[Met_help] [rt.rap.ucar.edu #59574] History for MET v2.0: PointStat not able to verify WDIR

Wed Dec 19 09:39:50 MST 2012

----------------------------------------------------------------
  Initial Request
----------------------------------------------------------------

To whom it may concern:

I am trying to use PointStat (MET version 2.0) to verify a variety of 
fields, one of which is wind direction (Grib code = 31). The User's Guide 
and the default PointStat config file indicate a table online (
http://www.nco.ncep.noaa.gov/pmb/docs/on388/table2.html) containing all 
the fields I should be able to verify. But, when I actually run PointStat, 
it exits with an error message saying that PointStat cannot be used to 
verify WDIR. Is this correct? Are there any known work-arounds?

Thanks for any help/advice.

Kiana

*~~*~~*~~*~~*~~*~~*~~*~~*~~*~~*~~*
Kiana Ross
MTS - Navigation and Geopositioning Department
Systems Engineering Division
The Aerospace Corporation

Email: Kiana.Ross at aero.org
Office: (310) 336-1141
Mailbox M4/944
*~~*~~*~~*~~*~~*~~*~~*~~*~~*~~*~~*

----------------------------------------------------------------
  Complete Ticket History
----------------------------------------------------------------

Subject: Re: [rt.rap.ucar.edu #59574] MET v2.0: PointStat not able to verify WDIR
From: John Halley Gotway
Time: Wed Dec 12 09:43:46 2012

Kiana,

I apologize for not getting back to you sooner.  I see that you're
asking about verifying wind direction using MET.  Wind direction is
trickier to verify than other scalar meteorological variables,
and the statistics that MET outputs are not necessarily appropriate
for evaluating wind direction.

That being said, let me point out a few things...

  - The most recent version of MET is 4.0, which was release on
6/25/2012.  There have been several bugfixes and improvement to MET
since version 2.0.  If possible, I'd strongly recommend using the
latest version.  If you choose to do so, be sure to also apply the
latest set of patches for it:
    http://www.dtcenter.org/met/users/support/known_issues/METv4.0/index.php

  - While the MET statistics tools, like point_stat and grid_stat, do
not evaluate wind direction directly, it can be done using a 2 step
process.  It's described in an online practical session as
part of the MET tutorial:
    http://www.dtcenter.org/met/users/support/OnLinePractical/OnLinePractical_112/point_stat/wind_dir.php

  - Basically, you use grid_stat or point_stat to create vector
partial sums lines (VL1L2) using one or more wind speed filtering
criteria.  Then you run a stat_analysis job to derive wind direction
errors from those lines.  The big drawback is that this wind direction
error is an aerial average, not point-by-point.  So it only works well
over smaller regions where the winds are largely uni-modal.

We realize that there is room for improvement when it comes to
evaluating wind direction in MET.  I've cc'ed Tressa Folwer on this
message.  She's a statistician and the MET project manager.  If you
have any suggestions for how we might improve wind direction
evaluation, we'd be happy to hear them.

Thanks,
John Halley Gotway
met_help at ucar.edu

On 12/10/2012 06:27 PM, Kiana L Ross via RT wrote:
>
> Mon Dec 10 18:27:45 2012: Request 59574 was acted upon.
> Transaction: Ticket created by Kiana.L.Ross at aero.org
>         Queue: met_help
>       Subject: MET v2.0: PointStat not able to verify WDIR
>         Owner: Nobody
>    Requestors: Kiana.L.Ross at aero.org
>        Status: new
>   Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=59574 >
>
>
> To whom it may concern:
>
> I am trying to use PointStat (MET version 2.0) to verify a variety
of
> fields, one of which is wind direction (Grib code = 31). The User's
Guide
> and the default PointStat config file indicate a table online (
> http://www.nco.ncep.noaa.gov/pmb/docs/on388/table2.html) containing
all
> the fields I should be able to verify. But, when I actually run
PointStat,
> it exits with an error message saying that PointStat cannot be used
to
> verify WDIR. Is this correct? Are there any known work-arounds?
>
> Thanks for any help/advice.
>
> Kiana
>
>
> *~~*~~*~~*~~*~~*~~*~~*~~*~~*~~*~~*
> Kiana Ross
> MTS - Navigation and Geopositioning Department
> Systems Engineering Division
> The Aerospace Corporation
>
> Email: Kiana.Ross at aero.org
> Office: (310) 336-1141
> Mailbox M4/944
> *~~*~~*~~*~~*~~*~~*~~*~~*~~*~~*~~*
>

------------------------------------------------
Subject: MET v2.0: PointStat not able to verify WDIR
From: Kiana L Ross
Time: Tue Dec 18 15:22:02 2012

John,

Thank you so much for such an informative response. I have one domain
that
is small enough so that the wind direction method you describe below
will
be worth trying. I would like to switch to MET version 4.0, but
unfortunately I don't have final say in the matter. I will pass along
that
version 4.0 is much more robust, however.

I have some additional questions regarding the vertical interpolation
that
PointStat does, but that is a different topic, so I will start a new
thread.

Thanks again,
Kiana

From:   "John Halley Gotway via RT" <met_help at ucar.edu>
To:     Kiana.L.Ross at aero.org,
Date:   12/12/2012 08:43 AM
Subject:        Re: [rt.rap.ucar.edu #59574] MET v2.0: PointStat not
able
to verify WDIR

Kiana,

I apologize for not getting back to you sooner.  I see that you're
asking
about verifying wind direction using MET.  Wind direction is trickier
to
verify than other scalar meteorological variables,
and the statistics that MET outputs are not necessarily appropriate
for
evaluating wind direction.

That being said, let me point out a few things...

  - The most recent version of MET is 4.0, which was release on
6/25/2012.
 There have been several bugfixes and improvement to MET since version
2.0.  If possible, I'd strongly recommend using the
latest version.  If you choose to do so, be sure to also apply the
latest
set of patches for it:

http://www.dtcenter.org/met/users/support/known_issues/METv4.0/index.php

  - While the MET statistics tools, like point_stat and grid_stat, do
not
evaluate wind direction directly, it can be done using a 2 step
process.
It's described in an online practical session as
part of the MET tutorial:

http://www.dtcenter.org/met/users/support/OnLinePractical/OnLinePractical_112/point_stat/wind_dir.php

  - Basically, you use grid_stat or point_stat to create vector
partial
sums lines (VL1L2) using one or more wind speed filtering criteria.
Then
you run a stat_analysis job to derive wind direction
errors from those lines.  The big drawback is that this wind direction
error is an aerial average, not point-by-point.  So it only works well
over smaller regions where the winds are largely uni-modal.

We realize that there is room for improvement when it comes to
evaluating
wind direction in MET.  I've cc'ed Tressa Folwer on this message.
She's a
statistician and the MET project manager.  If you
have any suggestions for how we might improve wind direction
evaluation,
we'd be happy to hear them.

Thanks,
John Halley Gotway
met_help at ucar.edu

On 12/10/2012 06:27 PM, Kiana L Ross via RT wrote:
>
> Mon Dec 10 18:27:45 2012: Request 59574 was acted upon.
> Transaction: Ticket created by Kiana.L.Ross at aero.org
>         Queue: met_help
>       Subject: MET v2.0: PointStat not able to verify WDIR
>         Owner: Nobody
>    Requestors: Kiana.L.Ross at aero.org
>        Status: new
>   Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=59574 >
>
>
> To whom it may concern:
>
> I am trying to use PointStat (MET version 2.0) to verify a variety
of
> fields, one of which is wind direction (Grib code = 31). The User's
Guide
> and the default PointStat config file indicate a table online (
> http://www.nco.ncep.noaa.gov/pmb/docs/on388/table2.html) containing
all
> the fields I should be able to verify. But, when I actually run
PointStat,
> it exits with an error message saying that PointStat cannot be used
to
> verify WDIR. Is this correct? Are there any known work-arounds?
>
> Thanks for any help/advice.
>
> Kiana
>
>
> *~~*~~*~~*~~*~~*~~*~~*~~*~~*~~*~~*
> Kiana Ross
> MTS - Navigation and Geopositioning Department
> Systems Engineering Division
> The Aerospace Corporation
>
> Email: Kiana.Ross at aero.org
> Office: (310) 336-1141
> Mailbox M4/944
> *~~*~~*~~*~~*~~*~~*~~*~~*~~*~~*~~*
>

------------------------------------------------