[Met_help] [rt.rap.ucar.edu #58413] History for Observations for Point-Stat (UNCLASSIFIED)
John Halley Gotway via RT
met_help at ucar.edu
Tue Oct 9 09:53:14 MDT 2012
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Classification: UNCLASSIFIED
Caveats: NONE
I am learning of a source of point observations for upper air and surface
which is an ascii file called OBS DOMAIN. We are trying to solve a problem
with erroneous data in the MADIS mesonet data and learned, through another
ARL colleague, about the OBS DOMAIN file which is used in WRF if observation
nudging is used. See the following link for a references and a sample of the
file.
http://www.mmm.ucar.edu/wrf/users/wrfv2/How_to_run_obs_fdda.html
The OBS DOMAIN file is subjected to extensive QC in order to obtain the best
set of observations for assimilation into the WRF. As such, I would like to
use the observations contained in it in Point-Stat.
The format does not appear to be MET specific nor netcdf.
Have you ever worked this file or data from this file as the source data to
provide input to Point-Stat?
Thanks.
R/
John
Mr John W. Raby, Meteorologist
U.S. Army Research Laboratory
White Sands Missile Range, NM 88002
(575) 678-2004 DSN 258-2004
FAX (575) 678-1230 DSN 258-1230
Email: john.w.raby2.civ at mail.mil
Classification: UNCLASSIFIED
Caveats: NONE
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #58413] Observations for Point-Stat (UNCLASSIFIED)
From: Paul Oldenburg
Time: Mon Sep 24 14:43:25 2012
John,
Is the OBS DOMAIN data that you mention the output of the conversion
script RT_fdda_reformat_obsnud.pl mentioned on the
web site you specified? If so, the input files to that script should
be in a format called LITTLE_R. We have developed
support in the MET tool ascii2nc for LITTLE_R, which we plan to
release with METv4.1 in January. If you have the
LITTLE_R formatted files, we could perhaps make arrangements to send
you the new version of ascii2nc to convert those
files, or do the conversion on your behalf if you send us the files.
Please let us know if you have the LITTLE_R files,
and please send us one or more if you do. We can take it from there.
Paul
On 09/21/2012 02:26 PM, Raby, John W USA CIV via RT wrote:
>
> Fri Sep 21 14:26:08 2012: Request 58413 was acted upon.
> Transaction: Ticket created by john.w.raby2.civ at mail.mil
> Queue: met_help
> Subject: Observations for Point-Stat (UNCLASSIFIED)
> Owner: Nobody
> Requestors: john.w.raby2.civ at mail.mil
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> I am learning of a source of point observations for upper air and
surface
> which is an ascii file called OBS DOMAIN. We are trying to solve a
problem
> with erroneous data in the MADIS mesonet data and learned, through
another
> ARL colleague, about the OBS DOMAIN file which is used in WRF if
observation
> nudging is used. See the following link for a references and a
sample of the
> file.
>
> http://www.mmm.ucar.edu/wrf/users/wrfv2/How_to_run_obs_fdda.html
>
> The OBS DOMAIN file is subjected to extensive QC in order to obtain
the best
> set of observations for assimilation into the WRF. As such, I would
like to
> use the observations contained in it in Point-Stat.
>
> The format does not appear to be MET specific nor netcdf.
>
> Have you ever worked this file or data from this file as the source
data to
> provide input to Point-Stat?
>
> Thanks.
>
> R/
> John
>
> Mr John W. Raby, Meteorologist
> U.S. Army Research Laboratory
> White Sands Missile Range, NM 88002
> (575) 678-2004 DSN 258-2004
> FAX (575) 678-1230 DSN 258-1230
> Email: john.w.raby2.civ at mail.mil
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
------------------------------------------------
Subject: Observations for Point-Stat (UNCLASSIFIED)
From: Raby, John W USA CIV
Time: Mon Sep 24 15:45:36 2012
Classification: UNCLASSIFIED
Caveats: NONE
Paul -
Thanks for sharing that info and your offer to help. I will go offline
to
share this with my colleagues who indeed do work with the LITTLE_R
format and
confirm the connection between the OBS DOMAIN file and the script you
describe. I will let you know soon.
Thanks.
R/
John
-----Original Message-----
From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
Sent: Monday, September 24, 2012 2:43 PM
To: Raby, John W CIV (US)
Subject: Re: [rt.rap.ucar.edu #58413] Observations for Point-Stat
(UNCLASSIFIED)
John,
Is the OBS DOMAIN data that you mention the output of the conversion
script
RT_fdda_reformat_obsnud.pl mentioned on the web site you specified?
If so,
the input files to that script should be in a format called LITTLE_R.
We have
developed support in the MET tool ascii2nc for LITTLE_R, which we plan
to
release with METv4.1 in January. If you have the LITTLE_R formatted
files, we
could perhaps make arrangements to send you the new version of
ascii2nc to
convert those files, or do the conversion on your behalf if you send
us the
files. Please let us know if you have the LITTLE_R files, and please
send us
one or more if you do. We can take it from there.
Paul
On 09/21/2012 02:26 PM, Raby, John W USA CIV via RT wrote:
>
> Fri Sep 21 14:26:08 2012: Request 58413 was acted upon.
> Transaction: Ticket created by john.w.raby2.civ at mail.mil
> Queue: met_help
> Subject: Observations for Point-Stat (UNCLASSIFIED)
> Owner: Nobody
> Requestors: john.w.raby2.civ at mail.mil
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> I am learning of a source of point observations for upper air and
surface
> which is an ascii file called OBS DOMAIN. We are trying to solve a
problem
> with erroneous data in the MADIS mesonet data and learned, through
another
> ARL colleague, about the OBS DOMAIN file which is used in WRF if
observation
> nudging is used. See the following link for a references and a
sample of the
> file.
>
> http://www.mmm.ucar.edu/wrf/users/wrfv2/How_to_run_obs_fdda.html
>
> The OBS DOMAIN file is subjected to extensive QC in order to obtain
the best
> set of observations for assimilation into the WRF. As such, I would
like to
> use the observations contained in it in Point-Stat.
>
> The format does not appear to be MET specific nor netcdf.
>
> Have you ever worked this file or data from this file as the source
data to
> provide input to Point-Stat?
>
> Thanks.
>
> R/
> John
>
> Mr John W. Raby, Meteorologist
> U.S. Army Research Laboratory
> White Sands Missile Range, NM 88002
> (575) 678-2004 DSN 258-2004
> FAX (575) 678-1230 DSN 258-1230
> Email: john.w.raby2.civ at mail.mil
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
Classification: UNCLASSIFIED
Caveats: NONE
------------------------------------------------
Subject: Observations for Point-Stat (UNCLASSIFIED)
From: brian.p.reen.ctr at mail.mil
Time: Tue Sep 25 08:22:49 2012
Classification: UNCLASSIFIED
Caveats: NONE
Paul,
John asked me to respond, since I'm familiar with the obs nudging file
formats. We are interested in obtaining the new version of ASCII2NC
since it
sounds like that could help us get our observation files in the format
MET
expects.
The OBS_DOMAIN files are either output from the script you mentioned,
or from
OBSGRID. As you noted, if we use RT_fdda_reformat_obsnud.pl, then we
could
simply use the LITTLE_R files that are fed to
RT_fdda_reformat_obsnud.pl.
However, OBSGRID provides QC of the data, so we would want to be able
to use
the QC'd data output from OBSGRID. In addition to outputting
OBS_DOMAIN
files, OBSGRID does produce LITTLE_R files so hopefully we can use
them in the
new version of ASCII2NC.
However, there are a few details we'll need to understand to determine
the
applicability of the new version of ASCII2NC to the LITTLE_R files
output from
OBSGRID. The U/V winds in the OBSGRID LITTLE_R file are rotated to
the model
grid (i.e., model-relative). Does ASCII2NC expect winds in the input
LITTLE_R
files to be model-relative or earth-relative? It would be helpful to
know if
ASCII2NC reads the U/V wind components or the wind speed / wind
direction
fields from LITTLE_R files, since the wind directions in the OBSGRID
LITTLE_R
files are incorrect. I should be able to fix OBSGRID so it produces
the
correct wind direction if needed. Finally, the observations in the
OBSGRID
LITTLE_R files are in hourly files, but within these files the
observations
are generally but not entirely in order of increasing time. Do I need
to
write code time-order the LITTLE_R files for ingestion by ASCII2NC?
The new version of ASCII2NC sounds like it could be very helpful. The
biggest
potential hindrance seems to be the model-relative / grid-relative
issue. The
other two details we should be able to work out once we know what
ASCII2NC is
expecting.
Brian
-----Original Message-----
From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
Sent: Monday, September 24, 2012 2:43 PM
To: Raby, John W CIV (US)
Subject: Re: [rt.rap.ucar.edu #58413] Observations for Point-Stat
(UNCLASSIFIED)
John,
Is the OBS DOMAIN data that you mention the output of the conversion
script
RT_fdda_reformat_obsnud.pl mentioned on the web site you specified?
If so,
the input files to that script should be in a format called LITTLE_R.
We have
developed support in the MET tool ascii2nc for LITTLE_R, which we plan
to
release with METv4.1 in January. If you have the LITTLE_R formatted
files, we
could perhaps make arrangements to send you the new version of
ascii2nc to
convert those files, or do the conversion on your behalf if you send
us the
files. Please let us know if you have the LITTLE_R files, and please
send us
one or more if you do. We can take it from there.
Paul
On 09/21/2012 02:26 PM, Raby, John W USA CIV via RT wrote:
>
> Fri Sep 21 14:26:08 2012: Request 58413 was acted upon.
> Transaction: Ticket created by john.w.raby2.civ at mail.mil
> Queue: met_help
> Subject: Observations for Point-Stat (UNCLASSIFIED)
> Owner: Nobody
> Requestors: john.w.raby2.civ at mail.mil
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> I am learning of a source of point observations for upper air and
surface
> which is an ascii file called OBS DOMAIN. We are trying to solve a
problem
> with erroneous data in the MADIS mesonet data and learned, through
another
> ARL colleague, about the OBS DOMAIN file which is used in WRF if
observation
> nudging is used. See the following link for a references and a
sample of the
> file.
>
> http://www.mmm.ucar.edu/wrf/users/wrfv2/How_to_run_obs_fdda.html
>
> The OBS DOMAIN file is subjected to extensive QC in order to obtain
the best
> set of observations for assimilation into the WRF. As such, I would
like to
> use the observations contained in it in Point-Stat.
>
> The format does not appear to be MET specific nor netcdf.
>
> Have you ever worked this file or data from this file as the source
data to
> provide input to Point-Stat?
>
> Thanks.
>
> R/
> John
>
> Mr John W. Raby, Meteorologist
> U.S. Army Research Laboratory
> White Sands Missile Range, NM 88002
> (575) 678-2004 DSN 258-2004
> FAX (575) 678-1230 DSN 258-1230
> Email: john.w.raby2.civ at mail.mil
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
Classification: UNCLASSIFIED
Caveats: NONE
Classification: UNCLASSIFIED
Caveats: NONE
Classification: UNCLASSIFIED
Caveats: NONE
Classification: UNCLASSIFIED
Caveats: NONE
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #58413] Observations for Point-Stat (UNCLASSIFIED)
From: John Halley Gotway
Time: Tue Sep 25 08:53:05 2012
Hello Brian,
This is John Halley Gotway. I'm another MET developer who worked on
recent enhancements to the ASCII2NC tool. I made these updates to
ASCII2NC in just the last 2 weeks in support of an upcoming
testing and evaluation project for the RAL RTFDDA system in which
we're involved. And the only little_r files I had available for
testing are from this project. While I'd be happy to send you guys
an alpha version of the updated ASCII2NC, I'd do so with the huge
caveat that it is not well tested. It is completely up to you make
sure you're getting out what you expect.
Regarding the timing of observations, that's fine. They don't need to
be in any particular order. The MET point_stat will ultimately be
reading the output of ASCII2NC and it doesn't expect the
observations to be in order by time.
However, the rotated winds are definitely an issue! MET assumes that
all point observations contain winds that are earth-relative. In
fact, when we find winds in model output that are grid-relative,
we automatically rotate them to being earth-relative so that they can
be compared with the earth-relative observations.
Just let me know how you'd like to proceed.
Thanks,
John
On 09/25/2012 08:22 AM, brian.p.reen.ctr at mail.mil via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> Paul,
> John asked me to respond, since I'm familiar with the obs nudging
file
> formats. We are interested in obtaining the new version of ASCII2NC
since it
> sounds like that could help us get our observation files in the
format MET
> expects.
>
> The OBS_DOMAIN files are either output from the script you
mentioned, or from
> OBSGRID. As you noted, if we use RT_fdda_reformat_obsnud.pl, then
we could
> simply use the LITTLE_R files that are fed to
RT_fdda_reformat_obsnud.pl.
> However, OBSGRID provides QC of the data, so we would want to be
able to use
> the QC'd data output from OBSGRID. In addition to outputting
OBS_DOMAIN
> files, OBSGRID does produce LITTLE_R files so hopefully we can use
them in the
> new version of ASCII2NC.
>
> However, there are a few details we'll need to understand to
determine the
> applicability of the new version of ASCII2NC to the LITTLE_R files
output from
> OBSGRID. The U/V winds in the OBSGRID LITTLE_R file are rotated to
the model
> grid (i.e., model-relative). Does ASCII2NC expect winds in the
input LITTLE_R
> files to be model-relative or earth-relative? It would be helpful
to know if
> ASCII2NC reads the U/V wind components or the wind speed / wind
direction
> fields from LITTLE_R files, since the wind directions in the OBSGRID
LITTLE_R
> files are incorrect. I should be able to fix OBSGRID so it produces
the
> correct wind direction if needed. Finally, the observations in the
OBSGRID
> LITTLE_R files are in hourly files, but within these files the
observations
> are generally but not entirely in order of increasing time. Do I
need to
> write code time-order the LITTLE_R files for ingestion by ASCII2NC?
>
> The new version of ASCII2NC sounds like it could be very helpful.
The biggest
> potential hindrance seems to be the model-relative / grid-relative
issue. The
> other two details we should be able to work out once we know what
ASCII2NC is
> expecting.
>
> Brian
>
> -----Original Message-----
> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
> Sent: Monday, September 24, 2012 2:43 PM
> To: Raby, John W CIV (US)
> Subject: Re: [rt.rap.ucar.edu #58413] Observations for Point-Stat
> (UNCLASSIFIED)
>
> John,
>
> Is the OBS DOMAIN data that you mention the output of the conversion
script
> RT_fdda_reformat_obsnud.pl mentioned on the web site you specified?
If so,
> the input files to that script should be in a format called
LITTLE_R. We have
> developed support in the MET tool ascii2nc for LITTLE_R, which we
plan to
> release with METv4.1 in January. If you have the LITTLE_R formatted
files, we
> could perhaps make arrangements to send you the new version of
ascii2nc to
> convert those files, or do the conversion on your behalf if you send
us the
> files. Please let us know if you have the LITTLE_R files, and
please send us
> one or more if you do. We can take it from there.
>
> Paul
>
>
> On 09/21/2012 02:26 PM, Raby, John W USA CIV via RT wrote:
>>
>> Fri Sep 21 14:26:08 2012: Request 58413 was acted upon.
>> Transaction: Ticket created by john.w.raby2.civ at mail.mil
>> Queue: met_help
>> Subject: Observations for Point-Stat (UNCLASSIFIED)
>> Owner: Nobody
>> Requestors: john.w.raby2.civ at mail.mil
>> Status: new
>> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>> I am learning of a source of point observations for upper air and
surface
>> which is an ascii file called OBS DOMAIN. We are trying to solve a
problem
>> with erroneous data in the MADIS mesonet data and learned, through
another
>> ARL colleague, about the OBS DOMAIN file which is used in WRF if
observation
>> nudging is used. See the following link for a references and a
sample of the
>> file.
>>
>> http://www.mmm.ucar.edu/wrf/users/wrfv2/How_to_run_obs_fdda.html
>>
>> The OBS DOMAIN file is subjected to extensive QC in order to obtain
the best
>> set of observations for assimilation into the WRF. As such, I would
like to
>> use the observations contained in it in Point-Stat.
>>
>> The format does not appear to be MET specific nor netcdf.
>>
>> Have you ever worked this file or data from this file as the source
data to
>> provide input to Point-Stat?
>>
>> Thanks.
>>
>> R/
>> John
>>
>> Mr John W. Raby, Meteorologist
>> U.S. Army Research Laboratory
>> White Sands Missile Range, NM 88002
>> (575) 678-2004 DSN 258-2004
>> FAX (575) 678-1230 DSN 258-1230
>> Email: john.w.raby2.civ at mail.mil
>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
------------------------------------------------
Subject: Observations for Point-Stat (UNCLASSIFIED)
From: Raby, John W USA CIV
Time: Tue Sep 25 09:09:28 2012
Classification: UNCLASSIFIED
Caveats: NONE
John -
Thanks for your response below and willingness to help us out on this
project.
I've cc'd Brian on this so he gets your response. He will be our
technical POC
for this project.
The payoff of this work would be that we could make use of Brian's
QC'd set of
observations which may solve a problem we've been having with the
MADIS
mesonet surface observations from the LA area where we are conducting
studies
having serious errors in some of the variables as well as duplicate
sets of
data from two different sensors being assigned to the same identifier.
Another
added benefit is that we use the same observations used in the FDDA
process
for validation with Point-Stat.
R/
John
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, September 25, 2012 8:53 AM
To: Raby, John W CIV (US)
Subject: Re: [rt.rap.ucar.edu #58413] Observations for Point-Stat
(UNCLASSIFIED)
Hello Brian,
This is John Halley Gotway. I'm another MET developer who worked on
recent
enhancements to the ASCII2NC tool. I made these updates to ASCII2NC
in just
the last 2 weeks in support of an upcoming testing and evaluation
project for
the RAL RTFDDA system in which we're involved. And the only little_r
files I
had available for testing are from this project. While I'd be happy
to send
you guys an alpha version of the updated ASCII2NC, I'd do so with the
huge
caveat that it is not well tested. It is completely up to you make
sure
you're getting out what you expect.
Regarding the timing of observations, that's fine. They don't need to
be in
any particular order. The MET point_stat will ultimately be reading
the
output of ASCII2NC and it doesn't expect the observations to be in
order by
time.
However, the rotated winds are definitely an issue! MET assumes that
all
point observations contain winds that are earth-relative. In fact,
when we
find winds in model output that are grid-relative, we automatically
rotate
them to being earth-relative so that they can be compared with the
earth-relative observations.
Just let me know how you'd like to proceed.
Thanks,
John
On 09/25/2012 08:22 AM, brian.p.reen.ctr at mail.mil via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> Paul,
> John asked me to respond, since I'm familiar with the obs nudging
file
> formats. We are interested in obtaining the new version of ASCII2NC
> since it sounds like that could help us get our observation files in
> the format MET expects.
>
> The OBS_DOMAIN files are either output from the script you
mentioned,
> or from OBSGRID. As you noted, if we use
RT_fdda_reformat_obsnud.pl,
> then we could simply use the LITTLE_R files that are fed to
> RT_fdda_reformat_obsnud.pl.
> However, OBSGRID provides QC of the data, so we would want to be
able
> to use the QC'd data output from OBSGRID. In addition to outputting
> OBS_DOMAIN files, OBSGRID does produce LITTLE_R files so hopefully
we
> can use them in the new version of ASCII2NC.
>
> However, there are a few details we'll need to understand to
determine
> the applicability of the new version of ASCII2NC to the LITTLE_R
files
> output from OBSGRID. The U/V winds in the OBSGRID LITTLE_R file are
> rotated to the model grid (i.e., model-relative). Does ASCII2NC
> expect winds in the input LITTLE_R files to be model-relative or
> earth-relative? It would be helpful to know if ASCII2NC reads the
U/V
> wind components or the wind speed / wind direction fields from
> LITTLE_R files, since the wind directions in the OBSGRID LITTLE_R
> files are incorrect. I should be able to fix OBSGRID so it produces
> the correct wind direction if needed. Finally, the observations in
> the OBSGRID LITTLE_R files are in hourly files, but within these
files
> the observations are generally but not entirely in order of
increasing time.
> Do I need to write code time-order the LITTLE_R files for ingestion
by
> ASCII2NC?
>
> The new version of ASCII2NC sounds like it could be very helpful.
The
> biggest potential hindrance seems to be the model-relative /
> grid-relative issue. The other two details we should be able to
work
> out once we know what ASCII2NC is expecting.
>
> Brian
>
> -----Original Message-----
> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
> Sent: Monday, September 24, 2012 2:43 PM
> To: Raby, John W CIV (US)
> Subject: Re: [rt.rap.ucar.edu #58413] Observations for Point-Stat
> (UNCLASSIFIED)
>
> John,
>
> Is the OBS DOMAIN data that you mention the output of the conversion
> script RT_fdda_reformat_obsnud.pl mentioned on the web site you
> specified? If so, the input files to that script should be in a
> format called LITTLE_R. We have developed support in the MET tool
> ascii2nc for LITTLE_R, which we plan to release with METv4.1 in
> January. If you have the LITTLE_R formatted files, we could perhaps
> make arrangements to send you the new version of ascii2nc to convert
> those files, or do the conversion on your behalf if you send us the
> files. Please let us know if you have the LITTLE_R files, and
please send
> us one or more if you do. We can take it from there.
>
> Paul
>
>
> On 09/21/2012 02:26 PM, Raby, John W USA CIV via RT wrote:
>>
>> Fri Sep 21 14:26:08 2012: Request 58413 was acted upon.
>> Transaction: Ticket created by john.w.raby2.civ at mail.mil
>> Queue: met_help
>> Subject: Observations for Point-Stat (UNCLASSIFIED)
>> Owner: Nobody
>> Requestors: john.w.raby2.civ at mail.mil
>> Status: new
>> Ticket <URL:
>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>> I am learning of a source of point observations for upper air and
>> surface which is an ascii file called OBS DOMAIN. We are trying to
>> solve a problem with erroneous data in the MADIS mesonet data and
>> learned, through another ARL colleague, about the OBS DOMAIN file
>> which is used in WRF if observation nudging is used. See the
>> following link for a references and a sample of the file.
>>
>> http://www.mmm.ucar.edu/wrf/users/wrfv2/How_to_run_obs_fdda.html
>>
>> The OBS DOMAIN file is subjected to extensive QC in order to obtain
>> the best set of observations for assimilation into the WRF. As
such,
>> I would like to use the observations contained in it in Point-Stat.
>>
>> The format does not appear to be MET specific nor netcdf.
>>
>> Have you ever worked this file or data from this file as the source
>> data to provide input to Point-Stat?
>>
>> Thanks.
>>
>> R/
>> John
>>
>> Mr John W. Raby, Meteorologist
>> U.S. Army Research Laboratory
>> White Sands Missile Range, NM 88002
>> (575) 678-2004 DSN 258-2004
>> FAX (575) 678-1230 DSN 258-1230
>> Email: john.w.raby2.civ at mail.mil
>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
Classification: UNCLASSIFIED
Caveats: NONE
------------------------------------------------
Subject: Observations for Point-Stat (UNCLASSIFIED)
From: Raby, John W USA CIV
Time: Tue Sep 25 15:20:04 2012
Classification: UNCLASSIFIED
Caveats: NONE
John -
Brian plans to modify the OBSGRID program to output a little_r file
which
contains earth-relative winds. I assume that this would be an input
file to
ASCII2NC. If you could make the alpha version you mentioned available,
I could
start setting up to do some testing here and share the results with
you if
you'd like. I'm not sure of the timing on getting the little_r file
from Brian
since he started looking into the modification this afternoon.
R/
John
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Tuesday, September 25, 2012 8:53 AM
To: Raby, John W CIV (US)
Subject: Re: [rt.rap.ucar.edu #58413] Observations for Point-Stat
(UNCLASSIFIED)
Hello Brian,
This is John Halley Gotway. I'm another MET developer who worked on
recent
enhancements to the ASCII2NC tool. I made these updates to ASCII2NC
in just
the last 2 weeks in support of an upcoming testing and evaluation
project for
the RAL RTFDDA system in which we're involved. And the only little_r
files I
had available for testing are from this project. While I'd be happy
to send
you guys an alpha version of the updated ASCII2NC, I'd do so with the
huge
caveat that it is not well tested. It is completely up to you make
sure
you're getting out what you expect.
Regarding the timing of observations, that's fine. They don't need to
be in
any particular order. The MET point_stat will ultimately be reading
the
output of ASCII2NC and it doesn't expect the observations to be in
order by
time.
However, the rotated winds are definitely an issue! MET assumes that
all
point observations contain winds that are earth-relative. In fact,
when we
find winds in model output that are grid-relative, we automatically
rotate
them to being earth-relative so that they can be compared with the
earth-relative observations.
Just let me know how you'd like to proceed.
Thanks,
John
On 09/25/2012 08:22 AM, brian.p.reen.ctr at mail.mil via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> Paul,
> John asked me to respond, since I'm familiar with the obs nudging
file
> formats. We are interested in obtaining the new version of ASCII2NC
> since it sounds like that could help us get our observation files in
> the format MET expects.
>
> The OBS_DOMAIN files are either output from the script you
mentioned,
> or from OBSGRID. As you noted, if we use
RT_fdda_reformat_obsnud.pl,
> then we could simply use the LITTLE_R files that are fed to
> RT_fdda_reformat_obsnud.pl.
> However, OBSGRID provides QC of the data, so we would want to be
able
> to use the QC'd data output from OBSGRID. In addition to outputting
> OBS_DOMAIN files, OBSGRID does produce LITTLE_R files so hopefully
we
> can use them in the new version of ASCII2NC.
>
> However, there are a few details we'll need to understand to
determine
> the applicability of the new version of ASCII2NC to the LITTLE_R
files
> output from OBSGRID. The U/V winds in the OBSGRID LITTLE_R file are
> rotated to the model grid (i.e., model-relative). Does ASCII2NC
> expect winds in the input LITTLE_R files to be model-relative or
> earth-relative? It would be helpful to know if ASCII2NC reads the
U/V
> wind components or the wind speed / wind direction fields from
> LITTLE_R files, since the wind directions in the OBSGRID LITTLE_R
> files are incorrect. I should be able to fix OBSGRID so it produces
> the correct wind direction if needed. Finally, the observations in
> the OBSGRID LITTLE_R files are in hourly files, but within these
files
> the observations are generally but not entirely in order of
increasing time.
> Do I need to write code time-order the LITTLE_R files for ingestion
by
> ASCII2NC?
>
> The new version of ASCII2NC sounds like it could be very helpful.
The
> biggest potential hindrance seems to be the model-relative /
> grid-relative issue. The other two details we should be able to
work
> out once we know what ASCII2NC is expecting.
>
> Brian
>
> -----Original Message-----
> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
> Sent: Monday, September 24, 2012 2:43 PM
> To: Raby, John W CIV (US)
> Subject: Re: [rt.rap.ucar.edu #58413] Observations for Point-Stat
> (UNCLASSIFIED)
>
> John,
>
> Is the OBS DOMAIN data that you mention the output of the conversion
> script RT_fdda_reformat_obsnud.pl mentioned on the web site you
> specified? If so, the input files to that script should be in a
> format called LITTLE_R. We have developed support in the MET tool
> ascii2nc for LITTLE_R, which we plan to release with METv4.1 in
> January. If you have the LITTLE_R formatted files, we could perhaps
> make arrangements to send you the new version of ascii2nc to convert
> those files, or do the conversion on your behalf if you send us the
> files. Please let us know if you have the LITTLE_R files, and
please send
> us one or more if you do. We can take it from there.
>
> Paul
>
>
> On 09/21/2012 02:26 PM, Raby, John W USA CIV via RT wrote:
>>
>> Fri Sep 21 14:26:08 2012: Request 58413 was acted upon.
>> Transaction: Ticket created by john.w.raby2.civ at mail.mil
>> Queue: met_help
>> Subject: Observations for Point-Stat (UNCLASSIFIED)
>> Owner: Nobody
>> Requestors: john.w.raby2.civ at mail.mil
>> Status: new
>> Ticket <URL:
>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>> I am learning of a source of point observations for upper air and
>> surface which is an ascii file called OBS DOMAIN. We are trying to
>> solve a problem with erroneous data in the MADIS mesonet data and
>> learned, through another ARL colleague, about the OBS DOMAIN file
>> which is used in WRF if observation nudging is used. See the
>> following link for a references and a sample of the file.
>>
>> http://www.mmm.ucar.edu/wrf/users/wrfv2/How_to_run_obs_fdda.html
>>
>> The OBS DOMAIN file is subjected to extensive QC in order to obtain
>> the best set of observations for assimilation into the WRF. As
such,
>> I would like to use the observations contained in it in Point-Stat.
>>
>> The format does not appear to be MET specific nor netcdf.
>>
>> Have you ever worked this file or data from this file as the source
>> data to provide input to Point-Stat?
>>
>> Thanks.
>>
>> R/
>> John
>>
>> Mr John W. Raby, Meteorologist
>> U.S. Army Research Laboratory
>> White Sands Missile Range, NM 88002
>> (575) 678-2004 DSN 258-2004
>> FAX (575) 678-1230 DSN 258-1230
>> Email: john.w.raby2.civ at mail.mil
>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
Classification: UNCLASSIFIED
Caveats: NONE
------------------------------------------------
Subject: RE: [rt.rap.ucar.edu #58413] Observations for Point-Stat (UNCLASSIFIED)
From: John Halley Gotway
Time: Tue Sep 25 15:51:59 2012
John,
OK. I should be able to package up and post a tar file to our
anonymous
ftp site tomorrow. I'll let you know when it's available. It'll
basically
be like a "patch" file that you place in the top-level METv4.0
directory
and unpack - I assume you're using METv4.0? Then, you'll need to
recompile MET and give it a shot.
Perhaps this will work out well - you'll be helping me out in the
testing
and you can let me know if anything looks fishy in the output.
Thanks,
John
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> John -
>
> Brian plans to modify the OBSGRID program to output a little_r file
which
> contains earth-relative winds. I assume that this would be an input
file
> to
> ASCII2NC. If you could make the alpha version you mentioned
available, I
> could
> start setting up to do some testing here and share the results with
you if
> you'd like. I'm not sure of the timing on getting the little_r file
from
> Brian
> since he started looking into the modification this afternoon.
>
> R/
> John
>
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, September 25, 2012 8:53 AM
> To: Raby, John W CIV (US)
> Subject: Re: [rt.rap.ucar.edu #58413] Observations for Point-Stat
> (UNCLASSIFIED)
>
> Hello Brian,
>
> This is John Halley Gotway. I'm another MET developer who worked on
> recent
> enhancements to the ASCII2NC tool. I made these updates to ASCII2NC
in
> just
> the last 2 weeks in support of an upcoming testing and evaluation
project
> for
> the RAL RTFDDA system in which we're involved. And the only
little_r
> files I
> had available for testing are from this project. While I'd be happy
to
> send
> you guys an alpha version of the updated ASCII2NC, I'd do so with
the huge
> caveat that it is not well tested. It is completely up to you make
sure
> you're getting out what you expect.
>
> Regarding the timing of observations, that's fine. They don't need
to be
> in
> any particular order. The MET point_stat will ultimately be reading
the
> output of ASCII2NC and it doesn't expect the observations to be in
order
> by
> time.
>
> However, the rotated winds are definitely an issue! MET assumes
that all
> point observations contain winds that are earth-relative. In fact,
when
> we
> find winds in model output that are grid-relative, we automatically
rotate
> them to being earth-relative so that they can be compared with the
> earth-relative observations.
>
> Just let me know how you'd like to proceed.
>
> Thanks,
> John
>
>
> On 09/25/2012 08:22 AM, brian.p.reen.ctr at mail.mil via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>> Paul,
>> John asked me to respond, since I'm familiar with the obs nudging
file
>> formats. We are interested in obtaining the new version of
ASCII2NC
>> since it sounds like that could help us get our observation files
in
>> the format MET expects.
>>
>> The OBS_DOMAIN files are either output from the script you
mentioned,
>> or from OBSGRID. As you noted, if we use
RT_fdda_reformat_obsnud.pl,
>> then we could simply use the LITTLE_R files that are fed to
>> RT_fdda_reformat_obsnud.pl.
>> However, OBSGRID provides QC of the data, so we would want to be
able
>> to use the QC'd data output from OBSGRID. In addition to
outputting
>> OBS_DOMAIN files, OBSGRID does produce LITTLE_R files so hopefully
we
>> can use them in the new version of ASCII2NC.
>>
>> However, there are a few details we'll need to understand to
determine
>> the applicability of the new version of ASCII2NC to the LITTLE_R
files
>> output from OBSGRID. The U/V winds in the OBSGRID LITTLE_R file
are
>> rotated to the model grid (i.e., model-relative). Does ASCII2NC
>> expect winds in the input LITTLE_R files to be model-relative or
>> earth-relative? It would be helpful to know if ASCII2NC reads the
U/V
>> wind components or the wind speed / wind direction fields from
>> LITTLE_R files, since the wind directions in the OBSGRID LITTLE_R
>> files are incorrect. I should be able to fix OBSGRID so it
produces
>> the correct wind direction if needed. Finally, the observations in
>> the OBSGRID LITTLE_R files are in hourly files, but within these
files
>> the observations are generally but not entirely in order of
increasing
>> time.
>> Do I need to write code time-order the LITTLE_R files for ingestion
by
>> ASCII2NC?
>>
>> The new version of ASCII2NC sounds like it could be very helpful.
The
>> biggest potential hindrance seems to be the model-relative /
>> grid-relative issue. The other two details we should be able to
work
>> out once we know what ASCII2NC is expecting.
>>
>> Brian
>>
>> -----Original Message-----
>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>> Sent: Monday, September 24, 2012 2:43 PM
>> To: Raby, John W CIV (US)
>> Subject: Re: [rt.rap.ucar.edu #58413] Observations for Point-Stat
>> (UNCLASSIFIED)
>>
>> John,
>>
>> Is the OBS DOMAIN data that you mention the output of the
conversion
>> script RT_fdda_reformat_obsnud.pl mentioned on the web site you
>> specified? If so, the input files to that script should be in a
>> format called LITTLE_R. We have developed support in the MET tool
>> ascii2nc for LITTLE_R, which we plan to release with METv4.1 in
>> January. If you have the LITTLE_R formatted files, we could
perhaps
>> make arrangements to send you the new version of ascii2nc to
convert
>> those files, or do the conversion on your behalf if you send us the
>> files. Please let us know if you have the LITTLE_R files, and
please
>> send
>> us one or more if you do. We can take it from there.
>>
>> Paul
>>
>>
>> On 09/21/2012 02:26 PM, Raby, John W USA CIV via RT wrote:
>>>
>>> Fri Sep 21 14:26:08 2012: Request 58413 was acted upon.
>>> Transaction: Ticket created by john.w.raby2.civ at mail.mil
>>> Queue: met_help
>>> Subject: Observations for Point-Stat (UNCLASSIFIED)
>>> Owner: Nobody
>>> Requestors: john.w.raby2.civ at mail.mil
>>> Status: new
>>> Ticket <URL:
>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>>>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>> I am learning of a source of point observations for upper air and
>>> surface which is an ascii file called OBS DOMAIN. We are trying to
>>> solve a problem with erroneous data in the MADIS mesonet data and
>>> learned, through another ARL colleague, about the OBS DOMAIN file
>>> which is used in WRF if observation nudging is used. See the
>>> following link for a references and a sample of the file.
>>>
>>> http://www.mmm.ucar.edu/wrf/users/wrfv2/How_to_run_obs_fdda.html
>>>
>>> The OBS DOMAIN file is subjected to extensive QC in order to
obtain
>>> the best set of observations for assimilation into the WRF. As
such,
>>> I would like to use the observations contained in it in Point-
Stat.
>>>
>>> The format does not appear to be MET specific nor netcdf.
>>>
>>> Have you ever worked this file or data from this file as the
source
>>> data to provide input to Point-Stat?
>>>
>>> Thanks.
>>>
>>> R/
>>> John
>>>
>>> Mr John W. Raby, Meteorologist
>>> U.S. Army Research Laboratory
>>> White Sands Missile Range, NM 88002
>>> (575) 678-2004 DSN 258-2004
>>> FAX (575) 678-1230 DSN 258-1230
>>> Email: john.w.raby2.civ at mail.mil
>>>
>>>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>>
>>>
>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
>
------------------------------------------------
Subject: Observations for Point-Stat (UNCLASSIFIED)
From: John Halley Gotway
Time: Wed Sep 26 10:40:11 2012
John,
OK, I've attached a tar file consisting of a small patch to ASCII2NC
for
little_r support. It's really just 3 source files that were updated.
First, please be sure you have the latest set of METv4.0 patches:
http://www.dtcenter.org/met/users/support/known_issues/METv4.0/index.php
Retrieve the patch file from that page, place it in the top-level
METv4.0
and then extract the files.
Second, place the attached tar file in the top-level METv4.0 directory
and
extract the files.
Third, recompile MET.
When you run ASCII2NC, you can now pass it a little_r file. You can
specify that it's little_r using the "-format little_r" command line
argument. But actually, you can omit that if you'd like because I
made
the code smart enough to interrogate the file and guess the format.
Please let me know what, if any, issues arise. Also, please closely
check
the output of the tool to make sure that its what you expect. As I
mentioned, this is development code that we have used extensively in
our
testing yet.
Hope it helps.
Thanks,
John
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> John -
>
> Brian plans to modify the OBSGRID program to output a little_r file
which
> contains earth-relative winds. I assume that this would be an input
file
> to
> ASCII2NC. If you could make the alpha version you mentioned
available, I
> could
> start setting up to do some testing here and share the results with
you if
> you'd like. I'm not sure of the timing on getting the little_r file
from
> Brian
> since he started looking into the modification this afternoon.
>
> R/
> John
>
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, September 25, 2012 8:53 AM
> To: Raby, John W CIV (US)
> Subject: Re: [rt.rap.ucar.edu #58413] Observations for Point-Stat
> (UNCLASSIFIED)
>
> Hello Brian,
>
> This is John Halley Gotway. I'm another MET developer who worked on
> recent
> enhancements to the ASCII2NC tool. I made these updates to ASCII2NC
in
> just
> the last 2 weeks in support of an upcoming testing and evaluation
project
> for
> the RAL RTFDDA system in which we're involved. And the only
little_r
> files I
> had available for testing are from this project. While I'd be happy
to
> send
> you guys an alpha version of the updated ASCII2NC, I'd do so with
the huge
> caveat that it is not well tested. It is completely up to you make
sure
> you're getting out what you expect.
>
> Regarding the timing of observations, that's fine. They don't need
to be
> in
> any particular order. The MET point_stat will ultimately be reading
the
> output of ASCII2NC and it doesn't expect the observations to be in
order
> by
> time.
>
> However, the rotated winds are definitely an issue! MET assumes
that all
> point observations contain winds that are earth-relative. In fact,
when
> we
> find winds in model output that are grid-relative, we automatically
rotate
> them to being earth-relative so that they can be compared with the
> earth-relative observations.
>
> Just let me know how you'd like to proceed.
>
> Thanks,
> John
>
>
> On 09/25/2012 08:22 AM, brian.p.reen.ctr at mail.mil via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>> Paul,
>> John asked me to respond, since I'm familiar with the obs nudging
file
>> formats. We are interested in obtaining the new version of
ASCII2NC
>> since it sounds like that could help us get our observation files
in
>> the format MET expects.
>>
>> The OBS_DOMAIN files are either output from the script you
mentioned,
>> or from OBSGRID. As you noted, if we use
RT_fdda_reformat_obsnud.pl,
>> then we could simply use the LITTLE_R files that are fed to
>> RT_fdda_reformat_obsnud.pl.
>> However, OBSGRID provides QC of the data, so we would want to be
able
>> to use the QC'd data output from OBSGRID. In addition to
outputting
>> OBS_DOMAIN files, OBSGRID does produce LITTLE_R files so hopefully
we
>> can use them in the new version of ASCII2NC.
>>
>> However, there are a few details we'll need to understand to
determine
>> the applicability of the new version of ASCII2NC to the LITTLE_R
files
>> output from OBSGRID. The U/V winds in the OBSGRID LITTLE_R file
are
>> rotated to the model grid (i.e., model-relative). Does ASCII2NC
>> expect winds in the input LITTLE_R files to be model-relative or
>> earth-relative? It would be helpful to know if ASCII2NC reads the
U/V
>> wind components or the wind speed / wind direction fields from
>> LITTLE_R files, since the wind directions in the OBSGRID LITTLE_R
>> files are incorrect. I should be able to fix OBSGRID so it
produces
>> the correct wind direction if needed. Finally, the observations in
>> the OBSGRID LITTLE_R files are in hourly files, but within these
files
>> the observations are generally but not entirely in order of
increasing
>> time.
>> Do I need to write code time-order the LITTLE_R files for ingestion
by
>> ASCII2NC?
>>
>> The new version of ASCII2NC sounds like it could be very helpful.
The
>> biggest potential hindrance seems to be the model-relative /
>> grid-relative issue. The other two details we should be able to
work
>> out once we know what ASCII2NC is expecting.
>>
>> Brian
>>
>> -----Original Message-----
>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>> Sent: Monday, September 24, 2012 2:43 PM
>> To: Raby, John W CIV (US)
>> Subject: Re: [rt.rap.ucar.edu #58413] Observations for Point-Stat
>> (UNCLASSIFIED)
>>
>> John,
>>
>> Is the OBS DOMAIN data that you mention the output of the
conversion
>> script RT_fdda_reformat_obsnud.pl mentioned on the web site you
>> specified? If so, the input files to that script should be in a
>> format called LITTLE_R. We have developed support in the MET tool
>> ascii2nc for LITTLE_R, which we plan to release with METv4.1 in
>> January. If you have the LITTLE_R formatted files, we could
perhaps
>> make arrangements to send you the new version of ascii2nc to
convert
>> those files, or do the conversion on your behalf if you send us the
>> files. Please let us know if you have the LITTLE_R files, and
please
>> send
>> us one or more if you do. We can take it from there.
>>
>> Paul
>>
>>
>> On 09/21/2012 02:26 PM, Raby, John W USA CIV via RT wrote:
>>>
>>> Fri Sep 21 14:26:08 2012: Request 58413 was acted upon.
>>> Transaction: Ticket created by john.w.raby2.civ at mail.mil
>>> Queue: met_help
>>> Subject: Observations for Point-Stat (UNCLASSIFIED)
>>> Owner: Nobody
>>> Requestors: john.w.raby2.civ at mail.mil
>>> Status: new
>>> Ticket <URL:
>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>>>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>> I am learning of a source of point observations for upper air and
>>> surface which is an ascii file called OBS DOMAIN. We are trying to
>>> solve a problem with erroneous data in the MADIS mesonet data and
>>> learned, through another ARL colleague, about the OBS DOMAIN file
>>> which is used in WRF if observation nudging is used. See the
>>> following link for a references and a sample of the file.
>>>
>>> http://www.mmm.ucar.edu/wrf/users/wrfv2/How_to_run_obs_fdda.html
>>>
>>> The OBS DOMAIN file is subjected to extensive QC in order to
obtain
>>> the best set of observations for assimilation into the WRF. As
such,
>>> I would like to use the observations contained in it in Point-
Stat.
>>>
>>> The format does not appear to be MET specific nor netcdf.
>>>
>>> Have you ever worked this file or data from this file as the
source
>>> data to provide input to Point-Stat?
>>>
>>> Thanks.
>>>
>>> R/
>>> John
>>>
>>> Mr John W. Raby, Meteorologist
>>> U.S. Army Research Laboratory
>>> White Sands Missile Range, NM 88002
>>> (575) 678-2004 DSN 258-2004
>>> FAX (575) 678-1230 DSN 258-1230
>>> Email: john.w.raby2.civ at mail.mil
>>>
>>>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>>
>>>
>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
>
------------------------------------------------
Subject: Observations for Point-Stat (UNCLASSIFIED)
From: Raby, John W USA CIV
Time: Wed Sep 26 10:54:16 2012
Classification: UNCLASSIFIED
Caveats: NONE
John -
Thanks for your cooperation on this project and the patch. We are
looking
forward to testing it and reporting results.
Our sysadm person is greasing the skids with his management to start
the
process of acquiring new Army approvals to load software on our RedHat
systems, so we may not have any results to report until this new
process is
completed. This is not because of anything from your end, just some
new ARL
procedures. Please bear with us.
We do have MET V4.0 compiled here as this was done before the new
approval
process was implemented. We will check the link for the latest
patches.
We understand that the code is developmental and not thoroughly tested
yet.
R/
John
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Wednesday, September 26, 2012 10:40 AM
To: Raby, John W CIV (US)
Subject: RE: [rt.rap.ucar.edu #58413] Observations for Point-Stat
(UNCLASSIFIED)
John,
OK, I've attached a tar file consisting of a small patch to ASCII2NC
for
little_r support. It's really just 3 source files that were updated.
First, please be sure you have the latest set of METv4.0 patches:
http://www.dtcenter.org/met/users/support/known_issues/METv4.0/index.php
Retrieve the patch file from that page, place it in the top-level
METv4.0 and
then extract the files.
Second, place the attached tar file in the top-level METv4.0 directory
and
extract the files.
Third, recompile MET.
When you run ASCII2NC, you can now pass it a little_r file. You can
specify
that it's little_r using the "-format little_r" command line argument.
But
actually, you can omit that if you'd like because I made the code
smart enough
to interrogate the file and guess the format.
Please let me know what, if any, issues arise. Also, please closely
check the
output of the tool to make sure that its what you expect. As I
mentioned,
this is development code that we have used extensively in our testing
yet.
Hope it helps.
Thanks,
John
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> John -
>
> Brian plans to modify the OBSGRID program to output a little_r file
> which contains earth-relative winds. I assume that this would be an
> input file to ASCII2NC. If you could make the alpha version you
> mentioned available, I could start setting up to do some testing
here
> and share the results with you if you'd like. I'm not sure of the
> timing on getting the little_r file from Brian since he started
> looking into the modification this afternoon.
>
> R/
> John
>
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Tuesday, September 25, 2012 8:53 AM
> To: Raby, John W CIV (US)
> Subject: Re: [rt.rap.ucar.edu #58413] Observations for Point-Stat
> (UNCLASSIFIED)
>
> Hello Brian,
>
> This is John Halley Gotway. I'm another MET developer who worked on
> recent enhancements to the ASCII2NC tool. I made these updates to
> ASCII2NC in just the last 2 weeks in support of an upcoming testing
> and evaluation project for the RAL RTFDDA system in which we're
> involved. And the only little_r files I had available for testing
are
> from this project. While I'd be happy to send you guys an alpha
> version of the updated ASCII2NC, I'd do so with the huge caveat that
> it is not well tested. It is completely up to you make sure you're
> getting out what you expect.
>
> Regarding the timing of observations, that's fine. They don't need
to
> be in any particular order. The MET point_stat will ultimately be
> reading the output of ASCII2NC and it doesn't expect the
observations
> to be in order by time.
>
> However, the rotated winds are definitely an issue! MET assumes
that
> all point observations contain winds that are earth-relative. In
> fact, when we find winds in model output that are grid-relative, we
> automatically rotate them to being earth-relative so that they can
be
> compared with the earth-relative observations.
>
> Just let me know how you'd like to proceed.
>
> Thanks,
> John
>
>
> On 09/25/2012 08:22 AM, brian.p.reen.ctr at mail.mil via RT wrote:
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>> Paul,
>> John asked me to respond, since I'm familiar with the obs nudging
>> file formats. We are interested in obtaining the new version of
>> ASCII2NC since it sounds like that could help us get our
observation
>> files in the format MET expects.
>>
>> The OBS_DOMAIN files are either output from the script you
mentioned,
>> or from OBSGRID. As you noted, if we use
RT_fdda_reformat_obsnud.pl,
>> then we could simply use the LITTLE_R files that are fed to
>> RT_fdda_reformat_obsnud.pl.
>> However, OBSGRID provides QC of the data, so we would want to be
able
>> to use the QC'd data output from OBSGRID. In addition to
outputting
>> OBS_DOMAIN files, OBSGRID does produce LITTLE_R files so hopefully
we
>> can use them in the new version of ASCII2NC.
>>
>> However, there are a few details we'll need to understand to
>> determine the applicability of the new version of ASCII2NC to the
>> LITTLE_R files output from OBSGRID. The U/V winds in the OBSGRID
>> LITTLE_R file are rotated to the model grid (i.e., model-relative).
>> Does ASCII2NC expect winds in the input LITTLE_R files to be
>> model-relative or earth-relative? It would be helpful to know if
>> ASCII2NC reads the U/V wind components or the wind speed / wind
>> direction fields from LITTLE_R files, since the wind directions in
>> the OBSGRID LITTLE_R files are incorrect. I should be able to fix
>> OBSGRID so it produces the correct wind direction if needed.
>> Finally, the observations in the OBSGRID LITTLE_R files are in
hourly
>> files, but within these files the observations are generally but
not
>> entirely in order of increasing time.
>> Do I need to write code time-order the LITTLE_R files for ingestion
>> by ASCII2NC?
>>
>> The new version of ASCII2NC sounds like it could be very helpful.
>> The biggest potential hindrance seems to be the model-relative /
>> grid-relative issue. The other two details we should be able to
work
>> out once we know what ASCII2NC is expecting.
>>
>> Brian
>>
>> -----Original Message-----
>> From: Paul Oldenburg via RT [mailto:met_help at ucar.edu]
>> Sent: Monday, September 24, 2012 2:43 PM
>> To: Raby, John W CIV (US)
>> Subject: Re: [rt.rap.ucar.edu #58413] Observations for Point-Stat
>> (UNCLASSIFIED)
>>
>> John,
>>
>> Is the OBS DOMAIN data that you mention the output of the
conversion
>> script RT_fdda_reformat_obsnud.pl mentioned on the web site you
>> specified? If so, the input files to that script should be in a
>> format called LITTLE_R. We have developed support in the MET tool
>> ascii2nc for LITTLE_R, which we plan to release with METv4.1 in
>> January. If you have the LITTLE_R formatted files, we could
perhaps
>> make arrangements to send you the new version of ascii2nc to
convert
>> those files, or do the conversion on your behalf if you send us the
>> files. Please let us know if you have the LITTLE_R files, and
please
>> send us one or more if you do. We can take it from there.
>>
>> Paul
>>
>>
>> On 09/21/2012 02:26 PM, Raby, John W USA CIV via RT wrote:
>>>
>>> Fri Sep 21 14:26:08 2012: Request 58413 was acted upon.
>>> Transaction: Ticket created by john.w.raby2.civ at mail.mil
>>> Queue: met_help
>>> Subject: Observations for Point-Stat (UNCLASSIFIED)
>>> Owner: Nobody
>>> Requestors: john.w.raby2.civ at mail.mil
>>> Status: new
>>> Ticket <URL:
>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>>>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>> I am learning of a source of point observations for upper air and
>>> surface which is an ascii file called OBS DOMAIN. We are trying to
>>> solve a problem with erroneous data in the MADIS mesonet data and
>>> learned, through another ARL colleague, about the OBS DOMAIN file
>>> which is used in WRF if observation nudging is used. See the
>>> following link for a references and a sample of the file.
>>>
>>> http://www.mmm.ucar.edu/wrf/users/wrfv2/How_to_run_obs_fdda.html
>>>
>>> The OBS DOMAIN file is subjected to extensive QC in order to
obtain
>>> the best set of observations for assimilation into the WRF. As
such,
>>> I would like to use the observations contained in it in Point-
Stat.
>>>
>>> The format does not appear to be MET specific nor netcdf.
>>>
>>> Have you ever worked this file or data from this file as the
source
>>> data to provide input to Point-Stat?
>>>
>>> Thanks.
>>>
>>> R/
>>> John
>>>
>>> Mr John W. Raby, Meteorologist
>>> U.S. Army Research Laboratory
>>> White Sands Missile Range, NM 88002
>>> (575) 678-2004 DSN 258-2004
>>> FAX (575) 678-1230 DSN 258-1230
>>> Email: john.w.raby2.civ at mail.mil
>>>
>>>
>>>
>>> Classification: UNCLASSIFIED
>>> Caveats: NONE
>>>
>>>
>>>
>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>> Classification: UNCLASSIFIED
>> Caveats: NONE
>>
>>
>>
>
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
>
Classification: UNCLASSIFIED
Caveats: NONE
------------------------------------------------
Subject: Observations for Point-Stat (UNCLASSIFIED)
From: Raby, John W USA CIV
Time: Tue Oct 09 07:00:23 2012
Classification: UNCLASSIFIED
Caveats: NONE
John -
No problem closing this for now. We have not installed the ASCII2NC
patch yet.
Once we do , I'll do some testing and let you know the results.
Thanks for bearing with us and for your assistance in investigating
whether we
can use this new source of observations.
R/
John
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Monday, October 08, 2012 8:48 AM
To: Raby, John W CIV (US)
Subject: [rt.rap.ucar.edu #58413] Resolved: Observations for Point-
Stat
(UNCLASSIFIED)
According to our records, your request has been resolved. If you have
any
further questions or concerns, please respond to this message.
Classification: UNCLASSIFIED
Caveats: NONE
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #58413] Resolved: Observations for Point-Stat (UNCLASSIFIED)
From: John Halley Gotway
Time: Tue Oct 09 09:53:01 2012
John,
Great, sounds good. If issues arise in the future, just send us a new
question.
Thanks,
John
On 10/09/2012 07:00 AM, Raby, John W USA CIV via RT wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=58413 >
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
> John -
>
> No problem closing this for now. We have not installed the ASCII2NC
patch yet.
> Once we do , I'll do some testing and let you know the results.
>
> Thanks for bearing with us and for your assistance in investigating
whether we
> can use this new source of observations.
>
> R/
> John
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Monday, October 08, 2012 8:48 AM
> To: Raby, John W CIV (US)
> Subject: [rt.rap.ucar.edu #58413] Resolved: Observations for Point-
Stat
> (UNCLASSIFIED)
>
> According to our records, your request has been resolved. If you
have any
> further questions or concerns, please respond to this message.
>
> Classification: UNCLASSIFIED
> Caveats: NONE
>
>
>
------------------------------------------------
More information about the Met_help
mailing list