[Met_help] [rt.rap.ucar.edu #68848] History for pb_with_netcdf_file
John Halley Gotway via RT
met_help at ucar.edu
Fri Oct 3 12:25:35 MDT 2014
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Good afternoon
Some years ago, I successfully tested METV2.0 for use MODE in verification of precipitation. I have used GRIB 1 files from our model and I have created with ncl-ncar functions the observations file(same grid, prjection, and so on) Everything was OK. Unfortunately, the lack of qualified staff in my team, I can not track changes in versions MET.
Recently I resumed checking MODE and I have installed the latest version METV4.1 .. All tests in scrips were OK. No pb during instalation.
I have run one on "my old scripts"(that went well in version 2.0) with te METv4.1 and I have a problem with netcdf file.
My question is:
- what is a recomanded version of netcf ?
What is wrong with my netcdf file ? How cad debug ?
Many thanks helping me,
Otilia
Here is a output of MODE
Wrote one record, skipped 0
*** Running MODE on APCP using a GRIB forecast and netCDF observation ***
DEBUG 1: Default Config File: /home/util/METv4.1/data/config/MODEConfig_default
DEBUG 1: Match Config File: /home/oper/data/METV4.1/CONFIG/MODEConfig_alaro
DEBUG 1: Merge Config File: /home/oper/data/METV4.1/CONFIG/MODEConfig_alaro
DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created new Met2dDataFile object of type "FileType_Gb1".
DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created new Met2dDataFile object of type "FileType_NcMet".
DEBUG 4: Latitude/Longitude Grid Data:
DEBUG 4: lat_ll = 43.5
DEBUG 4: lon_ll = -20.5
DEBUG 4: delta_lat = 0.1
DEBUG 4: delta_lon = 0.125
DEBUG 4: Nlat = 51
DEBUG 4: Nlon = 77
DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo object of type "FileType_Gb1".
DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo object of type "FileType_NcMet".
DEBUG 1: Forecast File: /home/oper/data/METV4.1/TRAV/PRECIP_030.grb
DEBUG 1: Observation File: /home/oper/data/METV4.1/TRAV/prec20140505.txt.1x1.nc
DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found range match for VarInfo "APCP/A24" in GRIB record 1 of GRIB file "/home/oper/data/METV4.1/TRAV/PRECIP_030.grb".
DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found 1 GRIB records matching VarInfo "APCP/A24" in GRIB file "/home/oper/data/METV4.1/TRAV/PRECIP_030.grb".
ERROR :
ERROR : process_fcst_obs_files() -> can't get data from file "/home/oper/data/METV4.1/TRAV/prec20140505.txt.1x1.nc"
ERROR :
National Meteorological Administration Mail Server
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: pb_with_netcdf_file
From: John Halley Gotway
Time: Thu Sep 04 09:36:37 2014
Otilia,
I see that you're having problems getting MET to read that NetCDF
file. It
would probably be easiest to just have you send me a copy of that
NetCDF
file (prec20140505.txt.1x1.nc). I'll take a look and let you know
where
the problem is. You can post that file to our anonymous ftp site
following
the instructions listed here:
http://www.dtcenter.org/met/users/support/met_help.php#ftp
Please let me know when you've posted that data file so I can go grab
it.
Thanks,
John Halley Gotway
met_help at ucar.edu
On Thu, Sep 4, 2014 at 3:50 AM, otilia.diaconu via RT
<met_help at ucar.edu>
wrote:
>
> Thu Sep 04 03:50:19 2014: Request 68848 was acted upon.
> Transaction: Ticket created by otilia.diaconu at meteoromania.ro
> Queue: met_help
> Subject: pb_with_netcdf_file
> Owner: Nobody
> Requestors: otilia.diaconu at meteoromania.ro
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=68848 >
>
>
> Good afternoon
> Some years ago, I successfully tested METV2.0 for use MODE in
verification
> of precipitation. I have used GRIB 1 files from our model and I have
> created with ncl-ncar functions the observations file(same grid,
prjection,
> and so on) Everything was OK. Unfortunately, the lack of qualified
staff in
> my team, I can not track changes in versions MET.
>
> Recently I resumed checking MODE and I have installed the latest
version
> METV4.1 .. All tests in scrips were OK. No pb during instalation.
>
> I have run one on "my old scripts"(that went well in version 2.0)
with te
> METv4.1 and I have a problem with netcdf file.
> My question is:
> - what is a recomanded version of netcf ?
>
>
> What is wrong with my netcdf file ? How cad debug ?
> Many thanks helping me,
> Otilia
>
> Here is a output of MODE
>
>
>
> Wrote one record, skipped 0
> *** Running MODE on APCP using a GRIB forecast and netCDF
observation ***
> DEBUG 1: Default Config File:
> /home/util/METv4.1/data/config/MODEConfig_default
> DEBUG 1: Match Config File:
/home/oper/data/METV4.1/CONFIG/MODEConfig_alaro
> DEBUG 1: Merge Config File:
/home/oper/data/METV4.1/CONFIG/MODEConfig_alaro
> DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created new
> Met2dDataFile object of type "FileType_Gb1".
> DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created new
> Met2dDataFile object of type "FileType_NcMet".
> DEBUG 4: Latitude/Longitude Grid Data:
> DEBUG 4: lat_ll = 43.5
> DEBUG 4: lon_ll = -20.5
> DEBUG 4: delta_lat = 0.1
> DEBUG 4: delta_lon = 0.125
> DEBUG 4: Nlat = 51
> DEBUG 4: Nlon = 77
> DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo
object of
> type "FileType_Gb1".
> DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo
object of
> type "FileType_NcMet".
> DEBUG 1: Forecast File: /home/oper/data/METV4.1/TRAV/PRECIP_030.grb
> DEBUG 1: Observation File: /home/oper/data/METV4.1/TRAV/
> prec20140505.txt.1x1.nc
> DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found range match
for
> VarInfo "APCP/A24" in GRIB record 1 of GRIB file
> "/home/oper/data/METV4.1/TRAV/PRECIP_030.grb".
> DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found 1 GRIB
records
> matching VarInfo "APCP/A24" in GRIB file
> "/home/oper/data/METV4.1/TRAV/PRECIP_030.grb".
> ERROR :
> ERROR : process_fcst_obs_files() -> can't get data from file
> "/home/oper/data/METV4.1/TRAV/prec20140505.txt.1x1.nc"
> ERROR :
> National Meteorological Administration Mail Server
>
>
------------------------------------------------
Subject: pb_with_netcdf_file
From: Otilia Diaconu
Time: Thu Sep 04 10:08:49 2014
Hi John,
Thank you,
I put the file in the directory "diaconu'
I have 2 files:
prec* is created using ncl-ncar and
radar* (created by a collegue of mine) with R. with second file the
error
is : Atribute not found . I see that in that file some atribute are
missing
..
Regards,
Otilia
<http://prec20140505.txt.1x1.nc>
On Thu, Sep 4, 2014 at 6:36 PM, John Halley Gotway via RT
<met_help at ucar.edu
> wrote:
> Otilia,
>
> I see that you're having problems getting MET to read that NetCDF
file. It
> would probably be easiest to just have you send me a copy of that
NetCDF
> file (prec20140505.txt.1x1.nc). I'll take a look and let you know
where
> the problem is. You can post that file to our anonymous ftp site
following
> the instructions listed here:
> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>
> Please let me know when you've posted that data file so I can go
grab it.
>
> Thanks,
> John Halley Gotway
> met_help at ucar.edu
>
>
> On Thu, Sep 4, 2014 at 3:50 AM, otilia.diaconu via RT
<met_help at ucar.edu>
> wrote:
>
> >
> > Thu Sep 04 03:50:19 2014: Request 68848 was acted upon.
> > Transaction: Ticket created by otilia.diaconu at meteoromania.ro
> > Queue: met_help
> > Subject: pb_with_netcdf_file
> > Owner: Nobody
> > Requestors: otilia.diaconu at meteoromania.ro
> > Status: new
> > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=68848 >
> >
> >
> > Good afternoon
> > Some years ago, I successfully tested METV2.0 for use MODE in
> verification
> > of precipitation. I have used GRIB 1 files from our model and I
have
> > created with ncl-ncar functions the observations file(same grid,
> prjection,
> > and so on) Everything was OK. Unfortunately, the lack of qualified
staff
> in
> > my team, I can not track changes in versions MET.
> >
> > Recently I resumed checking MODE and I have installed the latest
version
> > METV4.1 .. All tests in scrips were OK. No pb during
instalation.
> >
> > I have run one on "my old scripts"(that went well in version 2.0)
with te
> > METv4.1 and I have a problem with netcdf file.
> > My question is:
> > - what is a recomanded version of netcf ?
> >
> >
> > What is wrong with my netcdf file ? How cad debug ?
> > Many thanks helping me,
> > Otilia
> >
> > Here is a output of MODE
> >
> >
> >
> > Wrote one record, skipped 0
> > *** Running MODE on APCP using a GRIB forecast and netCDF
observation ***
> > DEBUG 1: Default Config File:
> > /home/util/METv4.1/data/config/MODEConfig_default
> > DEBUG 1: Match Config File:
> /home/oper/data/METV4.1/CONFIG/MODEConfig_alaro
> > DEBUG 1: Merge Config File:
> /home/oper/data/METV4.1/CONFIG/MODEConfig_alaro
> > DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created
new
> > Met2dDataFile object of type "FileType_Gb1".
> > DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created
new
> > Met2dDataFile object of type "FileType_NcMet".
> > DEBUG 4: Latitude/Longitude Grid Data:
> > DEBUG 4: lat_ll = 43.5
> > DEBUG 4: lon_ll = -20.5
> > DEBUG 4: delta_lat = 0.1
> > DEBUG 4: delta_lon = 0.125
> > DEBUG 4: Nlat = 51
> > DEBUG 4: Nlon = 77
> > DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo
object of
> > type "FileType_Gb1".
> > DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo
object of
> > type "FileType_NcMet".
> > DEBUG 1: Forecast File:
/home/oper/data/METV4.1/TRAV/PRECIP_030.grb
> > DEBUG 1: Observation File: /home/oper/data/METV4.1/TRAV/
> > prec20140505.txt.1x1.nc
> > DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found range match
for
> > VarInfo "APCP/A24" in GRIB record 1 of GRIB file
> > "/home/oper/data/METV4.1/TRAV/PRECIP_030.grb".
> > DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found 1 GRIB
records
> > matching VarInfo "APCP/A24" in GRIB file
> > "/home/oper/data/METV4.1/TRAV/PRECIP_030.grb".
> > ERROR :
> > ERROR : process_fcst_obs_files() -> can't get data from file
> > "/home/oper/data/METV4.1/TRAV/prec20140505.txt.1x1.nc"
> > ERROR :
> > National Meteorological Administration Mail Server
> >
> >
>
>
------------------------------------------------
Subject: pb_with_netcdf_file
From: Otilia Diaconu
Time: Thu Sep 04 10:14:06 2014
Hi John,
Thank you,
I put the file in the directory "diaconu'
I have 2 files:
prec* is created using ncl-ncar and
radar* (created by a collegue of mine) with R. with second file the
error
is : Atribute not found . I see that in that file some atribute are
missing
..
Regards,
Otilia
On Thu, Sep 4, 2014 at 6:36 PM, John Halley Gotway via RT
<met_help at ucar.edu
> wrote:
> Otilia,
>
> I see that you're having problems getting MET to read that NetCDF
file. It
> would probably be easiest to just have you send me a copy of that
NetCDF
> file (prec20140505.txt.1x1.nc). I'll take a look and let you know
where
> the problem is. You can post that file to our anonymous ftp site
following
> the instructions listed here:
> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>
> Please let me know when you've posted that data file so I can go
grab it.
>
> Thanks,
> John Halley Gotway
> met_help at ucar.edu
>
>
> On Thu, Sep 4, 2014 at 3:50 AM, otilia.diaconu via RT
<met_help at ucar.edu>
> wrote:
>
> >
> > Thu Sep 04 03:50:19 2014: Request 68848 was acted upon.
> > Transaction: Ticket created by otilia.diaconu at meteoromania.ro
> > Queue: met_help
> > Subject: pb_with_netcdf_file
> > Owner: Nobody
> > Requestors: otilia.diaconu at meteoromania.ro
> > Status: new
> > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=68848 >
> >
> >
> > Good afternoon
> > Some years ago, I successfully tested METV2.0 for use MODE in
> verification
> > of precipitation. I have used GRIB 1 files from our model and I
have
> > created with ncl-ncar functions the observations file(same grid,
> prjection,
> > and so on) Everything was OK. Unfortunately, the lack of qualified
staff
> in
> > my team, I can not track changes in versions MET.
> >
> > Recently I resumed checking MODE and I have installed the latest
version
> > METV4.1 .. All tests in scrips were OK. No pb during
instalation.
> >
> > I have run one on "my old scripts"(that went well in version 2.0)
with te
> > METv4.1 and I have a problem with netcdf file.
> > My question is:
> > - what is a recomanded version of netcf ?
> >
> >
> > What is wrong with my netcdf file ? How cad debug ?
> > Many thanks helping me,
> > Otilia
> >
> > Here is a output of MODE
> >
> >
> >
> > Wrote one record, skipped 0
> > *** Running MODE on APCP using a GRIB forecast and netCDF
observation ***
> > DEBUG 1: Default Config File:
> > /home/util/METv4.1/data/config/MODEConfig_default
> > DEBUG 1: Match Config File:
> /home/oper/data/METV4.1/CONFIG/MODEConfig_alaro
> > DEBUG 1: Merge Config File:
> /home/oper/data/METV4.1/CONFIG/MODEConfig_alaro
> > DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created
new
> > Met2dDataFile object of type "FileType_Gb1".
> > DEBUG 4: Met2dDataFileFactory::new_met_2d_data_file() -> created
new
> > Met2dDataFile object of type "FileType_NcMet".
> > DEBUG 4: Latitude/Longitude Grid Data:
> > DEBUG 4: lat_ll = 43.5
> > DEBUG 4: lon_ll = -20.5
> > DEBUG 4: delta_lat = 0.1
> > DEBUG 4: delta_lon = 0.125
> > DEBUG 4: Nlat = 51
> > DEBUG 4: Nlon = 77
> > DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo
object of
> > type "FileType_Gb1".
> > DEBUG 4: VarInfoFactory::new_var_info() -> created new VarInfo
object of
> > type "FileType_NcMet".
> > DEBUG 1: Forecast File:
/home/oper/data/METV4.1/TRAV/PRECIP_030.grb
> > DEBUG 1: Observation File: /home/oper/data/METV4.1/TRAV/
> > prec20140505.txt.1x1.nc
> > DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found range match
for
> > VarInfo "APCP/A24" in GRIB record 1 of GRIB file
> > "/home/oper/data/METV4.1/TRAV/PRECIP_030.grb".
> > DEBUG 3: MetGrib1DataFile::data_plane_array() -> Found 1 GRIB
records
> > matching VarInfo "APCP/A24" in GRIB file
> > "/home/oper/data/METV4.1/TRAV/PRECIP_030.grb".
> > ERROR :
> > ERROR : process_fcst_obs_files() -> can't get data from file
> > "/home/oper/data/METV4.1/TRAV/prec20140505.txt.1x1.nc"
> > ERROR :
> > National Meteorological Administration Mail Server
> >
> >
>
>
------------------------------------------------
Subject: pb_with_netcdf_file
From: Otilia Diaconu
Time: Thu Sep 04 10:17:01 2014
Hi John,
Thank you,
I put the file in the directory "diaconu'
I have 2 files:
prec* is created using ncl-ncar and
radar* (created by a collegue of mine) with R. with second file the
error
is : Atribute not found . I see that in that file some atribute are
missing
..
Regards,
------------------------------------------------
Subject: pb_with_netcdf_file
From: John Halley Gotway
Time: Thu Sep 04 14:07:36 2014
Otilia,
Thanks for sending those files.
I was able to run the prec20140505.txt.1x1.nc file through the MET
plot_data_plane tool to verify that MET can read it:
METv4.1/bin/plot_data_plane prec20140505.txt.1x1.nc
prec20140505.txt.1x1.ps 'name="APCP";level="(*,*)";'
I converted the resulting image into png format and have attached it
to
this message.
I suspect that the problem is in how you have the config file set up.
You're running MODE to compare a GRIB forecast file to a NetCDF
observation
file, and those need to be specified in different ways in the config
file.
Please try setting up your MODE config file like this:
// This is for the GRIB file
fcst = {
field = {
name = "APCP";
level = "A24";
};
... all the other settings go here ...
};
// This is for the NetCDF file
obs = {
field = {
name = "APCP";
level = "(*,*)";
};
... all the other settings go here ...
};
There is also a problem in the NetCDF APCP variable attributes -
you're
missing "init_time_ut" and "valid_time_ut". "ut" stands for unix
time,
which is the number of seconds since Jan 1, 1970. MET reads those
variable
attributes to get the timing information.
I see the following in the file now:
APCP:valid_time = "20140505_060000" ;
APCP:init_time = "20140505_000000" ;
You can convert those to unix time using the following commands:
date -ud ''2014-05-05' UTC '06:00:00'' +%s
1399269600
date -ud ''2014-05-05' UTC '00:00:00'' +%s
1399248000
Add them to that existing NetCDF file with this "ncatted" command:
ncatted -a init_time_ut,APCP,a,ll,1399248000 -a
valid_time_ut,APCP,a,ll,1399269600 prec20140505.txt.1x1.nc
Please try running MODE again with these changes. If you experience
more
problems, please send me your GRIB forecast file, NetCDF observation
file,
and MODE configuration file (via our ftp site).
Regarding the second file, ppradar_2014-05-05-18.nc, METv4.1 is not
able to
read it. It is a CF-compliant NetCDF file but METv4.1 does not
support
that. However, METv5.0 (which is due out tomorrow), does support
CF-compliant files. I was able to plot this with the METv5.0 version
of
plot_data_plane. But that file doesn't have any timing information
other
than in the file name, so it produced some warnings.
I'd suggest downloading METv5.0 when it is released tomorrow -
although I
think we'll send out the email about it next Monday.
Thanks,
John
On Thu, Sep 4, 2014 at 10:17 AM, Otilia Diaconu via RT
<met_help at ucar.edu>
wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=68848 >
>
> Hi John,
> Thank you,
> I put the file in the directory "diaconu'
> I have 2 files:
> prec* is created using ncl-ncar and
> radar* (created by a collegue of mine) with R. with second file the
error
> is : Atribute not found . I see that in that file some atribute are
missing
> ..
> Regards,
>
>
------------------------------------------------
Subject: pb_with_netcdf_file
From: Otilia Diaconu
Time: Thu Sep 04 14:28:02 2014
Thank you very much ,
I know the problem with time .. (changes in ncl-ncar new versions
since 5.2
used in 2010).. I will be able to set-up the correct time.
I will made the necessary corrections tomorrow ..
It's 23:22 here ..
I will download at install 5.0 version and I will try to use "radar"
files.. My collegue works to set-up in R the others atributes for
those
files.
Many thanks,
Regards
Otilia
On Thu, Sep 4, 2014 at 11:07 PM, John Halley Gotway via RT <
met_help at ucar.edu> wrote:
> Otilia,
>
> Thanks for sending those files.
>
> I was able to run the prec20140505.txt.1x1.nc file through the MET
> plot_data_plane tool to verify that MET can read it:
> METv4.1/bin/plot_data_plane prec20140505.txt.1x1.nc
> prec20140505.txt.1x1.ps 'name="APCP";level="(*,*)";'
> I converted the resulting image into png format and have attached it
to
> this message.
>
> I suspect that the problem is in how you have the config file set
up.
> You're running MODE to compare a GRIB forecast file to a NetCDF
observation
> file, and those need to be specified in different ways in the config
file.
> Please try setting up your MODE config file like this:
>
> // This is for the GRIB file
> fcst = {
> field = {
> name = "APCP";
> level = "A24";
> };
> ... all the other settings go here ...
> };
>
> // This is for the NetCDF file
> obs = {
> field = {
> name = "APCP";
> level = "(*,*)";
> };
> ... all the other settings go here ...
> };
>
> There is also a problem in the NetCDF APCP variable attributes -
you're
> missing "init_time_ut" and "valid_time_ut". "ut" stands for unix
time,
> which is the number of seconds since Jan 1, 1970. MET reads those
variable
> attributes to get the timing information.
>
> I see the following in the file now:
> APCP:valid_time = "20140505_060000" ;
> APCP:init_time = "20140505_000000" ;
>
> You can convert those to unix time using the following commands:
> date -ud ''2014-05-05' UTC '06:00:00'' +%s
> 1399269600
> date -ud ''2014-05-05' UTC '00:00:00'' +%s
> 1399248000
>
> Add them to that existing NetCDF file with this "ncatted" command:
> ncatted -a init_time_ut,APCP,a,ll,1399248000 -a
> valid_time_ut,APCP,a,ll,1399269600 prec20140505.txt.1x1.nc
>
> Please try running MODE again with these changes. If you experience
more
> problems, please send me your GRIB forecast file, NetCDF observation
file,
> and MODE configuration file (via our ftp site).
>
> Regarding the second file, ppradar_2014-05-05-18.nc, METv4.1 is not
able
> to
> read it. It is a CF-compliant NetCDF file but METv4.1 does not
support
> that. However, METv5.0 (which is due out tomorrow), does support
> CF-compliant files. I was able to plot this with the METv5.0
version of
> plot_data_plane. But that file doesn't have any timing information
other
> than in the file name, so it produced some warnings.
>
> I'd suggest downloading METv5.0 when it is released tomorrow -
although I
> think we'll send out the email about it next Monday.
>
> Thanks,
> John
>
>
>
>
>
> On Thu, Sep 4, 2014 at 10:17 AM, Otilia Diaconu via RT
<met_help at ucar.edu>
> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=68848 >
> >
> > Hi John,
> > Thank you,
> > I put the file in the directory "diaconu'
> > I have 2 files:
> > prec* is created using ncl-ncar and
> > radar* (created by a collegue of mine) with R. with second file
the
> error
> > is : Atribute not found . I see that in that file some atribute
are
> missing
> > ..
> > Regards,
> >
> >
>
>
------------------------------------------------
Subject: pb_with_netcdf_file
From: Otilia Diaconu
Time: Thu Sep 04 15:04:49 2014
Thank you very much ,
I know the problem with time .. (changes in ncl-ncar new versions
since 5.2
used in 2010).. I will be able to set-up the correct time.
I will made the necessary corrections tomorrow ..
It's 00:03 here ..
I will download at install 5.0 version and I will try to use "radar"
files.. My collegue works to set-up in R the others atributes for
those
files.
Many thanks,
Regards
Otilia
------------------------------------------------
Subject: pb_with_netcdf_file
From: Otilia Diaconu
Time: Thu Sep 04 15:10:10 2014
Soory if you recive this Email many times..I don't know what happens
.. It
seems to send-it to myself :(
======================
I'll try again :)
===============
Thank you very much ,
I know the problem with time .. (changes in ncl-ncar new versions
since 5.2
used in 2010).. I will be able to set-up the correct time.
I will made the necessary corrections tomorrow ..
It's 00:09 here ..
I will download at install 5.0 version and I will try to use "radar"
files.. My collegue works to set-up in R the others atributes for
those
files.
Many thanks,
Regards
Otilia
------------------------------------------------
Subject: pb_with_netcdf_file
From: Otilia Diaconu
Time: Fri Sep 05 03:46:48 2014
Hi John,
made the necessary changes to the configuration file and in "necdf
atributes". I have run METv4.1 .(.5.0 is not available at this time ),
but
The netcdf file cannot be read
*process_fcst_obs_files() -> can't get data from file*
prec20140505.txt.1x1.nc
I have put the Grib file and the Confifg file in directory "diaconu".
If
you have time to see..
Thank you. A good day and nice weekend,
Otilia
ᐧ
On Thu, Sep 4, 2014 at 11:28 PM, Otilia Diaconu via RT
<met_help at ucar.edu>
wrote:
> Thank you very much ,
> I know the problem with time .. (changes in ncl-ncar new versions
since 5.2
> used in 2010).. I will be able to set-up the correct time.
> I will made the necessary corrections tomorrow ..
> It's 23:22 here ..
> I will download at install 5.0 version and I will try to use "radar"
> files.. My collegue works to set-up in R the others atributes for
those
> files.
> Many thanks,
> Regards
> Otilia
>
>
> On Thu, Sep 4, 2014 at 11:07 PM, John Halley Gotway via RT <
> met_help at ucar.edu> wrote:
>
> > Otilia,
> >
> > Thanks for sending those files.
> >
> > I was able to run the prec20140505.txt.1x1.nc file through the MET
> > plot_data_plane tool to verify that MET can read it:
> > METv4.1/bin/plot_data_plane prec20140505.txt.1x1.nc
> > prec20140505.txt.1x1.ps 'name="APCP";level="(*,*)";'
> > I converted the resulting image into png format and have attached
it to
> > this message.
> >
> > I suspect that the problem is in how you have the config file set
up.
> > You're running MODE to compare a GRIB forecast file to a NetCDF
> observation
> > file, and those need to be specified in different ways in the
config
> file.
> > Please try setting up your MODE config file like this:
> >
> > // This is for the GRIB file
> > fcst = {
> > field = {
> > name = "APCP";
> > level = "A24";
> > };
> > ... all the other settings go here ...
> > };
> >
> > // This is for the NetCDF file
> > obs = {
> > field = {
> > name = "APCP";
> > level = "(*,*)";
> > };
> > ... all the other settings go here ...
> > };
> >
> > There is also a problem in the NetCDF APCP variable attributes -
you're
> > missing "init_time_ut" and "valid_time_ut". "ut" stands for unix
time,
> > which is the number of seconds since Jan 1, 1970. MET reads those
> variable
> > attributes to get the timing information.
> >
> > I see the following in the file now:
> > APCP:valid_time = "20140505_060000" ;
> > APCP:init_time = "20140505_000000" ;
> >
> > You can convert those to unix time using the following commands:
> > date -ud ''2014-05-05' UTC '06:00:00'' +%s
> > 1399269600
> > date -ud ''2014-05-05' UTC '00:00:00'' +%s
> > 1399248000
> >
> > Add them to that existing NetCDF file with this "ncatted" command:
> > ncatted -a init_time_ut,APCP,a,ll,1399248000 -a
> > valid_time_ut,APCP,a,ll,1399269600 prec20140505.txt.1x1.nc
> >
> > Please try running MODE again with these changes. If you
experience more
> > problems, please send me your GRIB forecast file, NetCDF
observation
> file,
> > and MODE configuration file (via our ftp site).
> >
> > Regarding the second file, ppradar_2014-05-05-18.nc, METv4.1 is
not able
> > to
> > read it. It is a CF-compliant NetCDF file but METv4.1 does not
support
> > that. However, METv5.0 (which is due out tomorrow), does support
> > CF-compliant files. I was able to plot this with the METv5.0
version of
> > plot_data_plane. But that file doesn't have any timing
information other
> > than in the file name, so it produced some warnings.
> >
> > I'd suggest downloading METv5.0 when it is released tomorrow -
although I
> > think we'll send out the email about it next Monday.
> >
> > Thanks,
> > John
> >
> >
> >
> >
> >
> > On Thu, Sep 4, 2014 at 10:17 AM, Otilia Diaconu via RT <
> met_help at ucar.edu>
> > wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=68848 >
> > >
> > > Hi John,
> > > Thank you,
> > > I put the file in the directory "diaconu'
> > > I have 2 files:
> > > prec* is created using ncl-ncar and
> > > radar* (created by a collegue of mine) with R. with second file
the
> > error
> > > is : Atribute not found . I see that in that file some atribute
are
> > missing
> > > ..
> > > Regards,
> > >
> > >
> >
> >
>
>
------------------------------------------------
Subject: pb_with_netcdf_file
From: Otilia Diaconu
Time: Tue Sep 09 22:24:58 2014
Hi John,
I successfully installed version 5.0 of MET. My colleague worked at
NetCDF
files observations to have a complete header. The header of obs files
is:
dimensions:
lon = 77 ;
lat = 51 ;
variables:
double lon(lon) ;
lon:units = "degrees_east" ;
lon:long_name = "lon" ;
double lat(lat) ;
lat:units = "degrees_north" ;
lat:long_name = "lat" ;
float APCP(lat, lon) ;
APCP:units = "kg/m^2" ;
APCP:_FillValue = -9999.f ;
APCP:missing_value = -9999. ;
APCP:long_name = "Total precipitation" ;
APCP:projection = "+init=epsg:4326 +proj=longlat
+datum=WGS84 +no_defs +ellps=WGS84 +towgs84=0,0,0" ;
APCP:projection_format = "PROJ.4" ;
APCP:min = -9999. ;
APCP:max = 20.3 ;
APCP:accum_time_sec = 43200 ;
APCP:accum_time = "12 hours" ;
APCP:valid_time = "20140505_180000" ;
APCP:init_time = "20140505_060000" ;
APCP:init_time_ut = 1399269600 ;
APCP:valid_time_ut = 1399312800 ;
// global attributes:
:Conventions = "CF-1.4" ;
:created_by = "R, packages ncdf and raster (version
2.3-0)"
;
:date = "2014-09-09 13:20:37" ;
:nco_openmp_thread_number = 1 ;
:Nlon = "77 grid_points" ;
:Nlat = "51 grid_points" ;
:delta_lon_deg = "0.125000 degrees" ;
:delta_lat_deg = "0.100000 degrees" ;
:lon_ll_deg = "20.200000 degrees_east" ;
:lat_ll_deg = "43.500000 degrees_north" ;
:Projection = "LatLon" ;
The MODE from 5.0 version works now but, even if the in the header I
have
the correct attribute* time_ut ,* the summary of "mode" doesn't
transform'it.
I have attached one immage with this !
Thank you in avance for your help,
Best regards,
Otilia Diaconu
ᐧ
On Fri, Sep 5, 2014 at 12:46 PM, Otilia Diaconu via RT
<met_help at ucar.edu>
wrote:
> Hi John,
> made the necessary changes to the configuration file and in "necdf
> atributes". I have run METv4.1 .(.5.0 is not available at this time
), but
> The netcdf file cannot be read
> *process_fcst_obs_files() -> can't get data from file*
> prec20140505.txt.1x1.nc
> I have put the Grib file and the Confifg file in directory
"diaconu". If
> you have time to see..
> Thank you. A good day and nice weekend,
> Otilia
> ᐧ
>
>
> On Thu, Sep 4, 2014 at 11:28 PM, Otilia Diaconu via RT
<met_help at ucar.edu>
> wrote:
>
> > Thank you very much ,
> > I know the problem with time .. (changes in ncl-ncar new versions
since
> 5.2
> > used in 2010).. I will be able to set-up the correct time.
> > I will made the necessary corrections tomorrow ..
> > It's 23:22 here ..
> > I will download at install 5.0 version and I will try to use
"radar"
> > files.. My collegue works to set-up in R the others atributes for
those
> > files.
> > Many thanks,
> > Regards
> > Otilia
> >
> >
> > On Thu, Sep 4, 2014 at 11:07 PM, John Halley Gotway via RT <
> > met_help at ucar.edu> wrote:
> >
> > > Otilia,
> > >
> > > Thanks for sending those files.
> > >
> > > I was able to run the prec20140505.txt.1x1.nc file through the
MET
> > > plot_data_plane tool to verify that MET can read it:
> > > METv4.1/bin/plot_data_plane prec20140505.txt.1x1.nc
> > > prec20140505.txt.1x1.ps 'name="APCP";level="(*,*)";'
> > > I converted the resulting image into png format and have
attached it to
> > > this message.
> > >
> > > I suspect that the problem is in how you have the config file
set up.
> > > You're running MODE to compare a GRIB forecast file to a NetCDF
> > observation
> > > file, and those need to be specified in different ways in the
config
> > file.
> > > Please try setting up your MODE config file like this:
> > >
> > > // This is for the GRIB file
> > > fcst = {
> > > field = {
> > > name = "APCP";
> > > level = "A24";
> > > };
> > > ... all the other settings go here ...
> > > };
> > >
> > > // This is for the NetCDF file
> > > obs = {
> > > field = {
> > > name = "APCP";
> > > level = "(*,*)";
> > > };
> > > ... all the other settings go here ...
> > > };
> > >
> > > There is also a problem in the NetCDF APCP variable attributes -
you're
> > > missing "init_time_ut" and "valid_time_ut". "ut" stands for
unix time,
> > > which is the number of seconds since Jan 1, 1970. MET reads
those
> > variable
> > > attributes to get the timing information.
> > >
> > > I see the following in the file now:
> > > APCP:valid_time = "20140505_060000" ;
> > > APCP:init_time = "20140505_000000" ;
> > >
> > > You can convert those to unix time using the following commands:
> > > date -ud ''2014-05-05' UTC '06:00:00'' +%s
> > > 1399269600
> > > date -ud ''2014-05-05' UTC '00:00:00'' +%s
> > > 1399248000
> > >
> > > Add them to that existing NetCDF file with this "ncatted"
command:
> > > ncatted -a init_time_ut,APCP,a,ll,1399248000 -a
> > > valid_time_ut,APCP,a,ll,1399269600 prec20140505.txt.1x1.nc
> > >
> > > Please try running MODE again with these changes. If you
experience
> more
> > > problems, please send me your GRIB forecast file, NetCDF
observation
> > file,
> > > and MODE configuration file (via our ftp site).
> > >
> > > Regarding the second file, ppradar_2014-05-05-18.nc, METv4.1 is
not
> able
> > > to
> > > read it. It is a CF-compliant NetCDF file but METv4.1 does not
support
> > > that. However, METv5.0 (which is due out tomorrow), does
support
> > > CF-compliant files. I was able to plot this with the METv5.0
version
> of
> > > plot_data_plane. But that file doesn't have any timing
information
> other
> > > than in the file name, so it produced some warnings.
> > >
> > > I'd suggest downloading METv5.0 when it is released tomorrow -
> although I
> > > think we'll send out the email about it next Monday.
> > >
> > > Thanks,
> > > John
> > >
> > >
> > >
> > >
> > >
> > > On Thu, Sep 4, 2014 at 10:17 AM, Otilia Diaconu via RT <
> > met_help at ucar.edu>
> > > wrote:
> > >
> > > >
> > > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=68848
>
> > > >
> > > > Hi John,
> > > > Thank you,
> > > > I put the file in the directory "diaconu'
> > > > I have 2 files:
> > > > prec* is created using ncl-ncar and
> > > > radar* (created by a collegue of mine) with R. with second
file the
> > > error
> > > > is : Atribute not found . I see that in that file some
atribute are
> > > missing
> > > > ..
> > > > Regards,
> > > >
> > > >
> > >
> > >
> >
> >
>
>
------------------------------------------------
Subject: pb_with_netcdf_file
From: John Halley Gotway
Time: Wed Sep 10 11:02:34 2014
Otilia,
Sorry for all this confusion, and I apologize for the delay in getting
back
with you. We were busy getting that release out the door.
The trouble here is that you're mixing file formats - you're using
elements
of the CF-compliant file and elements of the internal MET NetCDF file
format. When MET sees this attribute:
Conventions = "CF-1.4" ;
It interprets the file as being CF-compliant. But specifying timing
information for CF-compliant files is not done in the same way as MET
NetCDF files. That's why those variable attributes aren't having any
effect.
I've attached the output of ncdump for a CF-compliant NetCDF file that
includes timing info. You'd need to mimic that usage for MET to
interpret
it correctly.
Alternatively, you could go with the MET NetCDF file format and mimic
the
output of the pcp_combine tool.
When you decide which path you'd like to take, let me know what
additional
questions arise.
Thanks,
John
On Tue, Sep 9, 2014 at 10:24 PM, Otilia Diaconu via RT
<met_help at ucar.edu>
wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=68848 >
>
> Hi John,
> I successfully installed version 5.0 of MET. My colleague worked at
NetCDF
> files observations to have a complete header. The header of obs
files is:
>
> dimensions:
> lon = 77 ;
> lat = 51 ;
> variables:
> double lon(lon) ;
> lon:units = "degrees_east" ;
> lon:long_name = "lon" ;
> double lat(lat) ;
> lat:units = "degrees_north" ;
> lat:long_name = "lat" ;
> float APCP(lat, lon) ;
> APCP:units = "kg/m^2" ;
> APCP:_FillValue = -9999.f ;
> APCP:missing_value = -9999. ;
> APCP:long_name = "Total precipitation" ;
> APCP:projection = "+init=epsg:4326 +proj=longlat
> +datum=WGS84 +no_defs +ellps=WGS84 +towgs84=0,0,0" ;
> APCP:projection_format = "PROJ.4" ;
> APCP:min = -9999. ;
> APCP:max = 20.3 ;
> APCP:accum_time_sec = 43200 ;
> APCP:accum_time = "12 hours" ;
> APCP:valid_time = "20140505_180000" ;
> APCP:init_time = "20140505_060000" ;
> APCP:init_time_ut = 1399269600 ;
> APCP:valid_time_ut = 1399312800 ;
>
> // global attributes:
> :Conventions = "CF-1.4" ;
> :created_by = "R, packages ncdf and raster (version
2.3-0)"
> ;
> :date = "2014-09-09 13:20:37" ;
> :nco_openmp_thread_number = 1 ;
> :Nlon = "77 grid_points" ;
> :Nlat = "51 grid_points" ;
> :delta_lon_deg = "0.125000 degrees" ;
> :delta_lat_deg = "0.100000 degrees" ;
> :lon_ll_deg = "20.200000 degrees_east" ;
> :lat_ll_deg = "43.500000 degrees_north" ;
> :Projection = "LatLon" ;
> The MODE from 5.0 version works now but, even if the in the header I
have
> the correct attribute* time_ut ,* the summary of "mode" doesn't
> transform'it.
> I have attached one immage with this !
> Thank you in avance for your help,
> Best regards,
> Otilia Diaconu
> ᐧ
>
> On Fri, Sep 5, 2014 at 12:46 PM, Otilia Diaconu via RT
<met_help at ucar.edu>
> wrote:
>
> > Hi John,
> > made the necessary changes to the configuration file and in "necdf
> > atributes". I have run METv4.1 .(.5.0 is not available at this
time ),
> but
> > The netcdf file cannot be read
> > *process_fcst_obs_files() -> can't get data from file*
> > prec20140505.txt.1x1.nc
> > I have put the Grib file and the Confifg file in directory
"diaconu". If
> > you have time to see..
> > Thank you. A good day and nice weekend,
> > Otilia
> > ᐧ
> >
> >
> > On Thu, Sep 4, 2014 at 11:28 PM, Otilia Diaconu via RT <
> met_help at ucar.edu>
> > wrote:
> >
> > > Thank you very much ,
> > > I know the problem with time .. (changes in ncl-ncar new
versions since
> > 5.2
> > > used in 2010).. I will be able to set-up the correct time.
> > > I will made the necessary corrections tomorrow ..
> > > It's 23:22 here ..
> > > I will download at install 5.0 version and I will try to use
"radar"
> > > files.. My collegue works to set-up in R the others atributes
for those
> > > files.
> > > Many thanks,
> > > Regards
> > > Otilia
> > >
> > >
> > > On Thu, Sep 4, 2014 at 11:07 PM, John Halley Gotway via RT <
> > > met_help at ucar.edu> wrote:
> > >
> > > > Otilia,
> > > >
> > > > Thanks for sending those files.
> > > >
> > > > I was able to run the prec20140505.txt.1x1.nc file through the
MET
> > > > plot_data_plane tool to verify that MET can read it:
> > > > METv4.1/bin/plot_data_plane prec20140505.txt.1x1.nc
> > > > prec20140505.txt.1x1.ps 'name="APCP";level="(*,*)";'
> > > > I converted the resulting image into png format and have
attached it
> to
> > > > this message.
> > > >
> > > > I suspect that the problem is in how you have the config file
set up.
> > > > You're running MODE to compare a GRIB forecast file to a
NetCDF
> > > observation
> > > > file, and those need to be specified in different ways in the
config
> > > file.
> > > > Please try setting up your MODE config file like this:
> > > >
> > > > // This is for the GRIB file
> > > > fcst = {
> > > > field = {
> > > > name = "APCP";
> > > > level = "A24";
> > > > };
> > > > ... all the other settings go here ...
> > > > };
> > > >
> > > > // This is for the NetCDF file
> > > > obs = {
> > > > field = {
> > > > name = "APCP";
> > > > level = "(*,*)";
> > > > };
> > > > ... all the other settings go here ...
> > > > };
> > > >
> > > > There is also a problem in the NetCDF APCP variable attributes
-
> you're
> > > > missing "init_time_ut" and "valid_time_ut". "ut" stands for
unix
> time,
> > > > which is the number of seconds since Jan 1, 1970. MET reads
those
> > > variable
> > > > attributes to get the timing information.
> > > >
> > > > I see the following in the file now:
> > > > APCP:valid_time = "20140505_060000" ;
> > > > APCP:init_time = "20140505_000000" ;
> > > >
> > > > You can convert those to unix time using the following
commands:
> > > > date -ud ''2014-05-05' UTC '06:00:00'' +%s
> > > > 1399269600
> > > > date -ud ''2014-05-05' UTC '00:00:00'' +%s
> > > > 1399248000
> > > >
> > > > Add them to that existing NetCDF file with this "ncatted"
command:
> > > > ncatted -a init_time_ut,APCP,a,ll,1399248000 -a
> > > > valid_time_ut,APCP,a,ll,1399269600 prec20140505.txt.1x1.nc
> > > >
> > > > Please try running MODE again with these changes. If you
experience
> > more
> > > > problems, please send me your GRIB forecast file, NetCDF
observation
> > > file,
> > > > and MODE configuration file (via our ftp site).
> > > >
> > > > Regarding the second file, ppradar_2014-05-05-18.nc, METv4.1
is not
> > able
> > > > to
> > > > read it. It is a CF-compliant NetCDF file but METv4.1 does
not
> support
> > > > that. However, METv5.0 (which is due out tomorrow), does
support
> > > > CF-compliant files. I was able to plot this with the METv5.0
version
> > of
> > > > plot_data_plane. But that file doesn't have any timing
information
> > other
> > > > than in the file name, so it produced some warnings.
> > > >
> > > > I'd suggest downloading METv5.0 when it is released tomorrow -
> > although I
> > > > think we'll send out the email about it next Monday.
> > > >
> > > > Thanks,
> > > > John
> > > >
> > > >
> > > >
> > > >
> > > >
> > > > On Thu, Sep 4, 2014 at 10:17 AM, Otilia Diaconu via RT <
> > > met_help at ucar.edu>
> > > > wrote:
> > > >
> > > > >
> > > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=68848 >
> > > > >
> > > > > Hi John,
> > > > > Thank you,
> > > > > I put the file in the directory "diaconu'
> > > > > I have 2 files:
> > > > > prec* is created using ncl-ncar and
> > > > > radar* (created by a collegue of mine) with R. with second
file
> the
> > > > error
> > > > > is : Atribute not found . I see that in that file some
atribute are
> > > > missing
> > > > > ..
> > > > > Regards,
> > > > >
> > > > >
> > > >
> > > >
> > >
> > >
> >
> >
>
>
------------------------------------------------
Subject: pb_with_netcdf_file
From: Otilia Diaconu
Time: Wed Sep 10 11:15:21 2014
Hi John,
Thank you for the message and explanations. I have supposed that the
CF-compliants could be the pb..My collegue works with R and the netcdf
file
is created with R. I am working witht "ncl ncar" and my netcdf files
are OK
for MET. I haven't a header on the "good" file .. to see what the time
is
included. I'll try tomorrow with my collegue, and, of course, I will
informe you about it. May be I wiil try to read the file in "ncl" and
re-write in "classical' format !
Regards
Otilia
On Wed, Sep 10, 2014 at 8:02 PM, John Halley Gotway via RT <
met_help at ucar.edu> wrote:
> Otilia,
>
> Sorry for all this confusion, and I apologize for the delay in
getting back
> with you. We were busy getting that release out the door.
>
> The trouble here is that you're mixing file formats - you're using
elements
> of the CF-compliant file and elements of the internal MET NetCDF
file
> format. When MET sees this attribute:
> Conventions = "CF-1.4" ;
>
> It interprets the file as being CF-compliant. But specifying timing
> information for CF-compliant files is not done in the same way as
MET
> NetCDF files. That's why those variable attributes aren't having
any
> effect.
>
> I've attached the output of ncdump for a CF-compliant NetCDF file
that
> includes timing info. You'd need to mimic that usage for MET to
interpret
> it correctly.
>
> Alternatively, you could go with the MET NetCDF file format and
mimic the
> output of the pcp_combine tool.
>
> When you decide which path you'd like to take, let me know what
additional
> questions arise.
>
> Thanks,
> John
>
>
>
>
> On Tue, Sep 9, 2014 at 10:24 PM, Otilia Diaconu via RT
<met_help at ucar.edu>
> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=68848 >
> >
> > Hi John,
> > I successfully installed version 5.0 of MET. My colleague worked
at
> NetCDF
> > files observations to have a complete header. The header of obs
files is:
> >
> > dimensions:
> > lon = 77 ;
> > lat = 51 ;
> > variables:
> > double lon(lon) ;
> > lon:units = "degrees_east" ;
> > lon:long_name = "lon" ;
> > double lat(lat) ;
> > lat:units = "degrees_north" ;
> > lat:long_name = "lat" ;
> > float APCP(lat, lon) ;
> > APCP:units = "kg/m^2" ;
> > APCP:_FillValue = -9999.f ;
> > APCP:missing_value = -9999. ;
> > APCP:long_name = "Total precipitation" ;
> > APCP:projection = "+init=epsg:4326 +proj=longlat
> > +datum=WGS84 +no_defs +ellps=WGS84 +towgs84=0,0,0" ;
> > APCP:projection_format = "PROJ.4" ;
> > APCP:min = -9999. ;
> > APCP:max = 20.3 ;
> > APCP:accum_time_sec = 43200 ;
> > APCP:accum_time = "12 hours" ;
> > APCP:valid_time = "20140505_180000" ;
> > APCP:init_time = "20140505_060000" ;
> > APCP:init_time_ut = 1399269600 ;
> > APCP:valid_time_ut = 1399312800 ;
> >
> > // global attributes:
> > :Conventions = "CF-1.4" ;
> > :created_by = "R, packages ncdf and raster
(version
> 2.3-0)"
> > ;
> > :date = "2014-09-09 13:20:37" ;
> > :nco_openmp_thread_number = 1 ;
> > :Nlon = "77 grid_points" ;
> > :Nlat = "51 grid_points" ;
> > :delta_lon_deg = "0.125000 degrees" ;
> > :delta_lat_deg = "0.100000 degrees" ;
> > :lon_ll_deg = "20.200000 degrees_east" ;
> > :lat_ll_deg = "43.500000 degrees_north" ;
> > :Projection = "LatLon" ;
> > The MODE from 5.0 version works now but, even if the in the header
I have
> > the correct attribute* time_ut ,* the summary of "mode" doesn't
> > transform'it.
> > I have attached one immage with this !
> > Thank you in avance for your help,
> > Best regards,
> > Otilia Diaconu
> > ᐧ
> >
> > On Fri, Sep 5, 2014 at 12:46 PM, Otilia Diaconu via RT <
> met_help at ucar.edu>
> > wrote:
> >
> > > Hi John,
> > > made the necessary changes to the configuration file and in
"necdf
> > > atributes". I have run METv4.1 .(.5.0 is not available at this
time ),
> > but
> > > The netcdf file cannot be read
> > > *process_fcst_obs_files() -> can't get data from file*
> > > prec20140505.txt.1x1.nc
> > > I have put the Grib file and the Confifg file in directory
"diaconu".
> If
> > > you have time to see..
> > > Thank you. A good day and nice weekend,
> > > Otilia
> > > ᐧ
> > >
> > >
> > > On Thu, Sep 4, 2014 at 11:28 PM, Otilia Diaconu via RT <
> > met_help at ucar.edu>
> > > wrote:
> > >
> > > > Thank you very much ,
> > > > I know the problem with time .. (changes in ncl-ncar new
versions
> since
> > > 5.2
> > > > used in 2010).. I will be able to set-up the correct time.
> > > > I will made the necessary corrections tomorrow ..
> > > > It's 23:22 here ..
> > > > I will download at install 5.0 version and I will try to use
"radar"
> > > > files.. My collegue works to set-up in R the others atributes
for
> those
> > > > files.
> > > > Many thanks,
> > > > Regards
> > > > Otilia
> > > >
> > > >
> > > > On Thu, Sep 4, 2014 at 11:07 PM, John Halley Gotway via RT <
> > > > met_help at ucar.edu> wrote:
> > > >
> > > > > Otilia,
> > > > >
> > > > > Thanks for sending those files.
> > > > >
> > > > > I was able to run the prec20140505.txt.1x1.nc file through
the MET
> > > > > plot_data_plane tool to verify that MET can read it:
> > > > > METv4.1/bin/plot_data_plane prec20140505.txt.1x1.nc
> > > > > prec20140505.txt.1x1.ps 'name="APCP";level="(*,*)";'
> > > > > I converted the resulting image into png format and have
attached
> it
> > to
> > > > > this message.
> > > > >
> > > > > I suspect that the problem is in how you have the config
file set
> up.
> > > > > You're running MODE to compare a GRIB forecast file to a
NetCDF
> > > > observation
> > > > > file, and those need to be specified in different ways in
the
> config
> > > > file.
> > > > > Please try setting up your MODE config file like this:
> > > > >
> > > > > // This is for the GRIB file
> > > > > fcst = {
> > > > > field = {
> > > > > name = "APCP";
> > > > > level = "A24";
> > > > > };
> > > > > ... all the other settings go here ...
> > > > > };
> > > > >
> > > > > // This is for the NetCDF file
> > > > > obs = {
> > > > > field = {
> > > > > name = "APCP";
> > > > > level = "(*,*)";
> > > > > };
> > > > > ... all the other settings go here ...
> > > > > };
> > > > >
> > > > > There is also a problem in the NetCDF APCP variable
attributes -
> > you're
> > > > > missing "init_time_ut" and "valid_time_ut". "ut" stands for
unix
> > time,
> > > > > which is the number of seconds since Jan 1, 1970. MET reads
those
> > > > variable
> > > > > attributes to get the timing information.
> > > > >
> > > > > I see the following in the file now:
> > > > > APCP:valid_time = "20140505_060000" ;
> > > > > APCP:init_time = "20140505_000000" ;
> > > > >
> > > > > You can convert those to unix time using the following
commands:
> > > > > date -ud ''2014-05-05' UTC '06:00:00'' +%s
> > > > > 1399269600
> > > > > date -ud ''2014-05-05' UTC '00:00:00'' +%s
> > > > > 1399248000
> > > > >
> > > > > Add them to that existing NetCDF file with this "ncatted"
command:
> > > > > ncatted -a init_time_ut,APCP,a,ll,1399248000 -a
> > > > > valid_time_ut,APCP,a,ll,1399269600 prec20140505.txt.1x1.nc
> > > > >
> > > > > Please try running MODE again with these changes. If you
> experience
> > > more
> > > > > problems, please send me your GRIB forecast file, NetCDF
> observation
> > > > file,
> > > > > and MODE configuration file (via our ftp site).
> > > > >
> > > > > Regarding the second file, ppradar_2014-05-05-18.nc, METv4.1
is
> not
> > > able
> > > > > to
> > > > > read it. It is a CF-compliant NetCDF file but METv4.1 does
not
> > support
> > > > > that. However, METv5.0 (which is due out tomorrow), does
support
> > > > > CF-compliant files. I was able to plot this with the
METv5.0
> version
> > > of
> > > > > plot_data_plane. But that file doesn't have any timing
information
> > > other
> > > > > than in the file name, so it produced some warnings.
> > > > >
> > > > > I'd suggest downloading METv5.0 when it is released tomorrow
-
> > > although I
> > > > > think we'll send out the email about it next Monday.
> > > > >
> > > > > Thanks,
> > > > > John
> > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > On Thu, Sep 4, 2014 at 10:17 AM, Otilia Diaconu via RT <
> > > > met_help at ucar.edu>
> > > > > wrote:
> > > > >
> > > > > >
> > > > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=68848 >
> > > > > >
> > > > > > Hi John,
> > > > > > Thank you,
> > > > > > I put the file in the directory "diaconu'
> > > > > > I have 2 files:
> > > > > > prec* is created using ncl-ncar and
> > > > > > radar* (created by a collegue of mine) with R. with
second file
> > the
> > > > > error
> > > > > > is : Atribute not found . I see that in that file some
atribute
> are
> > > > > missing
> > > > > > ..
> > > > > > Regards,
> > > > > >
> > > > > >
> > > > >
> > > > >
> > > >
> > > >
> > >
> > >
> >
> >
>
>
------------------------------------------------
Subject: pb_with_netcdf_file
From: John Halley Gotway
Time: Wed Sep 10 13:33:11 2014
OK, I'll resolve this issue for now. If more questions or problems
arise,
please let us know.
Thanks,
John
On Wed, Sep 10, 2014 at 11:15 AM, Otilia Diaconu via RT
<met_help at ucar.edu>
wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=68848 >
>
> Hi John,
> Thank you for the message and explanations. I have supposed that the
> CF-compliants could be the pb..My collegue works with R and the
netcdf file
> is created with R. I am working witht "ncl ncar" and my netcdf files
are OK
> for MET. I haven't a header on the "good" file .. to see what the
time is
> included. I'll try tomorrow with my collegue, and, of course, I will
> informe you about it. May be I wiil try to read the file in "ncl"
and
> re-write in "classical' format !
> Regards
> Otilia
>
> On Wed, Sep 10, 2014 at 8:02 PM, John Halley Gotway via RT <
> met_help at ucar.edu> wrote:
>
> > Otilia,
> >
> > Sorry for all this confusion, and I apologize for the delay in
getting
> back
> > with you. We were busy getting that release out the door.
> >
> > The trouble here is that you're mixing file formats - you're using
> elements
> > of the CF-compliant file and elements of the internal MET NetCDF
file
> > format. When MET sees this attribute:
> > Conventions = "CF-1.4" ;
> >
> > It interprets the file as being CF-compliant. But specifying
timing
> > information for CF-compliant files is not done in the same way as
MET
> > NetCDF files. That's why those variable attributes aren't having
any
> > effect.
> >
> > I've attached the output of ncdump for a CF-compliant NetCDF file
that
> > includes timing info. You'd need to mimic that usage for MET to
> interpret
> > it correctly.
> >
> > Alternatively, you could go with the MET NetCDF file format and
mimic the
> > output of the pcp_combine tool.
> >
> > When you decide which path you'd like to take, let me know what
> additional
> > questions arise.
> >
> > Thanks,
> > John
> >
> >
> >
> >
> > On Tue, Sep 9, 2014 at 10:24 PM, Otilia Diaconu via RT <
> met_help at ucar.edu>
> > wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=68848 >
> > >
> > > Hi John,
> > > I successfully installed version 5.0 of MET. My colleague
worked at
> > NetCDF
> > > files observations to have a complete header. The header of obs
files
> is:
> > >
> > > dimensions:
> > > lon = 77 ;
> > > lat = 51 ;
> > > variables:
> > > double lon(lon) ;
> > > lon:units = "degrees_east" ;
> > > lon:long_name = "lon" ;
> > > double lat(lat) ;
> > > lat:units = "degrees_north" ;
> > > lat:long_name = "lat" ;
> > > float APCP(lat, lon) ;
> > > APCP:units = "kg/m^2" ;
> > > APCP:_FillValue = -9999.f ;
> > > APCP:missing_value = -9999. ;
> > > APCP:long_name = "Total precipitation" ;
> > > APCP:projection = "+init=epsg:4326 +proj=longlat
> > > +datum=WGS84 +no_defs +ellps=WGS84 +towgs84=0,0,0" ;
> > > APCP:projection_format = "PROJ.4" ;
> > > APCP:min = -9999. ;
> > > APCP:max = 20.3 ;
> > > APCP:accum_time_sec = 43200 ;
> > > APCP:accum_time = "12 hours" ;
> > > APCP:valid_time = "20140505_180000" ;
> > > APCP:init_time = "20140505_060000" ;
> > > APCP:init_time_ut = 1399269600 ;
> > > APCP:valid_time_ut = 1399312800 ;
> > >
> > > // global attributes:
> > > :Conventions = "CF-1.4" ;
> > > :created_by = "R, packages ncdf and raster
(version
> > 2.3-0)"
> > > ;
> > > :date = "2014-09-09 13:20:37" ;
> > > :nco_openmp_thread_number = 1 ;
> > > :Nlon = "77 grid_points" ;
> > > :Nlat = "51 grid_points" ;
> > > :delta_lon_deg = "0.125000 degrees" ;
> > > :delta_lat_deg = "0.100000 degrees" ;
> > > :lon_ll_deg = "20.200000 degrees_east" ;
> > > :lat_ll_deg = "43.500000 degrees_north" ;
> > > :Projection = "LatLon" ;
> > > The MODE from 5.0 version works now but, even if the in the
header I
> have
> > > the correct attribute* time_ut ,* the summary of "mode" doesn't
> > > transform'it.
> > > I have attached one immage with this !
> > > Thank you in avance for your help,
> > > Best regards,
> > > Otilia Diaconu
> > > ᐧ
> > >
> > > On Fri, Sep 5, 2014 at 12:46 PM, Otilia Diaconu via RT <
> > met_help at ucar.edu>
> > > wrote:
> > >
> > > > Hi John,
> > > > made the necessary changes to the configuration file and in
"necdf
> > > > atributes". I have run METv4.1 .(.5.0 is not available at this
time
> ),
> > > but
> > > > The netcdf file cannot be read
> > > > *process_fcst_obs_files() -> can't get data from file*
> > > > prec20140505.txt.1x1.nc
> > > > I have put the Grib file and the Confifg file in directory
"diaconu".
> > If
> > > > you have time to see..
> > > > Thank you. A good day and nice weekend,
> > > > Otilia
> > > > ᐧ
> > > >
> > > >
> > > > On Thu, Sep 4, 2014 at 11:28 PM, Otilia Diaconu via RT <
> > > met_help at ucar.edu>
> > > > wrote:
> > > >
> > > > > Thank you very much ,
> > > > > I know the problem with time .. (changes in ncl-ncar new
versions
> > since
> > > > 5.2
> > > > > used in 2010).. I will be able to set-up the correct time.
> > > > > I will made the necessary corrections tomorrow ..
> > > > > It's 23:22 here ..
> > > > > I will download at install 5.0 version and I will try to use
> "radar"
> > > > > files.. My collegue works to set-up in R the others
atributes for
> > those
> > > > > files.
> > > > > Many thanks,
> > > > > Regards
> > > > > Otilia
> > > > >
> > > > >
> > > > > On Thu, Sep 4, 2014 at 11:07 PM, John Halley Gotway via RT <
> > > > > met_help at ucar.edu> wrote:
> > > > >
> > > > > > Otilia,
> > > > > >
> > > > > > Thanks for sending those files.
> > > > > >
> > > > > > I was able to run the prec20140505.txt.1x1.nc file through
the
> MET
> > > > > > plot_data_plane tool to verify that MET can read it:
> > > > > > METv4.1/bin/plot_data_plane prec20140505.txt.1x1.nc
> > > > > > prec20140505.txt.1x1.ps 'name="APCP";level="(*,*)";'
> > > > > > I converted the resulting image into png format and have
attached
> > it
> > > to
> > > > > > this message.
> > > > > >
> > > > > > I suspect that the problem is in how you have the config
file set
> > up.
> > > > > > You're running MODE to compare a GRIB forecast file to a
NetCDF
> > > > > observation
> > > > > > file, and those need to be specified in different ways in
the
> > config
> > > > > file.
> > > > > > Please try setting up your MODE config file like this:
> > > > > >
> > > > > > // This is for the GRIB file
> > > > > > fcst = {
> > > > > > field = {
> > > > > > name = "APCP";
> > > > > > level = "A24";
> > > > > > };
> > > > > > ... all the other settings go here ...
> > > > > > };
> > > > > >
> > > > > > // This is for the NetCDF file
> > > > > > obs = {
> > > > > > field = {
> > > > > > name = "APCP";
> > > > > > level = "(*,*)";
> > > > > > };
> > > > > > ... all the other settings go here ...
> > > > > > };
> > > > > >
> > > > > > There is also a problem in the NetCDF APCP variable
attributes -
> > > you're
> > > > > > missing "init_time_ut" and "valid_time_ut". "ut" stands
for unix
> > > time,
> > > > > > which is the number of seconds since Jan 1, 1970. MET
reads
> those
> > > > > variable
> > > > > > attributes to get the timing information.
> > > > > >
> > > > > > I see the following in the file now:
> > > > > > APCP:valid_time = "20140505_060000" ;
> > > > > > APCP:init_time = "20140505_000000" ;
> > > > > >
> > > > > > You can convert those to unix time using the following
commands:
> > > > > > date -ud ''2014-05-05' UTC '06:00:00'' +%s
> > > > > > 1399269600
> > > > > > date -ud ''2014-05-05' UTC '00:00:00'' +%s
> > > > > > 1399248000
> > > > > >
> > > > > > Add them to that existing NetCDF file with this "ncatted"
> command:
> > > > > > ncatted -a init_time_ut,APCP,a,ll,1399248000 -a
> > > > > > valid_time_ut,APCP,a,ll,1399269600 prec20140505.txt.1x1.nc
> > > > > >
> > > > > > Please try running MODE again with these changes. If you
> > experience
> > > > more
> > > > > > problems, please send me your GRIB forecast file, NetCDF
> > observation
> > > > > file,
> > > > > > and MODE configuration file (via our ftp site).
> > > > > >
> > > > > > Regarding the second file, ppradar_2014-05-05-18.nc,
METv4.1 is
> > not
> > > > able
> > > > > > to
> > > > > > read it. It is a CF-compliant NetCDF file but METv4.1
does not
> > > support
> > > > > > that. However, METv5.0 (which is due out tomorrow), does
support
> > > > > > CF-compliant files. I was able to plot this with the
METv5.0
> > version
> > > > of
> > > > > > plot_data_plane. But that file doesn't have any timing
> information
> > > > other
> > > > > > than in the file name, so it produced some warnings.
> > > > > >
> > > > > > I'd suggest downloading METv5.0 when it is released
tomorrow -
> > > > although I
> > > > > > think we'll send out the email about it next Monday.
> > > > > >
> > > > > > Thanks,
> > > > > > John
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > On Thu, Sep 4, 2014 at 10:17 AM, Otilia Diaconu via RT <
> > > > > met_help at ucar.edu>
> > > > > > wrote:
> > > > > >
> > > > > > >
> > > > > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=68848
> >
> > > > > > >
> > > > > > > Hi John,
> > > > > > > Thank you,
> > > > > > > I put the file in the directory "diaconu'
> > > > > > > I have 2 files:
> > > > > > > prec* is created using ncl-ncar and
> > > > > > > radar* (created by a collegue of mine) with R. with
second
> file
> > > the
> > > > > > error
> > > > > > > is : Atribute not found . I see that in that file some
atribute
> > are
> > > > > > missing
> > > > > > > ..
> > > > > > > Regards,
> > > > > > >
> > > > > > >
> > > > > >
> > > > > >
> > > > >
> > > > >
> > > >
> > > >
> > >
> > >
> >
> >
>
>
------------------------------------------------
More information about the Met_help
mailing list