[Met_help] [rt.rap.ucar.edu #95177] History for Verifying models with few cycle times
George McCabe via RT
met_help at ucar.edu
Thu May 7 11:54:15 MDT 2020
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Hi, there,
I'd like to ask a quick question. What should be the METplus configuration
settings if we want to verify for every hour in a 48-hr forecast, but there
are only two cycles of model runs: 06Z and 12Z? Right now I have METplus
cycling through so many different time combinations of files that do not
exist, but that's because I am not too sure of how to set this type of
thing. Can you assist me in figuring out the proper settings in this
situation?
Thanks!
Perry
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: Verifying models with few cycle times
From: George McCabe
Time: Wed May 06 10:17:30 2020
Hi Perry,
If you are processing a single day, you could loop by INIT and only
run the
6Z and 12Z init times:
[config]
LOOP_BY = INIT
INIT_TIME_FMT = %Y%m%d%H
INIT_BEG = 2017020106
INIT_END = 2017020112
INIT_INCREMENT = 6H
LEAD_SEQ = begin_end_incr(0,48,1)
However, if you need to process more than 1 day, there isn't currently
a
good way to loop by INIT with uneven increments. There is
functionality we
want to implement that involves specifying times to exclude, so if
that
existed you could tell it to exclude 0H and 18H and it would skip
those
times.
If you are looping by VALID time, there is an INIT_SEQ variable that
can be
used to build the forecast lead list dynamically for each valid time
based
on the list of available init times. You will also have to set the max
forecast lead to use:
LOOP_BY = VALID
VALID_TIME_FMT = %Y%m%d%H
VALID_BEG = 2017020100
VALID_END = 2017020100
VALID_INCREMENT = 6H
INIT_SEQ = 6, 12
LEAD_SEQ_MAX = 48
The above configuration would produce the following (tested with
Example
wrapper):
Processing forecast lead 12 hours initialized at 2017-01-31 12Z and
valid
at 2017-02-01 00Z
Processing forecast lead 12 hours initialized at 2017-01-31 12Z and
valid
at 2017-02-01 00Z
Processing forecast lead 18 hours initialized at 2017-01-31 06Z and
valid
at 2017-02-01 00Z
Processing forecast lead 18 hours initialized at 2017-01-31 06Z and
valid
at 2017-02-01 00Z
Processing forecast lead 1 day 12 hours initialized at 2017-01-30 12Z
and
valid at 2017-02-01 00Z
Processing forecast lead 1 day 12 hours initialized at 2017-01-30 12Z
and
valid at 2017-02-01 00Z
Processing forecast lead 1 day 18 hours initialized at 2017-01-30 06Z
and
valid at 2017-02-01 00Z
Processing forecast lead 1 day 18 hours initialized at 2017-01-30 06Z
and
valid at 2017-02-01 00Z
More info on that here:
https://ncar.github.io/METplus/Users_Guide/systemconfiguration.html#init-
seq
Let me know if either of those solutions would work for you. If not,
we can
look closer at your case to see what would work.
Thanks,
George
On Wed, May 6, 2020 at 9:25 AM perry.shafran at noaa.gov via RT <
met_help at ucar.edu> wrote:
>
> Wed May 06 09:25:20 2020: Request 95177 was acted upon.
> Transaction: Ticket created by perry.shafran at noaa.gov
> Queue: met_help
> Subject: Verifying models with few cycle times
> Owner: Nobody
> Requestors: perry.shafran at noaa.gov
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=95177 >
>
>
> Hi, there,
>
> I'd like to ask a quick question. What should be the METplus
configuration
> settings if we want to verify for every hour in a 48-hr forecast,
but there
> are only two cycles of model runs: 06Z and 12Z? Right now I have
METplus
> cycling through so many different time combinations of files that do
not
> exist, but that's because I am not too sure of how to set this type
of
> thing. Can you assist me in figuring out the proper settings in
this
> situation?
>
> Thanks!
>
> Perry
>
>
--
George McCabe - Software Engineer III
National Center for Atmospheric Research
Research Applications Laboratory
303-497-2768
---
My working day may not be your working day. Please do not feel obliged
to
reply to this email outside of your normal working hours.
------------------------------------------------
Subject: Verifying models with few cycle times
From: perry.shafran at noaa.gov
Time: Wed May 06 11:25:01 2020
Hi, George,
Let me look at this and see if it will work. Seems like a reasonably
easy
way to do this.
Keep the ticket open until I get back to you.
Thanks!
Perry
On Wed, May 6, 2020 at 12:17 PM George McCabe via RT
<met_help at ucar.edu>
wrote:
> Hi Perry,
>
> If you are processing a single day, you could loop by INIT and only
run the
> 6Z and 12Z init times:
>
> [config]
> LOOP_BY = INIT
>
> INIT_TIME_FMT = %Y%m%d%H
>
> INIT_BEG = 2017020106
>
> INIT_END = 2017020112
>
> INIT_INCREMENT = 6H
>
> LEAD_SEQ = begin_end_incr(0,48,1)
>
> However, if you need to process more than 1 day, there isn't
currently a
> good way to loop by INIT with uneven increments. There is
functionality we
> want to implement that involves specifying times to exclude, so if
that
> existed you could tell it to exclude 0H and 18H and it would skip
those
> times.
>
> If you are looping by VALID time, there is an INIT_SEQ variable that
can be
> used to build the forecast lead list dynamically for each valid time
based
> on the list of available init times. You will also have to set the
max
> forecast lead to use:
>
> LOOP_BY = VALID
>
> VALID_TIME_FMT = %Y%m%d%H
>
> VALID_BEG = 2017020100
>
> VALID_END = 2017020100
>
> VALID_INCREMENT = 6H
>
> INIT_SEQ = 6, 12
>
> LEAD_SEQ_MAX = 48
>
> The above configuration would produce the following (tested with
Example
> wrapper):
>
> Processing forecast lead 12 hours initialized at 2017-01-31 12Z and
valid
> at 2017-02-01 00Z
> Processing forecast lead 12 hours initialized at 2017-01-31 12Z and
valid
> at 2017-02-01 00Z
> Processing forecast lead 18 hours initialized at 2017-01-31 06Z and
valid
> at 2017-02-01 00Z
> Processing forecast lead 18 hours initialized at 2017-01-31 06Z and
valid
> at 2017-02-01 00Z
> Processing forecast lead 1 day 12 hours initialized at 2017-01-30
12Z and
> valid at 2017-02-01 00Z
> Processing forecast lead 1 day 12 hours initialized at 2017-01-30
12Z and
> valid at 2017-02-01 00Z
> Processing forecast lead 1 day 18 hours initialized at 2017-01-30
06Z and
> valid at 2017-02-01 00Z
> Processing forecast lead 1 day 18 hours initialized at 2017-01-30
06Z and
> valid at 2017-02-01 00Z
>
> More info on that here:
>
>
https://ncar.github.io/METplus/Users_Guide/systemconfiguration.html#init-
seq
>
> Let me know if either of those solutions would work for you. If not,
we can
> look closer at your case to see what would work.
>
> Thanks,
> George
>
> On Wed, May 6, 2020 at 9:25 AM perry.shafran at noaa.gov via RT <
> met_help at ucar.edu> wrote:
>
> >
> > Wed May 06 09:25:20 2020: Request 95177 was acted upon.
> > Transaction: Ticket created by perry.shafran at noaa.gov
> > Queue: met_help
> > Subject: Verifying models with few cycle times
> > Owner: Nobody
> > Requestors: perry.shafran at noaa.gov
> > Status: new
> > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=95177 >
> >
> >
> > Hi, there,
> >
> > I'd like to ask a quick question. What should be the METplus
> configuration
> > settings if we want to verify for every hour in a 48-hr forecast,
but
> there
> > are only two cycles of model runs: 06Z and 12Z? Right now I have
> METplus
> > cycling through so many different time combinations of files that
do not
> > exist, but that's because I am not too sure of how to set this
type of
> > thing. Can you assist me in figuring out the proper settings in
this
> > situation?
> >
> > Thanks!
> >
> > Perry
> >
> >
>
> --
> George McCabe - Software Engineer III
> National Center for Atmospheric Research
> Research Applications Laboratory
> 303-497-2768
> ---
> My working day may not be your working day. Please do not feel
obliged to
> reply to this email outside of your normal working hours.
>
>
------------------------------------------------
Subject: Verifying models with few cycle times
From: perry.shafran at noaa.gov
Time: Wed May 06 12:22:03 2020
Hi, George,
For the valid time case, I entered exactly what you had suggested, but
tried to change the valid beg and end times to 00Z to 03Z. However,
since
the VALID_INCREMENT is 6H, it did not produce any output for valid
times of
01, 02, or 03Z.
So then I set VALID_INCREMENT to 1H, and it did the same thing as it
did
originally, look for data that had cycle times of 15Z, 16Z, etc,
cycles
that did not have data. This was even as INIT_SEQ was set to 6, 12.
What I want is to verify data hourly, but we only have two cycles
worth of
data. So for example, each hour, here are the valid forecasts,
because we
only have initial times of 06Z and 12Z:
00Z - forecast hours valid: 12, 18, 36, 42
01Z - forecast hours valid: 13, 19, 37, 43
02Z - forecast hours valid: 14, 20, 38, 44
03Z - forecast hours valid: 15, 21, 39, 45
04Z - forecast hours valid: 16, 22, 40, 46
05Z - forecast hours valid: 17, 23, 41, 47
06Z - forecast hours valid: 18, 24, 42, 48
07Z - forecast hours valid: 01, 19, 25, 43
08Z - forecast hours valid: 02, 20, 26, 44
09Z - forecast hours valid: 03, 21, 27, 45
10Z - forecast hours valid: 04, 22, 28, 46
11Z - forecast hours valid: 05, 23, 29, 47
12Z - forecast hours valid: 06, 24, 30, 48
13Z - forecast hours valid: 01, 07, 25, 31
14Z - forecast hours valid: 02, 08, 26, 32
15Z - forecast hours valid: 03, 09, 27, 33
16Z - forecast hours valid: 04, 10, 28, 34
17Z - forecast hours valid: 05, 11, 29, 35
18Z - forecast hours valid: 06, 12, 30, 36
19Z - forecast hours valid: 07, 13, 31, 37
20Z - forecast hours valid: 08, 14, 32, 38
21Z - forecast hours valid: 09, 15, 33, 39
22Z - forecast hours valid: 10, 16, 34, 40
23Z - forecast hours valid: 11, 17, 36, 41
That means, for example, at valid time 12Z there should only be
forecast
hours (valid lead times) (from previous cycles) of 06hr, 24 hr, 30 hr,
and
48 hr.
Does this help you understand what we need?
Thanks!
Perry
On Wed, May 6, 2020 at 12:17 PM George McCabe via RT
<met_help at ucar.edu>
wrote:
> Hi Perry,
>
> If you are processing a single day, you could loop by INIT and only
run the
> 6Z and 12Z init times:
>
> [config]
> LOOP_BY = INIT
>
> INIT_TIME_FMT = %Y%m%d%H
>
> INIT_BEG = 2017020106
>
> INIT_END = 2017020112
>
> INIT_INCREMENT = 6H
>
> LEAD_SEQ = begin_end_incr(0,48,1)
>
> However, if you need to process more than 1 day, there isn't
currently a
> good way to loop by INIT with uneven increments. There is
functionality we
> want to implement that involves specifying times to exclude, so if
that
> existed you could tell it to exclude 0H and 18H and it would skip
those
> times.
>
> If you are looping by VALID time, there is an INIT_SEQ variable that
can be
> used to build the forecast lead list dynamically for each valid time
based
> on the list of available init times. You will also have to set the
max
> forecast lead to use:
>
> LOOP_BY = VALID
>
> VALID_TIME_FMT = %Y%m%d%H
>
> VALID_BEG = 2017020100
>
> VALID_END = 2017020100
>
> VALID_INCREMENT = 6H
>
> INIT_SEQ = 6, 12
>
> LEAD_SEQ_MAX = 48
>
> The above configuration would produce the following (tested with
Example
> wrapper):
>
> Processing forecast lead 12 hours initialized at 2017-01-31 12Z and
valid
> at 2017-02-01 00Z
> Processing forecast lead 12 hours initialized at 2017-01-31 12Z and
valid
> at 2017-02-01 00Z
> Processing forecast lead 18 hours initialized at 2017-01-31 06Z and
valid
> at 2017-02-01 00Z
> Processing forecast lead 18 hours initialized at 2017-01-31 06Z and
valid
> at 2017-02-01 00Z
> Processing forecast lead 1 day 12 hours initialized at 2017-01-30
12Z and
> valid at 2017-02-01 00Z
> Processing forecast lead 1 day 12 hours initialized at 2017-01-30
12Z and
> valid at 2017-02-01 00Z
> Processing forecast lead 1 day 18 hours initialized at 2017-01-30
06Z and
> valid at 2017-02-01 00Z
> Processing forecast lead 1 day 18 hours initialized at 2017-01-30
06Z and
> valid at 2017-02-01 00Z
>
> More info on that here:
>
>
https://ncar.github.io/METplus/Users_Guide/systemconfiguration.html#init-
seq
>
> Let me know if either of those solutions would work for you. If not,
we can
> look closer at your case to see what would work.
>
> Thanks,
> George
>
> On Wed, May 6, 2020 at 9:25 AM perry.shafran at noaa.gov via RT <
> met_help at ucar.edu> wrote:
>
> >
> > Wed May 06 09:25:20 2020: Request 95177 was acted upon.
> > Transaction: Ticket created by perry.shafran at noaa.gov
> > Queue: met_help
> > Subject: Verifying models with few cycle times
> > Owner: Nobody
> > Requestors: perry.shafran at noaa.gov
> > Status: new
> > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=95177 >
> >
> >
> > Hi, there,
> >
> > I'd like to ask a quick question. What should be the METplus
> configuration
> > settings if we want to verify for every hour in a 48-hr forecast,
but
> there
> > are only two cycles of model runs: 06Z and 12Z? Right now I have
> METplus
> > cycling through so many different time combinations of files that
do not
> > exist, but that's because I am not too sure of how to set this
type of
> > thing. Can you assist me in figuring out the proper settings in
this
> > situation?
> >
> > Thanks!
> >
> > Perry
> >
> >
>
> --
> George McCabe - Software Engineer III
> National Center for Atmospheric Research
> Research Applications Laboratory
> 303-497-2768
> ---
> My working day may not be your working day. Please do not feel
obliged to
> reply to this email outside of your normal working hours.
>
>
------------------------------------------------
Subject: Verifying models with few cycle times
From: George McCabe
Time: Wed May 06 12:26:21 2020
Hi Perry,
Do you also have LEAD_SEQ set in your config? The value from this
takes
precedence over INIT_SEQ, so you will have to remove or comment it
out.
On Wed, May 6, 2020 at 12:22 PM perry.shafran at noaa.gov via RT <
met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=95177 >
>
> Hi, George,
>
> For the valid time case, I entered exactly what you had suggested,
but
> tried to change the valid beg and end times to 00Z to 03Z. However,
since
> the VALID_INCREMENT is 6H, it did not produce any output for valid
times of
> 01, 02, or 03Z.
>
> So then I set VALID_INCREMENT to 1H, and it did the same thing as it
did
> originally, look for data that had cycle times of 15Z, 16Z, etc,
cycles
> that did not have data. This was even as INIT_SEQ was set to 6, 12.
>
> What I want is to verify data hourly, but we only have two cycles
worth of
> data. So for example, each hour, here are the valid forecasts,
because we
> only have initial times of 06Z and 12Z:
>
> 00Z - forecast hours valid: 12, 18, 36, 42
> 01Z - forecast hours valid: 13, 19, 37, 43
> 02Z - forecast hours valid: 14, 20, 38, 44
> 03Z - forecast hours valid: 15, 21, 39, 45
> 04Z - forecast hours valid: 16, 22, 40, 46
> 05Z - forecast hours valid: 17, 23, 41, 47
> 06Z - forecast hours valid: 18, 24, 42, 48
> 07Z - forecast hours valid: 01, 19, 25, 43
> 08Z - forecast hours valid: 02, 20, 26, 44
> 09Z - forecast hours valid: 03, 21, 27, 45
> 10Z - forecast hours valid: 04, 22, 28, 46
> 11Z - forecast hours valid: 05, 23, 29, 47
> 12Z - forecast hours valid: 06, 24, 30, 48
> 13Z - forecast hours valid: 01, 07, 25, 31
> 14Z - forecast hours valid: 02, 08, 26, 32
> 15Z - forecast hours valid: 03, 09, 27, 33
> 16Z - forecast hours valid: 04, 10, 28, 34
> 17Z - forecast hours valid: 05, 11, 29, 35
> 18Z - forecast hours valid: 06, 12, 30, 36
> 19Z - forecast hours valid: 07, 13, 31, 37
> 20Z - forecast hours valid: 08, 14, 32, 38
> 21Z - forecast hours valid: 09, 15, 33, 39
> 22Z - forecast hours valid: 10, 16, 34, 40
> 23Z - forecast hours valid: 11, 17, 36, 41
>
> That means, for example, at valid time 12Z there should only be
forecast
> hours (valid lead times) (from previous cycles) of 06hr, 24 hr, 30
hr, and
> 48 hr.
>
> Does this help you understand what we need?
>
> Thanks!
>
> Perry
>
>
> On Wed, May 6, 2020 at 12:17 PM George McCabe via RT
<met_help at ucar.edu>
> wrote:
>
> > Hi Perry,
> >
> > If you are processing a single day, you could loop by INIT and
only run
> the
> > 6Z and 12Z init times:
> >
> > [config]
> > LOOP_BY = INIT
> >
> > INIT_TIME_FMT = %Y%m%d%H
> >
> > INIT_BEG = 2017020106
> >
> > INIT_END = 2017020112
> >
> > INIT_INCREMENT = 6H
> >
> > LEAD_SEQ = begin_end_incr(0,48,1)
> >
> > However, if you need to process more than 1 day, there isn't
currently a
> > good way to loop by INIT with uneven increments. There is
functionality
> we
> > want to implement that involves specifying times to exclude, so if
that
> > existed you could tell it to exclude 0H and 18H and it would skip
those
> > times.
> >
> > If you are looping by VALID time, there is an INIT_SEQ variable
that can
> be
> > used to build the forecast lead list dynamically for each valid
time
> based
> > on the list of available init times. You will also have to set the
max
> > forecast lead to use:
> >
> > LOOP_BY = VALID
> >
> > VALID_TIME_FMT = %Y%m%d%H
> >
> > VALID_BEG = 2017020100
> >
> > VALID_END = 2017020100
> >
> > VALID_INCREMENT = 6H
> >
> > INIT_SEQ = 6, 12
> >
> > LEAD_SEQ_MAX = 48
> >
> > The above configuration would produce the following (tested with
Example
> > wrapper):
> >
> > Processing forecast lead 12 hours initialized at 2017-01-31 12Z
and valid
> > at 2017-02-01 00Z
> > Processing forecast lead 12 hours initialized at 2017-01-31 12Z
and valid
> > at 2017-02-01 00Z
> > Processing forecast lead 18 hours initialized at 2017-01-31 06Z
and valid
> > at 2017-02-01 00Z
> > Processing forecast lead 18 hours initialized at 2017-01-31 06Z
and valid
> > at 2017-02-01 00Z
> > Processing forecast lead 1 day 12 hours initialized at 2017-01-30
12Z and
> > valid at 2017-02-01 00Z
> > Processing forecast lead 1 day 12 hours initialized at 2017-01-30
12Z and
> > valid at 2017-02-01 00Z
> > Processing forecast lead 1 day 18 hours initialized at 2017-01-30
06Z and
> > valid at 2017-02-01 00Z
> > Processing forecast lead 1 day 18 hours initialized at 2017-01-30
06Z and
> > valid at 2017-02-01 00Z
> >
> > More info on that here:
> >
> >
>
https://ncar.github.io/METplus/Users_Guide/systemconfiguration.html#init-
seq
> >
> > Let me know if either of those solutions would work for you. If
not, we
> can
> > look closer at your case to see what would work.
> >
> > Thanks,
> > George
> >
> > On Wed, May 6, 2020 at 9:25 AM perry.shafran at noaa.gov via RT <
> > met_help at ucar.edu> wrote:
> >
> > >
> > > Wed May 06 09:25:20 2020: Request 95177 was acted upon.
> > > Transaction: Ticket created by perry.shafran at noaa.gov
> > > Queue: met_help
> > > Subject: Verifying models with few cycle times
> > > Owner: Nobody
> > > Requestors: perry.shafran at noaa.gov
> > > Status: new
> > > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=95177
> >
> > >
> > >
> > > Hi, there,
> > >
> > > I'd like to ask a quick question. What should be the METplus
> > configuration
> > > settings if we want to verify for every hour in a 48-hr
forecast, but
> > there
> > > are only two cycles of model runs: 06Z and 12Z? Right now I
have
> > METplus
> > > cycling through so many different time combinations of files
that do
> not
> > > exist, but that's because I am not too sure of how to set this
type of
> > > thing. Can you assist me in figuring out the proper settings in
this
> > > situation?
> > >
> > > Thanks!
> > >
> > > Perry
> > >
> > >
> >
> > --
> > George McCabe - Software Engineer III
> > National Center for Atmospheric Research
> > Research Applications Laboratory
> > 303-497-2768
> > ---
> > My working day may not be your working day. Please do not feel
obliged to
> > reply to this email outside of your normal working hours.
> >
> >
>
>
--
George McCabe - Software Engineer III
National Center for Atmospheric Research
Research Applications Laboratory
303-497-2768
---
My working day may not be your working day. Please do not feel obliged
to
reply to this email outside of your normal working hours.
------------------------------------------------
Subject: Verifying models with few cycle times
From: perry.shafran at noaa.gov
Time: Wed May 06 12:28:41 2020
I think I removed it, but let me try it again just to double check.
Perry
On Wed, May 6, 2020 at 2:26 PM George McCabe via RT
<met_help at ucar.edu>
wrote:
> Hi Perry,
>
> Do you also have LEAD_SEQ set in your config? The value from this
takes
> precedence over INIT_SEQ, so you will have to remove or comment it
out.
>
> On Wed, May 6, 2020 at 12:22 PM perry.shafran at noaa.gov via RT <
> met_help at ucar.edu> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=95177 >
> >
> > Hi, George,
> >
> > For the valid time case, I entered exactly what you had suggested,
but
> > tried to change the valid beg and end times to 00Z to 03Z.
However,
> since
> > the VALID_INCREMENT is 6H, it did not produce any output for valid
times
> of
> > 01, 02, or 03Z.
> >
> > So then I set VALID_INCREMENT to 1H, and it did the same thing as
it did
> > originally, look for data that had cycle times of 15Z, 16Z, etc,
cycles
> > that did not have data. This was even as INIT_SEQ was set to 6,
12.
> >
> > What I want is to verify data hourly, but we only have two cycles
worth
> of
> > data. So for example, each hour, here are the valid forecasts,
because
> we
> > only have initial times of 06Z and 12Z:
> >
> > 00Z - forecast hours valid: 12, 18, 36, 42
> > 01Z - forecast hours valid: 13, 19, 37, 43
> > 02Z - forecast hours valid: 14, 20, 38, 44
> > 03Z - forecast hours valid: 15, 21, 39, 45
> > 04Z - forecast hours valid: 16, 22, 40, 46
> > 05Z - forecast hours valid: 17, 23, 41, 47
> > 06Z - forecast hours valid: 18, 24, 42, 48
> > 07Z - forecast hours valid: 01, 19, 25, 43
> > 08Z - forecast hours valid: 02, 20, 26, 44
> > 09Z - forecast hours valid: 03, 21, 27, 45
> > 10Z - forecast hours valid: 04, 22, 28, 46
> > 11Z - forecast hours valid: 05, 23, 29, 47
> > 12Z - forecast hours valid: 06, 24, 30, 48
> > 13Z - forecast hours valid: 01, 07, 25, 31
> > 14Z - forecast hours valid: 02, 08, 26, 32
> > 15Z - forecast hours valid: 03, 09, 27, 33
> > 16Z - forecast hours valid: 04, 10, 28, 34
> > 17Z - forecast hours valid: 05, 11, 29, 35
> > 18Z - forecast hours valid: 06, 12, 30, 36
> > 19Z - forecast hours valid: 07, 13, 31, 37
> > 20Z - forecast hours valid: 08, 14, 32, 38
> > 21Z - forecast hours valid: 09, 15, 33, 39
> > 22Z - forecast hours valid: 10, 16, 34, 40
> > 23Z - forecast hours valid: 11, 17, 36, 41
> >
> > That means, for example, at valid time 12Z there should only be
forecast
> > hours (valid lead times) (from previous cycles) of 06hr, 24 hr, 30
hr,
> and
> > 48 hr.
> >
> > Does this help you understand what we need?
> >
> > Thanks!
> >
> > Perry
> >
> >
> > On Wed, May 6, 2020 at 12:17 PM George McCabe via RT
<met_help at ucar.edu>
> > wrote:
> >
> > > Hi Perry,
> > >
> > > If you are processing a single day, you could loop by INIT and
only run
> > the
> > > 6Z and 12Z init times:
> > >
> > > [config]
> > > LOOP_BY = INIT
> > >
> > > INIT_TIME_FMT = %Y%m%d%H
> > >
> > > INIT_BEG = 2017020106
> > >
> > > INIT_END = 2017020112
> > >
> > > INIT_INCREMENT = 6H
> > >
> > > LEAD_SEQ = begin_end_incr(0,48,1)
> > >
> > > However, if you need to process more than 1 day, there isn't
currently
> a
> > > good way to loop by INIT with uneven increments. There is
functionality
> > we
> > > want to implement that involves specifying times to exclude, so
if that
> > > existed you could tell it to exclude 0H and 18H and it would
skip those
> > > times.
> > >
> > > If you are looping by VALID time, there is an INIT_SEQ variable
that
> can
> > be
> > > used to build the forecast lead list dynamically for each valid
time
> > based
> > > on the list of available init times. You will also have to set
the max
> > > forecast lead to use:
> > >
> > > LOOP_BY = VALID
> > >
> > > VALID_TIME_FMT = %Y%m%d%H
> > >
> > > VALID_BEG = 2017020100
> > >
> > > VALID_END = 2017020100
> > >
> > > VALID_INCREMENT = 6H
> > >
> > > INIT_SEQ = 6, 12
> > >
> > > LEAD_SEQ_MAX = 48
> > >
> > > The above configuration would produce the following (tested with
> Example
> > > wrapper):
> > >
> > > Processing forecast lead 12 hours initialized at 2017-01-31 12Z
and
> valid
> > > at 2017-02-01 00Z
> > > Processing forecast lead 12 hours initialized at 2017-01-31 12Z
and
> valid
> > > at 2017-02-01 00Z
> > > Processing forecast lead 18 hours initialized at 2017-01-31 06Z
and
> valid
> > > at 2017-02-01 00Z
> > > Processing forecast lead 18 hours initialized at 2017-01-31 06Z
and
> valid
> > > at 2017-02-01 00Z
> > > Processing forecast lead 1 day 12 hours initialized at 2017-01-
30 12Z
> and
> > > valid at 2017-02-01 00Z
> > > Processing forecast lead 1 day 12 hours initialized at 2017-01-
30 12Z
> and
> > > valid at 2017-02-01 00Z
> > > Processing forecast lead 1 day 18 hours initialized at 2017-01-
30 06Z
> and
> > > valid at 2017-02-01 00Z
> > > Processing forecast lead 1 day 18 hours initialized at 2017-01-
30 06Z
> and
> > > valid at 2017-02-01 00Z
> > >
> > > More info on that here:
> > >
> > >
> >
>
https://ncar.github.io/METplus/Users_Guide/systemconfiguration.html#init-
seq
> > >
> > > Let me know if either of those solutions would work for you. If
not, we
> > can
> > > look closer at your case to see what would work.
> > >
> > > Thanks,
> > > George
> > >
> > > On Wed, May 6, 2020 at 9:25 AM perry.shafran at noaa.gov via RT <
> > > met_help at ucar.edu> wrote:
> > >
> > > >
> > > > Wed May 06 09:25:20 2020: Request 95177 was acted upon.
> > > > Transaction: Ticket created by perry.shafran at noaa.gov
> > > > Queue: met_help
> > > > Subject: Verifying models with few cycle times
> > > > Owner: Nobody
> > > > Requestors: perry.shafran at noaa.gov
> > > > Status: new
> > > > Ticket <URL:
> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=95177
> > >
> > > >
> > > >
> > > > Hi, there,
> > > >
> > > > I'd like to ask a quick question. What should be the METplus
> > > configuration
> > > > settings if we want to verify for every hour in a 48-hr
forecast, but
> > > there
> > > > are only two cycles of model runs: 06Z and 12Z? Right now I
have
> > > METplus
> > > > cycling through so many different time combinations of files
that do
> > not
> > > > exist, but that's because I am not too sure of how to set this
type
> of
> > > > thing. Can you assist me in figuring out the proper settings
in this
> > > > situation?
> > > >
> > > > Thanks!
> > > >
> > > > Perry
> > > >
> > > >
> > >
> > > --
> > > George McCabe - Software Engineer III
> > > National Center for Atmospheric Research
> > > Research Applications Laboratory
> > > 303-497-2768
> > > ---
> > > My working day may not be your working day. Please do not feel
obliged
> to
> > > reply to this email outside of your normal working hours.
> > >
> > >
> >
> >
>
> --
> George McCabe - Software Engineer III
> National Center for Atmospheric Research
> Research Applications Laboratory
> 303-497-2768
> ---
> My working day may not be your working day. Please do not feel
obliged to
> reply to this email outside of your normal working hours.
>
>
------------------------------------------------
Subject: Verifying models with few cycle times
From: perry.shafran at noaa.gov
Time: Thu May 07 07:00:08 2020
Hi, George,
Yes, I think I got that to work!
Thanks!
Perry
On Wed, May 6, 2020 at 2:28 PM Perry Shafran - NOAA Affiliate <
perry.shafran at noaa.gov> wrote:
> I think I removed it, but let me try it again just to double check.
>
> Perry
>
> On Wed, May 6, 2020 at 2:26 PM George McCabe via RT
<met_help at ucar.edu>
> wrote:
>
>> Hi Perry,
>>
>> Do you also have LEAD_SEQ set in your config? The value from this
takes
>> precedence over INIT_SEQ, so you will have to remove or comment it
out.
>>
>> On Wed, May 6, 2020 at 12:22 PM perry.shafran at noaa.gov via RT <
>> met_help at ucar.edu> wrote:
>>
>> >
>> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=95177 >
>> >
>> > Hi, George,
>> >
>> > For the valid time case, I entered exactly what you had
suggested, but
>> > tried to change the valid beg and end times to 00Z to 03Z.
However,
>> since
>> > the VALID_INCREMENT is 6H, it did not produce any output for
valid
>> times of
>> > 01, 02, or 03Z.
>> >
>> > So then I set VALID_INCREMENT to 1H, and it did the same thing as
it did
>> > originally, look for data that had cycle times of 15Z, 16Z, etc,
cycles
>> > that did not have data. This was even as INIT_SEQ was set to 6,
12.
>> >
>> > What I want is to verify data hourly, but we only have two cycles
worth
>> of
>> > data. So for example, each hour, here are the valid forecasts,
because
>> we
>> > only have initial times of 06Z and 12Z:
>> >
>> > 00Z - forecast hours valid: 12, 18, 36, 42
>> > 01Z - forecast hours valid: 13, 19, 37, 43
>> > 02Z - forecast hours valid: 14, 20, 38, 44
>> > 03Z - forecast hours valid: 15, 21, 39, 45
>> > 04Z - forecast hours valid: 16, 22, 40, 46
>> > 05Z - forecast hours valid: 17, 23, 41, 47
>> > 06Z - forecast hours valid: 18, 24, 42, 48
>> > 07Z - forecast hours valid: 01, 19, 25, 43
>> > 08Z - forecast hours valid: 02, 20, 26, 44
>> > 09Z - forecast hours valid: 03, 21, 27, 45
>> > 10Z - forecast hours valid: 04, 22, 28, 46
>> > 11Z - forecast hours valid: 05, 23, 29, 47
>> > 12Z - forecast hours valid: 06, 24, 30, 48
>> > 13Z - forecast hours valid: 01, 07, 25, 31
>> > 14Z - forecast hours valid: 02, 08, 26, 32
>> > 15Z - forecast hours valid: 03, 09, 27, 33
>> > 16Z - forecast hours valid: 04, 10, 28, 34
>> > 17Z - forecast hours valid: 05, 11, 29, 35
>> > 18Z - forecast hours valid: 06, 12, 30, 36
>> > 19Z - forecast hours valid: 07, 13, 31, 37
>> > 20Z - forecast hours valid: 08, 14, 32, 38
>> > 21Z - forecast hours valid: 09, 15, 33, 39
>> > 22Z - forecast hours valid: 10, 16, 34, 40
>> > 23Z - forecast hours valid: 11, 17, 36, 41
>> >
>> > That means, for example, at valid time 12Z there should only be
forecast
>> > hours (valid lead times) (from previous cycles) of 06hr, 24 hr,
30 hr,
>> and
>> > 48 hr.
>> >
>> > Does this help you understand what we need?
>> >
>> > Thanks!
>> >
>> > Perry
>> >
>> >
>> > On Wed, May 6, 2020 at 12:17 PM George McCabe via RT
<met_help at ucar.edu
>> >
>> > wrote:
>> >
>> > > Hi Perry,
>> > >
>> > > If you are processing a single day, you could loop by INIT and
only
>> run
>> > the
>> > > 6Z and 12Z init times:
>> > >
>> > > [config]
>> > > LOOP_BY = INIT
>> > >
>> > > INIT_TIME_FMT = %Y%m%d%H
>> > >
>> > > INIT_BEG = 2017020106
>> > >
>> > > INIT_END = 2017020112
>> > >
>> > > INIT_INCREMENT = 6H
>> > >
>> > > LEAD_SEQ = begin_end_incr(0,48,1)
>> > >
>> > > However, if you need to process more than 1 day, there isn't
>> currently a
>> > > good way to loop by INIT with uneven increments. There is
>> functionality
>> > we
>> > > want to implement that involves specifying times to exclude, so
if
>> that
>> > > existed you could tell it to exclude 0H and 18H and it would
skip
>> those
>> > > times.
>> > >
>> > > If you are looping by VALID time, there is an INIT_SEQ variable
that
>> can
>> > be
>> > > used to build the forecast lead list dynamically for each valid
time
>> > based
>> > > on the list of available init times. You will also have to set
the max
>> > > forecast lead to use:
>> > >
>> > > LOOP_BY = VALID
>> > >
>> > > VALID_TIME_FMT = %Y%m%d%H
>> > >
>> > > VALID_BEG = 2017020100
>> > >
>> > > VALID_END = 2017020100
>> > >
>> > > VALID_INCREMENT = 6H
>> > >
>> > > INIT_SEQ = 6, 12
>> > >
>> > > LEAD_SEQ_MAX = 48
>> > >
>> > > The above configuration would produce the following (tested
with
>> Example
>> > > wrapper):
>> > >
>> > > Processing forecast lead 12 hours initialized at 2017-01-31 12Z
and
>> valid
>> > > at 2017-02-01 00Z
>> > > Processing forecast lead 12 hours initialized at 2017-01-31 12Z
and
>> valid
>> > > at 2017-02-01 00Z
>> > > Processing forecast lead 18 hours initialized at 2017-01-31 06Z
and
>> valid
>> > > at 2017-02-01 00Z
>> > > Processing forecast lead 18 hours initialized at 2017-01-31 06Z
and
>> valid
>> > > at 2017-02-01 00Z
>> > > Processing forecast lead 1 day 12 hours initialized at 2017-01-
30 12Z
>> and
>> > > valid at 2017-02-01 00Z
>> > > Processing forecast lead 1 day 12 hours initialized at 2017-01-
30 12Z
>> and
>> > > valid at 2017-02-01 00Z
>> > > Processing forecast lead 1 day 18 hours initialized at 2017-01-
30 06Z
>> and
>> > > valid at 2017-02-01 00Z
>> > > Processing forecast lead 1 day 18 hours initialized at 2017-01-
30 06Z
>> and
>> > > valid at 2017-02-01 00Z
>> > >
>> > > More info on that here:
>> > >
>> > >
>> >
>>
https://ncar.github.io/METplus/Users_Guide/systemconfiguration.html#init-
seq
>> > >
>> > > Let me know if either of those solutions would work for you. If
not,
>> we
>> > can
>> > > look closer at your case to see what would work.
>> > >
>> > > Thanks,
>> > > George
>> > >
>> > > On Wed, May 6, 2020 at 9:25 AM perry.shafran at noaa.gov via RT <
>> > > met_help at ucar.edu> wrote:
>> > >
>> > > >
>> > > > Wed May 06 09:25:20 2020: Request 95177 was acted upon.
>> > > > Transaction: Ticket created by perry.shafran at noaa.gov
>> > > > Queue: met_help
>> > > > Subject: Verifying models with few cycle times
>> > > > Owner: Nobody
>> > > > Requestors: perry.shafran at noaa.gov
>> > > > Status: new
>> > > > Ticket <URL:
>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=95177
>> > >
>> > > >
>> > > >
>> > > > Hi, there,
>> > > >
>> > > > I'd like to ask a quick question. What should be the METplus
>> > > configuration
>> > > > settings if we want to verify for every hour in a 48-hr
forecast,
>> but
>> > > there
>> > > > are only two cycles of model runs: 06Z and 12Z? Right now I
have
>> > > METplus
>> > > > cycling through so many different time combinations of files
that do
>> > not
>> > > > exist, but that's because I am not too sure of how to set
this type
>> of
>> > > > thing. Can you assist me in figuring out the proper settings
in
>> this
>> > > > situation?
>> > > >
>> > > > Thanks!
>> > > >
>> > > > Perry
>> > > >
>> > > >
>> > >
>> > > --
>> > > George McCabe - Software Engineer III
>> > > National Center for Atmospheric Research
>> > > Research Applications Laboratory
>> > > 303-497-2768
>> > > ---
>> > > My working day may not be your working day. Please do not feel
>> obliged to
>> > > reply to this email outside of your normal working hours.
>> > >
>> > >
>> >
>> >
>>
>> --
>> George McCabe - Software Engineer III
>> National Center for Atmospheric Research
>> Research Applications Laboratory
>> 303-497-2768
>> ---
>> My working day may not be your working day. Please do not feel
obliged to
>> reply to this email outside of your normal working hours.
>>
>>
------------------------------------------------
Subject: Verifying models with few cycle times
From: George McCabe
Time: Thu May 07 11:25:47 2020
Great to hear! A warning if both INIT_SEQ and LEAD_SEQ are set would
probably be useful. I'll make a note of that. I will close this
ticket, but
feel free to open a new one if you have any other issues.
- George
On Thu, May 7, 2020 at 7:00 AM perry.shafran at noaa.gov via RT <
met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=95177 >
>
> Hi, George,
>
> Yes, I think I got that to work!
>
> Thanks!
>
> Perry
>
> On Wed, May 6, 2020 at 2:28 PM Perry Shafran - NOAA Affiliate <
> perry.shafran at noaa.gov> wrote:
>
> > I think I removed it, but let me try it again just to double
check.
> >
> > Perry
> >
> > On Wed, May 6, 2020 at 2:26 PM George McCabe via RT
<met_help at ucar.edu>
> > wrote:
> >
> >> Hi Perry,
> >>
> >> Do you also have LEAD_SEQ set in your config? The value from this
takes
> >> precedence over INIT_SEQ, so you will have to remove or comment
it out.
> >>
> >> On Wed, May 6, 2020 at 12:22 PM perry.shafran at noaa.gov via RT <
> >> met_help at ucar.edu> wrote:
> >>
> >> >
> >> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=95177 >
> >> >
> >> > Hi, George,
> >> >
> >> > For the valid time case, I entered exactly what you had
suggested, but
> >> > tried to change the valid beg and end times to 00Z to 03Z.
However,
> >> since
> >> > the VALID_INCREMENT is 6H, it did not produce any output for
valid
> >> times of
> >> > 01, 02, or 03Z.
> >> >
> >> > So then I set VALID_INCREMENT to 1H, and it did the same thing
as it
> did
> >> > originally, look for data that had cycle times of 15Z, 16Z,
etc,
> cycles
> >> > that did not have data. This was even as INIT_SEQ was set to
6, 12.
> >> >
> >> > What I want is to verify data hourly, but we only have two
cycles
> worth
> >> of
> >> > data. So for example, each hour, here are the valid forecasts,
> because
> >> we
> >> > only have initial times of 06Z and 12Z:
> >> >
> >> > 00Z - forecast hours valid: 12, 18, 36, 42
> >> > 01Z - forecast hours valid: 13, 19, 37, 43
> >> > 02Z - forecast hours valid: 14, 20, 38, 44
> >> > 03Z - forecast hours valid: 15, 21, 39, 45
> >> > 04Z - forecast hours valid: 16, 22, 40, 46
> >> > 05Z - forecast hours valid: 17, 23, 41, 47
> >> > 06Z - forecast hours valid: 18, 24, 42, 48
> >> > 07Z - forecast hours valid: 01, 19, 25, 43
> >> > 08Z - forecast hours valid: 02, 20, 26, 44
> >> > 09Z - forecast hours valid: 03, 21, 27, 45
> >> > 10Z - forecast hours valid: 04, 22, 28, 46
> >> > 11Z - forecast hours valid: 05, 23, 29, 47
> >> > 12Z - forecast hours valid: 06, 24, 30, 48
> >> > 13Z - forecast hours valid: 01, 07, 25, 31
> >> > 14Z - forecast hours valid: 02, 08, 26, 32
> >> > 15Z - forecast hours valid: 03, 09, 27, 33
> >> > 16Z - forecast hours valid: 04, 10, 28, 34
> >> > 17Z - forecast hours valid: 05, 11, 29, 35
> >> > 18Z - forecast hours valid: 06, 12, 30, 36
> >> > 19Z - forecast hours valid: 07, 13, 31, 37
> >> > 20Z - forecast hours valid: 08, 14, 32, 38
> >> > 21Z - forecast hours valid: 09, 15, 33, 39
> >> > 22Z - forecast hours valid: 10, 16, 34, 40
> >> > 23Z - forecast hours valid: 11, 17, 36, 41
> >> >
> >> > That means, for example, at valid time 12Z there should only be
> forecast
> >> > hours (valid lead times) (from previous cycles) of 06hr, 24 hr,
30 hr,
> >> and
> >> > 48 hr.
> >> >
> >> > Does this help you understand what we need?
> >> >
> >> > Thanks!
> >> >
> >> > Perry
> >> >
> >> >
> >> > On Wed, May 6, 2020 at 12:17 PM George McCabe via RT <
> met_help at ucar.edu
> >> >
> >> > wrote:
> >> >
> >> > > Hi Perry,
> >> > >
> >> > > If you are processing a single day, you could loop by INIT
and only
> >> run
> >> > the
> >> > > 6Z and 12Z init times:
> >> > >
> >> > > [config]
> >> > > LOOP_BY = INIT
> >> > >
> >> > > INIT_TIME_FMT = %Y%m%d%H
> >> > >
> >> > > INIT_BEG = 2017020106
> >> > >
> >> > > INIT_END = 2017020112
> >> > >
> >> > > INIT_INCREMENT = 6H
> >> > >
> >> > > LEAD_SEQ = begin_end_incr(0,48,1)
> >> > >
> >> > > However, if you need to process more than 1 day, there isn't
> >> currently a
> >> > > good way to loop by INIT with uneven increments. There is
> >> functionality
> >> > we
> >> > > want to implement that involves specifying times to exclude,
so if
> >> that
> >> > > existed you could tell it to exclude 0H and 18H and it would
skip
> >> those
> >> > > times.
> >> > >
> >> > > If you are looping by VALID time, there is an INIT_SEQ
variable that
> >> can
> >> > be
> >> > > used to build the forecast lead list dynamically for each
valid time
> >> > based
> >> > > on the list of available init times. You will also have to
set the
> max
> >> > > forecast lead to use:
> >> > >
> >> > > LOOP_BY = VALID
> >> > >
> >> > > VALID_TIME_FMT = %Y%m%d%H
> >> > >
> >> > > VALID_BEG = 2017020100
> >> > >
> >> > > VALID_END = 2017020100
> >> > >
> >> > > VALID_INCREMENT = 6H
> >> > >
> >> > > INIT_SEQ = 6, 12
> >> > >
> >> > > LEAD_SEQ_MAX = 48
> >> > >
> >> > > The above configuration would produce the following (tested
with
> >> Example
> >> > > wrapper):
> >> > >
> >> > > Processing forecast lead 12 hours initialized at 2017-01-31
12Z and
> >> valid
> >> > > at 2017-02-01 00Z
> >> > > Processing forecast lead 12 hours initialized at 2017-01-31
12Z and
> >> valid
> >> > > at 2017-02-01 00Z
> >> > > Processing forecast lead 18 hours initialized at 2017-01-31
06Z and
> >> valid
> >> > > at 2017-02-01 00Z
> >> > > Processing forecast lead 18 hours initialized at 2017-01-31
06Z and
> >> valid
> >> > > at 2017-02-01 00Z
> >> > > Processing forecast lead 1 day 12 hours initialized at 2017-
01-30
> 12Z
> >> and
> >> > > valid at 2017-02-01 00Z
> >> > > Processing forecast lead 1 day 12 hours initialized at 2017-
01-30
> 12Z
> >> and
> >> > > valid at 2017-02-01 00Z
> >> > > Processing forecast lead 1 day 18 hours initialized at 2017-
01-30
> 06Z
> >> and
> >> > > valid at 2017-02-01 00Z
> >> > > Processing forecast lead 1 day 18 hours initialized at 2017-
01-30
> 06Z
> >> and
> >> > > valid at 2017-02-01 00Z
> >> > >
> >> > > More info on that here:
> >> > >
> >> > >
> >> >
> >>
>
https://ncar.github.io/METplus/Users_Guide/systemconfiguration.html#init-
seq
> >> > >
> >> > > Let me know if either of those solutions would work for you.
If not,
> >> we
> >> > can
> >> > > look closer at your case to see what would work.
> >> > >
> >> > > Thanks,
> >> > > George
> >> > >
> >> > > On Wed, May 6, 2020 at 9:25 AM perry.shafran at noaa.gov via RT
<
> >> > > met_help at ucar.edu> wrote:
> >> > >
> >> > > >
> >> > > > Wed May 06 09:25:20 2020: Request 95177 was acted upon.
> >> > > > Transaction: Ticket created by perry.shafran at noaa.gov
> >> > > > Queue: met_help
> >> > > > Subject: Verifying models with few cycle times
> >> > > > Owner: Nobody
> >> > > > Requestors: perry.shafran at noaa.gov
> >> > > > Status: new
> >> > > > Ticket <URL:
> >> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=95177
> >> > >
> >> > > >
> >> > > >
> >> > > > Hi, there,
> >> > > >
> >> > > > I'd like to ask a quick question. What should be the
METplus
> >> > > configuration
> >> > > > settings if we want to verify for every hour in a 48-hr
forecast,
> >> but
> >> > > there
> >> > > > are only two cycles of model runs: 06Z and 12Z? Right now
I have
> >> > > METplus
> >> > > > cycling through so many different time combinations of
files that
> do
> >> > not
> >> > > > exist, but that's because I am not too sure of how to set
this
> type
> >> of
> >> > > > thing. Can you assist me in figuring out the proper
settings in
> >> this
> >> > > > situation?
> >> > > >
> >> > > > Thanks!
> >> > > >
> >> > > > Perry
> >> > > >
> >> > > >
> >> > >
> >> > > --
> >> > > George McCabe - Software Engineer III
> >> > > National Center for Atmospheric Research
> >> > > Research Applications Laboratory
> >> > > 303-497-2768
> >> > > ---
> >> > > My working day may not be your working day. Please do not
feel
> >> obliged to
> >> > > reply to this email outside of your normal working hours.
> >> > >
> >> > >
> >> >
> >> >
> >>
> >> --
> >> George McCabe - Software Engineer III
> >> National Center for Atmospheric Research
> >> Research Applications Laboratory
> >> 303-497-2768
> >> ---
> >> My working day may not be your working day. Please do not feel
obliged
> to
> >> reply to this email outside of your normal working hours.
> >>
> >>
>
>
--
George McCabe - Software Engineer III
National Center for Atmospheric Research
Research Applications Laboratory
303-497-2768
---
My working day may not be your working day. Please do not feel obliged
to
reply to this email outside of your normal working hours.
------------------------------------------------
More information about the Met_help
mailing list