[Met_help] [rt.rap.ucar.edu #91976] History for Question about using Point_Stat with MADIS2NC and PB2NC data files
Howard Soh via RT
met_help at ucar.edu
Wed Sep 11 07:57:37 MDT 2019
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Hi,
I am using Met version 8.0 to compare my WRF runs over California to observations. I've been using PREPBUFR data and PB2NC to get my observations and then using point_stat to compare them to the UPP'ed WRF files, which has been working well.
Recently, I've been trying to use MADIS files from MESO to look at additional stations. I've run the MADIS2NC successfully for a select number of station ID's (using sid). So my next goal was to use both the PREPBUFR data (after running PB2NC) and the MADIS data (after running MADIS2NC) to compare my WRF output to both observation datasets.
I'm using the "-point_obs" option with point_stat to use multiple observation files for example:
point_stat wrf3DGRIB_d02_f003.grb pb2nc/PREPBUFR_GDAS_2019090800_f000_pb.nc -point_obs MESO_20190908_0300.nc configs/MET_configs/PointStatConfig -outdir . -v 2
My problem is that when I run my command, the MADIS data is not used, which I believe is due to my PointStatConfig file (point_stat config file) having the "obs_quality" setting. This setting, which I had as
obs_quality = [ "0", "1", "2", "9" ];
is necessary for my PB2NC'ed files to be used, but if any of them are used then my MADIS2NC'ed files don't get counted. But if I set the array as empty, then my MADIS2NC'ed files are used, but not my PB2NC'ed files.
So my question is does anyone know how I can incorporate both observations for comparison to my WRF run in the point_stat calculations? It seems as though I wish I could use two different Point_Stat config files, one for MADIS and one for PREPBUFR, and then merge them together to then both be used in the overall point_stat statistical calculations. Any insight would be extremely helpful. If needed I can supply example data files.
Thank you in advance for your help!
Best,
Jeremy
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: Question about using Point_Stat with MADIS2NC and PB2NC data files
From: Howard Soh
Time: Tue Sep 10 12:13:15 2019
In general, the empty list ("obs_quality = [];") is a solution. An
alternative is adding quality values from MESO data.
For example,
ncdump -v obs_qty_table MESO_20190908_0300.nc
obs_qty_table =
"V",
"S",
"C",
"Q",
"NA",
"X",
"B" ;
obs_quality = [ "0", "1", "2", "9" ];
==>
obs_quality = [ "0", "1", "2", "9", "V", "S", "C", "Q", "NA", "X",
"B" ];
Cheers,
Howard
FYI:
If you upload your configurations and input data files to our ftp
server, we can investigate why an empty list does not work.
How to upload files:
- Go to https://dtcenter.org/met/users/support/met_help.php
- Scroll down to "How To Send Us Data"
.
And let us know when the files are uploaded
On Mon Sep 09 16:35:36 2019, jdberman at albany.edu wrote:
> Hi,
>
> I am using Met version 8.0 to compare my WRF runs over California to
> observations. I've been using PREPBUFR data and PB2NC to get my
> observations and then using point_stat to compare them to the UPP'ed
> WRF files, which has been working well.
>
> Recently, I've been trying to use MADIS files from MESO to look at
> additional stations. I've run the MADIS2NC successfully for a select
> number of station ID's (using sid). So my next goal was to use both
> the PREPBUFR data (after running PB2NC) and the MADIS data (after
> running MADIS2NC) to compare my WRF output to both observation
> datasets.
>
> I'm using the "-point_obs" option with point_stat to use multiple
> observation files for example:
>
> point_stat wrf3DGRIB_d02_f003.grb
> pb2nc/PREPBUFR_GDAS_2019090800_f000_pb.nc -point_obs
> MESO_20190908_0300.nc configs/MET_configs/PointStatConfig -outdir .
-v
> 2
>
> My problem is that when I run my command, the MADIS data is not
used,
> which I believe is due to my PointStatConfig file (point_stat config
> file) having the "obs_quality" setting. This setting, which I had as
>
>
> obs_quality = [ "0", "1", "2", "9" ];
>
> is necessary for my PB2NC'ed files to be used, but if any of them
are
> used then my MADIS2NC'ed files don't get counted. But if I set the
> array as empty, then my MADIS2NC'ed files are used, but not my
> PB2NC'ed files.
>
> So my question is does anyone know how I can incorporate both
> observations for comparison to my WRF run in the point_stat
> calculations? It seems as though I wish I could use two different
> Point_Stat config files, one for MADIS and one for PREPBUFR, and
then
> merge them together to then both be used in the overall point_stat
> statistical calculations. Any insight would be extremely helpful. If
> needed I can supply example data files.
>
> Thank you in advance for your help!
>
> Best,
> Jeremy
------------------------------------------------
Subject: Question about using Point_Stat with MADIS2NC and PB2NC data files
From: Berman, Jeremy D
Time: Tue Sep 10 17:09:44 2019
Hi Howard,
Thanks! Including the quality values from MESO worked! Also, for some
strange reason, using the empty list "obs_quality = [];" wasn't
working. But then, I tried putting the MESO data first in the command,
like:
point_stat MESO_20190908_0300.nc -point_obs wrf3DGRIB_d02_f003.grb
pb2nc/PREPBUFR_GDAS_2019090800_f000_pb.nc
configs/MET_configs/PointStatConfig -outdir . -v 2
And then miraculously that worked, while if the MESO data was listed
second it did not use the MESO data. At least now it is working!
Thanks for your help!
Jeremy
________________________________
From: Howard Soh via RT <met_help at ucar.edu>
Sent: Tuesday, September 10, 2019 2:13:15 PM
To: Berman, Jeremy D <jdberman at albany.edu>
Subject: [rt.rap.ucar.edu #91976] Question about using Point_Stat with
MADIS2NC and PB2NC data files
In general, the empty list ("obs_quality = [];") is a solution. An
alternative is adding quality values from MESO data.
For example,
ncdump -v obs_qty_table MESO_20190908_0300.nc
obs_qty_table =
"V",
"S",
"C",
"Q",
"NA",
"X",
"B" ;
obs_quality = [ "0", "1", "2", "9" ];
==>
obs_quality = [ "0", "1", "2", "9", "V", "S", "C", "Q", "NA", "X",
"B" ];
Cheers,
Howard
FYI:
If you upload your configurations and input data files to our ftp
server, we can investigate why an empty list does not work.
How to upload files:
- Go to https://dtcenter.org/met/users/support/met_help.php
- Scroll down to "How To Send Us Data"
.
And let us know when the files are uploaded
On Mon Sep 09 16:35:36 2019, jdberman at albany.edu wrote:
> Hi,
>
> I am using Met version 8.0 to compare my WRF runs over California to
> observations. I've been using PREPBUFR data and PB2NC to get my
> observations and then using point_stat to compare them to the UPP'ed
> WRF files, which has been working well.
>
> Recently, I've been trying to use MADIS files from MESO to look at
> additional stations. I've run the MADIS2NC successfully for a select
> number of station ID's (using sid). So my next goal was to use both
> the PREPBUFR data (after running PB2NC) and the MADIS data (after
> running MADIS2NC) to compare my WRF output to both observation
> datasets.
>
> I'm using the "-point_obs" option with point_stat to use multiple
> observation files for example:
>
> point_stat wrf3DGRIB_d02_f003.grb
> pb2nc/PREPBUFR_GDAS_2019090800_f000_pb.nc -point_obs
> MESO_20190908_0300.nc configs/MET_configs/PointStatConfig -outdir .
-v
> 2
>
> My problem is that when I run my command, the MADIS data is not
used,
> which I believe is due to my PointStatConfig file (point_stat config
> file) having the "obs_quality" setting. This setting, which I had as
>
>
> obs_quality = [ "0", "1", "2", "9" ];
>
> is necessary for my PB2NC'ed files to be used, but if any of them
are
> used then my MADIS2NC'ed files don't get counted. But if I set the
> array as empty, then my MADIS2NC'ed files are used, but not my
> PB2NC'ed files.
>
> So my question is does anyone know how I can incorporate both
> observations for comparison to my WRF run in the point_stat
> calculations? It seems as though I wish I could use two different
> Point_Stat config files, one for MADIS and one for PREPBUFR, and
then
> merge them together to then both be used in the overall point_stat
> statistical calculations. Any insight would be extremely helpful. If
> needed I can supply example data files.
>
> Thank you in advance for your help!
>
> Best,
> Jeremy
------------------------------------------------
Subject: Question about using Point_Stat with MADIS2NC and PB2NC data files
From: Howard Soh
Time: Tue Sep 10 18:11:02 2019
I'm glad it works for you. I'm going to close this ticket.
If you increase the verbose level (-v 3), it will display how many
data were rejected and why.
Cheers,
Howard
On Tue Sep 10 17:09:44 2019, jdberman at albany.edu wrote:
> Hi Howard,
>
>
> Thanks! Including the quality values from MESO worked! Also, for
some
> strange reason, using the empty list "obs_quality = [];" wasn't
> working. But then, I tried putting the MESO data first in the
command,
> like:
>
>
> point_stat MESO_20190908_0300.nc -point_obs wrf3DGRIB_d02_f003.grb
> pb2nc/PREPBUFR_GDAS_2019090800_f000_pb.nc
> configs/MET_configs/PointStatConfig -outdir . -v 2
>
>
> And then miraculously that worked, while if the MESO data was listed
> second it did not use the MESO data. At least now it is working!
>
>
> Thanks for your help!
>
>
> Jeremy
>
> ________________________________
> From: Howard Soh via RT <met_help at ucar.edu>
> Sent: Tuesday, September 10, 2019 2:13:15 PM
> To: Berman, Jeremy D <jdberman at albany.edu>
> Subject: [rt.rap.ucar.edu #91976] Question about using Point_Stat
with
> MADIS2NC and PB2NC data files
>
> In general, the empty list ("obs_quality = [];") is a solution.
An
> alternative is adding quality values from MESO data.
>
> For example,
>
> ncdump -v obs_qty_table MESO_20190908_0300.nc
>
> obs_qty_table =
> "V",
> "S",
> "C",
> "Q",
> "NA",
> "X",
> "B" ;
>
> obs_quality = [ "0", "1", "2", "9" ];
> ==>
> obs_quality = [ "0", "1", "2", "9", "V", "S", "C", "Q", "NA",
"X",
> "B" ];
>
>
> Cheers,
> Howard
>
> FYI:
> If you upload your configurations and input data files to our ftp
> server, we can investigate why an empty list does not work.
>
> How to upload files:
> - Go to https://dtcenter.org/met/users/support/met_help.php
> - Scroll down to "How To Send Us Data"
> .
> And let us know when the files are uploaded
>
> On Mon Sep 09 16:35:36 2019, jdberman at albany.edu wrote:
> > Hi,
> >
> > I am using Met version 8.0 to compare my WRF runs over California
to
> > observations. I've been using PREPBUFR data and PB2NC to get my
> > observations and then using point_stat to compare them to the
UPP'ed
> > WRF files, which has been working well.
> >
> > Recently, I've been trying to use MADIS files from MESO to look at
> > additional stations. I've run the MADIS2NC successfully for a
select
> > number of station ID's (using sid). So my next goal was to use
both
> > the PREPBUFR data (after running PB2NC) and the MADIS data (after
> > running MADIS2NC) to compare my WRF output to both observation
> > datasets.
> >
> > I'm using the "-point_obs" option with point_stat to use multiple
> > observation files for example:
> >
> > point_stat wrf3DGRIB_d02_f003.grb
> > pb2nc/PREPBUFR_GDAS_2019090800_f000_pb.nc -point_obs
> > MESO_20190908_0300.nc configs/MET_configs/PointStatConfig -outdir
.
> > -v
> > 2
> >
> > My problem is that when I run my command, the MADIS data is not
used,
> > which I believe is due to my PointStatConfig file (point_stat
config
> > file) having the "obs_quality" setting. This setting, which I had
as
> >
> >
> > obs_quality = [ "0", "1", "2", "9" ];
> >
> > is necessary for my PB2NC'ed files to be used, but if any of them
are
> > used then my MADIS2NC'ed files don't get counted. But if I set the
> > array as empty, then my MADIS2NC'ed files are used, but not my
> > PB2NC'ed files.
> >
> > So my question is does anyone know how I can incorporate both
> > observations for comparison to my WRF run in the point_stat
> > calculations? It seems as though I wish I could use two different
> > Point_Stat config files, one for MADIS and one for PREPBUFR, and
then
> > merge them together to then both be used in the overall point_stat
> > statistical calculations. Any insight would be extremely helpful.
If
> > needed I can supply example data files.
> >
> > Thank you in advance for your help!
> >
> > Best,
> > Jeremy
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #91976] Question about using Point_Stat with MADIS2NC and PB2NC data files
From: Berman, Jeremy D
Time: Tue Sep 10 18:17:22 2019
Thanks for the tip, and thanks again! :)
Jeremy
> On Sep 10, 2019, at 8:11 PM, Howard Soh via RT <met_help at ucar.edu>
wrote:
>
> I'm glad it works for you. I'm going to close this ticket.
>
> If you increase the verbose level (-v 3), it will display how many
data were rejected and why.
>
> Cheers,
> Howard
>
>> On Tue Sep 10 17:09:44 2019, jdberman at albany.edu wrote:
>> Hi Howard,
>>
>>
>> Thanks! Including the quality values from MESO worked! Also, for
some
>> strange reason, using the empty list "obs_quality = [];" wasn't
>> working. But then, I tried putting the MESO data first in the
command,
>> like:
>>
>>
>> point_stat MESO_20190908_0300.nc -point_obs wrf3DGRIB_d02_f003.grb
>> pb2nc/PREPBUFR_GDAS_2019090800_f000_pb.nc
>> configs/MET_configs/PointStatConfig -outdir . -v 2
>>
>>
>> And then miraculously that worked, while if the MESO data was
listed
>> second it did not use the MESO data. At least now it is working!
>>
>>
>> Thanks for your help!
>>
>>
>> Jeremy
>>
>> ________________________________
>> From: Howard Soh via RT <met_help at ucar.edu>
>> Sent: Tuesday, September 10, 2019 2:13:15 PM
>> To: Berman, Jeremy D <jdberman at albany.edu>
>> Subject: [rt.rap.ucar.edu #91976] Question about using Point_Stat
with
>> MADIS2NC and PB2NC data files
>>
>> In general, the empty list ("obs_quality = [];") is a solution.
An
>> alternative is adding quality values from MESO data.
>>
>> For example,
>>
>> ncdump -v obs_qty_table MESO_20190908_0300.nc
>>
>> obs_qty_table =
>> "V",
>> "S",
>> "C",
>> "Q",
>> "NA",
>> "X",
>> "B" ;
>>
>> obs_quality = [ "0", "1", "2", "9" ];
>> ==>
>> obs_quality = [ "0", "1", "2", "9", "V", "S", "C", "Q", "NA",
"X",
>> "B" ];
>>
>>
>> Cheers,
>> Howard
>>
>> FYI:
>> If you upload your configurations and input data files to our ftp
>> server, we can investigate why an empty list does not work.
>>
>> How to upload files:
>> - Go to https://dtcenter.org/met/users/support/met_help.php
>> - Scroll down to "How To Send Us Data"
>> .
>> And let us know when the files are uploaded
>>
>>> On Mon Sep 09 16:35:36 2019, jdberman at albany.edu wrote:
>>> Hi,
>>>
>>> I am using Met version 8.0 to compare my WRF runs over California
to
>>> observations. I've been using PREPBUFR data and PB2NC to get my
>>> observations and then using point_stat to compare them to the
UPP'ed
>>> WRF files, which has been working well.
>>>
>>> Recently, I've been trying to use MADIS files from MESO to look at
>>> additional stations. I've run the MADIS2NC successfully for a
select
>>> number of station ID's (using sid). So my next goal was to use
both
>>> the PREPBUFR data (after running PB2NC) and the MADIS data (after
>>> running MADIS2NC) to compare my WRF output to both observation
>>> datasets.
>>>
>>> I'm using the "-point_obs" option with point_stat to use multiple
>>> observation files for example:
>>>
>>> point_stat wrf3DGRIB_d02_f003.grb
>>> pb2nc/PREPBUFR_GDAS_2019090800_f000_pb.nc -point_obs
>>> MESO_20190908_0300.nc configs/MET_configs/PointStatConfig -outdir
.
>>> -v
>>> 2
>>>
>>> My problem is that when I run my command, the MADIS data is not
used,
>>> which I believe is due to my PointStatConfig file (point_stat
config
>>> file) having the "obs_quality" setting. This setting, which I had
as
>>>
>>>
>>> obs_quality = [ "0", "1", "2", "9" ];
>>>
>>> is necessary for my PB2NC'ed files to be used, but if any of them
are
>>> used then my MADIS2NC'ed files don't get counted. But if I set the
>>> array as empty, then my MADIS2NC'ed files are used, but not my
>>> PB2NC'ed files.
>>>
>>> So my question is does anyone know how I can incorporate both
>>> observations for comparison to my WRF run in the point_stat
>>> calculations? It seems as though I wish I could use two different
>>> Point_Stat config files, one for MADIS and one for PREPBUFR, and
then
>>> merge them together to then both be used in the overall point_stat
>>> statistical calculations. Any insight would be extremely helpful.
If
>>> needed I can supply example data files.
>>>
>>> Thank you in advance for your help!
>>>
>>> Best,
>>> Jeremy
>
>
>
------------------------------------------------
More information about the Met_help
mailing list