "File
../out/pcp_combine/sample_fcst_12L_2005080712V_12A.nc generated
20130723_215844 UTC on host dakota by the MET pcp_combine tool" ;
:MET_version = "V4.0" ;
:MET_tool = "pcp_combine" ;
:RunCommand = "Sum: 4 files with accumulations of
030000." ;
:Projection = "Lambert Conformal" ;
:scale_lat_1 = "25.000000" ;
:scale_lat_2 = "25.000000" ;
:lat_pin = "12.190000" ;
:lon_pin = "-133.459000" ;
:x_pin = "0.000000" ;
:y_pin = "0.000000" ;
:lon_orient = "-95.000000" ;
:d_km = "40.635000" ;
:r_km = "6371.200000" ;
:nx = "185" ;
:ny = "129 grid_points" ;
}
On 11/26/2013 02:22 PM, Li, Qiang via RT wrote:
>
> Tue Nov 26 14:22:58 2013: Request 64437 was acted upon.
> Transaction: Ticket created by liqiang at ou.edu
> Queue: met_help
> Subject: thanks!
> Owner: Nobody
> Requestors: liqiang at ou.edu
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=64437 >
>
>
> john
> hello!
> as you send me the information , I have succeeded running my job
without the -config file.
> the run script like this:
> ./stat_analysis \
> ## -config ${inputdir}/statanalysis_aggregate_time${hour_forecast}
\
> -lookin
${datadir}/20130428/00Z_27km/metout/3km/time${hour_forecast}
${datadir}/20130513/00Z_27km/metout/3km/time${hour_forecast} \
> -lookin
${datadir}/20130523/12Z_27km/metout/3km/time${hour_forecast}
${datadir}/20130605/12Z_27km/metout/3km/time${hour_forecast} \
> -lookin
${datadir}/20130608/00Z_27km/metout/3km/time${hour_forecast}
${datadir}/20130629/12Z_27km/metout/3km/time${hour_forecast} \
> -lookin
${datadir}/20130909/12Z_27km/metout/3km/time${hour_forecast} \
> -fcst_lead ${ii}0000 -out ./3km_stat_analysis_${ii}_cts.txt \
> -v 2 \
> -job aggregate_stat -line_type MPR -out_line_type CTS
-out_fcst_thresh \>=10.0 -out_obs_thresh \>=10.0
>
> and I get the output result ,but when I add the -config, there is
also error information.
>
> I think this is just way to solve the question.
>
> by the way ask the another question.
> I want to verify the radar output of my storm case with the
Grid_tat library.
>
> I have wrote the observation(REFO.nc) and the forecast(REFC.nc) file
for NETCDF grid files.
> and I also repair the my GridStatConfig file(all the file are
attached)
> when I run ./grid_stat REFC.nc REFO.nc GridStatConfig
> the error information:
> DEBUG 1: Default Config File:
/home/liqiang/wrf33/METv4.0/data/config/GridStatConfig_default
> DEBUG 1: User Config File: GridStatConfig
> NetCDF: Attribute not found
>
> so I am not sure whether the Met can run radar data verification.
> another question I get my NETCDF file using the NCL, but the guider
said the NETCDF should output from the Pcp-Combine. I am not sure that
is fit enough. the radar just output one file a few minutes and not
need pcp-combine processing.
>
> last question about the grid file.
> like the guider tell the *,* specifies the two dimensions for the
gridded field , and I also set in my config file.
> but the write variable name in my NETCDF files according to the
on388 table .
> I am not sure all the above is available.
>
> and if I want to run radar verification using MET , could you give
me some suggestion.
>
>
> Thank you from soul !
>
>
> Qiang Li
>
>
>
>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #64437] thanks!
From: John Halley Gotway
Time: Mon Dec 09 11:18:05 2013
Qiang,
POD stands for Probability of Detection and that's the same as what's
call PODY (Prob of Detecting a Yes) in the MET output. PODY and POFD
are both categorical statistics, meaning that they're
defined based on an event (e.g. like precip > 0 mm). You can specify
one or more thresholds to be used when in the MET configuration files.
For example, suppose you're verifying 24-hour
precipitation and you'd like to compute statistics for 3 different
thresholds. Here's how you'd set up a Grid-Stat config file:
field = [
{
name = "APCP";
level = [ "A24" ];
cat_thresh = [ >0.0, >=5.0, >=10.0 ];
}
];
The "cat_thresh" stands for "categorical threshold". So I'm telling
Grid-Stat to compute categorical statistics for this field using the 3
different thresholds I've specified.
Hope that helps clarify.
Thanks,
John
On 12/06/2013 04:13 PM, Li, Qiang via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=64437 >
>
> thanks your helpful suggestion.
>
> because my radar data do not include the projection information and
your listed information, I think I should give the information for my
data first of all.
> I am not sure whether the NCL can do that.
>
> I wan to ask another question about Met
>
> I wan to plot the ROC image, and the guider I must get the hit
rate(POD) and false alarm rate(POFD) for each user-specified
threshold.
> I think the POD means the PODY.
> but I am not know the threshold meanings. how to set in my config
file.
>
> thanks!
>
> Qiang
> ________________________________________
> From: John Halley Gotway via RT [met_help at ucar.edu]
> Sent: Monday, December 02, 2013 4:03 PM
> To: Li, Qiang
> Subject: Re: [rt.rap.ucar.edu #64437] thanks!
>
> Qiang,
>
> I tried adding in the config file but was still unable to reproduce
the seg fault you're seeing.
>
> I took a look at the NetCDF files you sent and see why MET is having
a hard time reading them. MET expects to find grid definition
information in the global attributes section and timing information
> in the variable attributes. It's erroring out when trying to access
those attributes. In order it use gridded NetCDF files in MET, you'll
need to make them look like the output of the pcp_combine tool.
>
> Listed below is the header from one of the NetCDF files from
pcp_combine created by the METv4.0 test scripts. Here are the
required parts.
>
> (1) dimensions should be named "lat" and "lon"
> (2) the "lat" and "lon" variable are **NOT** required
> (3) gridded variables (e.g. APCP_12) must use the "lat" and "lon"
dimensions
> (4) gridded variables should include the attributes listed in the
example (for timing info, only the init_time_ut, valid_time_ut, and
accum_time_sec are actually used. "ut" stands for unix time,
> the number of seconds since Jan 1, 1970).
> (5) global attributes should include the grid/projection
information. This example shows how they're specified for a lambert
conformal projection. Let me know if you're using a different
> projection and I'll send you an example.
>
> Hope that helps get you going.
>
> Thanks,
> John Halley Gotway
> met_help at ucar.edu
>
> > ncdump -h
METv4.0/out/pcp_combine/sample_fcst_12L_2005080712V_12A.nc
>
> netcdf sample_fcst_12L_2005080712V_12A {
> dimensions:
> lat = 129 ;
> lon = 185 ;
> variables:
> float lat(lat, lon) ;
> lat:long_name = "latitude" ;
> lat:units = "degrees_north" ;
> lat:standard_name = "latitude" ;
> float lon(lat, lon) ;
> lon:long_name = "longitude" ;
> lon:units = "degrees_east" ;
> lon:standard_name = "longitude" ;
> float APCP_12(lat, lon) ;
> APCP_12:name = "APCP_12" ;
> APCP_12:long_name = "Total precipitation" ;
> APCP_12:level = "A12" ;
> APCP_12:units = "kg/m^2" ;
> APCP_12:_FillValue = -9999.f ;
> APCP_12:init_time = "20050807_000000" ;
> APCP_12:init_time_ut = 1123372800 ;
> APCP_12:valid_time = "20050807_120000" ;
> APCP_12:valid_time_ut = 1123416000 ;
> APCP_12:accum_time = "120000" ;
> APCP_12:accum_time_sec = 43200 ;
>
> // global attributes:
> :FileOrigins = "File
../out/pcp_combine/sample_fcst_12L_2005080712V_12A.nc generated
20130723_215844 UTC on host dakota by the MET pcp_combine tool" ;
> :MET_version = "V4.0" ;
> :MET_tool = "pcp_combine" ;
> :RunCommand = "Sum: 4 files with accumulations of
030000." ;
> :Projection = "Lambert Conformal" ;
> :scale_lat_1 = "25.000000" ;
> :scale_lat_2 = "25.000000" ;
> :lat_pin = "12.190000" ;
> :lon_pin = "-133.459000" ;
> :x_pin = "0.000000" ;
> :y_pin = "0.000000" ;
> :lon_orient = "-95.000000" ;
> :d_km = "40.635000" ;
> :r_km = "6371.200000" ;
> :nx = "185" ;
> :ny = "129 grid_points" ;
> }
>
>
> On 11/26/2013 02:22 PM, Li, Qiang via RT wrote:
>>
>> Tue Nov 26 14:22:58 2013: Request 64437 was acted upon.
>> Transaction: Ticket created by liqiang at ou.edu
>> Queue: met_help
>> Subject: thanks!
>> Owner: Nobody
>> Requestors: liqiang at ou.edu
>> Status: new
>> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=64437 >
>>
>>
>> john
>> hello!
>> as you send me the information , I have succeeded running my job
without the -config file.
>> the run script like this:
>> ./stat_analysis \
>> ## -config
${inputdir}/statanalysis_aggregate_time${hour_forecast} \
>> -lookin
${datadir}/20130428/00Z_27km/metout/3km/time${hour_forecast}
${datadir}/20130513/00Z_27km/metout/3km/time${hour_forecast} \
>> -lookin
${datadir}/20130523/12Z_27km/metout/3km/time${hour_forecast}
${datadir}/20130605/12Z_27km/metout/3km/time${hour_forecast} \
>> -lookin
${datadir}/20130608/00Z_27km/metout/3km/time${hour_forecast}
${datadir}/20130629/12Z_27km/metout/3km/time${hour_forecast} \
>> -lookin
${datadir}/20130909/12Z_27km/metout/3km/time${hour_forecast} \
>> -fcst_lead ${ii}0000 -out ./3km_stat_analysis_${ii}_cts.txt \
>> -v 2 \
>> -job aggregate_stat -line_type MPR -out_line_type CTS
-out_fcst_thresh \>=10.0 -out_obs_thresh \>=10.0
>>
>> and I get the output result ,but when I add the -config, there is
also error information.
>>
>> I think this is just way to solve the question.
>>
>> by the way ask the another question.
>> I want to verify the radar output of my storm case with the
Grid_tat library.
>>
>> I have wrote the observation(REFO.nc) and the forecast(REFC.nc)
file for NETCDF grid files.
>> and I also repair the my GridStatConfig file(all the file are
attached)
>> when I run ./grid_stat REFC.nc REFO.nc GridStatConfig
>> the error information:
>> DEBUG 1: Default Config File:
/home/liqiang/wrf33/METv4.0/data/config/GridStatConfig_default
>> DEBUG 1: User Config File: GridStatConfig
>> NetCDF: Attribute not found
>>
>> so I am not sure whether the Met can run radar data verification.
>> another question I get my NETCDF file using the NCL, but the guider
said the NETCDF should output from the Pcp-Combine. I am not sure that
is fit enough. the radar just output one file a few minutes and not
need pcp-combine processing.
>>
>> last question about the grid file.
>> like the guider tell the *,* specifies the two dimensions for the
gridded field , and I also set in my config file.
>> but the write variable name in my NETCDF files according to the
on388 table .
>> I am not sure all the above is available.
>>
>> and if I want to run radar verification using MET , could you give
me some suggestion.
>>
>>
>> Thank you from soul !
>>
>>
>> Qiang Li
>>
>>
>>
>>
>
>
------------------------------------------------
More information about the Met_help
mailing list