[Met_help] [rt.rap.ucar.edu #62240] History for A question on 'ERROR : process_fcst_climo_files'
John Halley Gotway via RT
met_help at ucar.edu
Fri Jul 19 09:18:34 MDT 2013
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Dear methelp,
I tried to use METv4.0 to get GSS score of APCP. I first run UPP with my WRF output and get a series of 'wrfprs' files. Then I downloaded 'NCEP ADP Global Upper Air and Surface' data to use as observations. I set the config in MET as below:
fcst = {
message_type = [ "ADPUPA", "ADPSFC"];
field = [
{
name = "APCP";
level = [ "A3" ];
cat_thresh = [ >0.0 ];
}
];
};
obs = fcst;
But I got the following error:
ERROR : process_fcst_climo_files() -> no fields matching APCP/A3 found in file: /../wrfprs_d01.000
Would you mind giving me any help to solve this problem?
Thanks and best regards,
Yue
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #62240] A question on 'ERROR : process_fcst_climo_files'
From: John Halley Gotway
Time: Mon Jul 15 11:32:59 2013
Yue,
Based on your description, it sounds like you're trying to use the
Point-Stat tool in MET to verify the GRIB output of the Unified Post-
Processor to the GDAS PREPBUFR point observations. And you've
configured Point-Stat to verify 3-hourly precipitation. There are a
couple of problems to explain here...
First, you're trying to verify 3-hourly precipitation (name = "APCP";
level = "A3";). So Point-Stat is looking through the input GRIB file
(wrfprs_d01.000) for a record of 3-hourly precip. However,
this is the output for the analysis time, and there is not 3-hourly
precip in it. Try running the following command: wgrib wrfprs_d01.000
| grep APCP
You'll see that there's no 3-houlry precip there.
If you pass Point-Stat the 3-hour forecast file, it should find a
record of 3-hourly precip: wrfprs_d01.003
If you're using the default setting for WRF-ARW, then the GRIB output
of UPP will contain precipitation accumulated over the entire model
run. But if you want to evaluated precipitation in 3-houlry
intervals, you'll need to run the pcp_combine tool to subtract them.
For example...
# Compute the 3-hour accumulation between hours 9 and 12
pcp_combine -subtract wrfprs_d01.012 12 wrfprs_d01.009 9
wrfprs_d01.APCP_09_12.nc
The second problem is that you're trying to retrieve precip point
observations from a PREPBUFR file. The PREPBUFR file contains
conventional obs, not precip. So even if you set it up properly,
you'd
get 0 matched pairs since there's no observations of precip in there.
I'd suggest starting by verifying things like temperature and relative
humidity - for which there are point observations available in the
PREPBUFR files.
Hope that helps.
Thanks,
John Halley Gotway
met_help at ucar.edu
On 07/13/2013 10:15 PM, Yue Zheng via RT wrote:
>
> Sat Jul 13 22:15:43 2013: Request 62240 was acted upon.
> Transaction: Ticket created by yue.zheng at outlook.com
> Queue: met_help
> Subject: A question on 'ERROR : process_fcst_climo_files'
> Owner: Nobody
> Requestors: yue.zheng at outlook.com
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=62240 >
>
>
> Dear methelp,
> I tried to use METv4.0 to get GSS score of APCP. I first run UPP
with my WRF output and get a series of 'wrfprs' files. Then I
downloaded 'NCEP ADP Global Upper Air and Surface' data to use as
observations. I set the config in MET as below:
> fcst = {
> message_type = [ "ADPUPA", "ADPSFC"];
> field = [
> {
> name = "APCP";
> level = [ "A3" ];
> cat_thresh = [ >0.0 ];
> }
> ];
> };
> obs = fcst;
>
> But I got the following error:
> ERROR : process_fcst_climo_files() -> no fields matching APCP/A3
found in file: /../wrfprs_d01.000
>
> Would you mind giving me any help to solve this problem?
> Thanks and best regards,
> Yue
>
>
------------------------------------------------
Subject: A question on 'ERROR : process_fcst_climo_files'
From: Yue Zheng
Time: Mon Jul 15 12:20:12 2013
Dear John,
Thank you so much for such detailed explaination!
Now I understand that I should calculate the 3-hourly accumulated
precip from the forecast files. I've verified surface variables such
as temperature, humidity, and wind speed with the PREPBUFR files
through ./point_stat. I also looked through MET user Guide, I think I
should use ./grid_stat to verify precipitation for my study. I would
like to get the GSS score for modeled 3-hourly (or 6 hourly)
precipitation, may I ask do you have any suggestions on what kind of
observation I could use?
Lots of thanks,
Best regards,
Yue
> Subject: Re: [rt.rap.ucar.edu #62240] A question on 'ERROR :
process_fcst_climo_files'
> From: met_help at ucar.edu
> To: yue.zheng at outlook.com
> Date: Mon, 15 Jul 2013 11:33:00 -0600
>
> Yue,
>
> Based on your description, it sounds like you're trying to use the
Point-Stat tool in MET to verify the GRIB output of the Unified Post-
Processor to the GDAS PREPBUFR point observations. And you've
> configured Point-Stat to verify 3-hourly precipitation. There are a
couple of problems to explain here...
>
> First, you're trying to verify 3-hourly precipitation (name =
"APCP"; level = "A3";). So Point-Stat is looking through the input
GRIB file (wrfprs_d01.000) for a record of 3-hourly precip. However,
> this is the output for the analysis time, and there is not 3-hourly
precip in it. Try running the following command: wgrib wrfprs_d01.000
| grep APCP
> You'll see that there's no 3-houlry precip there.
>
> If you pass Point-Stat the 3-hour forecast file, it should find a
record of 3-hourly precip: wrfprs_d01.003
>
> If you're using the default setting for WRF-ARW, then the GRIB
output of UPP will contain precipitation accumulated over the entire
model run. But if you want to evaluated precipitation in 3-houlry
> intervals, you'll need to run the pcp_combine tool to subtract them.
For example...
>
> # Compute the 3-hour accumulation between hours 9 and 12
> pcp_combine -subtract wrfprs_d01.012 12 wrfprs_d01.009 9
wrfprs_d01.APCP_09_12.nc
>
> The second problem is that you're trying to retrieve precip point
observations from a PREPBUFR file. The PREPBUFR file contains
conventional obs, not precip. So even if you set it up properly, you'd
> get 0 matched pairs since there's no observations of precip in
there.
>
> I'd suggest starting by verifying things like temperature and
relative humidity - for which there are point observations available
in the PREPBUFR files.
>
> Hope that helps.
>
> Thanks,
> John Halley Gotway
> met_help at ucar.edu
>
> On 07/13/2013 10:15 PM, Yue Zheng via RT wrote:
> >
> > Sat Jul 13 22:15:43 2013: Request 62240 was acted upon.
> > Transaction: Ticket created by yue.zheng at outlook.com
> > Queue: met_help
> > Subject: A question on 'ERROR : process_fcst_climo_files'
> > Owner: Nobody
> > Requestors: yue.zheng at outlook.com
> > Status: new
> > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=62240 >
> >
> >
> > Dear methelp,
> > I tried to use METv4.0 to get GSS score of APCP. I first run UPP
with my WRF output and get a series of 'wrfprs' files. Then I
downloaded 'NCEP ADP Global Upper Air and Surface' data to use as
observations. I set the config in MET as below:
> > fcst = {
> > message_type = [ "ADPUPA", "ADPSFC"];
> > field = [
> > {
> > name = "APCP";
> > level = [ "A3" ];
> > cat_thresh = [ >0.0 ];
> > }
> > ];
> > };
> > obs = fcst;
> >
> > But I got the following error:
> > ERROR : process_fcst_climo_files() -> no fields matching APCP/A3
found in file: /../wrfprs_d01.000
> >
> > Would you mind giving me any help to solve this problem?
> > Thanks and best regards,
> > Yue
> >
> >
>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #62240] A question on 'ERROR : process_fcst_climo_files'
From: John Halley Gotway
Time: Mon Jul 15 12:47:05 2013
Yue,
It depends on the location of your modelling domain. If it happens to
be over the continental United States, you could use the Stage II or
Stage IV precipitation analysis. If not, your options are
more limited. Could you send me one of your GRIB output files from
UPP, and then I could advise you on your options. You can just post
that file to our anonymous ftp site, following these instructions:
http://www.dtcenter.org/met/users/support/met_help.php#ftp
Thanks,
John
On 07/15/2013 12:20 PM, Yue Zheng via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=62240 >
>
> Dear John,
>
> Thank you so much for such detailed explaination!
>
> Now I understand that I should calculate the 3-hourly accumulated
precip from the forecast files. I've verified surface variables such
as temperature, humidity, and wind speed with the PREPBUFR files
through ./point_stat. I also looked through MET user Guide, I think I
should use ./grid_stat to verify precipitation for my study. I would
like to get the GSS score for modeled 3-hourly (or 6 hourly)
precipitation, may I ask do you have any suggestions on what kind of
observation I could use?
>
> Lots of thanks,
>
> Best regards,
>
> Yue
>
>
>
>
>
>> Subject: Re: [rt.rap.ucar.edu #62240] A question on 'ERROR :
process_fcst_climo_files'
>> From: met_help at ucar.edu
>> To: yue.zheng at outlook.com
>> Date: Mon, 15 Jul 2013 11:33:00 -0600
>>
>> Yue,
>>
>> Based on your description, it sounds like you're trying to use the
Point-Stat tool in MET to verify the GRIB output of the Unified Post-
Processor to the GDAS PREPBUFR point observations. And you've
>> configured Point-Stat to verify 3-hourly precipitation. There are a
couple of problems to explain here...
>>
>> First, you're trying to verify 3-hourly precipitation (name =
"APCP"; level = "A3";). So Point-Stat is looking through the input
GRIB file (wrfprs_d01.000) for a record of 3-hourly precip. However,
>> this is the output for the analysis time, and there is not 3-hourly
precip in it. Try running the following command: wgrib wrfprs_d01.000
| grep APCP
>> You'll see that there's no 3-houlry precip there.
>>
>> If you pass Point-Stat the 3-hour forecast file, it should find a
record of 3-hourly precip: wrfprs_d01.003
>>
>> If you're using the default setting for WRF-ARW, then the GRIB
output of UPP will contain precipitation accumulated over the entire
model run. But if you want to evaluated precipitation in 3-houlry
>> intervals, you'll need to run the pcp_combine tool to subtract
them. For example...
>>
>> # Compute the 3-hour accumulation between hours 9 and 12
>> pcp_combine -subtract wrfprs_d01.012 12 wrfprs_d01.009 9
wrfprs_d01.APCP_09_12.nc
>>
>> The second problem is that you're trying to retrieve precip point
observations from a PREPBUFR file. The PREPBUFR file contains
conventional obs, not precip. So even if you set it up properly, you'd
>> get 0 matched pairs since there's no observations of precip in
there.
>>
>> I'd suggest starting by verifying things like temperature and
relative humidity - for which there are point observations available
in the PREPBUFR files.
>>
>> Hope that helps.
>>
>> Thanks,
>> John Halley Gotway
>> met_help at ucar.edu
>>
>> On 07/13/2013 10:15 PM, Yue Zheng via RT wrote:
>>>
>>> Sat Jul 13 22:15:43 2013: Request 62240 was acted upon.
>>> Transaction: Ticket created by yue.zheng at outlook.com
>>> Queue: met_help
>>> Subject: A question on 'ERROR : process_fcst_climo_files'
>>> Owner: Nobody
>>> Requestors: yue.zheng at outlook.com
>>> Status: new
>>> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=62240 >
>>>
>>>
>>> Dear methelp,
>>> I tried to use METv4.0 to get GSS score of APCP. I first run UPP
with my WRF output and get a series of 'wrfprs' files. Then I
downloaded 'NCEP ADP Global Upper Air and Surface' data to use as
observations. I set the config in MET as below:
>>> fcst = {
>>> message_type = [ "ADPUPA", "ADPSFC"];
>>> field = [
>>> {
>>> name = "APCP";
>>> level = [ "A3" ];
>>> cat_thresh = [ >0.0 ];
>>> }
>>> ];
>>> };
>>> obs = fcst;
>>>
>>> But I got the following error:
>>> ERROR : process_fcst_climo_files() -> no fields matching APCP/A3
found in file: /../wrfprs_d01.000
>>>
>>> Would you mind giving me any help to solve this problem?
>>> Thanks and best regards,
>>> Yue
>>>
>>>
>>
>
>
------------------------------------------------
Subject: A question on 'ERROR : process_fcst_climo_files'
From: Yue Zheng
Time: Thu Jul 18 14:17:38 2013
Dear John,
I'm so sorry for my late reply.
I followed the instructions and ftp my GRIB output files
(WRFPRS_d01.003) from UPP to a folder named with my last name "zheng".
I have four domains in my research, and they are over US, Europe,
Africa, and India. I have the Stage IV MPE raw data over US, but I
haven't had a chance to run it with METv4.
May I ask your advice on the precipitation observations over these
domains?
I really appreciate for your so nice help!
Thanks again,
Best regards,
Yue
> Subject: Re: [rt.rap.ucar.edu #62240] A question on 'ERROR :
process_fcst_climo_files'
> From: met_help at ucar.edu
> To: yue.zheng at outlook.com
> Date: Mon, 15 Jul 2013 12:47:06 -0600
>
> Yue,
>
> It depends on the location of your modelling domain. If it happens
to be over the continental United States, you could use the Stage II
or Stage IV precipitation analysis. If not, your options are
> more limited. Could you send me one of your GRIB output files from
UPP, and then I could advise you on your options. You can just post
that file to our anonymous ftp site, following these instructions:
> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>
> Thanks,
> John
>
> On 07/15/2013 12:20 PM, Yue Zheng via RT wrote:
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=62240 >
> >
> > Dear John,
> >
> > Thank you so much for such detailed explaination!
> >
> > Now I understand that I should calculate the 3-hourly accumulated
precip from the forecast files. I've verified surface variables such
as temperature, humidity, and wind speed with the PREPBUFR files
through ./point_stat. I also looked through MET user Guide, I think I
should use ./grid_stat to verify precipitation for my study. I would
like to get the GSS score for modeled 3-hourly (or 6 hourly)
precipitation, may I ask do you have any suggestions on what kind of
observation I could use?
> >
> > Lots of thanks,
> >
> > Best regards,
> >
> > Yue
> >
> >
> >
> >
> >
> >> Subject: Re: [rt.rap.ucar.edu #62240] A question on 'ERROR :
process_fcst_climo_files'
> >> From: met_help at ucar.edu
> >> To: yue.zheng at outlook.com
> >> Date: Mon, 15 Jul 2013 11:33:00 -0600
> >>
> >> Yue,
> >>
> >> Based on your description, it sounds like you're trying to use
the Point-Stat tool in MET to verify the GRIB output of the Unified
Post-Processor to the GDAS PREPBUFR point observations. And you've
> >> configured Point-Stat to verify 3-hourly precipitation. There are
a couple of problems to explain here...
> >>
> >> First, you're trying to verify 3-hourly precipitation (name =
"APCP"; level = "A3";). So Point-Stat is looking through the input
GRIB file (wrfprs_d01.000) for a record of 3-hourly precip. However,
> >> this is the output for the analysis time, and there is not 3-
hourly precip in it. Try running the following command: wgrib
wrfprs_d01.000 | grep APCP
> >> You'll see that there's no 3-houlry precip there.
> >>
> >> If you pass Point-Stat the 3-hour forecast file, it should find a
record of 3-hourly precip: wrfprs_d01.003
> >>
> >> If you're using the default setting for WRF-ARW, then the GRIB
output of UPP will contain precipitation accumulated over the entire
model run. But if you want to evaluated precipitation in 3-houlry
> >> intervals, you'll need to run the pcp_combine tool to subtract
them. For example...
> >>
> >> # Compute the 3-hour accumulation between hours 9 and 12
> >> pcp_combine -subtract wrfprs_d01.012 12 wrfprs_d01.009 9
wrfprs_d01.APCP_09_12.nc
> >>
> >> The second problem is that you're trying to retrieve precip point
observations from a PREPBUFR file. The PREPBUFR file contains
conventional obs, not precip. So even if you set it up properly, you'd
> >> get 0 matched pairs since there's no observations of precip in
there.
> >>
> >> I'd suggest starting by verifying things like temperature and
relative humidity - for which there are point observations available
in the PREPBUFR files.
> >>
> >> Hope that helps.
> >>
> >> Thanks,
> >> John Halley Gotway
> >> met_help at ucar.edu
> >>
> >> On 07/13/2013 10:15 PM, Yue Zheng via RT wrote:
> >>>
> >>> Sat Jul 13 22:15:43 2013: Request 62240 was acted upon.
> >>> Transaction: Ticket created by yue.zheng at outlook.com
> >>> Queue: met_help
> >>> Subject: A question on 'ERROR : process_fcst_climo_files'
> >>> Owner: Nobody
> >>> Requestors: yue.zheng at outlook.com
> >>> Status: new
> >>> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=62240 >
> >>>
> >>>
> >>> Dear methelp,
> >>> I tried to use METv4.0 to get GSS score of APCP. I first run UPP
with my WRF output and get a series of 'wrfprs' files. Then I
downloaded 'NCEP ADP Global Upper Air and Surface' data to use as
observations. I set the config in MET as below:
> >>> fcst = {
> >>> message_type = [ "ADPUPA", "ADPSFC"];
> >>> field = [
> >>> {
> >>> name = "APCP";
> >>> level = [ "A3" ];
> >>> cat_thresh = [ >0.0 ];
> >>> }
> >>> ];
> >>> };
> >>> obs = fcst;
> >>>
> >>> But I got the following error:
> >>> ERROR : process_fcst_climo_files() -> no fields matching APCP/A3
found in file: /../wrfprs_d01.000
> >>>
> >>> Would you mind giving me any help to solve this problem?
> >>> Thanks and best regards,
> >>> Yue
> >>>
> >>>
> >>
> >
> >
>
------------------------------------------------
Subject: A question on 'ERROR : process_fcst_climo_files'
From: John Halley Gotway
Time: Thu Jul 18 16:09:15 2013
Yue,
Wow, you're really covering the globe with those 4 domains. Yes, you
could use StageIV data over the US domain. However, if you need to
inter-compare the results, that won't help much with the other
3 domains.
The only other option I could point you to is TRMM data, as described
on the MET downloads page:
http://www.dtcenter.org/met/users/downloads/observation_data.php
Let me go through the steps of how you'd get data for the Africa
domain, and you can mimic it for the other domains:
# You could plot that domain in MET as follows (I've converted the
resulting plot to png format and attached it)
METv4.1/bin/plot_data_plane Africa/WRFPRS_d01.003
Africa/WRFPRS_d01.003_APCP_03.ps 'name="APCP"; level="A03";'
convert -rotate 90 Africa/WRFPRS_d01.003_APCP_03.ps
Africa/WRFPRS_d01.003_APCP_03.png
# Determine the approx min/max of lat/lon of that domain
wgrib -d 1 -V Africa/WRFPRS_d01.003
rec 1:0:date 2006081006 PRMSL kpds5=2 kpds6=102 kpds7=0 levels=(0,0)
grid=255 MSL 3hr fcst:
PRMSL=Pressure reduced to MSL [Pa]
timerange 0 P1 3 P2 0 TimeU 1 nx 299 ny 299 GDS grid 3 num_in_ave
0 missing 0
center 7 subcenter 0 process 125 Table 2 scan: WE:SN winds(grid)
Lambert Conf: Lat1 -17.394000 Lon1 -20.461000 Lov 2.256000
Latin1 60.000000 Latin2 30.000000 LatSP 0.000000 LonSP 0.000000
North Pole (299 x 299) Dx 27.000000 Dy 27.000000 scan 64 mode
136
min/max data 99694.4 102813 num bits 16 BDS_Ref 9.96944e+06
DecScale 2 BinScale 3
Let's use the lat/lon ranges from -20 to 45.
# Retrieve 1/4 degree TRMM data for this lat/lon range.
Go here:
http://disc2.nascom.nasa.gov/Giovanni/tovas/realtime.3B42RT.shtml
- Set latitude range 20 S to 45 N.
- Set longitude range 20 W to 45 E.
- Select Accumulated Rainfall.
- Set the dates to August 10, 2006 at 03Z.
- Click the "ASCII Output" button.
- Save the resulting ASCII data.
# Run the data through the trmm2nc.R script:
# http://www.dtcenter.org/met/users/downloads/Rscripts/trmm2nc.R
Rscript trmm2nc.R Africa_data.txt Africa_data.nc
# Determine the domain that the TRMM data is on
ncdump -h Africa_data.nc
...
:Projection = "LatLon" ;
:lat_ll = "-25 degrees_north" ;
:lon_ll = "-50 degrees_east" ;
:delta_lat = "0.25 degrees" ;
:delta_lon = "0.25 degrees" ;
:Nlat = "301 grid_points" ;
:Nlon = "421 grid_points" ;
...
# Use copygb to regrid your GRIB file to the same domain
# Info on this here:
#
http://www.dtcenter.org/met/users/support/online_tutorial/METv4.0/copygb/run2.php
copygb -xg"255 0 421 301 -25000 -50000 128 50000 55000 250 250 64"
Africa/WRFPRS_d01.003 Africa/WRFPRS_d01.003_regrid
# Plot the new GRIB file
METv4.1/bin/plot_data_plane Africa/WRFPRS_d01.003_regrid
Africa/WRFPRS_d01.003_APCP_03_regrid.ps 'name="APCP"; level="A03";'
# Run Grid-Stat
METv4.1/bin/grid_stat Africa/WRFPRS_d01.003_regrid Africa_data.nc
GridStatConfig_APCP_03 -outdir out -v 2
That config file is attached.
Hope that helps get you going.
Thanks,
John Halley Gotway
met_help at ucar.edu
On 07/18/2013 02:17 PM, Yue Zheng via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=62240 >
>
> Dear John,
> I'm so sorry for my late reply.
> I followed the instructions and ftp my GRIB output files
(WRFPRS_d01.003) from UPP to a folder named with my last name "zheng".
I have four domains in my research, and they are over US, Europe,
Africa, and India. I have the Stage IV MPE raw data over US, but I
haven't had a chance to run it with METv4.
> May I ask your advice on the precipitation observations over these
domains?
> I really appreciate for your so nice help!
> Thanks again,
> Best regards,
> Yue
>
>
>> Subject: Re: [rt.rap.ucar.edu #62240] A question on 'ERROR :
process_fcst_climo_files'
>> From: met_help at ucar.edu
>> To: yue.zheng at outlook.com
>> Date: Mon, 15 Jul 2013 12:47:06 -0600
>>
>> Yue,
>>
>> It depends on the location of your modelling domain. If it happens
to be over the continental United States, you could use the Stage II
or Stage IV precipitation analysis. If not, your options are
>> more limited. Could you send me one of your GRIB output files from
UPP, and then I could advise you on your options. You can just post
that file to our anonymous ftp site, following these instructions:
>> http://www.dtcenter.org/met/users/support/met_help.php#ftp
>>
>> Thanks,
>> John
>>
>> On 07/15/2013 12:20 PM, Yue Zheng via RT wrote:
>>>
>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=62240 >
>>>
>>> Dear John,
>>>
>>> Thank you so much for such detailed explaination!
>>>
>>> Now I understand that I should calculate the 3-hourly accumulated
precip from the forecast files. I've verified surface variables such
as temperature, humidity, and wind speed with the PREPBUFR files
through ./point_stat. I also looked through MET user Guide, I think I
should use ./grid_stat to verify precipitation for my study. I would
like to get the GSS score for modeled 3-hourly (or 6 hourly)
precipitation, may I ask do you have any suggestions on what kind of
observation I could use?
>>>
>>> Lots of thanks,
>>>
>>> Best regards,
>>>
>>> Yue
>>>
>>>
>>>
>>>
>>>
>>>> Subject: Re: [rt.rap.ucar.edu #62240] A question on 'ERROR :
process_fcst_climo_files'
>>>> From: met_help at ucar.edu
>>>> To: yue.zheng at outlook.com
>>>> Date: Mon, 15 Jul 2013 11:33:00 -0600
>>>>
>>>> Yue,
>>>>
>>>> Based on your description, it sounds like you're trying to use
the Point-Stat tool in MET to verify the GRIB output of the Unified
Post-Processor to the GDAS PREPBUFR point observations. And you've
>>>> configured Point-Stat to verify 3-hourly precipitation. There are
a couple of problems to explain here...
>>>>
>>>> First, you're trying to verify 3-hourly precipitation (name =
"APCP"; level = "A3";). So Point-Stat is looking through the input
GRIB file (wrfprs_d01.000) for a record of 3-hourly precip. However,
>>>> this is the output for the analysis time, and there is not 3-
hourly precip in it. Try running the following command: wgrib
wrfprs_d01.000 | grep APCP
>>>> You'll see that there's no 3-houlry precip there.
>>>>
>>>> If you pass Point-Stat the 3-hour forecast file, it should find a
record of 3-hourly precip: wrfprs_d01.003
>>>>
>>>> If you're using the default setting for WRF-ARW, then the GRIB
output of UPP will contain precipitation accumulated over the entire
model run. But if you want to evaluated precipitation in 3-houlry
>>>> intervals, you'll need to run the pcp_combine tool to subtract
them. For example...
>>>>
>>>> # Compute the 3-hour accumulation between hours 9 and 12
>>>> pcp_combine -subtract wrfprs_d01.012 12 wrfprs_d01.009 9
wrfprs_d01.APCP_09_12.nc
>>>>
>>>> The second problem is that you're trying to retrieve precip point
observations from a PREPBUFR file. The PREPBUFR file contains
conventional obs, not precip. So even if you set it up properly, you'd
>>>> get 0 matched pairs since there's no observations of precip in
there.
>>>>
>>>> I'd suggest starting by verifying things like temperature and
relative humidity - for which there are point observations available
in the PREPBUFR files.
>>>>
>>>> Hope that helps.
>>>>
>>>> Thanks,
>>>> John Halley Gotway
>>>> met_help at ucar.edu
>>>>
>>>> On 07/13/2013 10:15 PM, Yue Zheng via RT wrote:
>>>>>
>>>>> Sat Jul 13 22:15:43 2013: Request 62240 was acted upon.
>>>>> Transaction: Ticket created by yue.zheng at outlook.com
>>>>> Queue: met_help
>>>>> Subject: A question on 'ERROR : process_fcst_climo_files'
>>>>> Owner: Nobody
>>>>> Requestors: yue.zheng at outlook.com
>>>>> Status: new
>>>>> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=62240 >
>>>>>
>>>>>
>>>>> Dear methelp,
>>>>> I tried to use METv4.0 to get GSS score of APCP. I first run UPP
with my WRF output and get a series of 'wrfprs' files. Then I
downloaded 'NCEP ADP Global Upper Air and Surface' data to use as
observations. I set the config in MET as below:
>>>>> fcst = {
>>>>> message_type = [ "ADPUPA", "ADPSFC"];
>>>>> field = [
>>>>> {
>>>>> name = "APCP";
>>>>> level = [ "A3" ];
>>>>> cat_thresh = [ >0.0 ];
>>>>> }
>>>>> ];
>>>>> };
>>>>> obs = fcst;
>>>>>
>>>>> But I got the following error:
>>>>> ERROR : process_fcst_climo_files() -> no fields matching APCP/A3
found in file: /../wrfprs_d01.000
>>>>>
>>>>> Would you mind giving me any help to solve this problem?
>>>>> Thanks and best regards,
>>>>> Yue
>>>>>
>>>>>
>>>>
>>>
>>>
>>
>
>
------------------------------------------------
Subject: A question on 'ERROR : process_fcst_climo_files'
From: John Halley Gotway
Time: Thu Jul 18 16:09:15 2013
////////////////////////////////////////////////////////////////////////////////
//
// Grid-Stat configuration file.
//
// For additional information, see the MET_BASE/data/config/README
file.
//
////////////////////////////////////////////////////////////////////////////////
//
// Output model name to be written
//
model = "Africa_Domain";
////////////////////////////////////////////////////////////////////////////////
//
// Forecast and observation fields to be verified
//
fcst = {
wind_thresh = [ NA ];
field = [
{
name = "APCP";
level = [ "A03" ];
cat_thresh = [ >0.0, >=5.0 ];
}
];
};
obs = {
wind_thresh = [ NA ];
field = [
{
name = "APCP_03";
level = [ "(*,*)" ];
cat_thresh = [ >0.0, >=5.0 ];
}
];
};
////////////////////////////////////////////////////////////////////////////////
//
// Verification masking regions
//
mask = {
grid = [ "FULL" ];
poly = [];
};
////////////////////////////////////////////////////////////////////////////////
//
// Confidence interval settings
//
ci_alpha = [ 0.10, 0.05 ];
boot = {
interval = PCTILE;
rep_prop = 1.0;
n_rep = 0;
rng = "mt19937";
seed = "1";
};
////////////////////////////////////////////////////////////////////////////////
//
// Interpolation methods
//
interp = {
field = FCST;
vld_thresh = 1.0;
type = [
{
method = UW_MEAN;
width = 1;
}
];
};
////////////////////////////////////////////////////////////////////////////////
//
// Neighborhood methods
//
nbrhd = {
vld_thresh = 1.0;
width = [ 3, 5 ];
cov_thresh = [ >=0.5 ];
}
////////////////////////////////////////////////////////////////////////////////
//
// Statistical output types
//
output_flag = {
fho = BOTH;
ctc = BOTH;
cts = BOTH;
mctc = NONE;
mcts = NONE;
cnt = BOTH;
sl1l2 = BOTH;
vl1l2 = BOTH;
pct = NONE;
pstd = NONE;
pjc = NONE;
prc = NONE;
nbrctc = BOTH;
nbrcts = BOTH;
nbrcnt = BOTH;
};
//
// NetCDF matched pairs output file
//
nc_pairs_flag = TRUE;
////////////////////////////////////////////////////////////////////////////////
rank_corr_flag = FALSE;
tmp_dir = "/tmp";
output_prefix = "APCP_03";
version = "V4.1";
////////////////////////////////////////////////////////////////////////////////
------------------------------------------------
Subject: A question on 'ERROR : process_fcst_climo_files'
From: Yue Zheng
Time: Thu Jul 18 18:40:48 2013
Dear John,
Thank you so much for the help!
I will use StageIV data for the US, and TRMM data for the other three
domains. I believe these will provide very useful results for my
research, and I will follow the steps from your advice.
I really appreciate your help!
Many thanks,
Best regards,
Yue
> Subject: Re: [rt.rap.ucar.edu #62240] A question on 'ERROR :
process_fcst_climo_files'
> From: met_help at ucar.edu
> To: yue.zheng at outlook.com
> Date: Thu, 18 Jul 2013 16:09:16 -0600
>
> Yue,
>
> Wow, you're really covering the globe with those 4 domains. Yes, you
could use StageIV data over the US domain. However, if you need to
inter-compare the results, that won't help much with the other
> 3 domains.
>
> The only other option I could point you to is TRMM data, as
described on the MET downloads page:
> http://www.dtcenter.org/met/users/downloads/observation_data.php
>
> Let me go through the steps of how you'd get data for the Africa
domain, and you can mimic it for the other domains:
>
> # You could plot that domain in MET as follows (I've converted the
resulting plot to png format and attached it)
> METv4.1/bin/plot_data_plane Africa/WRFPRS_d01.003
Africa/WRFPRS_d01.003_APCP_03.ps 'name="APCP"; level="A03";'
> convert -rotate 90 Africa/WRFPRS_d01.003_APCP_03.ps
Africa/WRFPRS_d01.003_APCP_03.png
>
> # Determine the approx min/max of lat/lon of that domain
> wgrib -d 1 -V Africa/WRFPRS_d01.003
> rec 1:0:date 2006081006 PRMSL kpds5=2 kpds6=102 kpds7=0 levels=(0,0)
grid=255 MSL 3hr fcst:
> PRMSL=Pressure reduced to MSL [Pa]
> timerange 0 P1 3 P2 0 TimeU 1 nx 299 ny 299 GDS grid 3 num_in_ave 0
missing 0
> center 7 subcenter 0 process 125 Table 2 scan: WE:SN winds(grid)
> Lambert Conf: Lat1 -17.394000 Lon1 -20.461000 Lov 2.256000
> Latin1 60.000000 Latin2 30.000000 LatSP 0.000000 LonSP 0.000000
> North Pole (299 x 299) Dx 27.000000 Dy 27.000000 scan 64 mode 136
> min/max data 99694.4 102813 num bits 16 BDS_Ref 9.96944e+06 DecScale
2 BinScale 3
>
> Let's use the lat/lon ranges from -20 to 45.
>
> # Retrieve 1/4 degree TRMM data for this lat/lon range.
> Go here:
http://disc2.nascom.nasa.gov/Giovanni/tovas/realtime.3B42RT.shtml
> - Set latitude range 20 S to 45 N.
> - Set longitude range 20 W to 45 E.
> - Select Accumulated Rainfall.
> - Set the dates to August 10, 2006 at 03Z.
> - Click the "ASCII Output" button.
> - Save the resulting ASCII data.
>
> # Run the data through the trmm2nc.R script:
> # http://www.dtcenter.org/met/users/downloads/Rscripts/trmm2nc.R
> Rscript trmm2nc.R Africa_data.txt Africa_data.nc
>
> # Determine the domain that the TRMM data is on
> ncdump -h Africa_data.nc
> ...
> :Projection = "LatLon" ;
> :lat_ll = "-25 degrees_north" ;
> :lon_ll = "-50 degrees_east" ;
> :delta_lat = "0.25 degrees" ;
> :delta_lon = "0.25 degrees" ;
> :Nlat = "301 grid_points" ;
> :Nlon = "421 grid_points" ;
> ...
>
> # Use copygb to regrid your GRIB file to the same domain
> # Info on this here:
> #
http://www.dtcenter.org/met/users/support/online_tutorial/METv4.0/copygb/run2.php
> copygb -xg"255 0 421 301 -25000 -50000 128 50000 55000 250 250 64"
Africa/WRFPRS_d01.003 Africa/WRFPRS_d01.003_regrid
>
> # Plot the new GRIB file
> METv4.1/bin/plot_data_plane Africa/WRFPRS_d01.003_regrid
Africa/WRFPRS_d01.003_APCP_03_regrid.ps 'name="APCP"; level="A03";'
>
> # Run Grid-Stat
> METv4.1/bin/grid_stat Africa/WRFPRS_d01.003_regrid Africa_data.nc
GridStatConfig_APCP_03 -outdir out -v 2
>
> That config file is attached.
>
> Hope that helps get you going.
>
> Thanks,
> John Halley Gotway
> met_help at ucar.edu
>
>
> On 07/18/2013 02:17 PM, Yue Zheng via RT wrote:
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=62240 >
> >
> > Dear John,
> > I'm so sorry for my late reply.
> > I followed the instructions and ftp my GRIB output files
(WRFPRS_d01.003) from UPP to a folder named with my last name "zheng".
I have four domains in my research, and they are over US, Europe,
Africa, and India. I have the Stage IV MPE raw data over US, but I
haven't had a chance to run it with METv4.
> > May I ask your advice on the precipitation observations over these
domains?
> > I really appreciate for your so nice help!
> > Thanks again,
> > Best regards,
> > Yue
> >
> >
> >> Subject: Re: [rt.rap.ucar.edu #62240] A question on 'ERROR :
process_fcst_climo_files'
> >> From: met_help at ucar.edu
> >> To: yue.zheng at outlook.com
> >> Date: Mon, 15 Jul 2013 12:47:06 -0600
> >>
> >> Yue,
> >>
> >> It depends on the location of your modelling domain. If it
happens to be over the continental United States, you could use the
Stage II or Stage IV precipitation analysis. If not, your options are
> >> more limited. Could you send me one of your GRIB output files
from UPP, and then I could advise you on your options. You can just
post that file to our anonymous ftp site, following these
instructions:
> >> http://www.dtcenter.org/met/users/support/met_help.php#ftp
> >>
> >> Thanks,
> >> John
> >>
> >> On 07/15/2013 12:20 PM, Yue Zheng via RT wrote:
> >>>
> >>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=62240 >
> >>>
> >>> Dear John,
> >>>
> >>> Thank you so much for such detailed explaination!
> >>>
> >>> Now I understand that I should calculate the 3-hourly
accumulated precip from the forecast files. I've verified surface
variables such as temperature, humidity, and wind speed with the
PREPBUFR files through ./point_stat. I also looked through MET user
Guide, I think I should use ./grid_stat to verify precipitation for my
study. I would like to get the GSS score for modeled 3-hourly (or 6
hourly) precipitation, may I ask do you have any suggestions on what
kind of observation I could use?
> >>>
> >>> Lots of thanks,
> >>>
> >>> Best regards,
> >>>
> >>> Yue
> >>>
> >>>
> >>>
> >>>
> >>>
> >>>> Subject: Re: [rt.rap.ucar.edu #62240] A question on 'ERROR :
process_fcst_climo_files'
> >>>> From: met_help at ucar.edu
> >>>> To: yue.zheng at outlook.com
> >>>> Date: Mon, 15 Jul 2013 11:33:00 -0600
> >>>>
> >>>> Yue,
> >>>>
> >>>> Based on your description, it sounds like you're trying to use
the Point-Stat tool in MET to verify the GRIB output of the Unified
Post-Processor to the GDAS PREPBUFR point observations. And you've
> >>>> configured Point-Stat to verify 3-hourly precipitation. There
are a couple of problems to explain here...
> >>>>
> >>>> First, you're trying to verify 3-hourly precipitation (name =
"APCP"; level = "A3";). So Point-Stat is looking through the input
GRIB file (wrfprs_d01.000) for a record of 3-hourly precip. However,
> >>>> this is the output for the analysis time, and there is not 3-
hourly precip in it. Try running the following command: wgrib
wrfprs_d01.000 | grep APCP
> >>>> You'll see that there's no 3-houlry precip there.
> >>>>
> >>>> If you pass Point-Stat the 3-hour forecast file, it should find
a record of 3-hourly precip: wrfprs_d01.003
> >>>>
> >>>> If you're using the default setting for WRF-ARW, then the GRIB
output of UPP will contain precipitation accumulated over the entire
model run. But if you want to evaluated precipitation in 3-houlry
> >>>> intervals, you'll need to run the pcp_combine tool to subtract
them. For example...
> >>>>
> >>>> # Compute the 3-hour accumulation between hours 9 and 12
> >>>> pcp_combine -subtract wrfprs_d01.012 12 wrfprs_d01.009 9
wrfprs_d01.APCP_09_12.nc
> >>>>
> >>>> The second problem is that you're trying to retrieve precip
point observations from a PREPBUFR file. The PREPBUFR file contains
conventional obs, not precip. So even if you set it up properly, you'd
> >>>> get 0 matched pairs since there's no observations of precip in
there.
> >>>>
> >>>> I'd suggest starting by verifying things like temperature and
relative humidity - for which there are point observations available
in the PREPBUFR files.
> >>>>
> >>>> Hope that helps.
> >>>>
> >>>> Thanks,
> >>>> John Halley Gotway
> >>>> met_help at ucar.edu
> >>>>
> >>>> On 07/13/2013 10:15 PM, Yue Zheng via RT wrote:
> >>>>>
> >>>>> Sat Jul 13 22:15:43 2013: Request 62240 was acted upon.
> >>>>> Transaction: Ticket created by yue.zheng at outlook.com
> >>>>> Queue: met_help
> >>>>> Subject: A question on 'ERROR : process_fcst_climo_files'
> >>>>> Owner: Nobody
> >>>>> Requestors: yue.zheng at outlook.com
> >>>>> Status: new
> >>>>> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=62240 >
> >>>>>
> >>>>>
> >>>>> Dear methelp,
> >>>>> I tried to use METv4.0 to get GSS score of APCP. I first run
UPP with my WRF output and get a series of 'wrfprs' files. Then I
downloaded 'NCEP ADP Global Upper Air and Surface' data to use as
observations. I set the config in MET as below:
> >>>>> fcst = {
> >>>>> message_type = [ "ADPUPA", "ADPSFC"];
> >>>>> field = [
> >>>>> {
> >>>>> name = "APCP";
> >>>>> level = [ "A3" ];
> >>>>> cat_thresh = [ >0.0 ];
> >>>>> }
> >>>>> ];
> >>>>> };
> >>>>> obs = fcst;
> >>>>>
> >>>>> But I got the following error:
> >>>>> ERROR : process_fcst_climo_files() -> no fields matching
APCP/A3 found in file: /../wrfprs_d01.000
> >>>>>
> >>>>> Would you mind giving me any help to solve this problem?
> >>>>> Thanks and best regards,
> >>>>> Yue
> >>>>>
> >>>>>
> >>>>
> >>>
> >>>
> >>
> >
> >
>
------------------------------------------------
More information about the Met_help
mailing list