[Met_help] [rt.rap.ucar.edu #80274] History for MET-TC: verifying OFCL forecasts
John Halley Gotway via RT
met_help at ucar.edu
Mon May 1 14:52:23 MDT 2017
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Hi John,
I ran into another interesting aspect of verifying OFCL forecasts... it
turns out that occasionally a special forecast will be issued after the
normal advisory time if there's reason to believe a significant update to
the original forecast is needed. This ends up taking on the "OFCL" tag in
the ATCF, and the original forecast is moved to "OFCO". Up until this
point, I've simply been verifying all "OFCL" forecasts, and this isn't
correct. When there's an "OFCO", it should replace the "OFCL" for that
forecast and be verified instead. Is there any way to do this automatically
through MET-TC?
An example forecast at 2016093000 for Hurricane Matthew (AL142016) is
attached.
Thanks,
Andy
--
Andrew Penny
Hurricane Model Diagnostician
SRG / National Hurricane Center
11691 SW 17th Street
Miami, FL 33165-2149
phone: 305.229.4457
email: andrew.penny at noaa.gov
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: MET-TC: verifying OFCL forecasts
From: John Halley Gotway
Time: Mon May 01 13:51:53 2017
Andy,
Sorry for the delay in responding. I was out of the office last week
teaching a MET tutorial at NCEP.
Thanks for explaining the logic that NHC uses for the naming of
updated
official forecasts. I just want to repeat back to you what I read to
make
sure it's crystal clear:
(1) NHC's operational forecast track is named using the AFCF ID
"OFCL".
(2) If for whatever reason, NHC updates the official forecast track,
they
rename the original "OFCL" track as "OFCO" and name the updated track
"OFCL".
(3) You would like to have the ability to verify only the original
"OFCL"
forecast tracks and ignore the updates.
One solution would be adding a new config file entry for the TC-Pairs
tool. When turned on...
(1) For each "OFCL" track, check to see if an "OFCO" track exists for
the
same storm/initialization time.
(2) If so, delete the existing "OFCL" forecast and rename the "OFCO"
track
as "OFCL".
Is that the functionality you're looking for? It isn't obvious to me
why
you'd want to do this. Why not just verify the updated forecast
instead of
the original one?
If my description above is correct, any suggestion on what we should
name
this config file option? Other config file switches to control
different
types of logic are named:
check_dup, interp12, and match_points
Another solution would be more general. In the config file, define an
original ATCF ID name and a new name. In our example, it'd be OFCO
and
OFCL. Any time the original (OFCO) exists, rename it to the new name
(OFCL) and over-write any existing tracks for the same case.
Do you prefer a specific true/false flag to mimic NHC logic or a more
general mapping of original ATCF ID to new ATCF ID?
Thanks,
John
On Wed, Apr 26, 2017 at 11:48 AM, Andrew Penny - NOAA Affiliate via RT
<
met_help at ucar.edu> wrote:
>
> Wed Apr 26 11:48:29 2017: Request 80274 was acted upon.
> Transaction: Ticket created by andrew.penny at noaa.gov
> Queue: met_help
> Subject: MET-TC: verifying OFCL forecasts
> Owner: Nobody
> Requestors: andrew.penny at noaa.gov
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=80274 >
>
>
> Hi John,
>
> I ran into another interesting aspect of verifying OFCL forecasts...
it
> turns out that occasionally a special forecast will be issued after
the
> normal advisory time if there's reason to believe a significant
update to
> the original forecast is needed. This ends up taking on the "OFCL"
tag in
> the ATCF, and the original forecast is moved to "OFCO". Up until
this
> point, I've simply been verifying all "OFCL" forecasts, and this
isn't
> correct. When there's an "OFCO", it should replace the "OFCL" for
that
> forecast and be verified instead. Is there any way to do this
automatically
> through MET-TC?
>
> An example forecast at 2016093000 for Hurricane Matthew (AL142016)
is
> attached.
>
> Thanks,
> Andy
>
> --
> Andrew Penny
> Hurricane Model Diagnostician
> SRG / National Hurricane Center
> 11691 SW 17th Street
> Miami, FL 33165-2149
> phone: 305.229.4457
> email: andrew.penny at noaa.gov
>
>
------------------------------------------------
Subject: MET-TC: verifying OFCL forecasts
From: Andrew Penny - NOAA Affiliate
Time: Mon May 01 14:41:49 2017
Hi John,
I'm not sure which option is best. The more general option might be
nice,
but I'm not sure there would be another reason to do this besides the
OFCO/OFCL example. So, perhaps the True/False option is best?
The reason we don't want to verify the updated forecasts is that it's
not a
fair comparison when other guidance is included, since the updated
forecast
was not available at the advisory time.
As far as naming goes, perhaps "use_special_OFCL" or something, with
the
default being false.
As always, thanks for your help!
Take care,
Andy
On Mon, May 1, 2017 at 7:51 PM, John Halley Gotway via RT
<met_help at ucar.edu
> wrote:
> Andy,
>
> Sorry for the delay in responding. I was out of the office last
week
> teaching a MET tutorial at NCEP.
>
> Thanks for explaining the logic that NHC uses for the naming of
updated
> official forecasts. I just want to repeat back to you what I read
to make
> sure it's crystal clear:
>
> (1) NHC's operational forecast track is named using the AFCF ID
"OFCL".
> (2) If for whatever reason, NHC updates the official forecast track,
they
> rename the original "OFCL" track as "OFCO" and name the updated
track
> "OFCL".
> (3) You would like to have the ability to verify only the original
"OFCL"
> forecast tracks and ignore the updates.
>
> One solution would be adding a new config file entry for the TC-
Pairs
> tool. When turned on...
> (1) For each "OFCL" track, check to see if an "OFCO" track exists
for the
> same storm/initialization time.
> (2) If so, delete the existing "OFCL" forecast and rename the "OFCO"
track
> as "OFCL".
>
> Is that the functionality you're looking for? It isn't obvious to
me why
> you'd want to do this. Why not just verify the updated forecast
instead of
> the original one?
>
> If my description above is correct, any suggestion on what we should
name
> this config file option? Other config file switches to control
different
> types of logic are named:
> check_dup, interp12, and match_points
>
> Another solution would be more general. In the config file, define
an
> original ATCF ID name and a new name. In our example, it'd be OFCO
and
> OFCL. Any time the original (OFCO) exists, rename it to the new
name
> (OFCL) and over-write any existing tracks for the same case.
>
> Do you prefer a specific true/false flag to mimic NHC logic or a
more
> general mapping of original ATCF ID to new ATCF ID?
>
> Thanks,
> John
>
>
> On Wed, Apr 26, 2017 at 11:48 AM, Andrew Penny - NOAA Affiliate via
RT <
> met_help at ucar.edu> wrote:
>
> >
> > Wed Apr 26 11:48:29 2017: Request 80274 was acted upon.
> > Transaction: Ticket created by andrew.penny at noaa.gov
> > Queue: met_help
> > Subject: MET-TC: verifying OFCL forecasts
> > Owner: Nobody
> > Requestors: andrew.penny at noaa.gov
> > Status: new
> > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=80274 >
> >
> >
> > Hi John,
> >
> > I ran into another interesting aspect of verifying OFCL
forecasts... it
> > turns out that occasionally a special forecast will be issued
after the
> > normal advisory time if there's reason to believe a significant
update to
> > the original forecast is needed. This ends up taking on the "OFCL"
tag in
> > the ATCF, and the original forecast is moved to "OFCO". Up until
this
> > point, I've simply been verifying all "OFCL" forecasts, and this
isn't
> > correct. When there's an "OFCO", it should replace the "OFCL" for
that
> > forecast and be verified instead. Is there any way to do this
> automatically
> > through MET-TC?
> >
> > An example forecast at 2016093000 for Hurricane Matthew (AL142016)
is
> > attached.
> >
> > Thanks,
> > Andy
> >
> > --
> > Andrew Penny
> > Hurricane Model Diagnostician
> > SRG / National Hurricane Center
> > 11691 SW 17th Street
> > Miami, FL 33165-2149
> > phone: 305.229.4457
> > email: andrew.penny at noaa.gov
> >
> >
>
>
--
Andrew Penny
Hurricane Model Diagnostician
SRG / National Hurricane Center
11691 SW 17th Street
Miami, FL 33165-2149
phone: 305.229.4457
email: andrew.penny at noaa.gov
------------------------------------------------
Subject: MET-TC: verifying OFCL forecasts
From: John Halley Gotway
Time: Mon May 01 14:46:15 2017
Andy,
Great, thanks for the info. I'll use it to define a development task
for
MET. The earliest this would be added would be the 6.1 release in the
next
6 months or so.
I'll go ahead and resolve this ticket.
Thanks,
John
On Mon, May 1, 2017 at 2:41 PM, Andrew Penny - NOAA Affiliate via RT <
met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=80274 >
>
> Hi John,
>
> I'm not sure which option is best. The more general option might be
nice,
> but I'm not sure there would be another reason to do this besides
the
> OFCO/OFCL example. So, perhaps the True/False option is best?
>
> The reason we don't want to verify the updated forecasts is that
it's not a
> fair comparison when other guidance is included, since the updated
forecast
> was not available at the advisory time.
>
> As far as naming goes, perhaps "use_special_OFCL" or something, with
the
> default being false.
>
> As always, thanks for your help!
>
> Take care,
> Andy
>
>
> On Mon, May 1, 2017 at 7:51 PM, John Halley Gotway via RT <
> met_help at ucar.edu
> > wrote:
>
> > Andy,
> >
> > Sorry for the delay in responding. I was out of the office last
week
> > teaching a MET tutorial at NCEP.
> >
> > Thanks for explaining the logic that NHC uses for the naming of
updated
> > official forecasts. I just want to repeat back to you what I read
to
> make
> > sure it's crystal clear:
> >
> > (1) NHC's operational forecast track is named using the AFCF ID
"OFCL".
> > (2) If for whatever reason, NHC updates the official forecast
track, they
> > rename the original "OFCL" track as "OFCO" and name the updated
track
> > "OFCL".
> > (3) You would like to have the ability to verify only the original
"OFCL"
> > forecast tracks and ignore the updates.
> >
> > One solution would be adding a new config file entry for the TC-
Pairs
> > tool. When turned on...
> > (1) For each "OFCL" track, check to see if an "OFCO" track exists
for the
> > same storm/initialization time.
> > (2) If so, delete the existing "OFCL" forecast and rename the
"OFCO"
> track
> > as "OFCL".
> >
> > Is that the functionality you're looking for? It isn't obvious to
me why
> > you'd want to do this. Why not just verify the updated forecast
instead
> of
> > the original one?
> >
> > If my description above is correct, any suggestion on what we
should name
> > this config file option? Other config file switches to control
different
> > types of logic are named:
> > check_dup, interp12, and match_points
> >
> > Another solution would be more general. In the config file,
define an
> > original ATCF ID name and a new name. In our example, it'd be
OFCO and
> > OFCL. Any time the original (OFCO) exists, rename it to the new
name
> > (OFCL) and over-write any existing tracks for the same case.
> >
> > Do you prefer a specific true/false flag to mimic NHC logic or a
more
> > general mapping of original ATCF ID to new ATCF ID?
> >
> > Thanks,
> > John
> >
> >
> > On Wed, Apr 26, 2017 at 11:48 AM, Andrew Penny - NOAA Affiliate
via RT <
> > met_help at ucar.edu> wrote:
> >
> > >
> > > Wed Apr 26 11:48:29 2017: Request 80274 was acted upon.
> > > Transaction: Ticket created by andrew.penny at noaa.gov
> > > Queue: met_help
> > > Subject: MET-TC: verifying OFCL forecasts
> > > Owner: Nobody
> > > Requestors: andrew.penny at noaa.gov
> > > Status: new
> > > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=80274
> >
> > >
> > >
> > > Hi John,
> > >
> > > I ran into another interesting aspect of verifying OFCL
forecasts... it
> > > turns out that occasionally a special forecast will be issued
after the
> > > normal advisory time if there's reason to believe a significant
update
> to
> > > the original forecast is needed. This ends up taking on the
"OFCL" tag
> in
> > > the ATCF, and the original forecast is moved to "OFCO". Up until
this
> > > point, I've simply been verifying all "OFCL" forecasts, and this
isn't
> > > correct. When there's an "OFCO", it should replace the "OFCL"
for that
> > > forecast and be verified instead. Is there any way to do this
> > automatically
> > > through MET-TC?
> > >
> > > An example forecast at 2016093000 for Hurricane Matthew
(AL142016) is
> > > attached.
> > >
> > > Thanks,
> > > Andy
> > >
> > > --
> > > Andrew Penny
> > > Hurricane Model Diagnostician
> > > SRG / National Hurricane Center
> > > 11691 SW 17th Street
> > > Miami, FL 33165-2149
> > > phone: 305.229.4457
> > > email: andrew.penny at noaa.gov
> > >
> > >
> >
> >
>
>
> --
> Andrew Penny
> Hurricane Model Diagnostician
> SRG / National Hurricane Center
> 11691 SW 17th Street
> Miami, FL 33165-2149
> phone: 305.229.4457
> email: andrew.penny at noaa.gov
>
>
------------------------------------------------
More information about the Met_help
mailing list