[Met_help] [rt.rap.ucar.edu #86346] History for tc_pairs for multiple storms

[John Halley Gotway via RT]

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

Hello,

I had run tc_pairs for individual storms. Now, I will like to combine all
the storms and created a track error figure from all of them.

Each individual storm have their own "aal.dat" and "bal.dat" filles as well
as the "tcst". Each storm is initialized on different days staring on 0z or
12z  with a lead time forecast of 120 hours.

Attempting on this, I combined all the ".tcst" into one (using "cat"
command) and did the plot using the new file with all the storms in it.


I had also tried using the configure file to create the ".tcst" with all
the storms in there.  Since I have multiple "aal.dat" and "bal.dat" files I
tried *.dat and specified the storm names in the configure file.

This is the command I ran attempting this:
tc_pairs -adeck
/mnt/lfs3/projects/gsd-fv3-hfip/Keren.Rosado/MET/data_Matthew_GF/tutorial/tc_data/aal*.dat
-bdeck
/mnt/lfs3/projects/gsd-fv3-hfip/Keren.Rosado/MET/data_Matthew_GF/tutorial/tc_data/bal*
-config
/mnt/lfs3/projects/gsd-fv3-hfip/Keren.Rosado/MET/tutorial_Matthew_GF/config/TCPairsConfig
-out
/mnt/lfs3/projects/gsd-fv3-hfip/Keren.Rosado/MET/tutorial_Matthew_GF/out/tc_pairs/tc_pairs_test

This approach also give me exactly the same results.

Are this two approaches correct to do what I want ? or is there another way
of doing it?


Thanks,

Keren


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

Subject: tc_pairs for multiple storms
From: John Halley Gotway
Time: Wed Jul 25 15:18:41 2018

Keren,

I see you have a question about setting up logic for running the MET-
TC
tools.  You really have a lot of flexibility to set it up how you'd
like.
There's no reason why all of the output needs to go into the same
.tcst
output.  The tc_stat and plot_tcmpr.R script can process multiple
.tcst
files at the same time.

In our use of MET-TC within the DTC, we usually:
(1) For each aal*.dat file, identify the corresponding bal*.dat file.
(2) Run tc_pairs once for each pair of aal/bal files.
(3) Run tc_stat and/or the plot_tcmpr.R script to process multiple
.tcst
output files at the same time.

Both the tc_stat and plot_tcmpr.R script use the "-lookin" command
line
option to specify the .tcst files to be read.  Let's say you have 3
output
files in a directory named "out": out/a.tcst, out/b.tcst, and
out/c.tcst.
The following 4 settings would all do the same thing:

(1) List the top-level directory and search recursively for .tcst
file:
-lookin out [ONLY WORKS IN TC-STAT, NOT PLOT_TCMPR.R]
(2) Use the -lookin option once to list multiple files: -lookin
out/a.tcst
out/b.tcst out/c.tcst
(3) Same thing but with wildcards: -lookin out/*.tcst
(4) Use the -lookin option multiple times: -lookin out/a.tcst -lookin
out/b.tcst -lookin out/c.tcst

So yes, you can definitely process multiple .tcst files in a single
call.

Hope that helps clarify.

Thanks,
John

On Wed, Jul 25, 2018 at 2:19 PM Keren Rosado - NOAA Affiliate via RT <
met_help at ucar.edu> wrote:

>
> Wed Jul 25 14:19:44 2018: Request 86346 was acted upon.
> Transaction: Ticket created by keren.rosado at noaa.gov
>        Queue: met_help
>      Subject: tc_pairs for multiple storms
>        Owner: Nobody
>   Requestors: keren.rosado at noaa.gov
>       Status: new
>  Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=86346 >
>
>
> Hello,
>
> I had run tc_pairs for individual storms. Now, I will like to
combine all
> the storms and created a track error figure from all of them.
>
> Each individual storm have their own "aal.dat" and "bal.dat" filles
as well
> as the "tcst". Each storm is initialized on different days staring
on 0z or
> 12z  with a lead time forecast of 120 hours.
>
> Attempting on this, I combined all the ".tcst" into one (using "cat"
> command) and did the plot using the new file with all the storms in
it.
>
>
> I had also tried using the configure file to create the ".tcst" with
all
> the storms in there.  Since I have multiple "aal.dat" and "bal.dat"
files I
> tried *.dat and specified the storm names in the configure file.
>
> This is the command I ran attempting this:
> tc_pairs -adeck
>
> /mnt/lfs3/projects/gsd-fv3-
hfip/Keren.Rosado/MET/data_Matthew_GF/tutorial/tc_data/aal*.dat
> -bdeck
>
> /mnt/lfs3/projects/gsd-fv3-
hfip/Keren.Rosado/MET/data_Matthew_GF/tutorial/tc_data/bal*
> -config
>
> /mnt/lfs3/projects/gsd-fv3-
hfip/Keren.Rosado/MET/tutorial_Matthew_GF/config/TCPairsConfig
> -out
>
> /mnt/lfs3/projects/gsd-fv3-
hfip/Keren.Rosado/MET/tutorial_Matthew_GF/out/tc_pairs/tc_pairs_test
>
> This approach also give me exactly the same results.
>
> Are this two approaches correct to do what I want ? or is there
another way
> of doing it?
>
>
> Thanks,
>
> Keren
>
>

------------------------------------------------
Subject: tc_pairs for multiple storms
From: Keren Rosado - NOAA Affiliate
Time: Wed Jul 25 15:38:42 2018

Thanks.

Keren

----
Keren Rosado, Ph.D.
NCAS-M Postdoctoral Fellow
NOAA ESRL
Office: (303) 497-5907

On Wed, Jul 25, 2018 at 3:18 PM, John Halley Gotway via RT <
met_help at ucar.edu> wrote:

> Keren,
>
> I see you have a question about setting up logic for running the
MET-TC
> tools.  You really have a lot of flexibility to set it up how you'd
like.
> There's no reason why all of the output needs to go into the same
.tcst
> output.  The tc_stat and plot_tcmpr.R script can process multiple
.tcst
> files at the same time.
>
> In our use of MET-TC within the DTC, we usually:
> (1) For each aal*.dat file, identify the corresponding bal*.dat
file.
> (2) Run tc_pairs once for each pair of aal/bal files.
> (3) Run tc_stat and/or the plot_tcmpr.R script to process multiple
.tcst
> output files at the same time.
>
> Both the tc_stat and plot_tcmpr.R script use the "-lookin" command
line
> option to specify the .tcst files to be read.  Let's say you have 3
output
> files in a directory named "out": out/a.tcst, out/b.tcst, and
out/c.tcst.
> The following 4 settings would all do the same thing:
>
> (1) List the top-level directory and search recursively for .tcst
file:
> -lookin out [ONLY WORKS IN TC-STAT, NOT PLOT_TCMPR.R]
> (2) Use the -lookin option once to list multiple files: -lookin
out/a.tcst
> out/b.tcst out/c.tcst
> (3) Same thing but with wildcards: -lookin out/*.tcst
> (4) Use the -lookin option multiple times: -lookin out/a.tcst
-lookin
> out/b.tcst -lookin out/c.tcst
>
> So yes, you can definitely process multiple .tcst files in a single
call.
>
> Hope that helps clarify.
>
> Thanks,
> John
>
> On Wed, Jul 25, 2018 at 2:19 PM Keren Rosado - NOAA Affiliate via RT
<
> met_help at ucar.edu> wrote:
>
> >
> > Wed Jul 25 14:19:44 2018: Request 86346 was acted upon.
> > Transaction: Ticket created by keren.rosado at noaa.gov
> >        Queue: met_help
> >      Subject: tc_pairs for multiple storms
> >        Owner: Nobody
> >   Requestors: keren.rosado at noaa.gov
> >       Status: new
> >  Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=86346 >
> >
> >
> > Hello,
> >
> > I had run tc_pairs for individual storms. Now, I will like to
combine all
> > the storms and created a track error figure from all of them.
> >
> > Each individual storm have their own "aal.dat" and "bal.dat"
filles as
> well
> > as the "tcst". Each storm is initialized on different days staring
on 0z
> or
> > 12z  with a lead time forecast of 120 hours.
> >
> > Attempting on this, I combined all the ".tcst" into one (using
"cat"
> > command) and did the plot using the new file with all the storms
in it.
> >
> >
> > I had also tried using the configure file to create the ".tcst"
with all
> > the storms in there.  Since I have multiple "aal.dat" and
"bal.dat"
> files I
> > tried *.dat and specified the storm names in the configure file.
> >
> > This is the command I ran attempting this:
> > tc_pairs -adeck
> >
> > /mnt/lfs3/projects/gsd-fv3-hfip/Keren.Rosado/MET/data_
> Matthew_GF/tutorial/tc_data/aal*.dat
> > -bdeck
> >
> > /mnt/lfs3/projects/gsd-fv3-hfip/Keren.Rosado/MET/data_
> Matthew_GF/tutorial/tc_data/bal*
> > -config
> >
> > /mnt/lfs3/projects/gsd-fv3-hfip/Keren.Rosado/MET/
> tutorial_Matthew_GF/config/TCPairsConfig
> > -out
> >
> > /mnt/lfs3/projects/gsd-fv3-hfip/Keren.Rosado/MET/
> tutorial_Matthew_GF/out/tc_pairs/tc_pairs_test
> >
> > This approach also give me exactly the same results.
> >
> > Are this two approaches correct to do what I want ? or is there
another
> way
> > of doing it?
> >
> >
> > Thanks,
> >
> > Keren
> >
> >
>
>

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