[Met_help] [rt.rap.ucar.edu #88802] History for HIRA Question
John Halley Gotway via RT
met_help at ucar.edu
Tue Jul 9 12:07:02 MDT 2019
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
John, quick question, when I run PTSTAT with HIRA for a specific threshold, in the stat output file I get the model data being between 0-1, the ob data being the actual value, and the climo data between 0-1. Is this correct or should the ob data be 0 or 1 to reflect the threshold? I an getting high Brier Scores which lead me to believe the ob data should be thresholded.
Thanks
Bob
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: HIRA Question
From: John Halley Gotway
Time: Mon Feb 11 09:41:53 2019
Bob,
Just catching up on met-help questions. I think I answered this in my
response last week, but yes, what you're seeing is how it was
designed.
The MPR line type produced by HiRA contains the same sort of data
that's
printed when we do probabilistic verification. The forecast and climo
are
the probability values between 0 and 1, and the observation is the raw
(unthresholded) value.
Thanks,
John
On Wed, Feb 6, 2019 at 12:03 PM robert.craig.2 at us.af.mil via RT <
met_help at ucar.edu> wrote:
>
> Wed Feb 06 12:02:46 2019: Request 88802 was acted upon.
> Transaction: Ticket created by robert.craig.2 at us.af.mil
> Queue: met_help
> Subject: HIRA Question
> Owner: Nobody
> Requestors: robert.craig.2 at us.af.mil
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=88802 >
>
>
> John, quick question, when I run PTSTAT with HIRA for a specific
> threshold, in the stat output file I get the model data being
between 0-1,
> the ob data being the actual value, and the climo data between 0-1.
Is
> this correct or should the ob data be 0 or 1 to reflect the
threshold? I
> an getting high Brier Scores which lead me to believe the ob data
should be
> thresholded.
>
> Thanks
> Bob
>
>
------------------------------------------------
Subject: RE: [Non-DoD Source] Re: [rt.rap.ucar.edu #88802] HIRA Question
From: robert.craig.2 at us.af.mil
Time: Mon Feb 11 09:48:49 2019
Yes, my problem was I wasn't setting -out_obs_thresh to the right
threshold. With that fix the Brier Scores are reasonable again.
By the way, do you have a beta version of MET 8.1 ready? The only way
we can get it is if you post it on ARL Safe. It would be nice move to
that version to reduce the work arounds.
Thanks
Bob
-----Original Message-----
From: John Halley Gotway via RT <met_help at ucar.edu>
Sent: Monday, February 11, 2019 10:42 AM
To: CRAIG, ROBERT J GS-12 USAF ACC 16 WS/WXN
<robert.craig.2 at us.af.mil>
Subject: [Non-DoD Source] Re: [rt.rap.ucar.edu #88802] HIRA Question
Bob,
Just catching up on met-help questions. I think I answered this in my
response last week, but yes, what you're seeing is how it was
designed.
The MPR line type produced by HiRA contains the same sort of data
that's printed when we do probabilistic verification. The forecast
and climo are the probability values between 0 and 1, and the
observation is the raw
(unthresholded) value.
Thanks,
John
On Wed, Feb 6, 2019 at 12:03 PM robert.craig.2 at us.af.mil via RT <
met_help at ucar.edu> wrote:
>
> Wed Feb 06 12:02:46 2019: Request 88802 was acted upon.
> Transaction: Ticket created by robert.craig.2 at us.af.mil
> Queue: met_help
> Subject: HIRA Question
> Owner: Nobody
> Requestors: robert.craig.2 at us.af.mil
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=88802
> >
>
>
> John, quick question, when I run PTSTAT with HIRA for a specific
> threshold, in the stat output file I get the model data being
between
> 0-1, the ob data being the actual value, and the climo data between
> 0-1. Is this correct or should the ob data be 0 or 1 to reflect the
> threshold? I an getting high Brier Scores which lead me to believe
> the ob data should be thresholded.
>
> Thanks
> Bob
>
>
------------------------------------------------
Subject: HIRA Question
From: John Halley Gotway
Time: Tue Feb 12 16:50:41 2019
Bob,
I tried to post the met-8.1_beta1 release tarball on ARL SAFE. While
I do
have a CAC card, it is through NOAA, not DoD. So ARL SAFE didn't
accept my
credentials. According to the site, I can post to it, as long as I
have a
"Request Code" from you. Can you please send me a request code?
Thanks,
John
On Mon, Feb 11, 2019 at 9:49 AM robert.craig.2 at us.af.mil via RT <
met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=88802 >
>
> Yes, my problem was I wasn't setting -out_obs_thresh to the right
> threshold. With that fix the Brier Scores are reasonable again.
>
> By the way, do you have a beta version of MET 8.1 ready? The only
way we
> can get it is if you post it on ARL Safe. It would be nice move to
that
> version to reduce the work arounds.
>
> Thanks
> Bob
>
>
> -----Original Message-----
> From: John Halley Gotway via RT <met_help at ucar.edu>
> Sent: Monday, February 11, 2019 10:42 AM
> To: CRAIG, ROBERT J GS-12 USAF ACC 16 WS/WXN
<robert.craig.2 at us.af.mil>
> Subject: [Non-DoD Source] Re: [rt.rap.ucar.edu #88802] HIRA Question
>
> Bob,
>
> Just catching up on met-help questions. I think I answered this in
my
> response last week, but yes, what you're seeing is how it was
designed.
> The MPR line type produced by HiRA contains the same sort of data
that's
> printed when we do probabilistic verification. The forecast and
climo are
> the probability values between 0 and 1, and the observation is the
raw
> (unthresholded) value.
>
> Thanks,
> John
>
> On Wed, Feb 6, 2019 at 12:03 PM robert.craig.2 at us.af.mil via RT <
> met_help at ucar.edu> wrote:
>
> >
> > Wed Feb 06 12:02:46 2019: Request 88802 was acted upon.
> > Transaction: Ticket created by robert.craig.2 at us.af.mil
> > Queue: met_help
> > Subject: HIRA Question
> > Owner: Nobody
> > Requestors: robert.craig.2 at us.af.mil
> > Status: new
> > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=88802
> > >
> >
> >
> > John, quick question, when I run PTSTAT with HIRA for a specific
> > threshold, in the stat output file I get the model data being
between
> > 0-1, the ob data being the actual value, and the climo data
between
> > 0-1. Is this correct or should the ob data be 0 or 1 to reflect
the
> > threshold? I an getting high Brier Scores which lead me to
believe
> > the ob data should be thresholded.
> >
> > Thanks
> > Bob
> >
> >
>
>
>
>
------------------------------------------------
Subject: HIRA Question
From: John Halley Gotway
Time: Wed Feb 13 10:29:52 2019
Bob,
Posted it about 30 minutes ago. Hope you got an email about it.
Please
let us know what questions you have about installing it on dev9.
On my end, I need to... run Fortify on this released version, out the
resulting PDF, upload the mbs file to Fortify-On-Demand, and invite AF
folks to see the results.
Thanks,
John
On Tue, Feb 12, 2019 at 4:50 PM John Halley Gotway <johnhg at ucar.edu>
wrote:
> Bob,
>
> I tried to post the met-8.1_beta1 release tarball on ARL SAFE.
While I do
> have a CAC card, it is through NOAA, not DoD. So ARL SAFE didn't
accept my
> credentials. According to the site, I can post to it, as long as I
have a
> "Request Code" from you. Can you please send me a request code?
>
> Thanks,
> John
>
> On Mon, Feb 11, 2019 at 9:49 AM robert.craig.2 at us.af.mil via RT <
> met_help at ucar.edu> wrote:
>
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=88802 >
>>
>> Yes, my problem was I wasn't setting -out_obs_thresh to the right
>> threshold. With that fix the Brier Scores are reasonable again.
>>
>> By the way, do you have a beta version of MET 8.1 ready? The only
way we
>> can get it is if you post it on ARL Safe. It would be nice move to
that
>> version to reduce the work arounds.
>>
>> Thanks
>> Bob
>>
>>
>> -----Original Message-----
>> From: John Halley Gotway via RT <met_help at ucar.edu>
>> Sent: Monday, February 11, 2019 10:42 AM
>> To: CRAIG, ROBERT J GS-12 USAF ACC 16 WS/WXN
<robert.craig.2 at us.af.mil>
>> Subject: [Non-DoD Source] Re: [rt.rap.ucar.edu #88802] HIRA
Question
>>
>> Bob,
>>
>> Just catching up on met-help questions. I think I answered this in
my
>> response last week, but yes, what you're seeing is how it was
designed.
>> The MPR line type produced by HiRA contains the same sort of data
that's
>> printed when we do probabilistic verification. The forecast and
climo are
>> the probability values between 0 and 1, and the observation is the
raw
>> (unthresholded) value.
>>
>> Thanks,
>> John
>>
>> On Wed, Feb 6, 2019 at 12:03 PM robert.craig.2 at us.af.mil via RT <
>> met_help at ucar.edu> wrote:
>>
>> >
>> > Wed Feb 06 12:02:46 2019: Request 88802 was acted upon.
>> > Transaction: Ticket created by robert.craig.2 at us.af.mil
>> > Queue: met_help
>> > Subject: HIRA Question
>> > Owner: Nobody
>> > Requestors: robert.craig.2 at us.af.mil
>> > Status: new
>> > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=88802
>> > >
>> >
>> >
>> > John, quick question, when I run PTSTAT with HIRA for a specific
>> > threshold, in the stat output file I get the model data being
between
>> > 0-1, the ob data being the actual value, and the climo data
between
>> > 0-1. Is this correct or should the ob data be 0 or 1 to reflect
the
>> > threshold? I an getting high Brier Scores which lead me to
believe
>> > the ob data should be thresholded.
>> >
>> > Thanks
>> > Bob
>> >
>> >
>>
>>
>>
>>
------------------------------------------------
More information about the Met_help
mailing list