[Met_help] [rt.rap.ucar.edu #42335] History for using stat-analysis tool with METV3.0

Tue Nov 23 13:40:06 MST 2010

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

  Dear Met_Help:

I am learning how to use stat-analysis tool with METV3.0.
I have .stat from point-stat tool and I want to only evaluate within 
map_poly shown in .stat. However, even when I run stat-analysis with 
-map_poly option followed by the poly filename, the results still 
include that of FULL domain and map_poly domain.  The stat-analysis 
output also doesn't show the map_poly option.  Could you tell me if I'm 
doing anything wrong?

I have attached a tar ball containing my config file 
(STATAnalysisConfig), shell script (nowcast_stat_analysis.sh), analysis 
output file (RZK_stat_analysis.out), and input .stat files 
(input_stat.tar). I've also included the results from dump_row option 
for the individual jobs.

Thank you.

Kyoko

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

Subject: Re: [rt.rap.ucar.edu #42335] using stat-analysis tool with METV3.0
From: John Halley Gotway
Time: Tue Nov 23 12:49:05 2010

Kyoko,

I think this is just a confusion in how to specify filtering
parameters for STAT-Analysis.  In you STAT output files you sent me,
when I look in the column for "VX_MASK", I see two unique values: FULL
and CASE20080809.  I assume that you're trying to run these STAT-
Analysis jobs using the only the data for "CASE20080809".  If that's
the case, you'd apply that filtering criteria using the job
command "-vx_mask CASE20080809".  That'll match only the lines with
that value in the VX_MASK column.

The "-mask_poly" option is only used when processing the matched pair
(MPR) lines.  It looks like we don't have any helpful comments in the
config file about this option.  Consider the job I just send
you in my previous email:
  stat_analysis -lookin . -job aggregate_stat -line_type MPR
-out_line_type CNT -vx_mask FULL -interp_mthd UW_MEAN -interp_pnts 1
-column_min FCST 1.0 -column_min OBS 1.0

If you were to add "-vx_mask ../data/poly/uswrp2010_case20080809.poly"
to this job command, it'd filter those MPR lines and only use the ones
that fall inside that polyline.

Hope that helps.

John

On 11/23/2010 11:59 AM, RAL HelpDesk {for Kyoko Ikeda} wrote:
>
> Tue Nov 23 11:59:40 2010: Request 42335 was acted upon.
> Transaction: Ticket created by kyoko
>        Queue: met_help
>      Subject: using stat-analysis tool with METV3.0
>        Owner: Nobody
>   Requestors: kyoko at ucar.edu
>       Status: new
>  Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=42335 >
>
>
>   Dear Met_Help:
>
> I am learning how to use stat-analysis tool with METV3.0.
> I have .stat from point-stat tool and I want to only evaluate within
> map_poly shown in .stat. However, even when I run stat-analysis with
> -map_poly option followed by the poly filename, the results still
> include that of FULL domain and map_poly domain.  The stat-analysis
> output also doesn't show the map_poly option.  Could you tell me if
I'm
> doing anything wrong?
>
> I have attached a tar ball containing my config file
> (STATAnalysisConfig), shell script (nowcast_stat_analysis.sh),
analysis
> output file (RZK_stat_analysis.out), and input .stat files
> (input_stat.tar). I've also included the results from dump_row
option
> for the individual jobs.
>
> Thank you.
>
> Kyoko
>
>
>

------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #42335] using stat-analysis tool with METV3.0
From: Kyoko Ikeda
Time: Tue Nov 23 13:39:00 2010

  Hi John,

The first option does the job I was trying to do.
THank you for your help!

Kyoko

On 11/23/10 12:49 PM, RAL HelpDesk {for John Halley Gotway} wrote:
> Kyoko,
>
> I think this is just a confusion in how to specify filtering
parameters for STAT-Analysis.  In you STAT output files you sent me,
when I look in the column for "VX_MASK", I see two unique values: FULL
> and CASE20080809.  I assume that you're trying to run these STAT-
Analysis jobs using the only the data for "CASE20080809".  If that's
the case, you'd apply that filtering criteria using the job
> command "-vx_mask CASE20080809".  That'll match only the lines with
that value in the VX_MASK column.
>
> The "-mask_poly" option is only used when processing the matched
pair (MPR) lines.  It looks like we don't have any helpful comments in
the config file about this option.  Consider the job I just send
> you in my previous email:
>    stat_analysis -lookin . -job aggregate_stat -line_type MPR
-out_line_type CNT -vx_mask FULL -interp_mthd UW_MEAN -interp_pnts 1
-column_min FCST 1.0 -column_min OBS 1.0
>
> If you were to add "-vx_mask
../data/poly/uswrp2010_case20080809.poly" to this job command, it'd
filter those MPR lines and only use the ones that fall inside that
polyline.
>
> Hope that helps.
>
> John
>
>
> On 11/23/2010 11:59 AM, RAL HelpDesk {for Kyoko Ikeda} wrote:
>> Tue Nov 23 11:59:40 2010: Request 42335 was acted upon.
>> Transaction: Ticket created by kyoko
>>         Queue: met_help
>>       Subject: using stat-analysis tool with METV3.0
>>         Owner: Nobody
>>    Requestors: kyoko at ucar.edu
>>        Status: new
>>   Ticket<URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=42335>
>>
>>
>>    Dear Met_Help:
>>
>> I am learning how to use stat-analysis tool with METV3.0.
>> I have .stat from point-stat tool and I want to only evaluate
within
>> map_poly shown in .stat. However, even when I run stat-analysis
with
>> -map_poly option followed by the poly filename, the results still
>> include that of FULL domain and map_poly domain.  The stat-analysis
>> output also doesn't show the map_poly option.  Could you tell me if
I'm
>> doing anything wrong?
>>
>> I have attached a tar ball containing my config file
>> (STATAnalysisConfig), shell script (nowcast_stat_analysis.sh),
analysis
>> output file (RZK_stat_analysis.out), and input .stat files
>> (input_stat.tar). I've also included the results from dump_row
option
>> for the individual jobs.
>>
>> Thank you.
>>
>> Kyoko
>>
>>
>>

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