[Met_help] [rt.rap.ucar.edu #55226] History for Error running MET mode tool
John Halley Gotway via RT
met_help at ucar.edu
Tue Apr 10 14:00:28 MDT 2012
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Hi, I have come as far as converting the observation data and forecast data
into the right format for MET mode tool.
There is still an error reported by MET during the generation of the output
".nc" file. Here's the message: *NetCDF: Numeric conversion not
representable.*
The full text is below. Appreciate if you could assist with this.
PS: the warning about timing is expected because I did not provide a proper
output time for my obs file.
Thanks,
Raizan
***WARNING***: process_fcst_obs_files() -> Forecast and observation valid
times do not match 20110807_180000 != -00011130_000000.
Forecast Field: PRATE at ZSFC
Observation Field: PRES at ZSFC
Identifying objects in the forecast and observation fields...
Computing contingency table statistics...
Identified: 0 forecast objects and 9 observation objects.
Performing merging (no merging) in the forecast field.
Performing merging (no merging) in the observation field.
Remaining: 0 forecast objects and 9 observation objects.
Performing matching between the forecast and observation fields.
Creating Contingency Table Statistics file:
../out/mode/mode_360000L_20110807_180000V_000000A_cts.txt
Creating Fcst-Obs Object Statistics file:
../out/mode/mode_360000L_20110807_180000V_000000A_obj.txt
Creating Object NetCDF file:
../out/mode/mode_360000L_20110807_180000V_000000A_obj.nc
*NetCDF: Numeric conversion not representable*
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #55226] Error running MET mode tool
From: John Halley Gotway
Time: Fri Mar 16 16:25:23 2012
Raizan,
Can you please send me the input data for MODE (forecast, observation,
and config file) so that I can run it here and reproduce your error?
Just follow the instructions here to post that data to our anonymous
ftp site:
http://www.dtcenter.org/met/users/support/met_help.php#ftp
Thanks,
John
On 03/16/2012 04:13 PM, Zan Rahmat via RT wrote:
>
> Fri Mar 16 16:13:49 2012: Request 55226 was acted upon.
> Transaction: Ticket created by raizan.rahmat at gmail.com
> Queue: met_help
> Subject: Error running MET mode tool
> Owner: Nobody
> Requestors: raizan.rahmat at gmail.com
> Status: new
> Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55226>
>
>
> Hi, I have come as far as converting the observation data and
forecast data
> into the right format for MET mode tool.
>
> There is still an error reported by MET during the generation of the
output
> ".nc" file. Here's the message: *NetCDF: Numeric conversion not
> representable.*
>
> The full text is below. Appreciate if you could assist with this.
>
> PS: the warning about timing is expected because I did not provide a
proper
> output time for my obs file.
>
> Thanks,
> Raizan
>
>
> ***WARNING***: process_fcst_obs_files() -> Forecast and observation
valid
> times do not match 20110807_180000 != -00011130_000000.
>
> Forecast Field: PRATE at ZSFC
> Observation Field: PRES at ZSFC
> Identifying objects in the forecast and observation fields...
> Computing contingency table statistics...
> Identified: 0 forecast objects and 9 observation objects.
> Performing merging (no merging) in the forecast field.
> Performing merging (no merging) in the observation field.
> Remaining: 0 forecast objects and 9 observation objects.
> Performing matching between the forecast and observation fields.
> Creating Contingency Table Statistics file:
> ../out/mode/mode_360000L_20110807_180000V_000000A_cts.txt
> Creating Fcst-Obs Object Statistics file:
> ../out/mode/mode_360000L_20110807_180000V_000000A_obj.txt
> Creating Object NetCDF file:
> ../out/mode/mode_360000L_20110807_180000V_000000A_obj.nc
> *NetCDF: Numeric conversion not representable*
------------------------------------------------
Subject: Error running MET mode tool
From: Zan Rahmat
Time: Fri Mar 16 16:48:52 2012
Dear John,
Thank you for the prompt reply. I have uploaded the 3 files.
I would just like to highlight that my obs variable has also not been
named
properly. When it was in netcdf format tha variable name was var1,
when I
converted it to grib using cdo, somehow it got named to PRES. I hope
this
is consequential to the output/outcome. What's in the obs file is
precipitation rate anyway.
Thanks,
Raizan
On 16 March 2012 22:25, John Halley Gotway via RT <met_help at ucar.edu>
wrote:
> Raizan,
>
> Can you please send me the input data for MODE (forecast,
observation, and
> config file) so that I can run it here and reproduce your error?
>
> Just follow the instructions here to post that data to our anonymous
ftp
> site:
> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>
> Thanks,
> John
>
> On 03/16/2012 04:13 PM, Zan Rahmat via RT wrote:
> >
> > Fri Mar 16 16:13:49 2012: Request 55226 was acted upon.
> > Transaction: Ticket created by raizan.rahmat at gmail.com
> > Queue: met_help
> > Subject: Error running MET mode tool
> > Owner: Nobody
> > Requestors: raizan.rahmat at gmail.com
> > Status: new
> > Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55226>
> >
> >
> > Hi, I have come as far as converting the observation data and
forecast
> data
> > into the right format for MET mode tool.
> >
> > There is still an error reported by MET during the generation of
the
> output
> > ".nc" file. Here's the message: *NetCDF: Numeric conversion not
> > representable.*
> >
> > The full text is below. Appreciate if you could assist with this.
> >
> > PS: the warning about timing is expected because I did not provide
a
> proper
> > output time for my obs file.
> >
> > Thanks,
> > Raizan
> >
> >
> > ***WARNING***: process_fcst_obs_files() -> Forecast and
observation
> valid
> > times do not match 20110807_180000 != -00011130_000000.
> >
> > Forecast Field: PRATE at ZSFC
> > Observation Field: PRES at ZSFC
> > Identifying objects in the forecast and observation fields...
> > Computing contingency table statistics...
> > Identified: 0 forecast objects and 9 observation objects.
> > Performing merging (no merging) in the forecast field.
> > Performing merging (no merging) in the observation field.
> > Remaining: 0 forecast objects and 9 observation objects.
> > Performing matching between the forecast and observation fields.
> > Creating Contingency Table Statistics file:
> > ../out/mode/mode_360000L_20110807_180000V_000000A_cts.txt
> > Creating Fcst-Obs Object Statistics file:
> > ../out/mode/mode_360000L_20110807_180000V_000000A_obj.txt
> > Creating Object NetCDF file:
> > ../out/mode/mode_360000L_20110807_180000V_000000A_obj.nc
> > *NetCDF: Numeric conversion not representable*
>
>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #55226] Error running MET mode tool
From: John Halley Gotway
Time: Fri Mar 16 17:08:20 2012
Raizan,
I was able to run with your data and reproduce the error message.
One option is to use the "-obj_plot" command line argument for MODE.
That disables the generation of the output NetCDF file that's causing
the error.
But the real problem is coming from your GRIB observation file. I
used "wgrib -V" on it and see the following output:
[johnhg at rambler]% wgrib -V ofile_struct_regridded.grib
Undefined parameter table (center 0-0 table 255), using NCEP-opn
rec 1:0:date 0000000000 PRES kpds5=1 kpds6=1 kpds7=0 levels=(0,0)
grid=255 sfc anl: bitmap: 481 undef
PRES=Pressure [Pa]
timerange 10 P1 0 P2 0 TimeU 2 nx 241 ny 241 GDS grid 0 num_in_ave
0 missing 0
center 0 subcenter 0 process 0 Table 255 scan: WE:SN winds(N/S)
latlon: lat 50.000000 to 53.133000 by 0.013000 nxny 58081
long -5.800000 to -0.739000 by 0.021000, (241 x 241) scan
64 mode 128 bdsgrid 1
min/max data 0 5908 num bits 13 BDS_Ref 0 DecScale 0 BinScale 0
The "date 0000000000" listed as all 0's is problematic. The MET code
is parsing that info and trying to interpret that as a time. It's
trying to write that time as a variable attribute in that
output NetCDF file, and that's the source of the error. If you can
encode the time information correctly in that GRIB file this error
should be resolved.
As far as sanity checking the contents of your observation GRIB file,
you may find it useful to view it using the IDV tool.
One last thing, in your config file, the "ZSFC" won't work - try using
"Z0" instead. Or try the following:
fcst_field = "PRES/Z0";
obs_field = "PRES/Z0";
That compares surface pressure from the two files.
Hope that helps.
Thanks,
John
On 03/16/2012 04:48 PM, Zan Rahmat via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55226>
>
> Dear John,
>
> Thank you for the prompt reply. I have uploaded the 3 files.
>
> I would just like to highlight that my obs variable has also not
been named
> properly. When it was in netcdf format tha variable name was var1,
when I
> converted it to grib using cdo, somehow it got named to PRES. I hope
this
> is consequential to the output/outcome. What's in the obs file is
> precipitation rate anyway.
>
> Thanks,
> Raizan
>
> On 16 March 2012 22:25, John Halley Gotway via RT<met_help at ucar.edu>
wrote:
>
>> Raizan,
>>
>> Can you please send me the input data for MODE (forecast,
observation, and
>> config file) so that I can run it here and reproduce your error?
>>
>> Just follow the instructions here to post that data to our
anonymous ftp
>> site:
>> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>>
>> Thanks,
>> John
>>
>> On 03/16/2012 04:13 PM, Zan Rahmat via RT wrote:
>>>
>>> Fri Mar 16 16:13:49 2012: Request 55226 was acted upon.
>>> Transaction: Ticket created by raizan.rahmat at gmail.com
>>> Queue: met_help
>>> Subject: Error running MET mode tool
>>> Owner: Nobody
>>> Requestors: raizan.rahmat at gmail.com
>>> Status: new
>>> Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55226>
>>>
>>>
>>> Hi, I have come as far as converting the observation data and
forecast
>> data
>>> into the right format for MET mode tool.
>>>
>>> There is still an error reported by MET during the generation of
the
>> output
>>> ".nc" file. Here's the message: *NetCDF: Numeric conversion not
>>> representable.*
>>>
>>> The full text is below. Appreciate if you could assist with this.
>>>
>>> PS: the warning about timing is expected because I did not provide
a
>> proper
>>> output time for my obs file.
>>>
>>> Thanks,
>>> Raizan
>>>
>>>
>>> ***WARNING***: process_fcst_obs_files() -> Forecast and
observation
>> valid
>>> times do not match 20110807_180000 != -00011130_000000.
>>>
>>> Forecast Field: PRATE at ZSFC
>>> Observation Field: PRES at ZSFC
>>> Identifying objects in the forecast and observation fields...
>>> Computing contingency table statistics...
>>> Identified: 0 forecast objects and 9 observation objects.
>>> Performing merging (no merging) in the forecast field.
>>> Performing merging (no merging) in the observation field.
>>> Remaining: 0 forecast objects and 9 observation objects.
>>> Performing matching between the forecast and observation fields.
>>> Creating Contingency Table Statistics file:
>>> ../out/mode/mode_360000L_20110807_180000V_000000A_cts.txt
>>> Creating Fcst-Obs Object Statistics file:
>>> ../out/mode/mode_360000L_20110807_180000V_000000A_obj.txt
>>> Creating Object NetCDF file:
>>> ../out/mode/mode_360000L_20110807_180000V_000000A_obj.nc
>>> *NetCDF: Numeric conversion not representable*
>>
>>
------------------------------------------------
Subject: Error running MET mode tool
From: Zan Rahmat
Time: Fri Mar 16 17:15:35 2012
Thank You John once again for the prompt reply!
I will give your suggestions a try and keep you posted.
Really appreciate your help there.
Warm Regards,
Raizan
On 16 March 2012 23:08, John Halley Gotway via RT <met_help at ucar.edu>
wrote:
> Raizan,
>
> I was able to run with your data and reproduce the error message.
>
> One option is to use the "-obj_plot" command line argument for MODE.
That
> disables the generation of the output NetCDF file that's causing the
error.
>
> But the real problem is coming from your GRIB observation file. I
used
> "wgrib -V" on it and see the following output:
>
> [johnhg at rambler]% wgrib -V ofile_struct_regridded.grib
>
> Undefined parameter table (center 0-0 table 255), using NCEP-opn
> rec 1:0:date 0000000000 PRES kpds5=1 kpds6=1 kpds7=0 levels=(0,0)
grid=255
> sfc anl: bitmap: 481 undef
> PRES=Pressure [Pa]
> timerange 10 P1 0 P2 0 TimeU 2 nx 241 ny 241 GDS grid 0
num_in_ave 0
> missing 0
> center 0 subcenter 0 process 0 Table 255 scan: WE:SN winds(N/S)
> latlon: lat 50.000000 to 53.133000 by 0.013000 nxny 58081
> long -5.800000 to -0.739000 by 0.021000, (241 x 241) scan
64
> mode 128 bdsgrid 1
> min/max data 0 5908 num bits 13 BDS_Ref 0 DecScale 0 BinScale 0
>
> The "date 0000000000" listed as all 0's is problematic. The MET
code is
> parsing that info and trying to interpret that as a time. It's
trying to
> write that time as a variable attribute in that
> output NetCDF file, and that's the source of the error. If you can
encode
> the time information correctly in that GRIB file this error should
be
> resolved.
>
> As far as sanity checking the contents of your observation GRIB
file, you
> may find it useful to view it using the IDV tool.
>
> One last thing, in your config file, the "ZSFC" won't work - try
using
> "Z0" instead. Or try the following:
> fcst_field = "PRES/Z0";
> obs_field = "PRES/Z0";
>
> That compares surface pressure from the two files.
>
> Hope that helps.
>
> Thanks,
> John
>
>
> On 03/16/2012 04:48 PM, Zan Rahmat via RT wrote:
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55226>
> >
> > Dear John,
> >
> > Thank you for the prompt reply. I have uploaded the 3 files.
> >
> > I would just like to highlight that my obs variable has also not
been
> named
> > properly. When it was in netcdf format tha variable name was var1,
when I
> > converted it to grib using cdo, somehow it got named to PRES. I
hope this
> > is consequential to the output/outcome. What's in the obs file is
> > precipitation rate anyway.
> >
> > Thanks,
> > Raizan
> >
> > On 16 March 2012 22:25, John Halley Gotway via
RT<met_help at ucar.edu>
> wrote:
> >
> >> Raizan,
> >>
> >> Can you please send me the input data for MODE (forecast,
observation,
> and
> >> config file) so that I can run it here and reproduce your error?
> >>
> >> Just follow the instructions here to post that data to our
anonymous ftp
> >> site:
> >> http://www.dtcenter.org/met/users/support/met_help.php#ftp
> >>
> >> Thanks,
> >> John
> >>
> >> On 03/16/2012 04:13 PM, Zan Rahmat via RT wrote:
> >>>
> >>> Fri Mar 16 16:13:49 2012: Request 55226 was acted upon.
> >>> Transaction: Ticket created by raizan.rahmat at gmail.com
> >>> Queue: met_help
> >>> Subject: Error running MET mode tool
> >>> Owner: Nobody
> >>> Requestors: raizan.rahmat at gmail.com
> >>> Status: new
> >>> Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55226
> >
> >>>
> >>>
> >>> Hi, I have come as far as converting the observation data and
forecast
> >> data
> >>> into the right format for MET mode tool.
> >>>
> >>> There is still an error reported by MET during the generation of
the
> >> output
> >>> ".nc" file. Here's the message: *NetCDF: Numeric conversion not
> >>> representable.*
> >>>
> >>> The full text is below. Appreciate if you could assist with
this.
> >>>
> >>> PS: the warning about timing is expected because I did not
provide a
> >> proper
> >>> output time for my obs file.
> >>>
> >>> Thanks,
> >>> Raizan
> >>>
> >>>
> >>> ***WARNING***: process_fcst_obs_files() -> Forecast and
observation
> >> valid
> >>> times do not match 20110807_180000 != -00011130_000000.
> >>>
> >>> Forecast Field: PRATE at ZSFC
> >>> Observation Field: PRES at ZSFC
> >>> Identifying objects in the forecast and observation fields...
> >>> Computing contingency table statistics...
> >>> Identified: 0 forecast objects and 9 observation objects.
> >>> Performing merging (no merging) in the forecast field.
> >>> Performing merging (no merging) in the observation field.
> >>> Remaining: 0 forecast objects and 9 observation objects.
> >>> Performing matching between the forecast and observation fields.
> >>> Creating Contingency Table Statistics file:
> >>> ../out/mode/mode_360000L_20110807_180000V_000000A_cts.txt
> >>> Creating Fcst-Obs Object Statistics file:
> >>> ../out/mode/mode_360000L_20110807_180000V_000000A_obj.txt
> >>> Creating Object NetCDF file:
> >>> ../out/mode/mode_360000L_20110807_180000V_000000A_obj.nc
> >>> *NetCDF: Numeric conversion not representable*
> >>
> >>
>
>
------------------------------------------------
Subject: Error running MET mode tool
From: Zan Rahmat
Time: Fri Mar 16 18:51:16 2012
Dear John,
I have assigned the .nc file a proper date and it works fine now.
Thanks!! That was really helpful.
Warm Regards,
Razian
On 16 March 2012 23:08, John Halley Gotway via RT <met_help at ucar.edu>
wrote:
> Raizan,
>
> I was able to run with your data and reproduce the error message.
>
> One option is to use the "-obj_plot" command line argument for MODE.
That
> disables the generation of the output NetCDF file that's causing the
error.
>
> But the real problem is coming from your GRIB observation file. I
used
> "wgrib -V" on it and see the following output:
>
> [johnhg at rambler]% wgrib -V ofile_struct_regridded.grib
>
> Undefined parameter table (center 0-0 table 255), using NCEP-opn
> rec 1:0:date 0000000000 PRES kpds5=1 kpds6=1 kpds7=0 levels=(0,0)
grid=255
> sfc anl: bitmap: 481 undef
> PRES=Pressure [Pa]
> timerange 10 P1 0 P2 0 TimeU 2 nx 241 ny 241 GDS grid 0
num_in_ave 0
> missing 0
> center 0 subcenter 0 process 0 Table 255 scan: WE:SN winds(N/S)
> latlon: lat 50.000000 to 53.133000 by 0.013000 nxny 58081
> long -5.800000 to -0.739000 by 0.021000, (241 x 241) scan
64
> mode 128 bdsgrid 1
> min/max data 0 5908 num bits 13 BDS_Ref 0 DecScale 0 BinScale 0
>
> The "date 0000000000" listed as all 0's is problematic. The MET
code is
> parsing that info and trying to interpret that as a time. It's
trying to
> write that time as a variable attribute in that
> output NetCDF file, and that's the source of the error. If you can
encode
> the time information correctly in that GRIB file this error should
be
> resolved.
>
> As far as sanity checking the contents of your observation GRIB
file, you
> may find it useful to view it using the IDV tool.
>
> One last thing, in your config file, the "ZSFC" won't work - try
using
> "Z0" instead. Or try the following:
> fcst_field = "PRES/Z0";
> obs_field = "PRES/Z0";
>
> That compares surface pressure from the two files.
>
> Hope that helps.
>
> Thanks,
> John
>
>
> On 03/16/2012 04:48 PM, Zan Rahmat via RT wrote:
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55226>
> >
> > Dear John,
> >
> > Thank you for the prompt reply. I have uploaded the 3 files.
> >
> > I would just like to highlight that my obs variable has also not
been
> named
> > properly. When it was in netcdf format tha variable name was var1,
when I
> > converted it to grib using cdo, somehow it got named to PRES. I
hope this
> > is consequential to the output/outcome. What's in the obs file is
> > precipitation rate anyway.
> >
> > Thanks,
> > Raizan
> >
> > On 16 March 2012 22:25, John Halley Gotway via
RT<met_help at ucar.edu>
> wrote:
> >
> >> Raizan,
> >>
> >> Can you please send me the input data for MODE (forecast,
observation,
> and
> >> config file) so that I can run it here and reproduce your error?
> >>
> >> Just follow the instructions here to post that data to our
anonymous ftp
> >> site:
> >> http://www.dtcenter.org/met/users/support/met_help.php#ftp
> >>
> >> Thanks,
> >> John
> >>
> >> On 03/16/2012 04:13 PM, Zan Rahmat via RT wrote:
> >>>
> >>> Fri Mar 16 16:13:49 2012: Request 55226 was acted upon.
> >>> Transaction: Ticket created by raizan.rahmat at gmail.com
> >>> Queue: met_help
> >>> Subject: Error running MET mode tool
> >>> Owner: Nobody
> >>> Requestors: raizan.rahmat at gmail.com
> >>> Status: new
> >>> Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=55226
> >
> >>>
> >>>
> >>> Hi, I have come as far as converting the observation data and
forecast
> >> data
> >>> into the right format for MET mode tool.
> >>>
> >>> There is still an error reported by MET during the generation of
the
> >> output
> >>> ".nc" file. Here's the message: *NetCDF: Numeric conversion not
> >>> representable.*
> >>>
> >>> The full text is below. Appreciate if you could assist with
this.
> >>>
> >>> PS: the warning about timing is expected because I did not
provide a
> >> proper
> >>> output time for my obs file.
> >>>
> >>> Thanks,
> >>> Raizan
> >>>
> >>>
> >>> ***WARNING***: process_fcst_obs_files() -> Forecast and
observation
> >> valid
> >>> times do not match 20110807_180000 != -00011130_000000.
> >>>
> >>> Forecast Field: PRATE at ZSFC
> >>> Observation Field: PRES at ZSFC
> >>> Identifying objects in the forecast and observation fields...
> >>> Computing contingency table statistics...
> >>> Identified: 0 forecast objects and 9 observation objects.
> >>> Performing merging (no merging) in the forecast field.
> >>> Performing merging (no merging) in the observation field.
> >>> Remaining: 0 forecast objects and 9 observation objects.
> >>> Performing matching between the forecast and observation fields.
> >>> Creating Contingency Table Statistics file:
> >>> ../out/mode/mode_360000L_20110807_180000V_000000A_cts.txt
> >>> Creating Fcst-Obs Object Statistics file:
> >>> ../out/mode/mode_360000L_20110807_180000V_000000A_obj.txt
> >>> Creating Object NetCDF file:
> >>> ../out/mode/mode_360000L_20110807_180000V_000000A_obj.nc
> >>> *NetCDF: Numeric conversion not representable*
> >>
> >>
>
>
------------------------------------------------
More information about the Met_help
mailing list