[Met_help] [rt.rap.ucar.edu #65367] History for MET-TC TK_ERR, MSLP_ERR?
John Halley Gotway via RT
met_help at ucar.edu
Thu Aug 27 12:21:43 MDT 2015
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Hi ,
I have a cursory knowledge of tc_pairs and tc_stats now, and am moving
on from the example in the tutorial. In that example, track error is
analyzed. This is done by simply specifying the TK_ERR column. Next I
would like to the same thing with MSLP error. Is there an equivalent
way to do this? I figured there would be a MSLP_ERR column, but
couldn't find it. . .
Dave
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #65367] MET-TC TK_ERR, MSLP_ERR?
From: John Halley Gotway
Time: Mon Feb 10 11:08:34 2014
Dave,
The .tcst files created by tc_pairs contain some "error" columns that
are computed by comparing the ADECK and BDECK track locations: TK_ERR,
ALTK_ERR, CRTK_ERR, X_ERR, and Y_ERR for track,
along-track, cross-track, east-west error, and north-south error.
Those types of errors are not trivial to compute, so tc_pairs does it
for you.
Other types of errors, like the difference of the mean sea level
pressures or differences in wind radii are trivial to compute. If we
had included all possible differences in the .tcst output line
that would have made an already long output line even longer.
Instead, tc_stat supports the use of the minus sign and/or "ABS".
For example, try the following...
-column AMSLP-BMSLP # For the MSLP error
-column ABS\(AMSLP-BMSLP\) # For the absolute MSLP error
-column ABS\(Y_ERR\) # For the absolute value of the north-
south track error
If you're running this on the command line, you'll need to escape the
'(' and ')' characters. If you use a config file, you shouldn't need
to.
So you can define arbitrary differences of the .tcst output columns
with/without absolute values. Hope that helps clarify.
Thanks,
John
On 02/07/2014 03:18 PM, David Ahijevych via RT wrote:
>
> Fri Feb 07 15:18:55 2014: Request 65367 was acted upon.
> Transaction: Ticket created by ahijevyc
> Queue: met_help
> Subject: MET-TC TK_ERR, MSLP_ERR?
> Owner: Nobody
> Requestors: ahijevyc at ucar.edu
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367 >
>
>
> Hi ,
> I have a cursory knowledge of tc_pairs and tc_stats now, and am
moving
> on from the example in the tutorial. In that example, track error
is
> analyzed. This is done by simply specifying the TK_ERR column.
Next I
> would like to the same thing with MSLP error. Is there an
equivalent
> way to do this? I figured there would be a MSLP_ERR column, but
> couldn't find it. . .
>
> Dave
>
>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #65367] MET-TC TK_ERR, MSLP_ERR?
From: David Ahijevych
Time: Mon Feb 10 13:23:26 2014
I think this is exactly what I was looking for - thanks for taking the
time to answer this!
On 2/10/14 11:08 AM, John Halley Gotway via RT wrote:
> Dave,
>
> The .tcst files created by tc_pairs contain some "error" columns
that are computed by comparing the ADECK and BDECK track locations:
TK_ERR, ALTK_ERR, CRTK_ERR, X_ERR, and Y_ERR for track,
> along-track, cross-track, east-west error, and north-south error.
Those types of errors are not trivial to compute, so tc_pairs does it
for you.
>
> Other types of errors, like the difference of the mean sea level
pressures or differences in wind radii are trivial to compute. If we
had included all possible differences in the .tcst output line
> that would have made an already long output line even longer.
Instead, tc_stat supports the use of the minus sign and/or "ABS".
>
> For example, try the following...
> -column AMSLP-BMSLP # For the MSLP error
> -column ABS\(AMSLP-BMSLP\) # For the absolute MSLP error
> -column ABS\(Y_ERR\) # For the absolute value of the
north-south track error
>
> If you're running this on the command line, you'll need to escape
the '(' and ')' characters. If you use a config file, you shouldn't
need to.
>
> So you can define arbitrary differences of the .tcst output columns
with/without absolute values. Hope that helps clarify.
>
> Thanks,
> John
>
> On 02/07/2014 03:18 PM, David Ahijevych via RT wrote:
>> Fri Feb 07 15:18:55 2014: Request 65367 was acted upon.
>> Transaction: Ticket created by ahijevyc
>> Queue: met_help
>> Subject: MET-TC TK_ERR, MSLP_ERR?
>> Owner: Nobody
>> Requestors: ahijevyc at ucar.edu
>> Status: new
>> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367 >
>>
>>
>> Hi ,
>> I have a cursory knowledge of tc_pairs and tc_stats now, and am
moving
>> on from the example in the tutorial. In that example, track error
is
>> analyzed. This is done by simply specifying the TK_ERR column.
Next I
>> would like to the same thing with MSLP error. Is there an
equivalent
>> way to do this? I figured there would be a MSLP_ERR column, but
>> couldn't find it. . .
>>
>> Dave
>>
>>
------------------------------------------------
Subject: MET-TC TK_ERR, MSLP_ERR?
From: David Ahijevych
Time: Wed Aug 05 11:14:42 2015
Hi everyone,
As far as I can tell, MET-TC software (which would be ideal for this)
doesn't calculate error statistics for radius of maximum wind (RMW).
It
does MSLP, VMAX, and distance to 34, 50, and 64kt winds in all 4
quadrants,
but no RMW. Is it possible to add this?
Thanks!
Dave
On Mon, Feb 10, 2014 at 1:23 PM, Dave Ahijevych <ahijevyc at ucar.edu>
wrote:
> I think this is exactly what I was looking for - thanks for taking
the
> time to answer this!
>
>
> On 2/10/14 11:08 AM, John Halley Gotway via RT wrote:
>
>> Dave,
>>
>> The .tcst files created by tc_pairs contain some "error" columns
that are
>> computed by comparing the ADECK and BDECK track locations: TK_ERR,
>> ALTK_ERR, CRTK_ERR, X_ERR, and Y_ERR for track,
>> along-track, cross-track, east-west error, and north-south error.
Those
>> types of errors are not trivial to compute, so tc_pairs does it for
you.
>>
>> Other types of errors, like the difference of the mean sea level
>> pressures or differences in wind radii are trivial to compute. If
we had
>> included all possible differences in the .tcst output line
>> that would have made an already long output line even longer.
Instead,
>> tc_stat supports the use of the minus sign and/or "ABS".
>>
>> For example, try the following...
>> -column AMSLP-BMSLP # For the MSLP error
>> -column ABS\(AMSLP-BMSLP\) # For the absolute MSLP error
>> -column ABS\(Y_ERR\) # For the absolute value of the
>> north-south track error
>>
>> If you're running this on the command line, you'll need to escape
the '('
>> and ')' characters. If you use a config file, you shouldn't need
to.
>>
>> So you can define arbitrary differences of the .tcst output columns
>> with/without absolute values. Hope that helps clarify.
>>
>> Thanks,
>> John
>>
>> On 02/07/2014 03:18 PM, David Ahijevych via RT wrote:
>>
>>> Fri Feb 07 15:18:55 2014: Request 65367 was acted upon.
>>> Transaction: Ticket created by ahijevyc
>>> Queue: met_help
>>> Subject: MET-TC TK_ERR, MSLP_ERR?
>>> Owner: Nobody
>>> Requestors: ahijevyc at ucar.edu
>>> Status: new
>>> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367
>>> >
>>>
>>>
>>>
>>> Hi ,
>>> I have a cursory knowledge of tc_pairs and tc_stats now, and am
moving
>>> on from the example in the tutorial. In that example, track error
is
>>> analyzed. This is done by simply specifying the TK_ERR column.
Next I
>>> would like to the same thing with MSLP error. Is there an
equivalent
>>> way to do this? I figured there would be a MSLP_ERR column, but
>>> couldn't find it. . .
>>>
>>> Dave
>>>
>>>
>>>
>
------------------------------------------------
Subject: MET-TC TK_ERR, MSLP_ERR?
From: John Halley Gotway
Time: Wed Aug 05 11:48:12 2015
Dave,
Good point. It looks like we stopped dumping out ADECK and BDECK
information after the 34, 50, and 64 kt wind radii. Here's the ATCF
file
format description:
http://www.nrlmry.navy.mil/atcf_web/docs/database/new/abrdeck.html
After the winds, there are several quantities that could be added:
RRP, MRD, GUSTS, EYE, SUBREGION, MAXSEAS, INITIALS, DIR, SPEED,
DETH,
SEAS, and then the seas radii (just like the wind info)
Listing values for ADECK and BDECK in the TCMPR output line would add
twice
as many columns to the output file.
Is it just radius of maximum winds you're after or would these other
quantities be useful as well?
I've CC'ed Tara Jensen (the MET project lead) and Kathryn Newman (MET-
TC
expert) to get their take on it.
Thanks,
John
On Wed, Aug 5, 2015 at 11:14 AM, David Ahijevych via RT
<met_help at ucar.edu>
wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367 >
>
> Hi everyone,
>
> As far as I can tell, MET-TC software (which would be ideal for
this)
> doesn't calculate error statistics for radius of maximum wind (RMW).
It
> does MSLP, VMAX, and distance to 34, 50, and 64kt winds in all 4
quadrants,
> but no RMW. Is it possible to add this?
> Thanks!
>
> Dave
>
>
> On Mon, Feb 10, 2014 at 1:23 PM, Dave Ahijevych <ahijevyc at ucar.edu>
wrote:
>
> > I think this is exactly what I was looking for - thanks for taking
the
> > time to answer this!
> >
> >
> > On 2/10/14 11:08 AM, John Halley Gotway via RT wrote:
> >
> >> Dave,
> >>
> >> The .tcst files created by tc_pairs contain some "error" columns
that
> are
> >> computed by comparing the ADECK and BDECK track locations:
TK_ERR,
> >> ALTK_ERR, CRTK_ERR, X_ERR, and Y_ERR for track,
> >> along-track, cross-track, east-west error, and north-south error.
Those
> >> types of errors are not trivial to compute, so tc_pairs does it
for you.
> >>
> >> Other types of errors, like the difference of the mean sea level
> >> pressures or differences in wind radii are trivial to compute.
If we
> had
> >> included all possible differences in the .tcst output line
> >> that would have made an already long output line even longer.
Instead,
> >> tc_stat supports the use of the minus sign and/or "ABS".
> >>
> >> For example, try the following...
> >> -column AMSLP-BMSLP # For the MSLP error
> >> -column ABS\(AMSLP-BMSLP\) # For the absolute MSLP error
> >> -column ABS\(Y_ERR\) # For the absolute value of the
> >> north-south track error
> >>
> >> If you're running this on the command line, you'll need to escape
the
> '('
> >> and ')' characters. If you use a config file, you shouldn't need
to.
> >>
> >> So you can define arbitrary differences of the .tcst output
columns
> >> with/without absolute values. Hope that helps clarify.
> >>
> >> Thanks,
> >> John
> >>
> >> On 02/07/2014 03:18 PM, David Ahijevych via RT wrote:
> >>
> >>> Fri Feb 07 15:18:55 2014: Request 65367 was acted upon.
> >>> Transaction: Ticket created by ahijevyc
> >>> Queue: met_help
> >>> Subject: MET-TC TK_ERR, MSLP_ERR?
> >>> Owner: Nobody
> >>> Requestors: ahijevyc at ucar.edu
> >>> Status: new
> >>> Ticket <URL:
> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367
> >>> >
> >>>
> >>>
> >>>
> >>> Hi ,
> >>> I have a cursory knowledge of tc_pairs and tc_stats now, and am
moving
> >>> on from the example in the tutorial. In that example, track
error is
> >>> analyzed. This is done by simply specifying the TK_ERR column.
Next I
> >>> would like to the same thing with MSLP error. Is there an
equivalent
> >>> way to do this? I figured there would be a MSLP_ERR column, but
> >>> couldn't find it. . .
> >>>
> >>> Dave
> >>>
> >>>
> >>>
> >
>
>
------------------------------------------------
Subject: MET-TC TK_ERR, MSLP_ERR?
From: David Ahijevych
Time: Wed Aug 05 13:45:30 2015
Hi John,
Radius of max winds is the main thing. It's much more useful than
anything
else.
RRP MRD GUSTS EYE DEPTH could be useful, but most models don't output
these
parameters.
I'd say MAXSEAS and SEAS are even less useful because the models I
look at
don't have wave height information to compare to.
And it seems pointless to compare fields like SUBREGION and INITIALS.
Dave
On Wed, Aug 5, 2015 at 11:48 AM, John Halley Gotway via RT <
met_help at ucar.edu> wrote:
> Dave,
>
> Good point. It looks like we stopped dumping out ADECK and BDECK
> information after the 34, 50, and 64 kt wind radii. Here's the ATCF
file
> format description:
>
http://www.nrlmry.navy.mil/atcf_web/docs/database/new/abrdeck.html
>
> After the winds, there are several quantities that could be added:
> RRP, MRD, GUSTS, EYE, SUBREGION, MAXSEAS, INITIALS, DIR, SPEED,
DETH,
> SEAS, and then the seas radii (just like the wind info)
>
> Listing values for ADECK and BDECK in the TCMPR output line would
add twice
> as many columns to the output file.
>
> Is it just radius of maximum winds you're after or would these other
> quantities be useful as well?
>
> I've CC'ed Tara Jensen (the MET project lead) and Kathryn Newman
(MET-TC
> expert) to get their take on it.
>
> Thanks,
> John
>
>
>
> On Wed, Aug 5, 2015 at 11:14 AM, David Ahijevych via RT
<met_help at ucar.edu
> >
> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367 >
> >
> > Hi everyone,
> >
> > As far as I can tell, MET-TC software (which would be ideal for
this)
> > doesn't calculate error statistics for radius of maximum wind
(RMW). It
> > does MSLP, VMAX, and distance to 34, 50, and 64kt winds in all 4
> quadrants,
> > but no RMW. Is it possible to add this?
> > Thanks!
> >
> > Dave
> >
> >
> > On Mon, Feb 10, 2014 at 1:23 PM, Dave Ahijevych
<ahijevyc at ucar.edu>
> wrote:
> >
> > > I think this is exactly what I was looking for - thanks for
taking the
> > > time to answer this!
> > >
> > >
> > > On 2/10/14 11:08 AM, John Halley Gotway via RT wrote:
> > >
> > >> Dave,
> > >>
> > >> The .tcst files created by tc_pairs contain some "error"
columns that
> > are
> > >> computed by comparing the ADECK and BDECK track locations:
TK_ERR,
> > >> ALTK_ERR, CRTK_ERR, X_ERR, and Y_ERR for track,
> > >> along-track, cross-track, east-west error, and north-south
error.
> Those
> > >> types of errors are not trivial to compute, so tc_pairs does it
for
> you.
> > >>
> > >> Other types of errors, like the difference of the mean sea
level
> > >> pressures or differences in wind radii are trivial to compute.
If we
> > had
> > >> included all possible differences in the .tcst output line
> > >> that would have made an already long output line even longer.
> Instead,
> > >> tc_stat supports the use of the minus sign and/or "ABS".
> > >>
> > >> For example, try the following...
> > >> -column AMSLP-BMSLP # For the MSLP error
> > >> -column ABS\(AMSLP-BMSLP\) # For the absolute MSLP error
> > >> -column ABS\(Y_ERR\) # For the absolute value of the
> > >> north-south track error
> > >>
> > >> If you're running this on the command line, you'll need to
escape the
> > '('
> > >> and ')' characters. If you use a config file, you shouldn't
need to.
> > >>
> > >> So you can define arbitrary differences of the .tcst output
columns
> > >> with/without absolute values. Hope that helps clarify.
> > >>
> > >> Thanks,
> > >> John
> > >>
> > >> On 02/07/2014 03:18 PM, David Ahijevych via RT wrote:
> > >>
> > >>> Fri Feb 07 15:18:55 2014: Request 65367 was acted upon.
> > >>> Transaction: Ticket created by ahijevyc
> > >>> Queue: met_help
> > >>> Subject: MET-TC TK_ERR, MSLP_ERR?
> > >>> Owner: Nobody
> > >>> Requestors: ahijevyc at ucar.edu
> > >>> Status: new
> > >>> Ticket <URL:
> > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367
> > >>> >
> > >>>
> > >>>
> > >>>
> > >>> Hi ,
> > >>> I have a cursory knowledge of tc_pairs and tc_stats now, and
am
> moving
> > >>> on from the example in the tutorial. In that example, track
error is
> > >>> analyzed. This is done by simply specifying the TK_ERR
column.
> Next I
> > >>> would like to the same thing with MSLP error. Is there an
equivalent
> > >>> way to do this? I figured there would be a MSLP_ERR column,
but
> > >>> couldn't find it. . .
> > >>>
> > >>> Dave
> > >>>
> > >>>
> > >>>
> > >
> >
> >
>
>
------------------------------------------------
Subject: MET-TC TK_ERR, MSLP_ERR?
From: David Ahijevych
Time: Wed Aug 05 13:49:47 2015
Oh, and DIR and SPEED could be useful. DIR error is a little tricky
to do
since it's a circular variable.
Dave
On Wed, Aug 5, 2015 at 1:45 PM, David Ahijevych <ahijevyc at ucar.edu>
wrote:
> Hi John,
>
> Radius of max winds is the main thing. It's much more useful than
> anything else.
> RRP MRD GUSTS EYE DEPTH could be useful, but most models don't
output
> these parameters.
>
> I'd say MAXSEAS and SEAS are even less useful because the models I
look at
> don't have wave height information to compare to.
>
> And it seems pointless to compare fields like SUBREGION and
INITIALS.
>
> Dave
>
>
> On Wed, Aug 5, 2015 at 11:48 AM, John Halley Gotway via RT <
> met_help at ucar.edu> wrote:
>
>> Dave,
>>
>> Good point. It looks like we stopped dumping out ADECK and BDECK
>> information after the 34, 50, and 64 kt wind radii. Here's the
ATCF file
>> format description:
>>
http://www.nrlmry.navy.mil/atcf_web/docs/database/new/abrdeck.html
>>
>> After the winds, there are several quantities that could be added:
>> RRP, MRD, GUSTS, EYE, SUBREGION, MAXSEAS, INITIALS, DIR, SPEED,
DETH,
>> SEAS, and then the seas radii (just like the wind info)
>>
>> Listing values for ADECK and BDECK in the TCMPR output line would
add
>> twice
>> as many columns to the output file.
>>
>> Is it just radius of maximum winds you're after or would these
other
>> quantities be useful as well?
>>
>> I've CC'ed Tara Jensen (the MET project lead) and Kathryn Newman
(MET-TC
>> expert) to get their take on it.
>>
>> Thanks,
>> John
>>
>>
>>
>> On Wed, Aug 5, 2015 at 11:14 AM, David Ahijevych via RT <
>> met_help at ucar.edu>
>> wrote:
>>
>> >
>> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367 >
>> >
>> > Hi everyone,
>> >
>> > As far as I can tell, MET-TC software (which would be ideal for
this)
>> > doesn't calculate error statistics for radius of maximum wind
(RMW). It
>> > does MSLP, VMAX, and distance to 34, 50, and 64kt winds in all 4
>> quadrants,
>> > but no RMW. Is it possible to add this?
>> > Thanks!
>> >
>> > Dave
>> >
>> >
>> > On Mon, Feb 10, 2014 at 1:23 PM, Dave Ahijevych
<ahijevyc at ucar.edu>
>> wrote:
>> >
>> > > I think this is exactly what I was looking for - thanks for
taking the
>> > > time to answer this!
>> > >
>> > >
>> > > On 2/10/14 11:08 AM, John Halley Gotway via RT wrote:
>> > >
>> > >> Dave,
>> > >>
>> > >> The .tcst files created by tc_pairs contain some "error"
columns that
>> > are
>> > >> computed by comparing the ADECK and BDECK track locations:
TK_ERR,
>> > >> ALTK_ERR, CRTK_ERR, X_ERR, and Y_ERR for track,
>> > >> along-track, cross-track, east-west error, and north-south
error.
>> Those
>> > >> types of errors are not trivial to compute, so tc_pairs does
it for
>> you.
>> > >>
>> > >> Other types of errors, like the difference of the mean sea
level
>> > >> pressures or differences in wind radii are trivial to compute.
If we
>> > had
>> > >> included all possible differences in the .tcst output line
>> > >> that would have made an already long output line even longer.
>> Instead,
>> > >> tc_stat supports the use of the minus sign and/or "ABS".
>> > >>
>> > >> For example, try the following...
>> > >> -column AMSLP-BMSLP # For the MSLP error
>> > >> -column ABS\(AMSLP-BMSLP\) # For the absolute MSLP error
>> > >> -column ABS\(Y_ERR\) # For the absolute value of
the
>> > >> north-south track error
>> > >>
>> > >> If you're running this on the command line, you'll need to
escape the
>> > '('
>> > >> and ')' characters. If you use a config file, you shouldn't
need to.
>> > >>
>> > >> So you can define arbitrary differences of the .tcst output
columns
>> > >> with/without absolute values. Hope that helps clarify.
>> > >>
>> > >> Thanks,
>> > >> John
>> > >>
>> > >> On 02/07/2014 03:18 PM, David Ahijevych via RT wrote:
>> > >>
>> > >>> Fri Feb 07 15:18:55 2014: Request 65367 was acted upon.
>> > >>> Transaction: Ticket created by ahijevyc
>> > >>> Queue: met_help
>> > >>> Subject: MET-TC TK_ERR, MSLP_ERR?
>> > >>> Owner: Nobody
>> > >>> Requestors: ahijevyc at ucar.edu
>> > >>> Status: new
>> > >>> Ticket <URL:
>> > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367
>> > >>> >
>> > >>>
>> > >>>
>> > >>>
>> > >>> Hi ,
>> > >>> I have a cursory knowledge of tc_pairs and tc_stats now, and
am
>> moving
>> > >>> on from the example in the tutorial. In that example, track
error
>> is
>> > >>> analyzed. This is done by simply specifying the TK_ERR
column.
>> Next I
>> > >>> would like to the same thing with MSLP error. Is there an
>> equivalent
>> > >>> way to do this? I figured there would be a MSLP_ERR column,
but
>> > >>> couldn't find it. . .
>> > >>>
>> > >>> Dave
>> > >>>
>> > >>>
>> > >>>
>> > >
>> >
>> >
>>
>>
>
------------------------------------------------
Subject: MET-TC TK_ERR, MSLP_ERR?
From: John Halley Gotway
Time: Wed Aug 05 14:42:41 2015
Dave,
Thanks for sending the details. I created a new development ticket to
track this request and included the notes you sent. We're working on
the
next release, 5.1, now and the question is whether we'll have
time/funding
to get these changes into 5.1 or if we'll have to wait until the next
release.
I'll talk to Tara about it at our development meeting tomorrow.
Thanks,
John
On Wed, Aug 5, 2015 at 1:49 PM, David Ahijevych via RT
<met_help at ucar.edu>
wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367 >
>
> Oh, and DIR and SPEED could be useful. DIR error is a little tricky
to do
> since it's a circular variable.
>
> Dave
>
>
> On Wed, Aug 5, 2015 at 1:45 PM, David Ahijevych <ahijevyc at ucar.edu>
wrote:
>
> > Hi John,
> >
> > Radius of max winds is the main thing. It's much more useful than
> > anything else.
> > RRP MRD GUSTS EYE DEPTH could be useful, but most models don't
output
> > these parameters.
> >
> > I'd say MAXSEAS and SEAS are even less useful because the models I
look
> at
> > don't have wave height information to compare to.
> >
> > And it seems pointless to compare fields like SUBREGION and
INITIALS.
> >
> > Dave
> >
> >
> > On Wed, Aug 5, 2015 at 11:48 AM, John Halley Gotway via RT <
> > met_help at ucar.edu> wrote:
> >
> >> Dave,
> >>
> >> Good point. It looks like we stopped dumping out ADECK and BDECK
> >> information after the 34, 50, and 64 kt wind radii. Here's the
ATCF
> file
> >> format description:
> >>
http://www.nrlmry.navy.mil/atcf_web/docs/database/new/abrdeck.html
> >>
> >> After the winds, there are several quantities that could be
added:
> >> RRP, MRD, GUSTS, EYE, SUBREGION, MAXSEAS, INITIALS, DIR,
SPEED, DETH,
> >> SEAS, and then the seas radii (just like the wind info)
> >>
> >> Listing values for ADECK and BDECK in the TCMPR output line would
add
> >> twice
> >> as many columns to the output file.
> >>
> >> Is it just radius of maximum winds you're after or would these
other
> >> quantities be useful as well?
> >>
> >> I've CC'ed Tara Jensen (the MET project lead) and Kathryn Newman
(MET-TC
> >> expert) to get their take on it.
> >>
> >> Thanks,
> >> John
> >>
> >>
> >>
> >> On Wed, Aug 5, 2015 at 11:14 AM, David Ahijevych via RT <
> >> met_help at ucar.edu>
> >> wrote:
> >>
> >> >
> >> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367 >
> >> >
> >> > Hi everyone,
> >> >
> >> > As far as I can tell, MET-TC software (which would be ideal for
this)
> >> > doesn't calculate error statistics for radius of maximum wind
(RMW).
> It
> >> > does MSLP, VMAX, and distance to 34, 50, and 64kt winds in all
4
> >> quadrants,
> >> > but no RMW. Is it possible to add this?
> >> > Thanks!
> >> >
> >> > Dave
> >> >
> >> >
> >> > On Mon, Feb 10, 2014 at 1:23 PM, Dave Ahijevych
<ahijevyc at ucar.edu>
> >> wrote:
> >> >
> >> > > I think this is exactly what I was looking for - thanks for
taking
> the
> >> > > time to answer this!
> >> > >
> >> > >
> >> > > On 2/10/14 11:08 AM, John Halley Gotway via RT wrote:
> >> > >
> >> > >> Dave,
> >> > >>
> >> > >> The .tcst files created by tc_pairs contain some "error"
columns
> that
> >> > are
> >> > >> computed by comparing the ADECK and BDECK track locations:
TK_ERR,
> >> > >> ALTK_ERR, CRTK_ERR, X_ERR, and Y_ERR for track,
> >> > >> along-track, cross-track, east-west error, and north-south
error.
> >> Those
> >> > >> types of errors are not trivial to compute, so tc_pairs does
it for
> >> you.
> >> > >>
> >> > >> Other types of errors, like the difference of the mean sea
level
> >> > >> pressures or differences in wind radii are trivial to
compute. If
> we
> >> > had
> >> > >> included all possible differences in the .tcst output line
> >> > >> that would have made an already long output line even
longer.
> >> Instead,
> >> > >> tc_stat supports the use of the minus sign and/or "ABS".
> >> > >>
> >> > >> For example, try the following...
> >> > >> -column AMSLP-BMSLP # For the MSLP error
> >> > >> -column ABS\(AMSLP-BMSLP\) # For the absolute MSLP
error
> >> > >> -column ABS\(Y_ERR\) # For the absolute value of
the
> >> > >> north-south track error
> >> > >>
> >> > >> If you're running this on the command line, you'll need to
escape
> the
> >> > '('
> >> > >> and ')' characters. If you use a config file, you shouldn't
need
> to.
> >> > >>
> >> > >> So you can define arbitrary differences of the .tcst output
columns
> >> > >> with/without absolute values. Hope that helps clarify.
> >> > >>
> >> > >> Thanks,
> >> > >> John
> >> > >>
> >> > >> On 02/07/2014 03:18 PM, David Ahijevych via RT wrote:
> >> > >>
> >> > >>> Fri Feb 07 15:18:55 2014: Request 65367 was acted upon.
> >> > >>> Transaction: Ticket created by ahijevyc
> >> > >>> Queue: met_help
> >> > >>> Subject: MET-TC TK_ERR, MSLP_ERR?
> >> > >>> Owner: Nobody
> >> > >>> Requestors: ahijevyc at ucar.edu
> >> > >>> Status: new
> >> > >>> Ticket <URL:
> >> > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367
> >> > >>> >
> >> > >>>
> >> > >>>
> >> > >>>
> >> > >>> Hi ,
> >> > >>> I have a cursory knowledge of tc_pairs and tc_stats now,
and am
> >> moving
> >> > >>> on from the example in the tutorial. In that example,
track error
> >> is
> >> > >>> analyzed. This is done by simply specifying the TK_ERR
column.
> >> Next I
> >> > >>> would like to the same thing with MSLP error. Is there an
> >> equivalent
> >> > >>> way to do this? I figured there would be a MSLP_ERR
column, but
> >> > >>> couldn't find it. . .
> >> > >>>
> >> > >>> Dave
> >> > >>>
> >> > >>>
> >> > >>>
> >> > >
> >> >
> >> >
> >>
> >>
> >
>
>
------------------------------------------------
Subject: MET-TC TK_ERR, MSLP_ERR?
From: John Halley Gotway
Time: Mon Aug 17 12:15:10 2015
Dave,
We have this defined as a development issue and will try to get it
into the
next release. But it all depends on whether we get approval to use a
small
pot of HWRF verification funding for this work.
Thanks,
John
On Wed, Aug 5, 2015 at 2:42 PM, John Halley Gotway <johnhg at ucar.edu>
wrote:
> Dave,
>
> Thanks for sending the details. I created a new development ticket
to
> track this request and included the notes you sent. We're working
on the
> next release, 5.1, now and the question is whether we'll have
time/funding
> to get these changes into 5.1 or if we'll have to wait until the
next
> release.
>
> I'll talk to Tara about it at our development meeting tomorrow.
>
> Thanks,
> John
>
> On Wed, Aug 5, 2015 at 1:49 PM, David Ahijevych via RT
<met_help at ucar.edu>
> wrote:
>
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367 >
>>
>> Oh, and DIR and SPEED could be useful. DIR error is a little
tricky to do
>> since it's a circular variable.
>>
>> Dave
>>
>>
>> On Wed, Aug 5, 2015 at 1:45 PM, David Ahijevych <ahijevyc at ucar.edu>
>> wrote:
>>
>> > Hi John,
>> >
>> > Radius of max winds is the main thing. It's much more useful
than
>> > anything else.
>> > RRP MRD GUSTS EYE DEPTH could be useful, but most models don't
output
>> > these parameters.
>> >
>> > I'd say MAXSEAS and SEAS are even less useful because the models
I look
>> at
>> > don't have wave height information to compare to.
>> >
>> > And it seems pointless to compare fields like SUBREGION and
INITIALS.
>> >
>> > Dave
>> >
>> >
>> > On Wed, Aug 5, 2015 at 11:48 AM, John Halley Gotway via RT <
>> > met_help at ucar.edu> wrote:
>> >
>> >> Dave,
>> >>
>> >> Good point. It looks like we stopped dumping out ADECK and
BDECK
>> >> information after the 34, 50, and 64 kt wind radii. Here's the
ATCF
>> file
>> >> format description:
>> >>
http://www.nrlmry.navy.mil/atcf_web/docs/database/new/abrdeck.html
>> >>
>> >> After the winds, there are several quantities that could be
added:
>> >> RRP, MRD, GUSTS, EYE, SUBREGION, MAXSEAS, INITIALS, DIR,
SPEED,
>> DETH,
>> >> SEAS, and then the seas radii (just like the wind info)
>> >>
>> >> Listing values for ADECK and BDECK in the TCMPR output line
would add
>> >> twice
>> >> as many columns to the output file.
>> >>
>> >> Is it just radius of maximum winds you're after or would these
other
>> >> quantities be useful as well?
>> >>
>> >> I've CC'ed Tara Jensen (the MET project lead) and Kathryn Newman
>> (MET-TC
>> >> expert) to get their take on it.
>> >>
>> >> Thanks,
>> >> John
>> >>
>> >>
>> >>
>> >> On Wed, Aug 5, 2015 at 11:14 AM, David Ahijevych via RT <
>> >> met_help at ucar.edu>
>> >> wrote:
>> >>
>> >> >
>> >> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367
>
>> >> >
>> >> > Hi everyone,
>> >> >
>> >> > As far as I can tell, MET-TC software (which would be ideal
for this)
>> >> > doesn't calculate error statistics for radius of maximum wind
(RMW).
>> It
>> >> > does MSLP, VMAX, and distance to 34, 50, and 64kt winds in all
4
>> >> quadrants,
>> >> > but no RMW. Is it possible to add this?
>> >> > Thanks!
>> >> >
>> >> > Dave
>> >> >
>> >> >
>> >> > On Mon, Feb 10, 2014 at 1:23 PM, Dave Ahijevych
<ahijevyc at ucar.edu>
>> >> wrote:
>> >> >
>> >> > > I think this is exactly what I was looking for - thanks for
taking
>> the
>> >> > > time to answer this!
>> >> > >
>> >> > >
>> >> > > On 2/10/14 11:08 AM, John Halley Gotway via RT wrote:
>> >> > >
>> >> > >> Dave,
>> >> > >>
>> >> > >> The .tcst files created by tc_pairs contain some "error"
columns
>> that
>> >> > are
>> >> > >> computed by comparing the ADECK and BDECK track locations:
TK_ERR,
>> >> > >> ALTK_ERR, CRTK_ERR, X_ERR, and Y_ERR for track,
>> >> > >> along-track, cross-track, east-west error, and north-south
error.
>> >> Those
>> >> > >> types of errors are not trivial to compute, so tc_pairs
does it
>> for
>> >> you.
>> >> > >>
>> >> > >> Other types of errors, like the difference of the mean sea
level
>> >> > >> pressures or differences in wind radii are trivial to
compute.
>> If we
>> >> > had
>> >> > >> included all possible differences in the .tcst output line
>> >> > >> that would have made an already long output line even
longer.
>> >> Instead,
>> >> > >> tc_stat supports the use of the minus sign and/or "ABS".
>> >> > >>
>> >> > >> For example, try the following...
>> >> > >> -column AMSLP-BMSLP # For the MSLP error
>> >> > >> -column ABS\(AMSLP-BMSLP\) # For the absolute MSLP
error
>> >> > >> -column ABS\(Y_ERR\) # For the absolute value of
the
>> >> > >> north-south track error
>> >> > >>
>> >> > >> If you're running this on the command line, you'll need to
escape
>> the
>> >> > '('
>> >> > >> and ')' characters. If you use a config file, you
shouldn't need
>> to.
>> >> > >>
>> >> > >> So you can define arbitrary differences of the .tcst output
>> columns
>> >> > >> with/without absolute values. Hope that helps clarify.
>> >> > >>
>> >> > >> Thanks,
>> >> > >> John
>> >> > >>
>> >> > >> On 02/07/2014 03:18 PM, David Ahijevych via RT wrote:
>> >> > >>
>> >> > >>> Fri Feb 07 15:18:55 2014: Request 65367 was acted upon.
>> >> > >>> Transaction: Ticket created by ahijevyc
>> >> > >>> Queue: met_help
>> >> > >>> Subject: MET-TC TK_ERR, MSLP_ERR?
>> >> > >>> Owner: Nobody
>> >> > >>> Requestors: ahijevyc at ucar.edu
>> >> > >>> Status: new
>> >> > >>> Ticket <URL:
>> >> > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367
>> >> > >>> >
>> >> > >>>
>> >> > >>>
>> >> > >>>
>> >> > >>> Hi ,
>> >> > >>> I have a cursory knowledge of tc_pairs and tc_stats now,
and am
>> >> moving
>> >> > >>> on from the example in the tutorial. In that example,
track
>> error
>> >> is
>> >> > >>> analyzed. This is done by simply specifying the TK_ERR
column.
>> >> Next I
>> >> > >>> would like to the same thing with MSLP error. Is there an
>> >> equivalent
>> >> > >>> way to do this? I figured there would be a MSLP_ERR
column, but
>> >> > >>> couldn't find it. . .
>> >> > >>>
>> >> > >>> Dave
>> >> > >>>
>> >> > >>>
>> >> > >>>
>> >> > >
>> >> >
>> >> >
>> >>
>> >>
>> >
>>
>>
>
------------------------------------------------
Subject: MET-TC TK_ERR, MSLP_ERR?
From: David Ahijevych
Time: Mon Aug 17 15:55:11 2015
Hi John,
Thanks for letting me know the status of getting RMW into MET.
I may have to come up with some hack-y way to compare RMW in the
meantime.
Chris D. and I are working on a paper about hurricane simulations in
MPAS
where we might analyze RMW. And as part of this new storm surge
project
I'm working on, I'm reading that the overall area of a storm is more
closely tied to storm surge than max wind!
Dave
On Mon, Aug 17, 2015 at 12:15 PM, John Halley Gotway via RT <
met_help at ucar.edu> wrote:
> Dave,
>
> We have this defined as a development issue and will try to get it
into the
> next release. But it all depends on whether we get approval to use
a small
> pot of HWRF verification funding for this work.
>
> Thanks,
> John
>
> On Wed, Aug 5, 2015 at 2:42 PM, John Halley Gotway <johnhg at ucar.edu>
> wrote:
>
> > Dave,
> >
> > Thanks for sending the details. I created a new development
ticket to
> > track this request and included the notes you sent. We're working
on the
> > next release, 5.1, now and the question is whether we'll have
> time/funding
> > to get these changes into 5.1 or if we'll have to wait until the
next
> > release.
> >
> > I'll talk to Tara about it at our development meeting tomorrow.
> >
> > Thanks,
> > John
> >
> > On Wed, Aug 5, 2015 at 1:49 PM, David Ahijevych via RT <
> met_help at ucar.edu>
> > wrote:
> >
> >>
> >> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367 >
> >>
> >> Oh, and DIR and SPEED could be useful. DIR error is a little
tricky to
> do
> >> since it's a circular variable.
> >>
> >> Dave
> >>
> >>
> >> On Wed, Aug 5, 2015 at 1:45 PM, David Ahijevych
<ahijevyc at ucar.edu>
> >> wrote:
> >>
> >> > Hi John,
> >> >
> >> > Radius of max winds is the main thing. It's much more useful
than
> >> > anything else.
> >> > RRP MRD GUSTS EYE DEPTH could be useful, but most models don't
output
> >> > these parameters.
> >> >
> >> > I'd say MAXSEAS and SEAS are even less useful because the
models I
> look
> >> at
> >> > don't have wave height information to compare to.
> >> >
> >> > And it seems pointless to compare fields like SUBREGION and
INITIALS.
> >> >
> >> > Dave
> >> >
> >> >
> >> > On Wed, Aug 5, 2015 at 11:48 AM, John Halley Gotway via RT <
> >> > met_help at ucar.edu> wrote:
> >> >
> >> >> Dave,
> >> >>
> >> >> Good point. It looks like we stopped dumping out ADECK and
BDECK
> >> >> information after the 34, 50, and 64 kt wind radii. Here's
the ATCF
> >> file
> >> >> format description:
> >> >>
> http://www.nrlmry.navy.mil/atcf_web/docs/database/new/abrdeck.html
> >> >>
> >> >> After the winds, there are several quantities that could be
added:
> >> >> RRP, MRD, GUSTS, EYE, SUBREGION, MAXSEAS, INITIALS, DIR,
SPEED,
> >> DETH,
> >> >> SEAS, and then the seas radii (just like the wind info)
> >> >>
> >> >> Listing values for ADECK and BDECK in the TCMPR output line
would add
> >> >> twice
> >> >> as many columns to the output file.
> >> >>
> >> >> Is it just radius of maximum winds you're after or would these
other
> >> >> quantities be useful as well?
> >> >>
> >> >> I've CC'ed Tara Jensen (the MET project lead) and Kathryn
Newman
> >> (MET-TC
> >> >> expert) to get their take on it.
> >> >>
> >> >> Thanks,
> >> >> John
> >> >>
> >> >>
> >> >>
> >> >> On Wed, Aug 5, 2015 at 11:14 AM, David Ahijevych via RT <
> >> >> met_help at ucar.edu>
> >> >> wrote:
> >> >>
> >> >> >
> >> >> > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367 >
> >> >> >
> >> >> > Hi everyone,
> >> >> >
> >> >> > As far as I can tell, MET-TC software (which would be ideal
for
> this)
> >> >> > doesn't calculate error statistics for radius of maximum
wind
> (RMW).
> >> It
> >> >> > does MSLP, VMAX, and distance to 34, 50, and 64kt winds in
all 4
> >> >> quadrants,
> >> >> > but no RMW. Is it possible to add this?
> >> >> > Thanks!
> >> >> >
> >> >> > Dave
> >> >> >
> >> >> >
> >> >> > On Mon, Feb 10, 2014 at 1:23 PM, Dave Ahijevych
<ahijevyc at ucar.edu
> >
> >> >> wrote:
> >> >> >
> >> >> > > I think this is exactly what I was looking for - thanks
for
> taking
> >> the
> >> >> > > time to answer this!
> >> >> > >
> >> >> > >
> >> >> > > On 2/10/14 11:08 AM, John Halley Gotway via RT wrote:
> >> >> > >
> >> >> > >> Dave,
> >> >> > >>
> >> >> > >> The .tcst files created by tc_pairs contain some "error"
columns
> >> that
> >> >> > are
> >> >> > >> computed by comparing the ADECK and BDECK track
locations:
> TK_ERR,
> >> >> > >> ALTK_ERR, CRTK_ERR, X_ERR, and Y_ERR for track,
> >> >> > >> along-track, cross-track, east-west error, and north-
south
> error.
> >> >> Those
> >> >> > >> types of errors are not trivial to compute, so tc_pairs
does it
> >> for
> >> >> you.
> >> >> > >>
> >> >> > >> Other types of errors, like the difference of the mean
sea level
> >> >> > >> pressures or differences in wind radii are trivial to
compute.
> >> If we
> >> >> > had
> >> >> > >> included all possible differences in the .tcst output
line
> >> >> > >> that would have made an already long output line even
longer.
> >> >> Instead,
> >> >> > >> tc_stat supports the use of the minus sign and/or "ABS".
> >> >> > >>
> >> >> > >> For example, try the following...
> >> >> > >> -column AMSLP-BMSLP # For the MSLP error
> >> >> > >> -column ABS\(AMSLP-BMSLP\) # For the absolute MSLP
error
> >> >> > >> -column ABS\(Y_ERR\) # For the absolute value
of the
> >> >> > >> north-south track error
> >> >> > >>
> >> >> > >> If you're running this on the command line, you'll need
to
> escape
> >> the
> >> >> > '('
> >> >> > >> and ')' characters. If you use a config file, you
shouldn't
> need
> >> to.
> >> >> > >>
> >> >> > >> So you can define arbitrary differences of the .tcst
output
> >> columns
> >> >> > >> with/without absolute values. Hope that helps clarify.
> >> >> > >>
> >> >> > >> Thanks,
> >> >> > >> John
> >> >> > >>
> >> >> > >> On 02/07/2014 03:18 PM, David Ahijevych via RT wrote:
> >> >> > >>
> >> >> > >>> Fri Feb 07 15:18:55 2014: Request 65367 was acted upon.
> >> >> > >>> Transaction: Ticket created by ahijevyc
> >> >> > >>> Queue: met_help
> >> >> > >>> Subject: MET-TC TK_ERR, MSLP_ERR?
> >> >> > >>> Owner: Nobody
> >> >> > >>> Requestors: ahijevyc at ucar.edu
> >> >> > >>> Status: new
> >> >> > >>> Ticket <URL:
> >> >> > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367
> >> >> > >>> >
> >> >> > >>>
> >> >> > >>>
> >> >> > >>>
> >> >> > >>> Hi ,
> >> >> > >>> I have a cursory knowledge of tc_pairs and tc_stats now,
and am
> >> >> moving
> >> >> > >>> on from the example in the tutorial. In that example,
track
> >> error
> >> >> is
> >> >> > >>> analyzed. This is done by simply specifying the TK_ERR
column.
> >> >> Next I
> >> >> > >>> would like to the same thing with MSLP error. Is there
an
> >> >> equivalent
> >> >> > >>> way to do this? I figured there would be a MSLP_ERR
column,
> but
> >> >> > >>> couldn't find it. . .
> >> >> > >>>
> >> >> > >>> Dave
> >> >> > >>>
> >> >> > >>>
> >> >> > >>>
> >> >> > >
> >> >> >
> >> >> >
> >> >>
> >> >>
> >> >
> >>
> >>
> >
>
>
------------------------------------------------
Subject: MET-TC TK_ERR, MSLP_ERR?
From: John Halley Gotway
Time: Tue Aug 25 16:16:22 2015
Dave,
I worked today on adding these columns to the output of tc_pairs for
the
version 5.1 release. Following the naming conventions listed in the
ATCF
documentation (
http://www.nrlmry.navy.mil/atcf_web/docs/database/new/abrdeck.html), I
added the following columns 16 columns:
ARADP BRADP ARRP BRRP AMRD BMRD AGUSTS BGUSTS AEYE BEYE ADIR BDIR
ASPEED
BSPEED ADEPTH BDEPTH
They are for the ADECK and BDECK values of RADP, RRP, MRD, GUSTS, EYE,
DIR,
SPEED, and DEPTH.
I see that in the test data we run, some of these values do show up in
the
BDECK output. However, they're all NA for the ADECK. Can you confirm
that
these values actually do show up in the model tracker output you have?
While the documentation lists these as integers, I'm writing them as
floats
to handle the computation of "consensus" forecasts. For example, a
consensus eyewall diameter of three diameters 10, 30, 25 would be
21.667.
Did I miss anything? Or do you have any other requests or
suggestions?
Thanks,
John
On Mon, Aug 17, 2015 at 3:55 PM, David Ahijevych via RT
<met_help at ucar.edu>
wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367 >
>
> Hi John,
> Thanks for letting me know the status of getting RMW into MET.
> I may have to come up with some hack-y way to compare RMW in the
meantime.
> Chris D. and I are working on a paper about hurricane simulations in
MPAS
> where we might analyze RMW. And as part of this new storm surge
project
> I'm working on, I'm reading that the overall area of a storm is more
> closely tied to storm surge than max wind!
>
> Dave
>
>
> On Mon, Aug 17, 2015 at 12:15 PM, John Halley Gotway via RT <
> met_help at ucar.edu> wrote:
>
> > Dave,
> >
> > We have this defined as a development issue and will try to get it
into
> the
> > next release. But it all depends on whether we get approval to
use a
> small
> > pot of HWRF verification funding for this work.
> >
> > Thanks,
> > John
> >
> > On Wed, Aug 5, 2015 at 2:42 PM, John Halley Gotway
<johnhg at ucar.edu>
> > wrote:
> >
> > > Dave,
> > >
> > > Thanks for sending the details. I created a new development
ticket to
> > > track this request and included the notes you sent. We're
working on
> the
> > > next release, 5.1, now and the question is whether we'll have
> > time/funding
> > > to get these changes into 5.1 or if we'll have to wait until the
next
> > > release.
> > >
> > > I'll talk to Tara about it at our development meeting tomorrow.
> > >
> > > Thanks,
> > > John
> > >
> > > On Wed, Aug 5, 2015 at 1:49 PM, David Ahijevych via RT <
> > met_help at ucar.edu>
> > > wrote:
> > >
> > >>
> > >> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367 >
> > >>
> > >> Oh, and DIR and SPEED could be useful. DIR error is a little
tricky
> to
> > do
> > >> since it's a circular variable.
> > >>
> > >> Dave
> > >>
> > >>
> > >> On Wed, Aug 5, 2015 at 1:45 PM, David Ahijevych
<ahijevyc at ucar.edu>
> > >> wrote:
> > >>
> > >> > Hi John,
> > >> >
> > >> > Radius of max winds is the main thing. It's much more useful
than
> > >> > anything else.
> > >> > RRP MRD GUSTS EYE DEPTH could be useful, but most models
don't
> output
> > >> > these parameters.
> > >> >
> > >> > I'd say MAXSEAS and SEAS are even less useful because the
models I
> > look
> > >> at
> > >> > don't have wave height information to compare to.
> > >> >
> > >> > And it seems pointless to compare fields like SUBREGION and
> INITIALS.
> > >> >
> > >> > Dave
> > >> >
> > >> >
> > >> > On Wed, Aug 5, 2015 at 11:48 AM, John Halley Gotway via RT <
> > >> > met_help at ucar.edu> wrote:
> > >> >
> > >> >> Dave,
> > >> >>
> > >> >> Good point. It looks like we stopped dumping out ADECK and
BDECK
> > >> >> information after the 34, 50, and 64 kt wind radii. Here's
the
> ATCF
> > >> file
> > >> >> format description:
> > >> >>
> > http://www.nrlmry.navy.mil/atcf_web/docs/database/new/abrdeck.html
> > >> >>
> > >> >> After the winds, there are several quantities that could be
added:
> > >> >> RRP, MRD, GUSTS, EYE, SUBREGION, MAXSEAS, INITIALS, DIR,
SPEED,
> > >> DETH,
> > >> >> SEAS, and then the seas radii (just like the wind info)
> > >> >>
> > >> >> Listing values for ADECK and BDECK in the TCMPR output line
would
> add
> > >> >> twice
> > >> >> as many columns to the output file.
> > >> >>
> > >> >> Is it just radius of maximum winds you're after or would
these
> other
> > >> >> quantities be useful as well?
> > >> >>
> > >> >> I've CC'ed Tara Jensen (the MET project lead) and Kathryn
Newman
> > >> (MET-TC
> > >> >> expert) to get their take on it.
> > >> >>
> > >> >> Thanks,
> > >> >> John
> > >> >>
> > >> >>
> > >> >>
> > >> >> On Wed, Aug 5, 2015 at 11:14 AM, David Ahijevych via RT <
> > >> >> met_help at ucar.edu>
> > >> >> wrote:
> > >> >>
> > >> >> >
> > >> >> > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367 >
> > >> >> >
> > >> >> > Hi everyone,
> > >> >> >
> > >> >> > As far as I can tell, MET-TC software (which would be
ideal for
> > this)
> > >> >> > doesn't calculate error statistics for radius of maximum
wind
> > (RMW).
> > >> It
> > >> >> > does MSLP, VMAX, and distance to 34, 50, and 64kt winds in
all 4
> > >> >> quadrants,
> > >> >> > but no RMW. Is it possible to add this?
> > >> >> > Thanks!
> > >> >> >
> > >> >> > Dave
> > >> >> >
> > >> >> >
> > >> >> > On Mon, Feb 10, 2014 at 1:23 PM, Dave Ahijevych <
> ahijevyc at ucar.edu
> > >
> > >> >> wrote:
> > >> >> >
> > >> >> > > I think this is exactly what I was looking for - thanks
for
> > taking
> > >> the
> > >> >> > > time to answer this!
> > >> >> > >
> > >> >> > >
> > >> >> > > On 2/10/14 11:08 AM, John Halley Gotway via RT wrote:
> > >> >> > >
> > >> >> > >> Dave,
> > >> >> > >>
> > >> >> > >> The .tcst files created by tc_pairs contain some
"error"
> columns
> > >> that
> > >> >> > are
> > >> >> > >> computed by comparing the ADECK and BDECK track
locations:
> > TK_ERR,
> > >> >> > >> ALTK_ERR, CRTK_ERR, X_ERR, and Y_ERR for track,
> > >> >> > >> along-track, cross-track, east-west error, and north-
south
> > error.
> > >> >> Those
> > >> >> > >> types of errors are not trivial to compute, so tc_pairs
does
> it
> > >> for
> > >> >> you.
> > >> >> > >>
> > >> >> > >> Other types of errors, like the difference of the mean
sea
> level
> > >> >> > >> pressures or differences in wind radii are trivial to
compute.
> > >> If we
> > >> >> > had
> > >> >> > >> included all possible differences in the .tcst output
line
> > >> >> > >> that would have made an already long output line even
longer.
> > >> >> Instead,
> > >> >> > >> tc_stat supports the use of the minus sign and/or
"ABS".
> > >> >> > >>
> > >> >> > >> For example, try the following...
> > >> >> > >> -column AMSLP-BMSLP # For the MSLP error
> > >> >> > >> -column ABS\(AMSLP-BMSLP\) # For the absolute MSLP
error
> > >> >> > >> -column ABS\(Y_ERR\) # For the absolute
value of
> the
> > >> >> > >> north-south track error
> > >> >> > >>
> > >> >> > >> If you're running this on the command line, you'll need
to
> > escape
> > >> the
> > >> >> > '('
> > >> >> > >> and ')' characters. If you use a config file, you
shouldn't
> > need
> > >> to.
> > >> >> > >>
> > >> >> > >> So you can define arbitrary differences of the .tcst
output
> > >> columns
> > >> >> > >> with/without absolute values. Hope that helps clarify.
> > >> >> > >>
> > >> >> > >> Thanks,
> > >> >> > >> John
> > >> >> > >>
> > >> >> > >> On 02/07/2014 03:18 PM, David Ahijevych via RT wrote:
> > >> >> > >>
> > >> >> > >>> Fri Feb 07 15:18:55 2014: Request 65367 was acted
upon.
> > >> >> > >>> Transaction: Ticket created by ahijevyc
> > >> >> > >>> Queue: met_help
> > >> >> > >>> Subject: MET-TC TK_ERR, MSLP_ERR?
> > >> >> > >>> Owner: Nobody
> > >> >> > >>> Requestors: ahijevyc at ucar.edu
> > >> >> > >>> Status: new
> > >> >> > >>> Ticket <URL:
> > >> >> > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367
> > >> >> > >>> >
> > >> >> > >>>
> > >> >> > >>>
> > >> >> > >>>
> > >> >> > >>> Hi ,
> > >> >> > >>> I have a cursory knowledge of tc_pairs and tc_stats
now, and
> am
> > >> >> moving
> > >> >> > >>> on from the example in the tutorial. In that example,
track
> > >> error
> > >> >> is
> > >> >> > >>> analyzed. This is done by simply specifying the
TK_ERR
> column.
> > >> >> Next I
> > >> >> > >>> would like to the same thing with MSLP error. Is
there an
> > >> >> equivalent
> > >> >> > >>> way to do this? I figured there would be a MSLP_ERR
column,
> > but
> > >> >> > >>> couldn't find it. . .
> > >> >> > >>>
> > >> >> > >>> Dave
> > >> >> > >>>
> > >> >> > >>>
> > >> >> > >>>
> > >> >> > >
> > >> >> >
> > >> >> >
> > >> >>
> > >> >>
> > >> >
> > >>
> > >>
> > >
> >
> >
>
>
------------------------------------------------
Subject: MET-TC TK_ERR, MSLP_ERR?
From: David Ahijevych
Time: Wed Aug 26 15:07:43 2015
Hi John,
Thanks!
Here's a line from my tracker output fort.66: It has RADP, RRP, MRD,
DIR,
and SPEED of 1008, 442, 100, 336, and 34 respectively, but no GUSTS
EYE or
DEPTH.
EP, 97E, 2015071200_F000_129N_1236W_97E, 2015071200, 03, MPAS, 000,
132N,
1230W, 19, 1004, XX, 34, NEQ, 0000, 0000, 0000, 0000, 1008, 442,
100,
-999, -9999, -9999, N, 336, 34, 124, 162, 91, 122
Maybe a different tracker can output GUSTS EYE or DEPTH, but not the
GFDL
vortex tracker. You could retain them as placeholders, but since GFDL
vortex tracker doesn't have them, I don't have any need for those 3
values.
Everything else sounds good. Can't think of any other suggestions.
Dave
On Tue, Aug 25, 2015 at 4:16 PM, John Halley Gotway via RT <
met_help at ucar.edu> wrote:
> Dave,
>
> I worked today on adding these columns to the output of tc_pairs for
the
> version 5.1 release. Following the naming conventions listed in the
ATCF
> documentation (
> http://www.nrlmry.navy.mil/atcf_web/docs/database/new/abrdeck.html),
I
> added the following columns 16 columns:
>
> ARADP BRADP ARRP BRRP AMRD BMRD AGUSTS BGUSTS AEYE BEYE ADIR BDIR
ASPEED
> BSPEED ADEPTH BDEPTH
>
> They are for the ADECK and BDECK values of RADP, RRP, MRD, GUSTS,
EYE, DIR,
> SPEED, and DEPTH.
>
> I see that in the test data we run, some of these values do show up
in the
> BDECK output. However, they're all NA for the ADECK. Can you
confirm that
> these values actually do show up in the model tracker output you
have?
>
> While the documentation lists these as integers, I'm writing them as
floats
> to handle the computation of "consensus" forecasts. For example, a
> consensus eyewall diameter of three diameters 10, 30, 25 would be
21.667.
>
> Did I miss anything? Or do you have any other requests or
suggestions?
>
> Thanks,
> John
>
> On Mon, Aug 17, 2015 at 3:55 PM, David Ahijevych via RT
<met_help at ucar.edu
> >
> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367 >
> >
> > Hi John,
> > Thanks for letting me know the status of getting RMW into MET.
> > I may have to come up with some hack-y way to compare RMW in the
> meantime.
> > Chris D. and I are working on a paper about hurricane simulations
in MPAS
> > where we might analyze RMW. And as part of this new storm surge
project
> > I'm working on, I'm reading that the overall area of a storm is
more
> > closely tied to storm surge than max wind!
> >
> > Dave
> >
> >
> > On Mon, Aug 17, 2015 at 12:15 PM, John Halley Gotway via RT <
> > met_help at ucar.edu> wrote:
> >
> > > Dave,
> > >
> > > We have this defined as a development issue and will try to get
it into
> > the
> > > next release. But it all depends on whether we get approval to
use a
> > small
> > > pot of HWRF verification funding for this work.
> > >
> > > Thanks,
> > > John
> > >
> > > On Wed, Aug 5, 2015 at 2:42 PM, John Halley Gotway
<johnhg at ucar.edu>
> > > wrote:
> > >
> > > > Dave,
> > > >
> > > > Thanks for sending the details. I created a new development
ticket
> to
> > > > track this request and included the notes you sent. We're
working on
> > the
> > > > next release, 5.1, now and the question is whether we'll have
> > > time/funding
> > > > to get these changes into 5.1 or if we'll have to wait until
the next
> > > > release.
> > > >
> > > > I'll talk to Tara about it at our development meeting
tomorrow.
> > > >
> > > > Thanks,
> > > > John
> > > >
> > > > On Wed, Aug 5, 2015 at 1:49 PM, David Ahijevych via RT <
> > > met_help at ucar.edu>
> > > > wrote:
> > > >
> > > >>
> > > >> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367
>
> > > >>
> > > >> Oh, and DIR and SPEED could be useful. DIR error is a little
tricky
> > to
> > > do
> > > >> since it's a circular variable.
> > > >>
> > > >> Dave
> > > >>
> > > >>
> > > >> On Wed, Aug 5, 2015 at 1:45 PM, David Ahijevych
<ahijevyc at ucar.edu>
> > > >> wrote:
> > > >>
> > > >> > Hi John,
> > > >> >
> > > >> > Radius of max winds is the main thing. It's much more
useful than
> > > >> > anything else.
> > > >> > RRP MRD GUSTS EYE DEPTH could be useful, but most models
don't
> > output
> > > >> > these parameters.
> > > >> >
> > > >> > I'd say MAXSEAS and SEAS are even less useful because the
models I
> > > look
> > > >> at
> > > >> > don't have wave height information to compare to.
> > > >> >
> > > >> > And it seems pointless to compare fields like SUBREGION and
> > INITIALS.
> > > >> >
> > > >> > Dave
> > > >> >
> > > >> >
> > > >> > On Wed, Aug 5, 2015 at 11:48 AM, John Halley Gotway via RT
<
> > > >> > met_help at ucar.edu> wrote:
> > > >> >
> > > >> >> Dave,
> > > >> >>
> > > >> >> Good point. It looks like we stopped dumping out ADECK
and BDECK
> > > >> >> information after the 34, 50, and 64 kt wind radii.
Here's the
> > ATCF
> > > >> file
> > > >> >> format description:
> > > >> >>
> > >
http://www.nrlmry.navy.mil/atcf_web/docs/database/new/abrdeck.html
> > > >> >>
> > > >> >> After the winds, there are several quantities that could
be
> added:
> > > >> >> RRP, MRD, GUSTS, EYE, SUBREGION, MAXSEAS, INITIALS,
DIR,
> SPEED,
> > > >> DETH,
> > > >> >> SEAS, and then the seas radii (just like the wind info)
> > > >> >>
> > > >> >> Listing values for ADECK and BDECK in the TCMPR output
line would
> > add
> > > >> >> twice
> > > >> >> as many columns to the output file.
> > > >> >>
> > > >> >> Is it just radius of maximum winds you're after or would
these
> > other
> > > >> >> quantities be useful as well?
> > > >> >>
> > > >> >> I've CC'ed Tara Jensen (the MET project lead) and Kathryn
Newman
> > > >> (MET-TC
> > > >> >> expert) to get their take on it.
> > > >> >>
> > > >> >> Thanks,
> > > >> >> John
> > > >> >>
> > > >> >>
> > > >> >>
> > > >> >> On Wed, Aug 5, 2015 at 11:14 AM, David Ahijevych via RT <
> > > >> >> met_help at ucar.edu>
> > > >> >> wrote:
> > > >> >>
> > > >> >> >
> > > >> >> > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367
> >
> > > >> >> >
> > > >> >> > Hi everyone,
> > > >> >> >
> > > >> >> > As far as I can tell, MET-TC software (which would be
ideal for
> > > this)
> > > >> >> > doesn't calculate error statistics for radius of maximum
wind
> > > (RMW).
> > > >> It
> > > >> >> > does MSLP, VMAX, and distance to 34, 50, and 64kt winds
in all
> 4
> > > >> >> quadrants,
> > > >> >> > but no RMW. Is it possible to add this?
> > > >> >> > Thanks!
> > > >> >> >
> > > >> >> > Dave
> > > >> >> >
> > > >> >> >
> > > >> >> > On Mon, Feb 10, 2014 at 1:23 PM, Dave Ahijevych <
> > ahijevyc at ucar.edu
> > > >
> > > >> >> wrote:
> > > >> >> >
> > > >> >> > > I think this is exactly what I was looking for -
thanks for
> > > taking
> > > >> the
> > > >> >> > > time to answer this!
> > > >> >> > >
> > > >> >> > >
> > > >> >> > > On 2/10/14 11:08 AM, John Halley Gotway via RT wrote:
> > > >> >> > >
> > > >> >> > >> Dave,
> > > >> >> > >>
> > > >> >> > >> The .tcst files created by tc_pairs contain some
"error"
> > columns
> > > >> that
> > > >> >> > are
> > > >> >> > >> computed by comparing the ADECK and BDECK track
locations:
> > > TK_ERR,
> > > >> >> > >> ALTK_ERR, CRTK_ERR, X_ERR, and Y_ERR for track,
> > > >> >> > >> along-track, cross-track, east-west error, and north-
south
> > > error.
> > > >> >> Those
> > > >> >> > >> types of errors are not trivial to compute, so
tc_pairs does
> > it
> > > >> for
> > > >> >> you.
> > > >> >> > >>
> > > >> >> > >> Other types of errors, like the difference of the
mean sea
> > level
> > > >> >> > >> pressures or differences in wind radii are trivial to
> compute.
> > > >> If we
> > > >> >> > had
> > > >> >> > >> included all possible differences in the .tcst output
line
> > > >> >> > >> that would have made an already long output line even
> longer.
> > > >> >> Instead,
> > > >> >> > >> tc_stat supports the use of the minus sign and/or
"ABS".
> > > >> >> > >>
> > > >> >> > >> For example, try the following...
> > > >> >> > >> -column AMSLP-BMSLP # For the MSLP error
> > > >> >> > >> -column ABS\(AMSLP-BMSLP\) # For the absolute
MSLP
> error
> > > >> >> > >> -column ABS\(Y_ERR\) # For the absolute
value of
> > the
> > > >> >> > >> north-south track error
> > > >> >> > >>
> > > >> >> > >> If you're running this on the command line, you'll
need to
> > > escape
> > > >> the
> > > >> >> > '('
> > > >> >> > >> and ')' characters. If you use a config file, you
shouldn't
> > > need
> > > >> to.
> > > >> >> > >>
> > > >> >> > >> So you can define arbitrary differences of the .tcst
output
> > > >> columns
> > > >> >> > >> with/without absolute values. Hope that helps
clarify.
> > > >> >> > >>
> > > >> >> > >> Thanks,
> > > >> >> > >> John
> > > >> >> > >>
> > > >> >> > >> On 02/07/2014 03:18 PM, David Ahijevych via RT wrote:
> > > >> >> > >>
> > > >> >> > >>> Fri Feb 07 15:18:55 2014: Request 65367 was acted
upon.
> > > >> >> > >>> Transaction: Ticket created by ahijevyc
> > > >> >> > >>> Queue: met_help
> > > >> >> > >>> Subject: MET-TC TK_ERR, MSLP_ERR?
> > > >> >> > >>> Owner: Nobody
> > > >> >> > >>> Requestors: ahijevyc at ucar.edu
> > > >> >> > >>> Status: new
> > > >> >> > >>> Ticket <URL:
> > > >> >> > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367
> > > >> >> > >>> >
> > > >> >> > >>>
> > > >> >> > >>>
> > > >> >> > >>>
> > > >> >> > >>> Hi ,
> > > >> >> > >>> I have a cursory knowledge of tc_pairs and tc_stats
now,
> and
> > am
> > > >> >> moving
> > > >> >> > >>> on from the example in the tutorial. In that
example,
> track
> > > >> error
> > > >> >> is
> > > >> >> > >>> analyzed. This is done by simply specifying the
TK_ERR
> > column.
> > > >> >> Next I
> > > >> >> > >>> would like to the same thing with MSLP error. Is
there an
> > > >> >> equivalent
> > > >> >> > >>> way to do this? I figured there would be a MSLP_ERR
> column,
> > > but
> > > >> >> > >>> couldn't find it. . .
> > > >> >> > >>>
> > > >> >> > >>> Dave
> > > >> >> > >>>
> > > >> >> > >>>
> > > >> >> > >>>
> > > >> >> > >
> > > >> >> >
> > > >> >> >
> > > >> >>
> > > >> >>
> > > >> >
> > > >>
> > > >>
> > > >
> > >
> > >
> >
> >
>
>
------------------------------------------------
Subject: MET-TC TK_ERR, MSLP_ERR?
From: John Halley Gotway
Time: Wed Aug 26 16:26:52 2015
Dave,
Thanks for the feedback.
We do see some values for GUSTS, EYE, and DEPTH in the BEST track
data.
It's conceivable that someone might want to filter their data based on
those values. So having the BEST track values in there makes sense.
We might as well leave slots for the ADECK values to, if nothing else,
keep
the layout of the file consistent. And perhaps other trackers would
populate them.
Thanks,
John
On Wed, Aug 26, 2015 at 3:07 PM, David Ahijevych via RT
<met_help at ucar.edu>
wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367 >
>
> Hi John,
> Thanks!
> Here's a line from my tracker output fort.66: It has RADP, RRP, MRD,
DIR,
> and SPEED of 1008, 442, 100, 336, and 34 respectively, but no GUSTS
EYE or
> DEPTH.
>
> EP, 97E, 2015071200_F000_129N_1236W_97E, 2015071200, 03, MPAS, 000,
132N,
> 1230W, 19, 1004, XX, 34, NEQ, 0000, 0000, 0000, 0000, 1008, 442,
100,
> -999, -9999, -9999, N, 336, 34, 124, 162, 91, 122
>
> Maybe a different tracker can output GUSTS EYE or DEPTH, but not the
GFDL
> vortex tracker. You could retain them as placeholders, but since
GFDL
> vortex tracker doesn't have them, I don't have any need for those 3
values.
>
> Everything else sounds good. Can't think of any other suggestions.
>
> Dave
>
>
> On Tue, Aug 25, 2015 at 4:16 PM, John Halley Gotway via RT <
> met_help at ucar.edu> wrote:
>
> > Dave,
> >
> > I worked today on adding these columns to the output of tc_pairs
for the
> > version 5.1 release. Following the naming conventions listed in
the ATCF
> > documentation (
> >
http://www.nrlmry.navy.mil/atcf_web/docs/database/new/abrdeck.html), I
> > added the following columns 16 columns:
> >
> > ARADP BRADP ARRP BRRP AMRD BMRD AGUSTS BGUSTS AEYE BEYE ADIR
BDIR
> ASPEED
> > BSPEED ADEPTH BDEPTH
> >
> > They are for the ADECK and BDECK values of RADP, RRP, MRD, GUSTS,
EYE,
> DIR,
> > SPEED, and DEPTH.
> >
> > I see that in the test data we run, some of these values do show
up in
> the
> > BDECK output. However, they're all NA for the ADECK. Can you
confirm
> that
> > these values actually do show up in the model tracker output you
have?
> >
> > While the documentation lists these as integers, I'm writing them
as
> floats
> > to handle the computation of "consensus" forecasts. For example,
a
> > consensus eyewall diameter of three diameters 10, 30, 25 would be
21.667.
> >
> > Did I miss anything? Or do you have any other requests or
suggestions?
> >
> > Thanks,
> > John
> >
> > On Mon, Aug 17, 2015 at 3:55 PM, David Ahijevych via RT <
> met_help at ucar.edu
> > >
> > wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367 >
> > >
> > > Hi John,
> > > Thanks for letting me know the status of getting RMW into MET.
> > > I may have to come up with some hack-y way to compare RMW in the
> > meantime.
> > > Chris D. and I are working on a paper about hurricane
simulations in
> MPAS
> > > where we might analyze RMW. And as part of this new storm surge
> project
> > > I'm working on, I'm reading that the overall area of a storm is
more
> > > closely tied to storm surge than max wind!
> > >
> > > Dave
> > >
> > >
> > > On Mon, Aug 17, 2015 at 12:15 PM, John Halley Gotway via RT <
> > > met_help at ucar.edu> wrote:
> > >
> > > > Dave,
> > > >
> > > > We have this defined as a development issue and will try to
get it
> into
> > > the
> > > > next release. But it all depends on whether we get approval
to use a
> > > small
> > > > pot of HWRF verification funding for this work.
> > > >
> > > > Thanks,
> > > > John
> > > >
> > > > On Wed, Aug 5, 2015 at 2:42 PM, John Halley Gotway
<johnhg at ucar.edu>
> > > > wrote:
> > > >
> > > > > Dave,
> > > > >
> > > > > Thanks for sending the details. I created a new development
ticket
> > to
> > > > > track this request and included the notes you sent. We're
working
> on
> > > the
> > > > > next release, 5.1, now and the question is whether we'll
have
> > > > time/funding
> > > > > to get these changes into 5.1 or if we'll have to wait until
the
> next
> > > > > release.
> > > > >
> > > > > I'll talk to Tara about it at our development meeting
tomorrow.
> > > > >
> > > > > Thanks,
> > > > > John
> > > > >
> > > > > On Wed, Aug 5, 2015 at 1:49 PM, David Ahijevych via RT <
> > > > met_help at ucar.edu>
> > > > > wrote:
> > > > >
> > > > >>
> > > > >> <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367 >
> > > > >>
> > > > >> Oh, and DIR and SPEED could be useful. DIR error is a
little
> tricky
> > > to
> > > > do
> > > > >> since it's a circular variable.
> > > > >>
> > > > >> Dave
> > > > >>
> > > > >>
> > > > >> On Wed, Aug 5, 2015 at 1:45 PM, David Ahijevych <
> ahijevyc at ucar.edu>
> > > > >> wrote:
> > > > >>
> > > > >> > Hi John,
> > > > >> >
> > > > >> > Radius of max winds is the main thing. It's much more
useful
> than
> > > > >> > anything else.
> > > > >> > RRP MRD GUSTS EYE DEPTH could be useful, but most models
don't
> > > output
> > > > >> > these parameters.
> > > > >> >
> > > > >> > I'd say MAXSEAS and SEAS are even less useful because the
> models I
> > > > look
> > > > >> at
> > > > >> > don't have wave height information to compare to.
> > > > >> >
> > > > >> > And it seems pointless to compare fields like SUBREGION
and
> > > INITIALS.
> > > > >> >
> > > > >> > Dave
> > > > >> >
> > > > >> >
> > > > >> > On Wed, Aug 5, 2015 at 11:48 AM, John Halley Gotway via
RT <
> > > > >> > met_help at ucar.edu> wrote:
> > > > >> >
> > > > >> >> Dave,
> > > > >> >>
> > > > >> >> Good point. It looks like we stopped dumping out ADECK
and
> BDECK
> > > > >> >> information after the 34, 50, and 64 kt wind radii.
Here's the
> > > ATCF
> > > > >> file
> > > > >> >> format description:
> > > > >> >>
> > > >
http://www.nrlmry.navy.mil/atcf_web/docs/database/new/abrdeck.html
> > > > >> >>
> > > > >> >> After the winds, there are several quantities that could
be
> > added:
> > > > >> >> RRP, MRD, GUSTS, EYE, SUBREGION, MAXSEAS, INITIALS,
DIR,
> > SPEED,
> > > > >> DETH,
> > > > >> >> SEAS, and then the seas radii (just like the wind info)
> > > > >> >>
> > > > >> >> Listing values for ADECK and BDECK in the TCMPR output
line
> would
> > > add
> > > > >> >> twice
> > > > >> >> as many columns to the output file.
> > > > >> >>
> > > > >> >> Is it just radius of maximum winds you're after or would
these
> > > other
> > > > >> >> quantities be useful as well?
> > > > >> >>
> > > > >> >> I've CC'ed Tara Jensen (the MET project lead) and
Kathryn
> Newman
> > > > >> (MET-TC
> > > > >> >> expert) to get their take on it.
> > > > >> >>
> > > > >> >> Thanks,
> > > > >> >> John
> > > > >> >>
> > > > >> >>
> > > > >> >>
> > > > >> >> On Wed, Aug 5, 2015 at 11:14 AM, David Ahijevych via RT
<
> > > > >> >> met_help at ucar.edu>
> > > > >> >> wrote:
> > > > >> >>
> > > > >> >> >
> > > > >> >> > <URL:
> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367
> > >
> > > > >> >> >
> > > > >> >> > Hi everyone,
> > > > >> >> >
> > > > >> >> > As far as I can tell, MET-TC software (which would be
ideal
> for
> > > > this)
> > > > >> >> > doesn't calculate error statistics for radius of
maximum wind
> > > > (RMW).
> > > > >> It
> > > > >> >> > does MSLP, VMAX, and distance to 34, 50, and 64kt
winds in
> all
> > 4
> > > > >> >> quadrants,
> > > > >> >> > but no RMW. Is it possible to add this?
> > > > >> >> > Thanks!
> > > > >> >> >
> > > > >> >> > Dave
> > > > >> >> >
> > > > >> >> >
> > > > >> >> > On Mon, Feb 10, 2014 at 1:23 PM, Dave Ahijevych <
> > > ahijevyc at ucar.edu
> > > > >
> > > > >> >> wrote:
> > > > >> >> >
> > > > >> >> > > I think this is exactly what I was looking for -
thanks for
> > > > taking
> > > > >> the
> > > > >> >> > > time to answer this!
> > > > >> >> > >
> > > > >> >> > >
> > > > >> >> > > On 2/10/14 11:08 AM, John Halley Gotway via RT
wrote:
> > > > >> >> > >
> > > > >> >> > >> Dave,
> > > > >> >> > >>
> > > > >> >> > >> The .tcst files created by tc_pairs contain some
"error"
> > > columns
> > > > >> that
> > > > >> >> > are
> > > > >> >> > >> computed by comparing the ADECK and BDECK track
locations:
> > > > TK_ERR,
> > > > >> >> > >> ALTK_ERR, CRTK_ERR, X_ERR, and Y_ERR for track,
> > > > >> >> > >> along-track, cross-track, east-west error, and
north-south
> > > > error.
> > > > >> >> Those
> > > > >> >> > >> types of errors are not trivial to compute, so
tc_pairs
> does
> > > it
> > > > >> for
> > > > >> >> you.
> > > > >> >> > >>
> > > > >> >> > >> Other types of errors, like the difference of the
mean sea
> > > level
> > > > >> >> > >> pressures or differences in wind radii are trivial
to
> > compute.
> > > > >> If we
> > > > >> >> > had
> > > > >> >> > >> included all possible differences in the .tcst
output line
> > > > >> >> > >> that would have made an already long output line
even
> > longer.
> > > > >> >> Instead,
> > > > >> >> > >> tc_stat supports the use of the minus sign and/or
"ABS".
> > > > >> >> > >>
> > > > >> >> > >> For example, try the following...
> > > > >> >> > >> -column AMSLP-BMSLP # For the MSLP
error
> > > > >> >> > >> -column ABS\(AMSLP-BMSLP\) # For the absolute
MSLP
> > error
> > > > >> >> > >> -column ABS\(Y_ERR\) # For the absolute
value
> of
> > > the
> > > > >> >> > >> north-south track error
> > > > >> >> > >>
> > > > >> >> > >> If you're running this on the command line, you'll
need to
> > > > escape
> > > > >> the
> > > > >> >> > '('
> > > > >> >> > >> and ')' characters. If you use a config file, you
> shouldn't
> > > > need
> > > > >> to.
> > > > >> >> > >>
> > > > >> >> > >> So you can define arbitrary differences of the
.tcst
> output
> > > > >> columns
> > > > >> >> > >> with/without absolute values. Hope that helps
clarify.
> > > > >> >> > >>
> > > > >> >> > >> Thanks,
> > > > >> >> > >> John
> > > > >> >> > >>
> > > > >> >> > >> On 02/07/2014 03:18 PM, David Ahijevych via RT
wrote:
> > > > >> >> > >>
> > > > >> >> > >>> Fri Feb 07 15:18:55 2014: Request 65367 was acted
upon.
> > > > >> >> > >>> Transaction: Ticket created by ahijevyc
> > > > >> >> > >>> Queue: met_help
> > > > >> >> > >>> Subject: MET-TC TK_ERR, MSLP_ERR?
> > > > >> >> > >>> Owner: Nobody
> > > > >> >> > >>> Requestors: ahijevyc at ucar.edu
> > > > >> >> > >>> Status: new
> > > > >> >> > >>> Ticket <URL:
> > > > >> >> >
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=65367
> > > > >> >> > >>> >
> > > > >> >> > >>>
> > > > >> >> > >>>
> > > > >> >> > >>>
> > > > >> >> > >>> Hi ,
> > > > >> >> > >>> I have a cursory knowledge of tc_pairs and
tc_stats now,
> > and
> > > am
> > > > >> >> moving
> > > > >> >> > >>> on from the example in the tutorial. In that
example,
> > track
> > > > >> error
> > > > >> >> is
> > > > >> >> > >>> analyzed. This is done by simply specifying the
TK_ERR
> > > column.
> > > > >> >> Next I
> > > > >> >> > >>> would like to the same thing with MSLP error. Is
there
> an
> > > > >> >> equivalent
> > > > >> >> > >>> way to do this? I figured there would be a
MSLP_ERR
> > column,
> > > > but
> > > > >> >> > >>> couldn't find it. . .
> > > > >> >> > >>>
> > > > >> >> > >>> Dave
> > > > >> >> > >>>
> > > > >> >> > >>>
> > > > >> >> > >>>
> > > > >> >> > >
> > > > >> >> >
> > > > >> >> >
> > > > >> >>
> > > > >> >>
> > > > >> >
> > > > >>
> > > > >>
> > > > >
> > > >
> > > >
> > >
> > >
> >
> >
>
>
------------------------------------------------
More information about the Met_help
mailing list