[Met_help] [rt.rap.ucar.edu #88209] History for Question about Mode Time-Domain 2D output files

[John Halley Gotway via RT]

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

Hello,

Before I start, I want to thank the met_help office (especially John 
Gotway) for the quick help over the last few weeks with my many MTD 
questions. Hopefully this is my final question. This question is 
regarding the 2d output files for MTD (mtd_..._2d.txt).

Does the 2d output file contain attribute information for the object 
_clusters_, in addition to the attributes for the simple objects? I 
noticed that the 2d output file follows the format of first outputting 
lines for each simple forecast object (at each valid time), followed by 
outputting lines for each simple observed object. However, I also 
noticed that after these two blocks, two more blocks (again with 
forecast objects first followed by observation objects) occur where the 
object_id always matches the object_cat. When adding up the area in the 
simple objects (first two blocks), they match the total area for the 
corresponding objects in the last two blocks. Thus, I believe these 
attributes in the final two blocks are for the observed and forecast 
clusters. However, it is a little confusing to me because these blocks 
still show object_ids such as F001 and O002 which indicates simple 
objects. Thus, I wanted to confirm this information with you.

I have attached an example of my 2d output file. The block of interest 
that I believe corresponds to the forecast cluster attributes starts on 
line 765.

Thanks!
Sam Degelia



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

Subject: Question about Mode Time-Domain 2D output files
From: Minna Win
Time: Mon Dec 17 16:37:47 2018

Hello,
It looks like you have some questions about Mode Time Domain.  John is
out
of the office for the next two weeks, but I am going to see if our
other
resident MODE/MTD expert, Randy Bullock can assist you with your
questions.

Thanks for your patience.

Regards,
Minna
---------------
Minna Win
NCAR
Research Applications Lab
Phone: 303-497-8423
Fax:   303-497-8401



On Mon, Dec 17, 2018 at 10:37 PM Samuel Degelia via RT
<met_help at ucar.edu>
wrote:

>
> Mon Dec 17 15:37:42 2018: Request 88209 was acted upon.
> Transaction: Ticket created by sdegelia at gmail.com
>        Queue: met_help
>      Subject: Question about Mode Time-Domain 2D output files
>        Owner: Nobody
>   Requestors: sdegelia at gmail.com
>       Status: new
>  Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=88209 >
>
>
> Hello,
>
> Before I start, I want to thank the met_help office (especially John
> Gotway) for the quick help over the last few weeks with my many MTD
> questions. Hopefully this is my final question. This question is
> regarding the 2d output files for MTD (mtd_..._2d.txt).
>
> Does the 2d output file contain attribute information for the object
> _clusters_, in addition to the attributes for the simple objects? I
> noticed that the 2d output file follows the format of first
outputting
> lines for each simple forecast object (at each valid time), followed
by
> outputting lines for each simple observed object. However, I also
> noticed that after these two blocks, two more blocks (again with
> forecast objects first followed by observation objects) occur where
the
> object_id always matches the object_cat. When adding up the area in
the
> simple objects (first two blocks), they match the total area for the
> corresponding objects in the last two blocks. Thus, I believe these
> attributes in the final two blocks are for the observed and forecast
> clusters. However, it is a little confusing to me because these
blocks
> still show object_ids such as F001 and O002 which indicates simple
> objects. Thus, I wanted to confirm this information with you.
>
> I have attached an example of my 2d output file. The block of
interest
> that I believe corresponds to the forecast cluster attributes starts
on
> line 765.
>
> Thanks!
> Sam Degelia
>
>
>

------------------------------------------------
Subject: Question about Mode Time-Domain 2D output files
From: Randy Bullock
Time: Thu Dec 20 09:46:13 2018

Hello Samuel -

You are correct in supposing that the final two blocks in the 2D text
output file are information for object clusters.  In addition, I agree
with your statement that listing simple object_id's for cluster
objects is confusing.  I'm not sure what else we can do though,
really.  The mapping, if you will, from simple objects to cluster
objects is many-to-one.  That is, there are usually many simple
objects that make up a single cluster object.  That means that for any
simple object, the question "What cluster object does this correspond
to, if any?" has a nice, simple answer.  But asking "What simple
objects went into making this cluster object?" has a complicated
answer, usually a list of simple object numbers, which we can't really
put into a single output column in the 2D text output table.  Another
option would be to simply put "NA" in the object_id column, I guess.

The whole issue of MTD output, what with 3D vs. 2D, Fcst vs. Obs,
Simple vs. cluster, Single vs pair, is so complex that we spent a lot
of time trying to figure out how to write out the information in a
clear way.  It may well be that we were less that 100% successful in
that.  Any suggestions you might have along those lines would be
appreciated.

On Mon Dec 17 15:37:42 2018, sdegelia at gmail.com wrote:
> Hello,
>
> Before I start, I want to thank the met_help office (especially John
> Gotway) for the quick help over the last few weeks with my many MTD
> questions. Hopefully this is my final question. This question is
> regarding the 2d output files for MTD (mtd_..._2d.txt).
>
> Does the 2d output file contain attribute information for the object
> _clusters_, in addition to the attributes for the simple objects? I
> noticed that the 2d output file follows the format of first
outputting
> lines for each simple forecast object (at each valid time), followed
by
> outputting lines for each simple observed object. However, I also
> noticed that after these two blocks, two more blocks (again with
> forecast objects first followed by observation objects) occur where
the
> object_id always matches the object_cat. When adding up the area in
the
> simple objects (first two blocks), they match the total area for the
> corresponding objects in the last two blocks. Thus, I believe these
> attributes in the final two blocks are for the observed and forecast
> clusters. However, it is a little confusing to me because these
blocks
> still show object_ids such as F001 and O002 which indicates simple
> objects. Thus, I wanted to confirm this information with you.
>
> I have attached an example of my 2d output file. The block of
interest
> that I believe corresponds to the forecast cluster attributes starts
on
> line 765.
>
> Thanks!
> Sam Degelia
>



------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #88209] Question about Mode Time-Domain 2D output files
From: Samuel Degelia
Time: Fri Dec 21 09:44:05 2018

Randy,

Thank you very much for your response. I am glad to learn that these
are the 2D cluster attributes. I was actually considering running MODE
after MTD to try and get those data. But now I won’t have to do that.

I understand that there isn’t really a great way to describe these
data in the 2D text file. All the information that I need is there, I
just wasn’t sure if that was exactly what I needed until after our
email discussion. Maybe just a quick note in the online documentation
to state that the cluster attributes are listed after the simple
objects. I think showing “NA” in the object ID column would also work.

Thanks again for your help,
Sam Degelia

> On Dec 20, 2018, at 11:46 AM, Randy Bullock via RT
<met_help at ucar.edu> wrote:
>
> Hello Samuel -
>
> You are correct in supposing that the final two blocks in the 2D
text output file are information for object clusters.  In addition, I
agree with your statement that listing simple object_id's for cluster
objects is confusing.  I'm not sure what else we can do though,
really.  The mapping, if you will, from simple objects to cluster
objects is many-to-one.  That is, there are usually many simple
objects that make up a single cluster object.  That means that for any
simple object, the question "What cluster object does this correspond
to, if any?" has a nice, simple answer.  But asking "What simple
objects went into making this cluster object?" has a complicated
answer, usually a list of simple object numbers, which we can't really
put into a single output column in the 2D text output table.  Another
option would be to simply put "NA" in the object_id column, I guess.
>
> The whole issue of MTD output, what with 3D vs. 2D, Fcst vs. Obs,
Simple vs. cluster, Single vs pair, is so complex that we spent a lot
of time trying to figure out how to write out the information in a
clear way.  It may well be that we were less that 100% successful in
that.  Any suggestions you might have along those lines would be
appreciated.
>
>> On Mon Dec 17 15:37:42 2018, sdegelia at gmail.com wrote:
>> Hello,
>>
>> Before I start, I want to thank the met_help office (especially
John
>> Gotway) for the quick help over the last few weeks with my many MTD
>> questions. Hopefully this is my final question. This question is
>> regarding the 2d output files for MTD (mtd_..._2d.txt).
>>
>> Does the 2d output file contain attribute information for the
object
>> _clusters_, in addition to the attributes for the simple objects? I
>> noticed that the 2d output file follows the format of first
outputting
>> lines for each simple forecast object (at each valid time),
followed by
>> outputting lines for each simple observed object. However, I also
>> noticed that after these two blocks, two more blocks (again with
>> forecast objects first followed by observation objects) occur where
the
>> object_id always matches the object_cat. When adding up the area in
the
>> simple objects (first two blocks), they match the total area for
the
>> corresponding objects in the last two blocks. Thus, I believe these
>> attributes in the final two blocks are for the observed and
forecast
>> clusters. However, it is a little confusing to me because these
blocks
>> still show object_ids such as F001 and O002 which indicates simple
>> objects. Thus, I wanted to confirm this information with you.
>>
>> I have attached an example of my 2d output file. The block of
interest
>> that I believe corresponds to the forecast cluster attributes
starts on
>> line 765.
>>
>> Thanks!
>> Sam Degelia
>>
>
>
>


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