[Met_help] [rt.rap.ucar.edu #93123] History for computing vector wind RMS
John Halley Gotway via RT
met_help at ucar.edu
Mon Nov 18 10:30:27 MST 2019
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
John,
Can you advise the easiest way to compute vector wind RMS using MET. E.g.
for each region, I want to calculate sqrt(U_error^2+ V_error^2).
Thanks
Elizabeth Satterfield
Editor, Monthly Weather Review
Chair, AMS Committee on Probability and Statistics
Meteorologist, Code 7531
Marine Meteorology Division
U.S. Naval Research Laboratory
T 831.656.4056 F 831.656.4026
<http://www.nrl.navy.mil/> www.nrl.navy.mil
__________________________________________________________________
FOR OFFICIAL USE ONLY - PRIVACY SENSITIVE
Any misuse or unauthorized disclosure may result in both civil and criminal
penalties.
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: computing vector wind RMS
From: John Halley Gotway
Time: Thu Nov 14 09:43:09 2019
Liz,
You don't send many questions to met_help, but when you do, they're
usually
zingers, as this one is! That is to say, you ask good questions.
Unfortunately, the short answer is no, there's no good way of using
met-8.1
to compute sqrt(U_error^2+ V_error^2). MET does have the ability to
compute vector partial sums (VL1L2 line type) and vector continuous
statistics (VCNT line type). Their contents are described in tables
7.17
and 7.19 of the MET User's Guide (
https://dtcenter.org/sites/default/files/community-code/met/docs/user-
guide/MET_Users_Guide_v8.1.2.pdf).
And the vector stats are defined in Appendix G.
In practice though, there is currently no way of getting VL1L2 output
when
using Python embedding. Additionally, the logic for deriving it from
GRIB
and NetCDF input files is a bit fragile as well. There are a handful
of
variables in MET for which special logic is defined... namely U-wind,
V-wind, wind speed, wind direction, precipitation, and specific
humidity.
The MET libraries for reading gridded data contain logic for
determining
what datasets qualify as being these, usually by their name. But
there's
currently no way of overriding this logic. And in the case of Python
embedding, they're always set to false.
We should enhance MET to provide a way of overriding, or for Python
embedding, setting this logic. We already have a somewhat similar
GitHub
issue for letting the user override other metadata, like the name of
the
data, level string, units string, timing information, and grid
definition (
https://github.com/NCAR/MET/issues/1020). This word would basically
be
adding options for setting flags to specify these data types.
Once MET knows which fields contain the U and V components of wind, it
processes them together to do 2 things...
(1) It rotates U and V from being grid-relative to earth-relative, but
only
if necessary. This is defined by a flag in the GRIB1/2 files. For
Python
embedding, we'd need another flag to define "wind_is_grid_relative =
TRUE
or FALSE".
(2) After rotating grid to earth-relative, it processes the U and V
vectors
to compute vector partial sums and statistics.
Let me talk to the MET development team to find out how these
enhancements
slot into our development timeline.
Thanks,
John
On Wed, Nov 13, 2019 at 4:13 PM Satterfield, Dr. Elizabeth via RT <
met_help at ucar.edu> wrote:
>
> Wed Nov 13 16:13:52 2019: Request 93123 was acted upon.
> Transaction: Ticket created by Elizabeth.Satterfield at nrlmry.navy.mil
> Queue: met_help
> Subject: computing vector wind RMS
> Owner: Nobody
> Requestors: Elizabeth.Satterfield at nrlmry.navy.mil
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=93123 >
>
>
> John,
>
>
>
> Can you advise the easiest way to compute vector wind RMS using MET.
E.g.
> for each region, I want to calculate sqrt(U_error^2+ V_error^2).
>
>
>
> Thanks
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> Elizabeth Satterfield
> Editor, Monthly Weather Review
>
> Chair, AMS Committee on Probability and Statistics
>
> Meteorologist, Code 7531
> Marine Meteorology Division
> U.S. Naval Research Laboratory
> T 831.656.4056 F 831.656.4026
> <http://www.nrl.navy.mil/> www.nrl.navy.mil
>
> __________________________________________________________________
> FOR OFFICIAL USE ONLY - PRIVACY SENSITIVE
> Any misuse or unauthorized disclosure may result in both civil and
criminal
> penalties.
>
>
>
>
>
>
>
------------------------------------------------
Subject: computing vector wind RMS
From: Satterfield, Dr. Elizabeth
Time: Thu Nov 14 09:49:23 2019
HI John,
Yes -- I experimented with the VL1L2 output and found that it would
not work
with Python embedding. It would be great if this capability could be
added.
Please keep me posted.
Liz
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Thursday, November 14, 2019 8:43 AM
To: Satterfield, Dr. Elizabeth
Subject: Re: [rt.rap.ucar.edu #93123] computing vector wind RMS
Liz,
You don't send many questions to met_help, but when you do, they're
usually
zingers, as this one is! That is to say, you ask good questions.
Unfortunately, the short answer is no, there's no good way of using
met-8.1
to compute sqrt(U_error^2+ V_error^2). MET does have the ability to
compute vector partial sums (VL1L2 line type) and vector continuous
statistics (VCNT line type). Their contents are described in tables
7.17
and 7.19 of the MET User's Guide (
https://dtcenter.org/sites/default/files/community-code/met/docs/user-
guide/MET_Users_Guide_v8.1.2.pdf).
And the vector stats are defined in Appendix G.
In practice though, there is currently no way of getting VL1L2 output
when
using Python embedding. Additionally, the logic for deriving it from
GRIB
and NetCDF input files is a bit fragile as well. There are a handful
of
variables in MET for which special logic is defined... namely U-wind,
V-wind, wind speed, wind direction, precipitation, and specific
humidity.
The MET libraries for reading gridded data contain logic for
determining
what datasets qualify as being these, usually by their name. But
there's
currently no way of overriding this logic. And in the case of Python
embedding, they're always set to false.
We should enhance MET to provide a way of overriding, or for Python
embedding, setting this logic. We already have a somewhat similar
GitHub
issue for letting the user override other metadata, like the name of
the
data, level string, units string, timing information, and grid
definition (
https://github.com/NCAR/MET/issues/1020). This word would basically
be
adding options for setting flags to specify these data types.
Once MET knows which fields contain the U and V components of wind, it
processes them together to do 2 things...
(1) It rotates U and V from being grid-relative to earth-relative, but
only
if necessary. This is defined by a flag in the GRIB1/2 files. For
Python
embedding, we'd need another flag to define "wind_is_grid_relative =
TRUE
or FALSE".
(2) After rotating grid to earth-relative, it processes the U and V
vectors
to compute vector partial sums and statistics.
Let me talk to the MET development team to find out how these
enhancements
slot into our development timeline.
Thanks,
John
On Wed, Nov 13, 2019 at 4:13 PM Satterfield, Dr. Elizabeth via RT <
met_help at ucar.edu> wrote:
>
> Wed Nov 13 16:13:52 2019: Request 93123 was acted upon.
> Transaction: Ticket created by Elizabeth.Satterfield at nrlmry.navy.mil
> Queue: met_help
> Subject: computing vector wind RMS
> Owner: Nobody
> Requestors: Elizabeth.Satterfield at nrlmry.navy.mil
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=93123 >
>
>
> John,
>
>
>
> Can you advise the easiest way to compute vector wind RMS using MET.
E.g.
> for each region, I want to calculate sqrt(U_error^2+ V_error^2).
>
>
>
> Thanks
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> Elizabeth Satterfield
> Editor, Monthly Weather Review
>
> Chair, AMS Committee on Probability and Statistics
>
> Meteorologist, Code 7531
> Marine Meteorology Division
> U.S. Naval Research Laboratory
> T 831.656.4056 F 831.656.4026
> <http://www.nrl.navy.mil/> www.nrl.navy.mil
>
> __________________________________________________________________
> FOR OFFICIAL USE ONLY - PRIVACY SENSITIVE
> Any misuse or unauthorized disclosure may result in both civil and
criminal
> penalties.
>
>
>
>
>
>
>
------------------------------------------------
Subject: computing vector wind RMS
From: John Halley Gotway
Time: Thu Nov 14 10:58:43 2019
Liz,
I updated this GitHub issue with more details:
https://github.com/NCAR/MET/issues/1020
We're working on development for met-9.0, but this task is behind
several
other high priority issues. So it's likely we won't be able to
include it
in met-9.0. But Tara tells me that there is NRL funding in the works
that
might help us address this.
Thanks,
John
On Thu, Nov 14, 2019 at 9:49 AM Satterfield, Dr. Elizabeth via RT <
met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=93123 >
>
> HI John,
>
> Yes -- I experimented with the VL1L2 output and found that it would
not
> work
> with Python embedding. It would be great if this capability could
be
> added.
> Please keep me posted.
>
> Liz
>
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Thursday, November 14, 2019 8:43 AM
> To: Satterfield, Dr. Elizabeth
> Subject: Re: [rt.rap.ucar.edu #93123] computing vector wind RMS
>
> Liz,
>
> You don't send many questions to met_help, but when you do, they're
usually
> zingers, as this one is! That is to say, you ask good questions.
>
> Unfortunately, the short answer is no, there's no good way of using
met-8.1
> to compute sqrt(U_error^2+ V_error^2). MET does have the ability to
> compute vector partial sums (VL1L2 line type) and vector continuous
> statistics (VCNT line type). Their contents are described in tables
7.17
> and 7.19 of the MET User's Guide (
>
> https://dtcenter.org/sites/default/files/community-
code/met/docs/user-guide/MET_Users_Guide_v8.1.2.pdf
> ).
> And the vector stats are defined in Appendix G.
>
> In practice though, there is currently no way of getting VL1L2
output when
> using Python embedding. Additionally, the logic for deriving it
from GRIB
> and NetCDF input files is a bit fragile as well. There are a
handful of
> variables in MET for which special logic is defined... namely U-
wind,
> V-wind, wind speed, wind direction, precipitation, and specific
humidity.
> The MET libraries for reading gridded data contain logic for
determining
> what datasets qualify as being these, usually by their name. But
there's
> currently no way of overriding this logic. And in the case of
Python
> embedding, they're always set to false.
>
> We should enhance MET to provide a way of overriding, or for Python
> embedding, setting this logic. We already have a somewhat similar
GitHub
> issue for letting the user override other metadata, like the name of
the
> data, level string, units string, timing information, and grid
definition (
> https://github.com/NCAR/MET/issues/1020). This word would basically
be
> adding options for setting flags to specify these data types.
>
> Once MET knows which fields contain the U and V components of wind,
it
> processes them together to do 2 things...
> (1) It rotates U and V from being grid-relative to earth-relative,
but only
> if necessary. This is defined by a flag in the GRIB1/2 files. For
Python
> embedding, we'd need another flag to define "wind_is_grid_relative =
TRUE
> or FALSE".
> (2) After rotating grid to earth-relative, it processes the U and V
vectors
> to compute vector partial sums and statistics.
>
> Let me talk to the MET development team to find out how these
enhancements
> slot into our development timeline.
>
> Thanks,
> John
>
>
>
>
>
> On Wed, Nov 13, 2019 at 4:13 PM Satterfield, Dr. Elizabeth via RT <
> met_help at ucar.edu> wrote:
>
> >
> > Wed Nov 13 16:13:52 2019: Request 93123 was acted upon.
> > Transaction: Ticket created by
Elizabeth.Satterfield at nrlmry.navy.mil
> > Queue: met_help
> > Subject: computing vector wind RMS
> > Owner: Nobody
> > Requestors: Elizabeth.Satterfield at nrlmry.navy.mil
> > Status: new
> > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=93123 >
> >
> >
> > John,
> >
> >
> >
> > Can you advise the easiest way to compute vector wind RMS using
MET.
> E.g.
> > for each region, I want to calculate sqrt(U_error^2+ V_error^2).
> >
> >
> >
> > Thanks
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> > Elizabeth Satterfield
> > Editor, Monthly Weather Review
> >
> > Chair, AMS Committee on Probability and Statistics
> >
> > Meteorologist, Code 7531
> > Marine Meteorology Division
> > U.S. Naval Research Laboratory
> > T 831.656.4056 F 831.656.4026
> > <http://www.nrl.navy.mil/> www.nrl.navy.mil
> >
> > __________________________________________________________________
> > FOR OFFICIAL USE ONLY - PRIVACY SENSITIVE
> > Any misuse or unauthorized disclosure may result in both civil and
> criminal
> > penalties.
> >
> >
> >
> >
> >
> >
> >
>
>
>
------------------------------------------------
Subject: computing vector wind RMS
From: Satterfield, Dr. Elizabeth
Time: Thu Nov 14 11:18:26 2019
Good to know. Yes, there are funds expected from NRL, hopefully we
can add
that in as a milestone.
Thanks,
Liz
-----Original Message-----
From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
Sent: Thursday, November 14, 2019 9:59 AM
To: Satterfield, Dr. Elizabeth
Subject: Re: [rt.rap.ucar.edu #93123] computing vector wind RMS
Liz,
I updated this GitHub issue with more details:
https://github.com/NCAR/MET/issues/1020
We're working on development for met-9.0, but this task is behind
several
other high priority issues. So it's likely we won't be able to
include it
in met-9.0. But Tara tells me that there is NRL funding in the works
that
might help us address this.
Thanks,
John
On Thu, Nov 14, 2019 at 9:49 AM Satterfield, Dr. Elizabeth via RT <
met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=93123 >
>
> HI John,
>
> Yes -- I experimented with the VL1L2 output and found that it would
not
> work
> with Python embedding. It would be great if this capability could
be
> added.
> Please keep me posted.
>
> Liz
>
>
> -----Original Message-----
> From: John Halley Gotway via RT [mailto:met_help at ucar.edu]
> Sent: Thursday, November 14, 2019 8:43 AM
> To: Satterfield, Dr. Elizabeth
> Subject: Re: [rt.rap.ucar.edu #93123] computing vector wind RMS
>
> Liz,
>
> You don't send many questions to met_help, but when you do, they're
usually
> zingers, as this one is! That is to say, you ask good questions.
>
> Unfortunately, the short answer is no, there's no good way of using
met-8.1
> to compute sqrt(U_error^2+ V_error^2). MET does have the ability to
> compute vector partial sums (VL1L2 line type) and vector continuous
> statistics (VCNT line type). Their contents are described in tables
7.17
> and 7.19 of the MET User's Guide (
>
> https://dtcenter.org/sites/default/files/community-
code/met/docs/user-guide/MET_Users_Guide_v8.1.2.pdf
> ).
> And the vector stats are defined in Appendix G.
>
> In practice though, there is currently no way of getting VL1L2
output when
> using Python embedding. Additionally, the logic for deriving it
from GRIB
> and NetCDF input files is a bit fragile as well. There are a
handful of
> variables in MET for which special logic is defined... namely U-
wind,
> V-wind, wind speed, wind direction, precipitation, and specific
humidity.
> The MET libraries for reading gridded data contain logic for
determining
> what datasets qualify as being these, usually by their name. But
there's
> currently no way of overriding this logic. And in the case of
Python
> embedding, they're always set to false.
>
> We should enhance MET to provide a way of overriding, or for Python
> embedding, setting this logic. We already have a somewhat similar
GitHub
> issue for letting the user override other metadata, like the name of
the
> data, level string, units string, timing information, and grid
definition (
> https://github.com/NCAR/MET/issues/1020). This word would basically
be
> adding options for setting flags to specify these data types.
>
> Once MET knows which fields contain the U and V components of wind,
it
> processes them together to do 2 things...
> (1) It rotates U and V from being grid-relative to earth-relative,
but only
> if necessary. This is defined by a flag in the GRIB1/2 files. For
Python
> embedding, we'd need another flag to define "wind_is_grid_relative =
TRUE
> or FALSE".
> (2) After rotating grid to earth-relative, it processes the U and V
vectors
> to compute vector partial sums and statistics.
>
> Let me talk to the MET development team to find out how these
enhancements
> slot into our development timeline.
>
> Thanks,
> John
>
>
>
>
>
> On Wed, Nov 13, 2019 at 4:13 PM Satterfield, Dr. Elizabeth via RT <
> met_help at ucar.edu> wrote:
>
> >
> > Wed Nov 13 16:13:52 2019: Request 93123 was acted upon.
> > Transaction: Ticket created by
Elizabeth.Satterfield at nrlmry.navy.mil
> > Queue: met_help
> > Subject: computing vector wind RMS
> > Owner: Nobody
> > Requestors: Elizabeth.Satterfield at nrlmry.navy.mil
> > Status: new
> > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=93123 >
> >
> >
> > John,
> >
> >
> >
> > Can you advise the easiest way to compute vector wind RMS using
MET.
> E.g.
> > for each region, I want to calculate sqrt(U_error^2+ V_error^2).
> >
> >
> >
> > Thanks
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> >
> > Elizabeth Satterfield
> > Editor, Monthly Weather Review
> >
> > Chair, AMS Committee on Probability and Statistics
> >
> > Meteorologist, Code 7531
> > Marine Meteorology Division
> > U.S. Naval Research Laboratory
> > T 831.656.4056 F 831.656.4026
> > <http://www.nrl.navy.mil/> www.nrl.navy.mil
> >
> > __________________________________________________________________
> > FOR OFFICIAL USE ONLY - PRIVACY SENSITIVE
> > Any misuse or unauthorized disclosure may result in both civil and
> criminal
> > penalties.
> >
> >
> >
> >
> >
> >
> >
>
>
>
------------------------------------------------
More information about the Met_help
mailing list