[Met_help] [rt.rap.ucar.edu #71923] History for another problem with gridstat
John Halley Gotway via RT
met_help at ucar.edu
Wed May 20 09:19:13 MDT 2015
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Hi John. I have another stange problem. I had run grid stat for another date 22 April2011. I got all outputs nicely. but when I try to visualise the OBS output using the ncview, it looks very strtange. I am sending the NC file for your reference.
also when i saw the data in the OBS file, it shows me some very strange values of precipitation. Absolutely unrealistic.
-rw-rw-r-- 1 aditya aditya 286376 May 13 22:03 grid_stat_270000L_20110423_030000V.stat-rw-rw-r-- 1 aditya aditya 2750 May 13 22:03 grid_stat_270000L_20110423_030000V_fho.txt-rw-rw-r-- 1 aditya aditya 2750 May 13 22:03 grid_stat_270000L_20110423_030000V_ctc.txt-rw-rw-r-- 1 aditya aditya 268 May 13 22:03 grid_stat_270000L_20110423_030000V_vl1l2.txt-rw-rw-r-- 1 aditya aditya 882 May 13 22:03 grid_stat_270000L_20110423_030000V_sl1l2.txt-rw-rw-r-- 1 aditya aditya 324 May 13 22:03 grid_stat_270000L_20110423_030000V_pstd.txt-rw-rw-r-- 1 aditya aditya 229 May 13 22:03 grid_stat_270000L_20110423_030000V_prc.txt-rw-rw-r-- 1 aditya aditya 229 May 13 22:03 grid_stat_270000L_20110423_030000V_pjc.txt-rw-rw-r-- 1 aditya aditya 229 May 13 22:03 grid_stat_270000L_20110423_030000V_pct.txt-rw-rw-r-- 1 aditya aditya 20250 May 13 22:03 grid_stat_270000L_20110423_030000V_nbrctc.txt-rw-rw-r-- 1 aditya aditya 998 May 13 22:03 grid_stat_270000L_2011042!
3_030000V_mcts.txt-rw-rw-r-- 1 aditya aditya 2310 May 13 22:03 grid_stat_270000L_20110423_030000V_mctc.txt-rw-rw-r-- 1 aditya aditya 11396 May 13 22:03 grid_stat_270000L_20110423_030000V_cts.txt-rw-rw-r-- 1 aditya aditya 3046 May 13 22:03 grid_stat_270000L_20110423_030000V_cnt.txt-rw-rw-r-- 1 aditya aditya 84240 May 13 22:03 grid_stat_270000L_20110423_030000V_nbrcts.txt-rw-rw-r-- 1 aditya aditya 21644 May 13 22:03 grid_stat_270000L_20110423_030000V_pairs.nc-rw-rw-r-- 1 aditya aditya 31428 May 13 22:03 grid_stat_270000L_20110423_030000V_nbrcnt.txtaditya at agniilap:~/geeta/installs/met/RUN_MET-22apr2011/out-22apr2011$
pls point me to right direction. geeta
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: another problem with gridstat
From: John Halley Gotway
Time: Wed May 13 12:04:54 2015
Geeta,
I agree. Those data values do look bad. I strongly suspect that this
isn't a problem caused by grid_stat. Instead I'm guessing that
observation
data you're passing to grid_stat is bad. I ran the following command
to
visualize the grid_stat pairs output and the resulting image is
attached.
The values in this data file range from 0 to 4.5e30!
met-5.0/bin/plot_data_plane
grid_stat_270000L_20110423_030000V_pairs.nc
obs.ps 'name="OBS_APCP_A24_FULL"; level="(*,*)";'
Please take a look at the observation data you're passing to grid_stat
to
see if it's corrupted in some way.
Thanks,
John
On Wed, May 13, 2015 at 11:12 AM, Geeta Geeta via RT
<met_help at ucar.edu>
wrote:
>
> Wed May 13 11:12:32 2015: Request 71923 was acted upon.
> Transaction: Ticket created by geeta124 at hotmail.com
> Queue: met_help
> Subject: another problem with gridstat
> Owner: Nobody
> Requestors: geeta124 at hotmail.com
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
>
>
> Hi John. I have another stange problem. I had run grid stat for
another
> date 22 April2011. I got all outputs nicely. but when I try to
visualise
> the OBS output using the ncview, it looks very strtange. I am
sending the
> NC file for your reference.
> also when i saw the data in the OBS file, it shows me some very
strange
> values of precipitation. Absolutely unrealistic.
>
> -rw-rw-r-- 1 aditya aditya 286376 May 13 22:03
> grid_stat_270000L_20110423_030000V.stat-rw-rw-r-- 1 aditya aditya
2750
> May 13 22:03 grid_stat_270000L_20110423_030000V_fho.txt-rw-rw-r-- 1
aditya
> aditya 2750 May 13 22:03
> grid_stat_270000L_20110423_030000V_ctc.txt-rw-rw-r-- 1 aditya aditya
> 268 May 13 22:03 grid_stat_270000L_20110423_030000V_vl1l2.txt-rw-rw-
r-- 1
> aditya aditya 882 May 13 22:03
> grid_stat_270000L_20110423_030000V_sl1l2.txt-rw-rw-r-- 1 aditya
aditya
> 324 May 13 22:03 grid_stat_270000L_20110423_030000V_pstd.txt-rw-rw-
r-- 1
> aditya aditya 229 May 13 22:03
> grid_stat_270000L_20110423_030000V_prc.txt-rw-rw-r-- 1 aditya aditya
> 229 May 13 22:03 grid_stat_270000L_20110423_030000V_pjc.txt-rw-rw-
r-- 1
> aditya aditya 229 May 13 22:03
> grid_stat_270000L_20110423_030000V_pct.txt-rw-rw-r-- 1 aditya aditya
> 20250 May 13 22:03 grid_stat_270000L_20110423_030000V_nbrctc.txt-rw-
rw-r--
> 1 aditya aditya 998 May 13 22:03 grid_stat_270000L_2011042!
> 3_030000V_mcts.txt-rw-rw-r-- 1 aditya aditya 2310 May 13 22:03
> grid_stat_270000L_20110423_030000V_mctc.txt-rw-rw-r-- 1 aditya
aditya
> 11396 May 13 22:03 grid_stat_270000L_20110423_030000V_cts.txt-rw-rw-
r-- 1
> aditya aditya 3046 May 13 22:03
> grid_stat_270000L_20110423_030000V_cnt.txt-rw-rw-r-- 1 aditya aditya
> 84240 May 13 22:03 grid_stat_270000L_20110423_030000V_nbrcts.txt-rw-
rw-r--
> 1 aditya aditya 21644 May 13 22:03
> grid_stat_270000L_20110423_030000V_pairs.nc-rw-rw-r-- 1 aditya
aditya
> 31428 May 13 22:03
> grid_stat_270000L_20110423_030000V_nbrcnt.txtaditya at agniilap
> :~/geeta/installs/met/RUN_MET-22apr2011/out-22apr2011$
> pls point me to right direction. geeta
>
>
>
------------------------------------------------
Subject: another problem with gridstat
From: Geeta Geeta
Time: Wed May 13 15:23:07 2015
hi John. I just now did whole process of making Observed data again.
Now sending you new NC file.
But the aggregated data of 8 observations (06UTC of 22 APR2011 - 03UTC
of 23 April 2011) looks very different from the plotted data on TRMM
website.
> Subject: Re: [rt.rap.ucar.edu #71923] another problem with gridstat
> From: met_help at ucar.edu
> To: geeta124 at hotmail.com
> Date: Wed, 13 May 2015 12:04:54 -0600
>
> Geeta,
>
> I agree. Those data values do look bad. I strongly suspect that
this
> isn't a problem caused by grid_stat. Instead I'm guessing that
observation
> data you're passing to grid_stat is bad. I ran the following
command to
> visualize the grid_stat pairs output and the resulting image is
attached.
> The values in this data file range from 0 to 4.5e30!
> met-5.0/bin/plot_data_plane
grid_stat_270000L_20110423_030000V_pairs.nc
> obs.ps 'name="OBS_APCP_A24_FULL"; level="(*,*)";'
>
> Please take a look at the observation data you're passing to
grid_stat to
> see if it's corrupted in some way.
>
> Thanks,
> John
>
> On Wed, May 13, 2015 at 11:12 AM, Geeta Geeta via RT
<met_help at ucar.edu>
> wrote:
>
> >
> > Wed May 13 11:12:32 2015: Request 71923 was acted upon.
> > Transaction: Ticket created by geeta124 at hotmail.com
> > Queue: met_help
> > Subject: another problem with gridstat
> > Owner: Nobody
> > Requestors: geeta124 at hotmail.com
> > Status: new
> > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
> >
> >
> > Hi John. I have another stange problem. I had run grid stat for
another
> > date 22 April2011. I got all outputs nicely. but when I try to
visualise
> > the OBS output using the ncview, it looks very strtange. I am
sending the
> > NC file for your reference.
> > also when i saw the data in the OBS file, it shows me some very
strange
> > values of precipitation. Absolutely unrealistic.
> >
> > -rw-rw-r-- 1 aditya aditya 286376 May 13 22:03
> > grid_stat_270000L_20110423_030000V.stat-rw-rw-r-- 1 aditya aditya
2750
> > May 13 22:03 grid_stat_270000L_20110423_030000V_fho.txt-rw-rw-r--
1 aditya
> > aditya 2750 May 13 22:03
> > grid_stat_270000L_20110423_030000V_ctc.txt-rw-rw-r-- 1 aditya
aditya
> > 268 May 13 22:03 grid_stat_270000L_20110423_030000V_vl1l2.txt-rw-
rw-r-- 1
> > aditya aditya 882 May 13 22:03
> > grid_stat_270000L_20110423_030000V_sl1l2.txt-rw-rw-r-- 1 aditya
aditya
> > 324 May 13 22:03 grid_stat_270000L_20110423_030000V_pstd.txt-rw-
rw-r-- 1
> > aditya aditya 229 May 13 22:03
> > grid_stat_270000L_20110423_030000V_prc.txt-rw-rw-r-- 1 aditya
aditya
> > 229 May 13 22:03 grid_stat_270000L_20110423_030000V_pjc.txt-rw-rw-
r-- 1
> > aditya aditya 229 May 13 22:03
> > grid_stat_270000L_20110423_030000V_pct.txt-rw-rw-r-- 1 aditya
aditya
> > 20250 May 13 22:03 grid_stat_270000L_20110423_030000V_nbrctc.txt-
rw-rw-r--
> > 1 aditya aditya 998 May 13 22:03 grid_stat_270000L_2011042!
> > 3_030000V_mcts.txt-rw-rw-r-- 1 aditya aditya 2310 May 13
22:03
> > grid_stat_270000L_20110423_030000V_mctc.txt-rw-rw-r-- 1 aditya
aditya
> > 11396 May 13 22:03 grid_stat_270000L_20110423_030000V_cts.txt-rw-
rw-r-- 1
> > aditya aditya 3046 May 13 22:03
> > grid_stat_270000L_20110423_030000V_cnt.txt-rw-rw-r-- 1 aditya
aditya
> > 84240 May 13 22:03 grid_stat_270000L_20110423_030000V_nbrcts.txt-
rw-rw-r--
> > 1 aditya aditya 21644 May 13 22:03
> > grid_stat_270000L_20110423_030000V_pairs.nc-rw-rw-r-- 1 aditya
aditya
> > 31428 May 13 22:03
> > grid_stat_270000L_20110423_030000V_nbrcnt.txtaditya at agniilap
> > :~/geeta/installs/met/RUN_MET-22apr2011/out-22apr2011$
> > pls point me to right direction. geeta
> >
> >
> >
>
------------------------------------------------
Subject: another problem with gridstat
From: John Halley Gotway
Time: Wed May 13 15:48:20 2015
Geeta,
OK, the NetCDF matched pairs file you sent looks reasonable to me.
You are
comparing it to some plot from the TRMM website and say that they
differ.
Can you please send me the plot to which you're comparing this data as
well
as the link you used to grab it?
I can't really comment on this until I understand the discrepancy
you're
seeing.
Thanks,
John
On Wed, May 13, 2015 at 3:23 PM, Geeta Geeta via RT
<met_help at ucar.edu>
wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
>
> hi John. I just now did whole process of making Observed data again.
Now
> sending you new NC file.
> But the aggregated data of 8 observations (06UTC of 22 APR2011 -
03UTC of
> 23 April 2011) looks very different from the plotted data on TRMM
website.
>
> > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > From: met_help at ucar.edu
> > To: geeta124 at hotmail.com
> > Date: Wed, 13 May 2015 12:04:54 -0600
> >
> > Geeta,
> >
> > I agree. Those data values do look bad. I strongly suspect that
this
> > isn't a problem caused by grid_stat. Instead I'm guessing that
> observation
> > data you're passing to grid_stat is bad. I ran the following
command to
> > visualize the grid_stat pairs output and the resulting image is
attached.
> > The values in this data file range from 0 to 4.5e30!
> > met-5.0/bin/plot_data_plane
> grid_stat_270000L_20110423_030000V_pairs.nc
> > obs.ps 'name="OBS_APCP_A24_FULL"; level="(*,*)";'
> >
> > Please take a look at the observation data you're passing to
grid_stat to
> > see if it's corrupted in some way.
> >
> > Thanks,
> > John
> >
> > On Wed, May 13, 2015 at 11:12 AM, Geeta Geeta via RT
<met_help at ucar.edu>
> > wrote:
> >
> > >
> > > Wed May 13 11:12:32 2015: Request 71923 was acted upon.
> > > Transaction: Ticket created by geeta124 at hotmail.com
> > > Queue: met_help
> > > Subject: another problem with gridstat
> > > Owner: Nobody
> > > Requestors: geeta124 at hotmail.com
> > > Status: new
> > > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
> >
> > >
> > >
> > > Hi John. I have another stange problem. I had run grid stat for
another
> > > date 22 April2011. I got all outputs nicely. but when I try to
> visualise
> > > the OBS output using the ncview, it looks very strtange. I am
sending
> the
> > > NC file for your reference.
> > > also when i saw the data in the OBS file, it shows me some very
strange
> > > values of precipitation. Absolutely unrealistic.
> > >
> > > -rw-rw-r-- 1 aditya aditya 286376 May 13 22:03
> > > grid_stat_270000L_20110423_030000V.stat-rw-rw-r-- 1 aditya
aditya
> 2750
> > > May 13 22:03 grid_stat_270000L_20110423_030000V_fho.txt-rw-rw-
r-- 1
> aditya
> > > aditya 2750 May 13 22:03
> > > grid_stat_270000L_20110423_030000V_ctc.txt-rw-rw-r-- 1 aditya
aditya
> > > 268 May 13 22:03
> grid_stat_270000L_20110423_030000V_vl1l2.txt-rw-rw-r-- 1
> > > aditya aditya 882 May 13 22:03
> > > grid_stat_270000L_20110423_030000V_sl1l2.txt-rw-rw-r-- 1 aditya
aditya
> > > 324 May 13 22:03 grid_stat_270000L_20110423_030000V_pstd.txt-rw-
rw-r--
> 1
> > > aditya aditya 229 May 13 22:03
> > > grid_stat_270000L_20110423_030000V_prc.txt-rw-rw-r-- 1 aditya
aditya
> > > 229 May 13 22:03 grid_stat_270000L_20110423_030000V_pjc.txt-rw-
rw-r-- 1
> > > aditya aditya 229 May 13 22:03
> > > grid_stat_270000L_20110423_030000V_pct.txt-rw-rw-r-- 1 aditya
aditya
> > > 20250 May 13 22:03
> grid_stat_270000L_20110423_030000V_nbrctc.txt-rw-rw-r--
> > > 1 aditya aditya 998 May 13 22:03 grid_stat_270000L_2011042!
> > > 3_030000V_mcts.txt-rw-rw-r-- 1 aditya aditya 2310 May 13
22:03
> > > grid_stat_270000L_20110423_030000V_mctc.txt-rw-rw-r-- 1 aditya
aditya
> > > 11396 May 13 22:03
> grid_stat_270000L_20110423_030000V_cts.txt-rw-rw-r-- 1
> > > aditya aditya 3046 May 13 22:03
> > > grid_stat_270000L_20110423_030000V_cnt.txt-rw-rw-r-- 1 aditya
aditya
> > > 84240 May 13 22:03
> grid_stat_270000L_20110423_030000V_nbrcts.txt-rw-rw-r--
> > > 1 aditya aditya 21644 May 13 22:03
> > > grid_stat_270000L_20110423_030000V_pairs.nc-rw-rw-r-- 1 aditya
aditya
> > > 31428 May 13 22:03
> > > grid_stat_270000L_20110423_030000V_nbrcnt.txtaditya at agniilap
> > > :~/geeta/installs/met/RUN_MET-22apr2011/out-22apr2011$
> > > pls point me to right direction. geeta
> > >
> > >
> > >
> >
>
>
------------------------------------------------
Subject: another problem with gridstat
From: Geeta Geeta
Time: Fri May 15 10:37:40 2015
Hi John.
I will download the image of 3B42 TRMM and send u. I am unable to open
the page now.
I have one more query reg the FSS. I am trying to apply FSS for some
thunderstorm simulations.
so essentially, it will tell me if the frequency of FCST and OBS are
close???
Can I apply this technique for evaluation of the MONSOON rainfall for
a season for entire southern India???. so for this I will be taking a
very large domain. I expect that in this case FSS will be better as
compared to that of the thunderstorm cases???
Will this be a viable idea???.
I need your feedback reg this.
Geeta
> Subject: Re: [rt.rap.ucar.edu #71923] another problem with gridstat
> From: met_help at ucar.edu
> To: geeta124 at hotmail.com
> Date: Wed, 13 May 2015 15:48:20 -0600
>
> Geeta,
>
> OK, the NetCDF matched pairs file you sent looks reasonable to me.
You are
> comparing it to some plot from the TRMM website and say that they
differ.
> Can you please send me the plot to which you're comparing this data
as well
> as the link you used to grab it?
>
> I can't really comment on this until I understand the discrepancy
you're
> seeing.
>
> Thanks,
> John
>
>
> On Wed, May 13, 2015 at 3:23 PM, Geeta Geeta via RT
<met_help at ucar.edu>
> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
> >
> > hi John. I just now did whole process of making Observed data
again. Now
> > sending you new NC file.
> > But the aggregated data of 8 observations (06UTC of 22 APR2011 -
03UTC of
> > 23 April 2011) looks very different from the plotted data on TRMM
website.
> >
> > > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > > From: met_help at ucar.edu
> > > To: geeta124 at hotmail.com
> > > Date: Wed, 13 May 2015 12:04:54 -0600
> > >
> > > Geeta,
> > >
> > > I agree. Those data values do look bad. I strongly suspect
that this
> > > isn't a problem caused by grid_stat. Instead I'm guessing that
> > observation
> > > data you're passing to grid_stat is bad. I ran the following
command to
> > > visualize the grid_stat pairs output and the resulting image is
attached.
> > > The values in this data file range from 0 to 4.5e30!
> > > met-5.0/bin/plot_data_plane
> > grid_stat_270000L_20110423_030000V_pairs.nc
> > > obs.ps 'name="OBS_APCP_A24_FULL"; level="(*,*)";'
> > >
> > > Please take a look at the observation data you're passing to
grid_stat to
> > > see if it's corrupted in some way.
> > >
> > > Thanks,
> > > John
> > >
> > > On Wed, May 13, 2015 at 11:12 AM, Geeta Geeta via RT
<met_help at ucar.edu>
> > > wrote:
> > >
> > > >
> > > > Wed May 13 11:12:32 2015: Request 71923 was acted upon.
> > > > Transaction: Ticket created by geeta124 at hotmail.com
> > > > Queue: met_help
> > > > Subject: another problem with gridstat
> > > > Owner: Nobody
> > > > Requestors: geeta124 at hotmail.com
> > > > Status: new
> > > > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
> > >
> > > >
> > > >
> > > > Hi John. I have another stange problem. I had run grid stat
for another
> > > > date 22 April2011. I got all outputs nicely. but when I try to
> > visualise
> > > > the OBS output using the ncview, it looks very strtange. I am
sending
> > the
> > > > NC file for your reference.
> > > > also when i saw the data in the OBS file, it shows me some
very strange
> > > > values of precipitation. Absolutely unrealistic.
> > > >
> > > > -rw-rw-r-- 1 aditya aditya 286376 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V.stat-rw-rw-r-- 1 aditya
aditya
> > 2750
> > > > May 13 22:03 grid_stat_270000L_20110423_030000V_fho.txt-rw-rw-
r-- 1
> > aditya
> > > > aditya 2750 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_ctc.txt-rw-rw-r-- 1 aditya
aditya
> > > > 268 May 13 22:03
> > grid_stat_270000L_20110423_030000V_vl1l2.txt-rw-rw-r-- 1
> > > > aditya aditya 882 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_sl1l2.txt-rw-rw-r-- 1
aditya aditya
> > > > 324 May 13 22:03 grid_stat_270000L_20110423_030000V_pstd.txt-
rw-rw-r--
> > 1
> > > > aditya aditya 229 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_prc.txt-rw-rw-r-- 1 aditya
aditya
> > > > 229 May 13 22:03 grid_stat_270000L_20110423_030000V_pjc.txt-
rw-rw-r-- 1
> > > > aditya aditya 229 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_pct.txt-rw-rw-r-- 1 aditya
aditya
> > > > 20250 May 13 22:03
> > grid_stat_270000L_20110423_030000V_nbrctc.txt-rw-rw-r--
> > > > 1 aditya aditya 998 May 13 22:03
grid_stat_270000L_2011042!
> > > > 3_030000V_mcts.txt-rw-rw-r-- 1 aditya aditya 2310 May 13
22:03
> > > > grid_stat_270000L_20110423_030000V_mctc.txt-rw-rw-r-- 1 aditya
aditya
> > > > 11396 May 13 22:03
> > grid_stat_270000L_20110423_030000V_cts.txt-rw-rw-r-- 1
> > > > aditya aditya 3046 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_cnt.txt-rw-rw-r-- 1 aditya
aditya
> > > > 84240 May 13 22:03
> > grid_stat_270000L_20110423_030000V_nbrcts.txt-rw-rw-r--
> > > > 1 aditya aditya 21644 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_pairs.nc-rw-rw-r-- 1 aditya
aditya
> > > > 31428 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_nbrcnt.txtaditya at agniilap
> > > > :~/geeta/installs/met/RUN_MET-22apr2011/out-22apr2011$
> > > > pls point me to right direction. geeta
> > > >
> > > >
> > > >
> > >
> >
> >
>
------------------------------------------------
Subject: another problem with gridstat
From: Geeta Geeta
Time: Fri May 15 10:39:11 2015
Hi John.
I will download the image of 3B42 TRMM and send u. I am unable to open
the page now.
I have one more query reg the FSS. I am trying to apply FSS for some
thunderstorm simulations.
so essentially, it will tell me if the frequency of FCST and OBS are
close???
Can I apply this technique for evaluation of the MONSOON rainfall for
a season (122 days in India) for entire southern India???. so for
this I will be taking a very large domain. I expect that in this case
FSS will be better as compared to that of the thunderstorm cases???
Will this be a viable idea???.
I need your feedback reg this.
Geeta
> Subject: Re: [rt.rap.ucar.edu #71923] another problem with gridstat
> From: met_help at ucar.edu
> To: geeta124 at hotmail.com
> Date: Wed, 13 May 2015 15:48:20 -0600
>
> Geeta,
>
> OK, the NetCDF matched pairs file you sent looks reasonable to me.
You are
> comparing it to some plot from the TRMM website and say that they
differ.
> Can you please send me the plot to which you're comparing this data
as well
> as the link you used to grab it?
>
> I can't really comment on this until I understand the discrepancy
you're
> seeing.
>
> Thanks,
> John
>
>
> On Wed, May 13, 2015 at 3:23 PM, Geeta Geeta via RT
<met_help at ucar.edu>
> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
> >
> > hi John. I just now did whole process of making Observed data
again. Now
> > sending you new NC file.
> > But the aggregated data of 8 observations (06UTC of 22 APR2011 -
03UTC of
> > 23 April 2011) looks very different from the plotted data on TRMM
website.
> >
> > > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > > From: met_help at ucar.edu
> > > To: geeta124 at hotmail.com
> > > Date: Wed, 13 May 2015 12:04:54 -0600
> > >
> > > Geeta,
> > >
> > > I agree. Those data values do look bad. I strongly suspect
that this
> > > isn't a problem caused by grid_stat. Instead I'm guessing that
> > observation
> > > data you're passing to grid_stat is bad. I ran the following
command to
> > > visualize the grid_stat pairs output and the resulting image is
attached.
> > > The values in this data file range from 0 to 4.5e30!
> > > met-5.0/bin/plot_data_plane
> > grid_stat_270000L_20110423_030000V_pairs.nc
> > > obs.ps 'name="OBS_APCP_A24_FULL"; level="(*,*)";'
> > >
> > > Please take a look at the observation data you're passing to
grid_stat to
> > > see if it's corrupted in some way.
> > >
> > > Thanks,
> > > John
> > >
> > > On Wed, May 13, 2015 at 11:12 AM, Geeta Geeta via RT
<met_help at ucar.edu>
> > > wrote:
> > >
> > > >
> > > > Wed May 13 11:12:32 2015: Request 71923 was acted upon.
> > > > Transaction: Ticket created by geeta124 at hotmail.com
> > > > Queue: met_help
> > > > Subject: another problem with gridstat
> > > > Owner: Nobody
> > > > Requestors: geeta124 at hotmail.com
> > > > Status: new
> > > > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
> > >
> > > >
> > > >
> > > > Hi John. I have another stange problem. I had run grid stat
for another
> > > > date 22 April2011. I got all outputs nicely. but when I try to
> > visualise
> > > > the OBS output using the ncview, it looks very strtange. I am
sending
> > the
> > > > NC file for your reference.
> > > > also when i saw the data in the OBS file, it shows me some
very strange
> > > > values of precipitation. Absolutely unrealistic.
> > > >
> > > > -rw-rw-r-- 1 aditya aditya 286376 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V.stat-rw-rw-r-- 1 aditya
aditya
> > 2750
> > > > May 13 22:03 grid_stat_270000L_20110423_030000V_fho.txt-rw-rw-
r-- 1
> > aditya
> > > > aditya 2750 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_ctc.txt-rw-rw-r-- 1 aditya
aditya
> > > > 268 May 13 22:03
> > grid_stat_270000L_20110423_030000V_vl1l2.txt-rw-rw-r-- 1
> > > > aditya aditya 882 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_sl1l2.txt-rw-rw-r-- 1
aditya aditya
> > > > 324 May 13 22:03 grid_stat_270000L_20110423_030000V_pstd.txt-
rw-rw-r--
> > 1
> > > > aditya aditya 229 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_prc.txt-rw-rw-r-- 1 aditya
aditya
> > > > 229 May 13 22:03 grid_stat_270000L_20110423_030000V_pjc.txt-
rw-rw-r-- 1
> > > > aditya aditya 229 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_pct.txt-rw-rw-r-- 1 aditya
aditya
> > > > 20250 May 13 22:03
> > grid_stat_270000L_20110423_030000V_nbrctc.txt-rw-rw-r--
> > > > 1 aditya aditya 998 May 13 22:03
grid_stat_270000L_2011042!
> > > > 3_030000V_mcts.txt-rw-rw-r-- 1 aditya aditya 2310 May 13
22:03
> > > > grid_stat_270000L_20110423_030000V_mctc.txt-rw-rw-r-- 1 aditya
aditya
> > > > 11396 May 13 22:03
> > grid_stat_270000L_20110423_030000V_cts.txt-rw-rw-r-- 1
> > > > aditya aditya 3046 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_cnt.txt-rw-rw-r-- 1 aditya
aditya
> > > > 84240 May 13 22:03
> > grid_stat_270000L_20110423_030000V_nbrcts.txt-rw-rw-r--
> > > > 1 aditya aditya 21644 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_pairs.nc-rw-rw-r-- 1 aditya
aditya
> > > > 31428 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_nbrcnt.txtaditya at agniilap
> > > > :~/geeta/installs/met/RUN_MET-22apr2011/out-22apr2011$
> > > > pls point me to right direction. geeta
> > > >
> > > >
> > > >
> > >
> >
> >
>
------------------------------------------------
Subject: another problem with gridstat
From: Geeta Geeta
Time: Fri May 15 10:52:09 2015
Hi John.
Just want to confirm a few points.
Suppose I calculate FSS for two seasons. say Summer (92 days) and
monsoon (122 days) in India.
I plot FSS vs spatial Scale for both seasons.
1. How can aggregate the FSS for entire season . I mean I will have
one UNIQUE plot of FSS vs scale for a day ???. How will I aggregate???
2. Calculated FSS has to be greater than 0.5 at a particular spatial
scale to considered useful FCST by the model. Is this statement
correct????
3. the season having larger FSS means that MODEL has more skill as
compared to the other season???
Can U pls clarify.
Geeta
From: geeta124 at hotmail.com
To: met_help at ucar.edu
Subject: RE: [rt.rap.ucar.edu #71923] another problem with gridstat
Date: Fri, 15 May 2015 22:09:07 +0530
Hi John.
I will download the image of 3B42 TRMM and send u. I am unable to open
the page now.
I have one more query reg the FSS. I am trying to apply FSS for some
thunderstorm simulations.
so essentially, it will tell me if the frequency of FCST and OBS are
close???
Can I apply this technique for evaluation of the MONSOON rainfall for
a season (122 days in India) for entire southern India???. so for
this I will be taking a very large domain. I expect that in this case
FSS will be better as compared to that of the thunderstorm cases???
Will this be a viable idea???.
I need your feedback reg this.
Geeta
> Subject: Re: [rt.rap.ucar.edu #71923] another problem with gridstat
> From: met_help at ucar.edu
> To: geeta124 at hotmail.com
> Date: Wed, 13 May 2015 15:48:20 -0600
>
> Geeta,
>
> OK, the NetCDF matched pairs file you sent looks reasonable to me.
You are
> comparing it to some plot from the TRMM website and say that they
differ.
> Can you please send me the plot to which you're comparing this data
as well
> as the link you used to grab it?
>
> I can't really comment on this until I understand the discrepancy
you're
> seeing.
>
> Thanks,
> John
>
>
> On Wed, May 13, 2015 at 3:23 PM, Geeta Geeta via RT
<met_help at ucar.edu>
> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
> >
> > hi John. I just now did whole process of making Observed data
again. Now
> > sending you new NC file.
> > But the aggregated data of 8 observations (06UTC of 22 APR2011 -
03UTC of
> > 23 April 2011) looks very different from the plotted data on TRMM
website.
> >
> > > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > > From: met_help at ucar.edu
> > > To: geeta124 at hotmail.com
> > > Date: Wed, 13 May 2015 12:04:54 -0600
> > >
> > > Geeta,
> > >
> > > I agree. Those data values do look bad. I strongly suspect
that this
> > > isn't a problem caused by grid_stat. Instead I'm guessing that
> > observation
> > > data you're passing to grid_stat is bad. I ran the following
command to
> > > visualize the grid_stat pairs output and the resulting image is
attached.
> > > The values in this data file range from 0 to 4.5e30!
> > > met-5.0/bin/plot_data_plane
> > grid_stat_270000L_20110423_030000V_pairs.nc
> > > obs.ps 'name="OBS_APCP_A24_FULL"; level="(*,*)";'
> > >
> > > Please take a look at the observation data you're passing to
grid_stat to
> > > see if it's corrupted in some way.
> > >
> > > Thanks,
> > > John
> > >
> > > On Wed, May 13, 2015 at 11:12 AM, Geeta Geeta via RT
<met_help at ucar.edu>
> > > wrote:
> > >
> > > >
> > > > Wed May 13 11:12:32 2015: Request 71923 was acted upon.
> > > > Transaction: Ticket created by geeta124 at hotmail.com
> > > > Queue: met_help
> > > > Subject: another problem with gridstat
> > > > Owner: Nobody
> > > > Requestors: geeta124 at hotmail.com
> > > > Status: new
> > > > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
> > >
> > > >
> > > >
> > > > Hi John. I have another stange problem. I had run grid stat
for another
> > > > date 22 April2011. I got all outputs nicely. but when I try to
> > visualise
> > > > the OBS output using the ncview, it looks very strtange. I am
sending
> > the
> > > > NC file for your reference.
> > > > also when i saw the data in the OBS file, it shows me some
very strange
> > > > values of precipitation. Absolutely unrealistic.
> > > >
> > > > -rw-rw-r-- 1 aditya aditya 286376 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V.stat-rw-rw-r-- 1 aditya
aditya
> > 2750
> > > > May 13 22:03 grid_stat_270000L_20110423_030000V_fho.txt-rw-rw-
r-- 1
> > aditya
> > > > aditya 2750 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_ctc.txt-rw-rw-r-- 1 aditya
aditya
> > > > 268 May 13 22:03
> > grid_stat_270000L_20110423_030000V_vl1l2.txt-rw-rw-r-- 1
> > > > aditya aditya 882 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_sl1l2.txt-rw-rw-r-- 1
aditya aditya
> > > > 324 May 13 22:03 grid_stat_270000L_20110423_030000V_pstd.txt-
rw-rw-r--
> > 1
> > > > aditya aditya 229 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_prc.txt-rw-rw-r-- 1 aditya
aditya
> > > > 229 May 13 22:03 grid_stat_270000L_20110423_030000V_pjc.txt-
rw-rw-r-- 1
> > > > aditya aditya 229 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_pct.txt-rw-rw-r-- 1 aditya
aditya
> > > > 20250 May 13 22:03
> > grid_stat_270000L_20110423_030000V_nbrctc.txt-rw-rw-r--
> > > > 1 aditya aditya 998 May 13 22:03
grid_stat_270000L_2011042!
> > > > 3_030000V_mcts.txt-rw-rw-r-- 1 aditya aditya 2310 May 13
22:03
> > > > grid_stat_270000L_20110423_030000V_mctc.txt-rw-rw-r-- 1 aditya
aditya
> > > > 11396 May 13 22:03
> > grid_stat_270000L_20110423_030000V_cts.txt-rw-rw-r-- 1
> > > > aditya aditya 3046 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_cnt.txt-rw-rw-r-- 1 aditya
aditya
> > > > 84240 May 13 22:03
> > grid_stat_270000L_20110423_030000V_nbrcts.txt-rw-rw-r--
> > > > 1 aditya aditya 21644 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_pairs.nc-rw-rw-r-- 1 aditya
aditya
> > > > 31428 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_nbrcnt.txtaditya at agniilap
> > > > :~/geeta/installs/met/RUN_MET-22apr2011/out-22apr2011$
> > > > pls point me to right direction. geeta
> > > >
> > > >
> > > >
> > >
> >
> >
>
------------------------------------------------
Subject: another problem with gridstat
From: John Halley Gotway
Time: Fri May 15 12:11:40 2015
Geeta,
Sure, you can compute the fractions skill score for comparing forecast
and
observed thunderstorms. When computing FSS, you first threshold the
fields
to define events and non-events. Then you look at successively larger
and
larger areas around each grid point to see how the forecast event
frequency
compares to the observed event frequency. Seems like applying that to
thunderstorms would be reasonable.
Also, applying it to rainfall (and monsoons) would be fine as well.
Keep
in mind that Grid-Stat is the tool that computes FSS. And you run
Grid-Stat once for each evaluation time. If you're evaluating once
per
day, you'd run Grid-Stat 122 times for the 122 days of monsoon. And
you'd
end up with 122 FSS values. You could look at those as a time series,
or
you could use the Stat-Analysis tool to aggregate them together into a
single FSS value, like this:
stat_analysis -job aggregate -line_type NBRCNT -lookin
out/grid_stat
It all sounds like a reasonable approach to me. Just be sure to pick
thresholds (for the thunderstorms and monsoons) that capture the
"events"
that you're interested in studying.
Thanks,
John
On Fri, May 15, 2015 at 10:52 AM, Geeta Geeta via RT
<met_help at ucar.edu>
wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
>
> Hi John.
> Just want to confirm a few points.
> Suppose I calculate FSS for two seasons. say Summer (92 days) and
monsoon
> (122 days) in India.
> I plot FSS vs spatial Scale for both seasons.
>
> 1. How can aggregate the FSS for entire season . I mean I will have
one
> UNIQUE plot of FSS vs scale for a day ???. How will I aggregate???
> 2. Calculated FSS has to be greater than 0.5 at a particular spatial
scale
> to considered useful FCST by the model. Is this statement
correct????
> 3. the season having larger FSS means that MODEL has more skill as
> compared to the other season???
>
> Can U pls clarify.
> Geeta
>
> From: geeta124 at hotmail.com
> To: met_help at ucar.edu
> Subject: RE: [rt.rap.ucar.edu #71923] another problem with gridstat
> Date: Fri, 15 May 2015 22:09:07 +0530
>
>
>
>
> Hi John.
> I will download the image of 3B42 TRMM and send u. I am unable to
open the
> page now.
>
> I have one more query reg the FSS. I am trying to apply FSS for some
> thunderstorm simulations.
> so essentially, it will tell me if the frequency of FCST and OBS are
> close???
>
> Can I apply this technique for evaluation of the MONSOON rainfall
for a
> season (122 days in India) for entire southern India???. so for
this I
> will be taking a very large domain. I expect that in this case FSS
will be
> better as compared to that of the thunderstorm cases???
>
> Will this be a viable idea???.
> I need your feedback reg this.
>
> Geeta
>
>
> > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > From: met_help at ucar.edu
> > To: geeta124 at hotmail.com
> > Date: Wed, 13 May 2015 15:48:20 -0600
> >
> > Geeta,
> >
> > OK, the NetCDF matched pairs file you sent looks reasonable to me.
You
> are
> > comparing it to some plot from the TRMM website and say that they
differ.
> > Can you please send me the plot to which you're comparing this
data as
> well
> > as the link you used to grab it?
> >
> > I can't really comment on this until I understand the discrepancy
you're
> > seeing.
> >
> > Thanks,
> > John
> >
> >
> > On Wed, May 13, 2015 at 3:23 PM, Geeta Geeta via RT
<met_help at ucar.edu>
> > wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
> > >
> > > hi John. I just now did whole process of making Observed data
again.
> Now
> > > sending you new NC file.
> > > But the aggregated data of 8 observations (06UTC of 22 APR2011 -
03UTC
> of
> > > 23 April 2011) looks very different from the plotted data on
TRMM
> website.
> > >
> > > > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > > > From: met_help at ucar.edu
> > > > To: geeta124 at hotmail.com
> > > > Date: Wed, 13 May 2015 12:04:54 -0600
> > > >
> > > > Geeta,
> > > >
> > > > I agree. Those data values do look bad. I strongly suspect
that
> this
> > > > isn't a problem caused by grid_stat. Instead I'm guessing
that
> > > observation
> > > > data you're passing to grid_stat is bad. I ran the following
> command to
> > > > visualize the grid_stat pairs output and the resulting image
is
> attached.
> > > > The values in this data file range from 0 to 4.5e30!
> > > > met-5.0/bin/plot_data_plane
> > > grid_stat_270000L_20110423_030000V_pairs.nc
> > > > obs.ps 'name="OBS_APCP_A24_FULL"; level="(*,*)";'
> > > >
> > > > Please take a look at the observation data you're passing to
> grid_stat to
> > > > see if it's corrupted in some way.
> > > >
> > > > Thanks,
> > > > John
> > > >
> > > > On Wed, May 13, 2015 at 11:12 AM, Geeta Geeta via RT <
> met_help at ucar.edu>
> > > > wrote:
> > > >
> > > > >
> > > > > Wed May 13 11:12:32 2015: Request 71923 was acted upon.
> > > > > Transaction: Ticket created by geeta124 at hotmail.com
> > > > > Queue: met_help
> > > > > Subject: another problem with gridstat
> > > > > Owner: Nobody
> > > > > Requestors: geeta124 at hotmail.com
> > > > > Status: new
> > > > > Ticket <URL:
> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
> > > >
> > > > >
> > > > >
> > > > > Hi John. I have another stange problem. I had run grid stat
for
> another
> > > > > date 22 April2011. I got all outputs nicely. but when I try
to
> > > visualise
> > > > > the OBS output using the ncview, it looks very strtange. I
am
> sending
> > > the
> > > > > NC file for your reference.
> > > > > also when i saw the data in the OBS file, it shows me some
very
> strange
> > > > > values of precipitation. Absolutely unrealistic.
> > > > >
> > > > > -rw-rw-r-- 1 aditya aditya 286376 May 13 22:03
> > > > > grid_stat_270000L_20110423_030000V.stat-rw-rw-r-- 1 aditya
aditya
> > > 2750
> > > > > May 13 22:03 grid_stat_270000L_20110423_030000V_fho.txt-rw-
rw-r-- 1
> > > aditya
> > > > > aditya 2750 May 13 22:03
> > > > > grid_stat_270000L_20110423_030000V_ctc.txt-rw-rw-r-- 1
aditya
> aditya
> > > > > 268 May 13 22:03
> > > grid_stat_270000L_20110423_030000V_vl1l2.txt-rw-rw-r-- 1
> > > > > aditya aditya 882 May 13 22:03
> > > > > grid_stat_270000L_20110423_030000V_sl1l2.txt-rw-rw-r-- 1
aditya
> aditya
> > > > > 324 May 13 22:03
> grid_stat_270000L_20110423_030000V_pstd.txt-rw-rw-r--
> > > 1
> > > > > aditya aditya 229 May 13 22:03
> > > > > grid_stat_270000L_20110423_030000V_prc.txt-rw-rw-r-- 1
aditya
> aditya
> > > > > 229 May 13 22:03
> grid_stat_270000L_20110423_030000V_pjc.txt-rw-rw-r-- 1
> > > > > aditya aditya 229 May 13 22:03
> > > > > grid_stat_270000L_20110423_030000V_pct.txt-rw-rw-r-- 1
aditya
> aditya
> > > > > 20250 May 13 22:03
> > > grid_stat_270000L_20110423_030000V_nbrctc.txt-rw-rw-r--
> > > > > 1 aditya aditya 998 May 13 22:03
grid_stat_270000L_2011042!
> > > > > 3_030000V_mcts.txt-rw-rw-r-- 1 aditya aditya 2310 May
13 22:03
> > > > > grid_stat_270000L_20110423_030000V_mctc.txt-rw-rw-r-- 1
aditya
> aditya
> > > > > 11396 May 13 22:03
> > > grid_stat_270000L_20110423_030000V_cts.txt-rw-rw-r-- 1
> > > > > aditya aditya 3046 May 13 22:03
> > > > > grid_stat_270000L_20110423_030000V_cnt.txt-rw-rw-r-- 1
aditya
> aditya
> > > > > 84240 May 13 22:03
> > > grid_stat_270000L_20110423_030000V_nbrcts.txt-rw-rw-r--
> > > > > 1 aditya aditya 21644 May 13 22:03
> > > > > grid_stat_270000L_20110423_030000V_pairs.nc-rw-rw-r-- 1
aditya
> aditya
> > > > > 31428 May 13 22:03
> > > > > grid_stat_270000L_20110423_030000V_nbrcnt.txtaditya at agniilap
> > > > > :~/geeta/installs/met/RUN_MET-22apr2011/out-22apr2011$
> > > > > pls point me to right direction. geeta
> > > > >
> > > > >
> > > > >
> > > >
> > >
> > >
> >
>
>
------------------------------------------------
Subject: another problem with gridstat
From: Geeta Geeta
Time: Fri May 15 12:23:42 2015
thank s John.
also pls comment about Useful FCST aswell. and a few other things I
had asked in my mail.
> Subject: Re: [rt.rap.ucar.edu #71923] another problem with gridstat
> From: met_help at ucar.edu
> To: geeta124 at hotmail.com
> Date: Fri, 15 May 2015 12:11:40 -0600
>
> Geeta,
>
> Sure, you can compute the fractions skill score for comparing
forecast and
> observed thunderstorms. When computing FSS, you first threshold the
fields
> to define events and non-events. Then you look at successively
larger and
> larger areas around each grid point to see how the forecast event
frequency
> compares to the observed event frequency. Seems like applying that
to
> thunderstorms would be reasonable.
>
> Also, applying it to rainfall (and monsoons) would be fine as well.
Keep
> in mind that Grid-Stat is the tool that computes FSS. And you run
> Grid-Stat once for each evaluation time. If you're evaluating once
per
> day, you'd run Grid-Stat 122 times for the 122 days of monsoon. And
you'd
> end up with 122 FSS values. You could look at those as a time
series, or
> you could use the Stat-Analysis tool to aggregate them together into
a
> single FSS value, like this:
>
> stat_analysis -job aggregate -line_type NBRCNT -lookin
out/grid_stat
>
> It all sounds like a reasonable approach to me. Just be sure to
pick
> thresholds (for the thunderstorms and monsoons) that capture the
"events"
> that you're interested in studying.
>
> Thanks,
> John
>
> On Fri, May 15, 2015 at 10:52 AM, Geeta Geeta via RT
<met_help at ucar.edu>
> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
> >
> > Hi John.
> > Just want to confirm a few points.
> > Suppose I calculate FSS for two seasons. say Summer (92 days) and
monsoon
> > (122 days) in India.
> > I plot FSS vs spatial Scale for both seasons.
> >
> > 1. How can aggregate the FSS for entire season . I mean I will
have one
> > UNIQUE plot of FSS vs scale for a day ???. How will I aggregate???
> > 2. Calculated FSS has to be greater than 0.5 at a particular
spatial scale
> > to considered useful FCST by the model. Is this statement
correct????
> > 3. the season having larger FSS means that MODEL has more skill
as
> > compared to the other season???
> >
> > Can U pls clarify.
> > Geeta
> >
> > From: geeta124 at hotmail.com
> > To: met_help at ucar.edu
> > Subject: RE: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > Date: Fri, 15 May 2015 22:09:07 +0530
> >
> >
> >
> >
> > Hi John.
> > I will download the image of 3B42 TRMM and send u. I am unable to
open the
> > page now.
> >
> > I have one more query reg the FSS. I am trying to apply FSS for
some
> > thunderstorm simulations.
> > so essentially, it will tell me if the frequency of FCST and OBS
are
> > close???
> >
> > Can I apply this technique for evaluation of the MONSOON rainfall
for a
> > season (122 days in India) for entire southern India???. so for
this I
> > will be taking a very large domain. I expect that in this case FSS
will be
> > better as compared to that of the thunderstorm cases???
> >
> > Will this be a viable idea???.
> > I need your feedback reg this.
> >
> > Geeta
> >
> >
> > > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > > From: met_help at ucar.edu
> > > To: geeta124 at hotmail.com
> > > Date: Wed, 13 May 2015 15:48:20 -0600
> > >
> > > Geeta,
> > >
> > > OK, the NetCDF matched pairs file you sent looks reasonable to
me. You
> > are
> > > comparing it to some plot from the TRMM website and say that
they differ.
> > > Can you please send me the plot to which you're comparing this
data as
> > well
> > > as the link you used to grab it?
> > >
> > > I can't really comment on this until I understand the
discrepancy you're
> > > seeing.
> > >
> > > Thanks,
> > > John
> > >
> > >
> > > On Wed, May 13, 2015 at 3:23 PM, Geeta Geeta via RT
<met_help at ucar.edu>
> > > wrote:
> > >
> > > >
> > > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
>
> > > >
> > > > hi John. I just now did whole process of making Observed data
again.
> > Now
> > > > sending you new NC file.
> > > > But the aggregated data of 8 observations (06UTC of 22 APR2011
- 03UTC
> > of
> > > > 23 April 2011) looks very different from the plotted data on
TRMM
> > website.
> > > >
> > > > > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > > > > From: met_help at ucar.edu
> > > > > To: geeta124 at hotmail.com
> > > > > Date: Wed, 13 May 2015 12:04:54 -0600
> > > > >
> > > > > Geeta,
> > > > >
> > > > > I agree. Those data values do look bad. I strongly suspect
that
> > this
> > > > > isn't a problem caused by grid_stat. Instead I'm guessing
that
> > > > observation
> > > > > data you're passing to grid_stat is bad. I ran the
following
> > command to
> > > > > visualize the grid_stat pairs output and the resulting image
is
> > attached.
> > > > > The values in this data file range from 0 to 4.5e30!
> > > > > met-5.0/bin/plot_data_plane
> > > > grid_stat_270000L_20110423_030000V_pairs.nc
> > > > > obs.ps 'name="OBS_APCP_A24_FULL"; level="(*,*)";'
> > > > >
> > > > > Please take a look at the observation data you're passing to
> > grid_stat to
> > > > > see if it's corrupted in some way.
> > > > >
> > > > > Thanks,
> > > > > John
> > > > >
> > > > > On Wed, May 13, 2015 at 11:12 AM, Geeta Geeta via RT <
> > met_help at ucar.edu>
> > > > > wrote:
> > > > >
> > > > > >
> > > > > > Wed May 13 11:12:32 2015: Request 71923 was acted upon.
> > > > > > Transaction: Ticket created by geeta124 at hotmail.com
> > > > > > Queue: met_help
> > > > > > Subject: another problem with gridstat
> > > > > > Owner: Nobody
> > > > > > Requestors: geeta124 at hotmail.com
> > > > > > Status: new
> > > > > > Ticket <URL:
> > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
> > > > >
> > > > > >
> > > > > >
> > > > > > Hi John. I have another stange problem. I had run grid
stat for
> > another
> > > > > > date 22 April2011. I got all outputs nicely. but when I
try to
> > > > visualise
> > > > > > the OBS output using the ncview, it looks very strtange. I
am
> > sending
> > > > the
> > > > > > NC file for your reference.
> > > > > > also when i saw the data in the OBS file, it shows me some
very
> > strange
> > > > > > values of precipitation. Absolutely unrealistic.
> > > > > >
> > > > > > -rw-rw-r-- 1 aditya aditya 286376 May 13 22:03
> > > > > > grid_stat_270000L_20110423_030000V.stat-rw-rw-r-- 1 aditya
aditya
> > > > 2750
> > > > > > May 13 22:03 grid_stat_270000L_20110423_030000V_fho.txt-
rw-rw-r-- 1
> > > > aditya
> > > > > > aditya 2750 May 13 22:03
> > > > > > grid_stat_270000L_20110423_030000V_ctc.txt-rw-rw-r-- 1
aditya
> > aditya
> > > > > > 268 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_vl1l2.txt-rw-rw-r-- 1
> > > > > > aditya aditya 882 May 13 22:03
> > > > > > grid_stat_270000L_20110423_030000V_sl1l2.txt-rw-rw-r-- 1
aditya
> > aditya
> > > > > > 324 May 13 22:03
> > grid_stat_270000L_20110423_030000V_pstd.txt-rw-rw-r--
> > > > 1
> > > > > > aditya aditya 229 May 13 22:03
> > > > > > grid_stat_270000L_20110423_030000V_prc.txt-rw-rw-r-- 1
aditya
> > aditya
> > > > > > 229 May 13 22:03
> > grid_stat_270000L_20110423_030000V_pjc.txt-rw-rw-r-- 1
> > > > > > aditya aditya 229 May 13 22:03
> > > > > > grid_stat_270000L_20110423_030000V_pct.txt-rw-rw-r-- 1
aditya
> > aditya
> > > > > > 20250 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_nbrctc.txt-rw-rw-r--
> > > > > > 1 aditya aditya 998 May 13 22:03
grid_stat_270000L_2011042!
> > > > > > 3_030000V_mcts.txt-rw-rw-r-- 1 aditya aditya 2310 May
13 22:03
> > > > > > grid_stat_270000L_20110423_030000V_mctc.txt-rw-rw-r-- 1
aditya
> > aditya
> > > > > > 11396 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_cts.txt-rw-rw-r-- 1
> > > > > > aditya aditya 3046 May 13 22:03
> > > > > > grid_stat_270000L_20110423_030000V_cnt.txt-rw-rw-r-- 1
aditya
> > aditya
> > > > > > 84240 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_nbrcts.txt-rw-rw-r--
> > > > > > 1 aditya aditya 21644 May 13 22:03
> > > > > > grid_stat_270000L_20110423_030000V_pairs.nc-rw-rw-r-- 1
aditya
> > aditya
> > > > > > 31428 May 13 22:03
> > > > > >
grid_stat_270000L_20110423_030000V_nbrcnt.txtaditya at agniilap
> > > > > > :~/geeta/installs/met/RUN_MET-22apr2011/out-22apr2011$
> > > > > > pls point me to right direction. geeta
> > > > > >
> > > > > >
> > > > > >
> > > > >
> > > >
> > > >
> > >
> >
> >
>
------------------------------------------------
Subject: another problem with gridstat
From: John Halley Gotway
Time: Fri May 15 13:27:16 2015
Geeta,
MET version 5.0 includes some additions to the NBRCNT line type... it
contains values for FSS, UFSS, and AFSS which are Fractions Skill
Score,
Uniform FSS, and Asymptotic FSS.
UFSS is computed as 0.5 + base_rate / 2.0. In your email, you stated
that
FSS is only considered useful if it's > 0.5. I believe you're
actually
supposed to compare it to UFSS, which is always a little bit greater
than
0.5. AFSS is computed as (2.0 * frate * orate) / (frate*frate +
orate*orate), where frate and orate are the fcst and obs rates
respectively. I believe that AFSS gives you the maximum FSS value
possible
given the number for forecast and observed events.
I believe I already answered your question about aggregating. You can
use
the stat_analysis tool to do this. I'd suggest using the
-fcst_init_beg
and -fcst_init_end or -fcst_valid_beg and -fcst_valid_end options for
stat_analysis to define the beginning/ending dates for the Summer and
Monsoon seasons.
As for interpreting FSS, yes, the higher the better. FSS is always
between
0 and 1, and the closer to 1, the better the performance.
Thanks,
John
On Fri, May 15, 2015 at 12:23 PM, Geeta Geeta via RT
<met_help at ucar.edu>
wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
>
> thank s John.
> also pls comment about Useful FCST aswell. and a few other things I
had
> asked in my mail.
>
>
> > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > From: met_help at ucar.edu
> > To: geeta124 at hotmail.com
> > Date: Fri, 15 May 2015 12:11:40 -0600
> >
> > Geeta,
> >
> > Sure, you can compute the fractions skill score for comparing
forecast
> and
> > observed thunderstorms. When computing FSS, you first threshold
the
> fields
> > to define events and non-events. Then you look at successively
larger
> and
> > larger areas around each grid point to see how the forecast event
> frequency
> > compares to the observed event frequency. Seems like applying
that to
> > thunderstorms would be reasonable.
> >
> > Also, applying it to rainfall (and monsoons) would be fine as
well. Keep
> > in mind that Grid-Stat is the tool that computes FSS. And you run
> > Grid-Stat once for each evaluation time. If you're evaluating
once per
> > day, you'd run Grid-Stat 122 times for the 122 days of monsoon.
And
> you'd
> > end up with 122 FSS values. You could look at those as a time
series, or
> > you could use the Stat-Analysis tool to aggregate them together
into a
> > single FSS value, like this:
> >
> > stat_analysis -job aggregate -line_type NBRCNT -lookin
out/grid_stat
> >
> > It all sounds like a reasonable approach to me. Just be sure to
pick
> > thresholds (for the thunderstorms and monsoons) that capture the
"events"
> > that you're interested in studying.
> >
> > Thanks,
> > John
> >
> > On Fri, May 15, 2015 at 10:52 AM, Geeta Geeta via RT
<met_help at ucar.edu>
> > wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
> > >
> > > Hi John.
> > > Just want to confirm a few points.
> > > Suppose I calculate FSS for two seasons. say Summer (92 days)
and
> monsoon
> > > (122 days) in India.
> > > I plot FSS vs spatial Scale for both seasons.
> > >
> > > 1. How can aggregate the FSS for entire season . I mean I will
have one
> > > UNIQUE plot of FSS vs scale for a day ???. How will I
aggregate???
> > > 2. Calculated FSS has to be greater than 0.5 at a particular
spatial
> scale
> > > to considered useful FCST by the model. Is this statement
correct????
> > > 3. the season having larger FSS means that MODEL has more skill
as
> > > compared to the other season???
> > >
> > > Can U pls clarify.
> > > Geeta
> > >
> > > From: geeta124 at hotmail.com
> > > To: met_help at ucar.edu
> > > Subject: RE: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > > Date: Fri, 15 May 2015 22:09:07 +0530
> > >
> > >
> > >
> > >
> > > Hi John.
> > > I will download the image of 3B42 TRMM and send u. I am unable
to open
> the
> > > page now.
> > >
> > > I have one more query reg the FSS. I am trying to apply FSS for
some
> > > thunderstorm simulations.
> > > so essentially, it will tell me if the frequency of FCST and OBS
are
> > > close???
> > >
> > > Can I apply this technique for evaluation of the MONSOON
rainfall for a
> > > season (122 days in India) for entire southern India???. so for
this I
> > > will be taking a very large domain. I expect that in this case
FSS
> will be
> > > better as compared to that of the thunderstorm cases???
> > >
> > > Will this be a viable idea???.
> > > I need your feedback reg this.
> > >
> > > Geeta
> > >
> > >
> > > > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > > > From: met_help at ucar.edu
> > > > To: geeta124 at hotmail.com
> > > > Date: Wed, 13 May 2015 15:48:20 -0600
> > > >
> > > > Geeta,
> > > >
> > > > OK, the NetCDF matched pairs file you sent looks reasonable to
me.
> You
> > > are
> > > > comparing it to some plot from the TRMM website and say that
they
> differ.
> > > > Can you please send me the plot to which you're comparing this
data
> as
> > > well
> > > > as the link you used to grab it?
> > > >
> > > > I can't really comment on this until I understand the
discrepancy
> you're
> > > > seeing.
> > > >
> > > > Thanks,
> > > > John
> > > >
> > > >
> > > > On Wed, May 13, 2015 at 3:23 PM, Geeta Geeta via RT <
> met_help at ucar.edu>
> > > > wrote:
> > > >
> > > > >
> > > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
> > > > >
> > > > > hi John. I just now did whole process of making Observed
data
> again.
> > > Now
> > > > > sending you new NC file.
> > > > > But the aggregated data of 8 observations (06UTC of 22
APR2011 -
> 03UTC
> > > of
> > > > > 23 April 2011) looks very different from the plotted data on
TRMM
> > > website.
> > > > >
> > > > > > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
> gridstat
> > > > > > From: met_help at ucar.edu
> > > > > > To: geeta124 at hotmail.com
> > > > > > Date: Wed, 13 May 2015 12:04:54 -0600
> > > > > >
> > > > > > Geeta,
> > > > > >
> > > > > > I agree. Those data values do look bad. I strongly
suspect that
> > > this
> > > > > > isn't a problem caused by grid_stat. Instead I'm guessing
that
> > > > > observation
> > > > > > data you're passing to grid_stat is bad. I ran the
following
> > > command to
> > > > > > visualize the grid_stat pairs output and the resulting
image is
> > > attached.
> > > > > > The values in this data file range from 0 to 4.5e30!
> > > > > > met-5.0/bin/plot_data_plane
> > > > > grid_stat_270000L_20110423_030000V_pairs.nc
> > > > > > obs.ps 'name="OBS_APCP_A24_FULL"; level="(*,*)";'
> > > > > >
> > > > > > Please take a look at the observation data you're passing
to
> > > grid_stat to
> > > > > > see if it's corrupted in some way.
> > > > > >
> > > > > > Thanks,
> > > > > > John
> > > > > >
> > > > > > On Wed, May 13, 2015 at 11:12 AM, Geeta Geeta via RT <
> > > met_help at ucar.edu>
> > > > > > wrote:
> > > > > >
> > > > > > >
> > > > > > > Wed May 13 11:12:32 2015: Request 71923 was acted upon.
> > > > > > > Transaction: Ticket created by geeta124 at hotmail.com
> > > > > > > Queue: met_help
> > > > > > > Subject: another problem with gridstat
> > > > > > > Owner: Nobody
> > > > > > > Requestors: geeta124 at hotmail.com
> > > > > > > Status: new
> > > > > > > Ticket <URL:
> > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
> > > > > >
> > > > > > >
> > > > > > >
> > > > > > > Hi John. I have another stange problem. I had run grid
stat for
> > > another
> > > > > > > date 22 April2011. I got all outputs nicely. but when I
try to
> > > > > visualise
> > > > > > > the OBS output using the ncview, it looks very strtange.
I am
> > > sending
> > > > > the
> > > > > > > NC file for your reference.
> > > > > > > also when i saw the data in the OBS file, it shows me
some very
> > > strange
> > > > > > > values of precipitation. Absolutely unrealistic.
> > > > > > >
> > > > > > > -rw-rw-r-- 1 aditya aditya 286376 May 13 22:03
> > > > > > > grid_stat_270000L_20110423_030000V.stat-rw-rw-r-- 1
aditya
> aditya
> > > > > 2750
> > > > > > > May 13 22:03
> grid_stat_270000L_20110423_030000V_fho.txt-rw-rw-r-- 1
> > > > > aditya
> > > > > > > aditya 2750 May 13 22:03
> > > > > > > grid_stat_270000L_20110423_030000V_ctc.txt-rw-rw-r-- 1
aditya
> > > aditya
> > > > > > > 268 May 13 22:03
> > > > > grid_stat_270000L_20110423_030000V_vl1l2.txt-rw-rw-r-- 1
> > > > > > > aditya aditya 882 May 13 22:03
> > > > > > > grid_stat_270000L_20110423_030000V_sl1l2.txt-rw-rw-r-- 1
aditya
> > > aditya
> > > > > > > 324 May 13 22:03
> > > grid_stat_270000L_20110423_030000V_pstd.txt-rw-rw-r--
> > > > > 1
> > > > > > > aditya aditya 229 May 13 22:03
> > > > > > > grid_stat_270000L_20110423_030000V_prc.txt-rw-rw-r-- 1
aditya
> > > aditya
> > > > > > > 229 May 13 22:03
> > > grid_stat_270000L_20110423_030000V_pjc.txt-rw-rw-r-- 1
> > > > > > > aditya aditya 229 May 13 22:03
> > > > > > > grid_stat_270000L_20110423_030000V_pct.txt-rw-rw-r-- 1
aditya
> > > aditya
> > > > > > > 20250 May 13 22:03
> > > > > grid_stat_270000L_20110423_030000V_nbrctc.txt-rw-rw-r--
> > > > > > > 1 aditya aditya 998 May 13 22:03
> grid_stat_270000L_2011042!
> > > > > > > 3_030000V_mcts.txt-rw-rw-r-- 1 aditya aditya 2310
May 13
> 22:03
> > > > > > > grid_stat_270000L_20110423_030000V_mctc.txt-rw-rw-r-- 1
aditya
> > > aditya
> > > > > > > 11396 May 13 22:03
> > > > > grid_stat_270000L_20110423_030000V_cts.txt-rw-rw-r-- 1
> > > > > > > aditya aditya 3046 May 13 22:03
> > > > > > > grid_stat_270000L_20110423_030000V_cnt.txt-rw-rw-r-- 1
aditya
> > > aditya
> > > > > > > 84240 May 13 22:03
> > > > > grid_stat_270000L_20110423_030000V_nbrcts.txt-rw-rw-r--
> > > > > > > 1 aditya aditya 21644 May 13 22:03
> > > > > > > grid_stat_270000L_20110423_030000V_pairs.nc-rw-rw-r-- 1
aditya
> > > aditya
> > > > > > > 31428 May 13 22:03
> > > > > > >
grid_stat_270000L_20110423_030000V_nbrcnt.txtaditya at agniilap
> > > > > > > :~/geeta/installs/met/RUN_MET-22apr2011/out-22apr2011$
> > > > > > > pls point me to right direction. geeta
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > > >
> > > >
> > >
> > >
> >
>
>
------------------------------------------------
Subject: another problem with gridstat
From: Geeta Geeta
Time: Sat May 16 12:22:38 2015
hi John.
sorry for another query.
At present i am taking the thresolds (>=0.1, >=2.5 ) etc for
converting FCST and OBS fileds to binary.
My question is if I have to use 90 th percentile, 75th percentile etc
as the the thresholds, then how will I go about doing that????
what I should do. ???
Geeta
> Subject: Re: [rt.rap.ucar.edu #71923] another problem with gridstat
> From: met_help at ucar.edu
> To: geeta124 at hotmail.com
> Date: Fri, 15 May 2015 13:27:17 -0600
>
> Geeta,
>
> MET version 5.0 includes some additions to the NBRCNT line type...
it
> contains values for FSS, UFSS, and AFSS which are Fractions Skill
Score,
> Uniform FSS, and Asymptotic FSS.
>
> UFSS is computed as 0.5 + base_rate / 2.0. In your email, you
stated that
> FSS is only considered useful if it's > 0.5. I believe you're
actually
> supposed to compare it to UFSS, which is always a little bit greater
than
> 0.5. AFSS is computed as (2.0 * frate * orate) / (frate*frate +
> orate*orate), where frate and orate are the fcst and obs rates
> respectively. I believe that AFSS gives you the maximum FSS value
possible
> given the number for forecast and observed events.
>
> I believe I already answered your question about aggregating. You
can use
> the stat_analysis tool to do this. I'd suggest using the
-fcst_init_beg
> and -fcst_init_end or -fcst_valid_beg and -fcst_valid_end options
for
> stat_analysis to define the beginning/ending dates for the Summer
and
> Monsoon seasons.
>
> As for interpreting FSS, yes, the higher the better. FSS is always
between
> 0 and 1, and the closer to 1, the better the performance.
>
> Thanks,
> John
>
>
>
> On Fri, May 15, 2015 at 12:23 PM, Geeta Geeta via RT
<met_help at ucar.edu>
> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
> >
> > thank s John.
> > also pls comment about Useful FCST aswell. and a few other things
I had
> > asked in my mail.
> >
> >
> > > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > > From: met_help at ucar.edu
> > > To: geeta124 at hotmail.com
> > > Date: Fri, 15 May 2015 12:11:40 -0600
> > >
> > > Geeta,
> > >
> > > Sure, you can compute the fractions skill score for comparing
forecast
> > and
> > > observed thunderstorms. When computing FSS, you first threshold
the
> > fields
> > > to define events and non-events. Then you look at successively
larger
> > and
> > > larger areas around each grid point to see how the forecast
event
> > frequency
> > > compares to the observed event frequency. Seems like applying
that to
> > > thunderstorms would be reasonable.
> > >
> > > Also, applying it to rainfall (and monsoons) would be fine as
well. Keep
> > > in mind that Grid-Stat is the tool that computes FSS. And you
run
> > > Grid-Stat once for each evaluation time. If you're evaluating
once per
> > > day, you'd run Grid-Stat 122 times for the 122 days of monsoon.
And
> > you'd
> > > end up with 122 FSS values. You could look at those as a time
series, or
> > > you could use the Stat-Analysis tool to aggregate them together
into a
> > > single FSS value, like this:
> > >
> > > stat_analysis -job aggregate -line_type NBRCNT -lookin
out/grid_stat
> > >
> > > It all sounds like a reasonable approach to me. Just be sure to
pick
> > > thresholds (for the thunderstorms and monsoons) that capture the
"events"
> > > that you're interested in studying.
> > >
> > > Thanks,
> > > John
> > >
> > > On Fri, May 15, 2015 at 10:52 AM, Geeta Geeta via RT
<met_help at ucar.edu>
> > > wrote:
> > >
> > > >
> > > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
>
> > > >
> > > > Hi John.
> > > > Just want to confirm a few points.
> > > > Suppose I calculate FSS for two seasons. say Summer (92 days)
and
> > monsoon
> > > > (122 days) in India.
> > > > I plot FSS vs spatial Scale for both seasons.
> > > >
> > > > 1. How can aggregate the FSS for entire season . I mean I will
have one
> > > > UNIQUE plot of FSS vs scale for a day ???. How will I
aggregate???
> > > > 2. Calculated FSS has to be greater than 0.5 at a particular
spatial
> > scale
> > > > to considered useful FCST by the model. Is this statement
correct????
> > > > 3. the season having larger FSS means that MODEL has more
skill as
> > > > compared to the other season???
> > > >
> > > > Can U pls clarify.
> > > > Geeta
> > > >
> > > > From: geeta124 at hotmail.com
> > > > To: met_help at ucar.edu
> > > > Subject: RE: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > > > Date: Fri, 15 May 2015 22:09:07 +0530
> > > >
> > > >
> > > >
> > > >
> > > > Hi John.
> > > > I will download the image of 3B42 TRMM and send u. I am unable
to open
> > the
> > > > page now.
> > > >
> > > > I have one more query reg the FSS. I am trying to apply FSS
for some
> > > > thunderstorm simulations.
> > > > so essentially, it will tell me if the frequency of FCST and
OBS are
> > > > close???
> > > >
> > > > Can I apply this technique for evaluation of the MONSOON
rainfall for a
> > > > season (122 days in India) for entire southern India???. so
for this I
> > > > will be taking a very large domain. I expect that in this case
FSS
> > will be
> > > > better as compared to that of the thunderstorm cases???
> > > >
> > > > Will this be a viable idea???.
> > > > I need your feedback reg this.
> > > >
> > > > Geeta
> > > >
> > > >
> > > > > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > > > > From: met_help at ucar.edu
> > > > > To: geeta124 at hotmail.com
> > > > > Date: Wed, 13 May 2015 15:48:20 -0600
> > > > >
> > > > > Geeta,
> > > > >
> > > > > OK, the NetCDF matched pairs file you sent looks reasonable
to me.
> > You
> > > > are
> > > > > comparing it to some plot from the TRMM website and say that
they
> > differ.
> > > > > Can you please send me the plot to which you're comparing
this data
> > as
> > > > well
> > > > > as the link you used to grab it?
> > > > >
> > > > > I can't really comment on this until I understand the
discrepancy
> > you're
> > > > > seeing.
> > > > >
> > > > > Thanks,
> > > > > John
> > > > >
> > > > >
> > > > > On Wed, May 13, 2015 at 3:23 PM, Geeta Geeta via RT <
> > met_help at ucar.edu>
> > > > > wrote:
> > > > >
> > > > > >
> > > > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
> > > > > >
> > > > > > hi John. I just now did whole process of making Observed
data
> > again.
> > > > Now
> > > > > > sending you new NC file.
> > > > > > But the aggregated data of 8 observations (06UTC of 22
APR2011 -
> > 03UTC
> > > > of
> > > > > > 23 April 2011) looks very different from the plotted data
on TRMM
> > > > website.
> > > > > >
> > > > > > > Subject: Re: [rt.rap.ucar.edu #71923] another problem
with
> > gridstat
> > > > > > > From: met_help at ucar.edu
> > > > > > > To: geeta124 at hotmail.com
> > > > > > > Date: Wed, 13 May 2015 12:04:54 -0600
> > > > > > >
> > > > > > > Geeta,
> > > > > > >
> > > > > > > I agree. Those data values do look bad. I strongly
suspect that
> > > > this
> > > > > > > isn't a problem caused by grid_stat. Instead I'm
guessing that
> > > > > > observation
> > > > > > > data you're passing to grid_stat is bad. I ran the
following
> > > > command to
> > > > > > > visualize the grid_stat pairs output and the resulting
image is
> > > > attached.
> > > > > > > The values in this data file range from 0 to 4.5e30!
> > > > > > > met-5.0/bin/plot_data_plane
> > > > > > grid_stat_270000L_20110423_030000V_pairs.nc
> > > > > > > obs.ps 'name="OBS_APCP_A24_FULL"; level="(*,*)";'
> > > > > > >
> > > > > > > Please take a look at the observation data you're
passing to
> > > > grid_stat to
> > > > > > > see if it's corrupted in some way.
> > > > > > >
> > > > > > > Thanks,
> > > > > > > John
> > > > > > >
> > > > > > > On Wed, May 13, 2015 at 11:12 AM, Geeta Geeta via RT <
> > > > met_help at ucar.edu>
> > > > > > > wrote:
> > > > > > >
> > > > > > > >
> > > > > > > > Wed May 13 11:12:32 2015: Request 71923 was acted
upon.
> > > > > > > > Transaction: Ticket created by geeta124 at hotmail.com
> > > > > > > > Queue: met_help
> > > > > > > > Subject: another problem with gridstat
> > > > > > > > Owner: Nobody
> > > > > > > > Requestors: geeta124 at hotmail.com
> > > > > > > > Status: new
> > > > > > > > Ticket <URL:
> > > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
> > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > > > Hi John. I have another stange problem. I had run grid
stat for
> > > > another
> > > > > > > > date 22 April2011. I got all outputs nicely. but when
I try to
> > > > > > visualise
> > > > > > > > the OBS output using the ncview, it looks very
strtange. I am
> > > > sending
> > > > > > the
> > > > > > > > NC file for your reference.
> > > > > > > > also when i saw the data in the OBS file, it shows me
some very
> > > > strange
> > > > > > > > values of precipitation. Absolutely unrealistic.
> > > > > > > >
> > > > > > > > -rw-rw-r-- 1 aditya aditya 286376 May 13 22:03
> > > > > > > > grid_stat_270000L_20110423_030000V.stat-rw-rw-r-- 1
aditya
> > aditya
> > > > > > 2750
> > > > > > > > May 13 22:03
> > grid_stat_270000L_20110423_030000V_fho.txt-rw-rw-r-- 1
> > > > > > aditya
> > > > > > > > aditya 2750 May 13 22:03
> > > > > > > > grid_stat_270000L_20110423_030000V_ctc.txt-rw-rw-r-- 1
aditya
> > > > aditya
> > > > > > > > 268 May 13 22:03
> > > > > > grid_stat_270000L_20110423_030000V_vl1l2.txt-rw-rw-r-- 1
> > > > > > > > aditya aditya 882 May 13 22:03
> > > > > > > > grid_stat_270000L_20110423_030000V_sl1l2.txt-rw-rw-r--
1 aditya
> > > > aditya
> > > > > > > > 324 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_pstd.txt-rw-rw-r--
> > > > > > 1
> > > > > > > > aditya aditya 229 May 13 22:03
> > > > > > > > grid_stat_270000L_20110423_030000V_prc.txt-rw-rw-r-- 1
aditya
> > > > aditya
> > > > > > > > 229 May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_pjc.txt-rw-rw-r-- 1
> > > > > > > > aditya aditya 229 May 13 22:03
> > > > > > > > grid_stat_270000L_20110423_030000V_pct.txt-rw-rw-r-- 1
aditya
> > > > aditya
> > > > > > > > 20250 May 13 22:03
> > > > > > grid_stat_270000L_20110423_030000V_nbrctc.txt-rw-rw-r--
> > > > > > > > 1 aditya aditya 998 May 13 22:03
> > grid_stat_270000L_2011042!
> > > > > > > > 3_030000V_mcts.txt-rw-rw-r-- 1 aditya aditya 2310
May 13
> > 22:03
> > > > > > > > grid_stat_270000L_20110423_030000V_mctc.txt-rw-rw-r--
1 aditya
> > > > aditya
> > > > > > > > 11396 May 13 22:03
> > > > > > grid_stat_270000L_20110423_030000V_cts.txt-rw-rw-r-- 1
> > > > > > > > aditya aditya 3046 May 13 22:03
> > > > > > > > grid_stat_270000L_20110423_030000V_cnt.txt-rw-rw-r-- 1
aditya
> > > > aditya
> > > > > > > > 84240 May 13 22:03
> > > > > > grid_stat_270000L_20110423_030000V_nbrcts.txt-rw-rw-r--
> > > > > > > > 1 aditya aditya 21644 May 13 22:03
> > > > > > > > grid_stat_270000L_20110423_030000V_pairs.nc-rw-rw-r--
1 aditya
> > > > aditya
> > > > > > > > 31428 May 13 22:03
> > > > > > > >
grid_stat_270000L_20110423_030000V_nbrcnt.txtaditya at agniilap
> > > > > > > > :~/geeta/installs/met/RUN_MET-22apr2011/out-22apr2011$
> > > > > > > > pls point me to right direction. geeta
> > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > > >
> > > > >
> > > >
> > > >
> > >
> >
> >
>
------------------------------------------------
Subject: another problem with gridstat
From: John Halley Gotway
Time: Mon May 18 09:44:10 2015
Geeta,
MET does not currently support percentile thresholding. However, we
have
it defined as a development task and plan to add it in the next year.
For
the time being, you'd have to compute the percentile value outside of
MET
and use that to define your threshold.
Thanks,
John
On Sat, May 16, 2015 at 12:22 PM, Geeta Geeta via RT
<met_help at ucar.edu>
wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
>
> hi John.
> sorry for another query.
> At present i am taking the thresolds (>=0.1, >=2.5 ) etc for
converting
> FCST and OBS fileds to binary.
> My question is if I have to use 90 th percentile, 75th percentile
etc as
> the the thresholds, then how will I go about doing that????
>
> what I should do. ???
> Geeta
>
> > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > From: met_help at ucar.edu
> > To: geeta124 at hotmail.com
> > Date: Fri, 15 May 2015 13:27:17 -0600
> >
> > Geeta,
> >
> > MET version 5.0 includes some additions to the NBRCNT line type...
it
> > contains values for FSS, UFSS, and AFSS which are Fractions Skill
Score,
> > Uniform FSS, and Asymptotic FSS.
> >
> > UFSS is computed as 0.5 + base_rate / 2.0. In your email, you
stated
> that
> > FSS is only considered useful if it's > 0.5. I believe you're
actually
> > supposed to compare it to UFSS, which is always a little bit
greater than
> > 0.5. AFSS is computed as (2.0 * frate * orate) / (frate*frate +
> > orate*orate), where frate and orate are the fcst and obs rates
> > respectively. I believe that AFSS gives you the maximum FSS value
> possible
> > given the number for forecast and observed events.
> >
> > I believe I already answered your question about aggregating. You
can
> use
> > the stat_analysis tool to do this. I'd suggest using the
-fcst_init_beg
> > and -fcst_init_end or -fcst_valid_beg and -fcst_valid_end options
for
> > stat_analysis to define the beginning/ending dates for the Summer
and
> > Monsoon seasons.
> >
> > As for interpreting FSS, yes, the higher the better. FSS is
always
> between
> > 0 and 1, and the closer to 1, the better the performance.
> >
> > Thanks,
> > John
> >
> >
> >
> > On Fri, May 15, 2015 at 12:23 PM, Geeta Geeta via RT
<met_help at ucar.edu>
> > wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
> > >
> > > thank s John.
> > > also pls comment about Useful FCST aswell. and a few other
things I had
> > > asked in my mail.
> > >
> > >
> > > > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > > > From: met_help at ucar.edu
> > > > To: geeta124 at hotmail.com
> > > > Date: Fri, 15 May 2015 12:11:40 -0600
> > > >
> > > > Geeta,
> > > >
> > > > Sure, you can compute the fractions skill score for comparing
> forecast
> > > and
> > > > observed thunderstorms. When computing FSS, you first
threshold the
> > > fields
> > > > to define events and non-events. Then you look at
successively
> larger
> > > and
> > > > larger areas around each grid point to see how the forecast
event
> > > frequency
> > > > compares to the observed event frequency. Seems like applying
that
> to
> > > > thunderstorms would be reasonable.
> > > >
> > > > Also, applying it to rainfall (and monsoons) would be fine as
well.
> Keep
> > > > in mind that Grid-Stat is the tool that computes FSS. And you
run
> > > > Grid-Stat once for each evaluation time. If you're evaluating
once
> per
> > > > day, you'd run Grid-Stat 122 times for the 122 days of
monsoon. And
> > > you'd
> > > > end up with 122 FSS values. You could look at those as a time
> series, or
> > > > you could use the Stat-Analysis tool to aggregate them
together into
> a
> > > > single FSS value, like this:
> > > >
> > > > stat_analysis -job aggregate -line_type NBRCNT -lookin
> out/grid_stat
> > > >
> > > > It all sounds like a reasonable approach to me. Just be sure
to pick
> > > > thresholds (for the thunderstorms and monsoons) that capture
the
> "events"
> > > > that you're interested in studying.
> > > >
> > > > Thanks,
> > > > John
> > > >
> > > > On Fri, May 15, 2015 at 10:52 AM, Geeta Geeta via RT <
> met_help at ucar.edu>
> > > > wrote:
> > > >
> > > > >
> > > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
> > > > >
> > > > > Hi John.
> > > > > Just want to confirm a few points.
> > > > > Suppose I calculate FSS for two seasons. say Summer (92
days) and
> > > monsoon
> > > > > (122 days) in India.
> > > > > I plot FSS vs spatial Scale for both seasons.
> > > > >
> > > > > 1. How can aggregate the FSS for entire season . I mean I
will
> have one
> > > > > UNIQUE plot of FSS vs scale for a day ???. How will I
aggregate???
> > > > > 2. Calculated FSS has to be greater than 0.5 at a particular
> spatial
> > > scale
> > > > > to considered useful FCST by the model. Is this statement
> correct????
> > > > > 3. the season having larger FSS means that MODEL has more
skill as
> > > > > compared to the other season???
> > > > >
> > > > > Can U pls clarify.
> > > > > Geeta
> > > > >
> > > > > From: geeta124 at hotmail.com
> > > > > To: met_help at ucar.edu
> > > > > Subject: RE: [rt.rap.ucar.edu #71923] another problem with
> gridstat
> > > > > Date: Fri, 15 May 2015 22:09:07 +0530
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > Hi John.
> > > > > I will download the image of 3B42 TRMM and send u. I am
unable to
> open
> > > the
> > > > > page now.
> > > > >
> > > > > I have one more query reg the FSS. I am trying to apply FSS
for
> some
> > > > > thunderstorm simulations.
> > > > > so essentially, it will tell me if the frequency of FCST and
OBS
> are
> > > > > close???
> > > > >
> > > > > Can I apply this technique for evaluation of the MONSOON
rainfall
> for a
> > > > > season (122 days in India) for entire southern India???. so
for
> this I
> > > > > will be taking a very large domain. I expect that in this
case FSS
> > > will be
> > > > > better as compared to that of the thunderstorm cases???
> > > > >
> > > > > Will this be a viable idea???.
> > > > > I need your feedback reg this.
> > > > >
> > > > > Geeta
> > > > >
> > > > >
> > > > > > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
> gridstat
> > > > > > From: met_help at ucar.edu
> > > > > > To: geeta124 at hotmail.com
> > > > > > Date: Wed, 13 May 2015 15:48:20 -0600
> > > > > >
> > > > > > Geeta,
> > > > > >
> > > > > > OK, the NetCDF matched pairs file you sent looks
reasonable to
> me.
> > > You
> > > > > are
> > > > > > comparing it to some plot from the TRMM website and say
that they
> > > differ.
> > > > > > Can you please send me the plot to which you're comparing
this
> data
> > > as
> > > > > well
> > > > > > as the link you used to grab it?
> > > > > >
> > > > > > I can't really comment on this until I understand the
discrepancy
> > > you're
> > > > > > seeing.
> > > > > >
> > > > > > Thanks,
> > > > > > John
> > > > > >
> > > > > >
> > > > > > On Wed, May 13, 2015 at 3:23 PM, Geeta Geeta via RT <
> > > met_help at ucar.edu>
> > > > > > wrote:
> > > > > >
> > > > > > >
> > > > > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
> >
> > > > > > >
> > > > > > > hi John. I just now did whole process of making Observed
data
> > > again.
> > > > > Now
> > > > > > > sending you new NC file.
> > > > > > > But the aggregated data of 8 observations (06UTC of 22
APR2011
> -
> > > 03UTC
> > > > > of
> > > > > > > 23 April 2011) looks very different from the plotted
data on
> TRMM
> > > > > website.
> > > > > > >
> > > > > > > > Subject: Re: [rt.rap.ucar.edu #71923] another problem
with
> > > gridstat
> > > > > > > > From: met_help at ucar.edu
> > > > > > > > To: geeta124 at hotmail.com
> > > > > > > > Date: Wed, 13 May 2015 12:04:54 -0600
> > > > > > > >
> > > > > > > > Geeta,
> > > > > > > >
> > > > > > > > I agree. Those data values do look bad. I strongly
suspect
> that
> > > > > this
> > > > > > > > isn't a problem caused by grid_stat. Instead I'm
guessing
> that
> > > > > > > observation
> > > > > > > > data you're passing to grid_stat is bad. I ran the
following
> > > > > command to
> > > > > > > > visualize the grid_stat pairs output and the resulting
image
> is
> > > > > attached.
> > > > > > > > The values in this data file range from 0 to 4.5e30!
> > > > > > > > met-5.0/bin/plot_data_plane
> > > > > > > grid_stat_270000L_20110423_030000V_pairs.nc
> > > > > > > > obs.ps 'name="OBS_APCP_A24_FULL"; level="(*,*)";'
> > > > > > > >
> > > > > > > > Please take a look at the observation data you're
passing to
> > > > > grid_stat to
> > > > > > > > see if it's corrupted in some way.
> > > > > > > >
> > > > > > > > Thanks,
> > > > > > > > John
> > > > > > > >
> > > > > > > > On Wed, May 13, 2015 at 11:12 AM, Geeta Geeta via RT <
> > > > > met_help at ucar.edu>
> > > > > > > > wrote:
> > > > > > > >
> > > > > > > > >
> > > > > > > > > Wed May 13 11:12:32 2015: Request 71923 was acted
upon.
> > > > > > > > > Transaction: Ticket created by geeta124 at hotmail.com
> > > > > > > > > Queue: met_help
> > > > > > > > > Subject: another problem with gridstat
> > > > > > > > > Owner: Nobody
> > > > > > > > > Requestors: geeta124 at hotmail.com
> > > > > > > > > Status: new
> > > > > > > > > Ticket <URL:
> > > > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
> > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > Hi John. I have another stange problem. I had run
grid
> stat for
> > > > > another
> > > > > > > > > date 22 April2011. I got all outputs nicely. but
when I
> try to
> > > > > > > visualise
> > > > > > > > > the OBS output using the ncview, it looks very
strtange. I
> am
> > > > > sending
> > > > > > > the
> > > > > > > > > NC file for your reference.
> > > > > > > > > also when i saw the data in the OBS file, it shows
me some
> very
> > > > > strange
> > > > > > > > > values of precipitation. Absolutely unrealistic.
> > > > > > > > >
> > > > > > > > > -rw-rw-r-- 1 aditya aditya 286376 May 13 22:03
> > > > > > > > > grid_stat_270000L_20110423_030000V.stat-rw-rw-r-- 1
aditya
> > > aditya
> > > > > > > 2750
> > > > > > > > > May 13 22:03
> > > grid_stat_270000L_20110423_030000V_fho.txt-rw-rw-r-- 1
> > > > > > > aditya
> > > > > > > > > aditya 2750 May 13 22:03
> > > > > > > > > grid_stat_270000L_20110423_030000V_ctc.txt-rw-rw-r--
1
> aditya
> > > > > aditya
> > > > > > > > > 268 May 13 22:03
> > > > > > > grid_stat_270000L_20110423_030000V_vl1l2.txt-rw-rw-r-- 1
> > > > > > > > > aditya aditya 882 May 13 22:03
> > > > > > > > > grid_stat_270000L_20110423_030000V_sl1l2.txt-rw-rw-
r-- 1
> aditya
> > > > > aditya
> > > > > > > > > 324 May 13 22:03
> > > > > grid_stat_270000L_20110423_030000V_pstd.txt-rw-rw-r--
> > > > > > > 1
> > > > > > > > > aditya aditya 229 May 13 22:03
> > > > > > > > > grid_stat_270000L_20110423_030000V_prc.txt-rw-rw-r--
1
> aditya
> > > > > aditya
> > > > > > > > > 229 May 13 22:03
> > > > > grid_stat_270000L_20110423_030000V_pjc.txt-rw-rw-r-- 1
> > > > > > > > > aditya aditya 229 May 13 22:03
> > > > > > > > > grid_stat_270000L_20110423_030000V_pct.txt-rw-rw-r--
1
> aditya
> > > > > aditya
> > > > > > > > > 20250 May 13 22:03
> > > > > > > grid_stat_270000L_20110423_030000V_nbrctc.txt-rw-rw-r--
> > > > > > > > > 1 aditya aditya 998 May 13 22:03
> > > grid_stat_270000L_2011042!
> > > > > > > > > 3_030000V_mcts.txt-rw-rw-r-- 1 aditya aditya
2310 May
> 13
> > > 22:03
> > > > > > > > > grid_stat_270000L_20110423_030000V_mctc.txt-rw-rw-
r-- 1
> aditya
> > > > > aditya
> > > > > > > > > 11396 May 13 22:03
> > > > > > > grid_stat_270000L_20110423_030000V_cts.txt-rw-rw-r-- 1
> > > > > > > > > aditya aditya 3046 May 13 22:03
> > > > > > > > > grid_stat_270000L_20110423_030000V_cnt.txt-rw-rw-r--
1
> aditya
> > > > > aditya
> > > > > > > > > 84240 May 13 22:03
> > > > > > > grid_stat_270000L_20110423_030000V_nbrcts.txt-rw-rw-r--
> > > > > > > > > 1 aditya aditya 21644 May 13 22:03
> > > > > > > > > grid_stat_270000L_20110423_030000V_pairs.nc-rw-rw-
r-- 1
> aditya
> > > > > aditya
> > > > > > > > > 31428 May 13 22:03
> > > > > > > > >
> grid_stat_270000L_20110423_030000V_nbrcnt.txtaditya at agniilap
> > > > > > > > > :~/geeta/installs/met/RUN_MET-22apr2011/out-
22apr2011$
> > > > > > > > > pls point me to right direction. geeta
> > > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > > >
> > > >
> > >
> > >
> >
>
>
------------------------------------------------
Subject: another problem with gridstat
From: Geeta Geeta
Time: Tue May 19 04:23:48 2015
Thanks john. Can U give an idea so as to how to do it???. I can try.
Geeta
> Subject: Re: [rt.rap.ucar.edu #71923] another problem with gridstat
> From: met_help at ucar.edu
> To: geeta124 at hotmail.com
> Date: Mon, 18 May 2015 09:44:10 -0600
>
> Geeta,
>
> MET does not currently support percentile thresholding. However, we
have
> it defined as a development task and plan to add it in the next
year. For
> the time being, you'd have to compute the percentile value outside
of MET
> and use that to define your threshold.
>
> Thanks,
> John
>
> On Sat, May 16, 2015 at 12:22 PM, Geeta Geeta via RT
<met_help at ucar.edu>
> wrote:
>
> >
> > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
> >
> > hi John.
> > sorry for another query.
> > At present i am taking the thresolds (>=0.1, >=2.5 ) etc for
converting
> > FCST and OBS fileds to binary.
> > My question is if I have to use 90 th percentile, 75th percentile
etc as
> > the the thresholds, then how will I go about doing that????
> >
> > what I should do. ???
> > Geeta
> >
> > > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > > From: met_help at ucar.edu
> > > To: geeta124 at hotmail.com
> > > Date: Fri, 15 May 2015 13:27:17 -0600
> > >
> > > Geeta,
> > >
> > > MET version 5.0 includes some additions to the NBRCNT line
type... it
> > > contains values for FSS, UFSS, and AFSS which are Fractions
Skill Score,
> > > Uniform FSS, and Asymptotic FSS.
> > >
> > > UFSS is computed as 0.5 + base_rate / 2.0. In your email, you
stated
> > that
> > > FSS is only considered useful if it's > 0.5. I believe you're
actually
> > > supposed to compare it to UFSS, which is always a little bit
greater than
> > > 0.5. AFSS is computed as (2.0 * frate * orate) / (frate*frate +
> > > orate*orate), where frate and orate are the fcst and obs rates
> > > respectively. I believe that AFSS gives you the maximum FSS
value
> > possible
> > > given the number for forecast and observed events.
> > >
> > > I believe I already answered your question about aggregating.
You can
> > use
> > > the stat_analysis tool to do this. I'd suggest using the
-fcst_init_beg
> > > and -fcst_init_end or -fcst_valid_beg and -fcst_valid_end
options for
> > > stat_analysis to define the beginning/ending dates for the
Summer and
> > > Monsoon seasons.
> > >
> > > As for interpreting FSS, yes, the higher the better. FSS is
always
> > between
> > > 0 and 1, and the closer to 1, the better the performance.
> > >
> > > Thanks,
> > > John
> > >
> > >
> > >
> > > On Fri, May 15, 2015 at 12:23 PM, Geeta Geeta via RT
<met_help at ucar.edu>
> > > wrote:
> > >
> > > >
> > > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
>
> > > >
> > > > thank s John.
> > > > also pls comment about Useful FCST aswell. and a few other
things I had
> > > > asked in my mail.
> > > >
> > > >
> > > > > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > > > > From: met_help at ucar.edu
> > > > > To: geeta124 at hotmail.com
> > > > > Date: Fri, 15 May 2015 12:11:40 -0600
> > > > >
> > > > > Geeta,
> > > > >
> > > > > Sure, you can compute the fractions skill score for
comparing
> > forecast
> > > > and
> > > > > observed thunderstorms. When computing FSS, you first
threshold the
> > > > fields
> > > > > to define events and non-events. Then you look at
successively
> > larger
> > > > and
> > > > > larger areas around each grid point to see how the forecast
event
> > > > frequency
> > > > > compares to the observed event frequency. Seems like
applying that
> > to
> > > > > thunderstorms would be reasonable.
> > > > >
> > > > > Also, applying it to rainfall (and monsoons) would be fine
as well.
> > Keep
> > > > > in mind that Grid-Stat is the tool that computes FSS. And
you run
> > > > > Grid-Stat once for each evaluation time. If you're
evaluating once
> > per
> > > > > day, you'd run Grid-Stat 122 times for the 122 days of
monsoon. And
> > > > you'd
> > > > > end up with 122 FSS values. You could look at those as a
time
> > series, or
> > > > > you could use the Stat-Analysis tool to aggregate them
together into
> > a
> > > > > single FSS value, like this:
> > > > >
> > > > > stat_analysis -job aggregate -line_type NBRCNT -lookin
> > out/grid_stat
> > > > >
> > > > > It all sounds like a reasonable approach to me. Just be
sure to pick
> > > > > thresholds (for the thunderstorms and monsoons) that capture
the
> > "events"
> > > > > that you're interested in studying.
> > > > >
> > > > > Thanks,
> > > > > John
> > > > >
> > > > > On Fri, May 15, 2015 at 10:52 AM, Geeta Geeta via RT <
> > met_help at ucar.edu>
> > > > > wrote:
> > > > >
> > > > > >
> > > > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
> > > > > >
> > > > > > Hi John.
> > > > > > Just want to confirm a few points.
> > > > > > Suppose I calculate FSS for two seasons. say Summer (92
days) and
> > > > monsoon
> > > > > > (122 days) in India.
> > > > > > I plot FSS vs spatial Scale for both seasons.
> > > > > >
> > > > > > 1. How can aggregate the FSS for entire season . I mean I
will
> > have one
> > > > > > UNIQUE plot of FSS vs scale for a day ???. How will I
aggregate???
> > > > > > 2. Calculated FSS has to be greater than 0.5 at a
particular
> > spatial
> > > > scale
> > > > > > to considered useful FCST by the model. Is this statement
> > correct????
> > > > > > 3. the season having larger FSS means that MODEL has more
skill as
> > > > > > compared to the other season???
> > > > > >
> > > > > > Can U pls clarify.
> > > > > > Geeta
> > > > > >
> > > > > > From: geeta124 at hotmail.com
> > > > > > To: met_help at ucar.edu
> > > > > > Subject: RE: [rt.rap.ucar.edu #71923] another problem with
> > gridstat
> > > > > > Date: Fri, 15 May 2015 22:09:07 +0530
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > Hi John.
> > > > > > I will download the image of 3B42 TRMM and send u. I am
unable to
> > open
> > > > the
> > > > > > page now.
> > > > > >
> > > > > > I have one more query reg the FSS. I am trying to apply
FSS for
> > some
> > > > > > thunderstorm simulations.
> > > > > > so essentially, it will tell me if the frequency of FCST
and OBS
> > are
> > > > > > close???
> > > > > >
> > > > > > Can I apply this technique for evaluation of the MONSOON
rainfall
> > for a
> > > > > > season (122 days in India) for entire southern India???.
so for
> > this I
> > > > > > will be taking a very large domain. I expect that in this
case FSS
> > > > will be
> > > > > > better as compared to that of the thunderstorm cases???
> > > > > >
> > > > > > Will this be a viable idea???.
> > > > > > I need your feedback reg this.
> > > > > >
> > > > > > Geeta
> > > > > >
> > > > > >
> > > > > > > Subject: Re: [rt.rap.ucar.edu #71923] another problem
with
> > gridstat
> > > > > > > From: met_help at ucar.edu
> > > > > > > To: geeta124 at hotmail.com
> > > > > > > Date: Wed, 13 May 2015 15:48:20 -0600
> > > > > > >
> > > > > > > Geeta,
> > > > > > >
> > > > > > > OK, the NetCDF matched pairs file you sent looks
reasonable to
> > me.
> > > > You
> > > > > > are
> > > > > > > comparing it to some plot from the TRMM website and say
that they
> > > > differ.
> > > > > > > Can you please send me the plot to which you're
comparing this
> > data
> > > > as
> > > > > > well
> > > > > > > as the link you used to grab it?
> > > > > > >
> > > > > > > I can't really comment on this until I understand the
discrepancy
> > > > you're
> > > > > > > seeing.
> > > > > > >
> > > > > > > Thanks,
> > > > > > > John
> > > > > > >
> > > > > > >
> > > > > > > On Wed, May 13, 2015 at 3:23 PM, Geeta Geeta via RT <
> > > > met_help at ucar.edu>
> > > > > > > wrote:
> > > > > > >
> > > > > > > >
> > > > > > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
> > >
> > > > > > > >
> > > > > > > > hi John. I just now did whole process of making
Observed data
> > > > again.
> > > > > > Now
> > > > > > > > sending you new NC file.
> > > > > > > > But the aggregated data of 8 observations (06UTC of 22
APR2011
> > -
> > > > 03UTC
> > > > > > of
> > > > > > > > 23 April 2011) looks very different from the plotted
data on
> > TRMM
> > > > > > website.
> > > > > > > >
> > > > > > > > > Subject: Re: [rt.rap.ucar.edu #71923] another
problem with
> > > > gridstat
> > > > > > > > > From: met_help at ucar.edu
> > > > > > > > > To: geeta124 at hotmail.com
> > > > > > > > > Date: Wed, 13 May 2015 12:04:54 -0600
> > > > > > > > >
> > > > > > > > > Geeta,
> > > > > > > > >
> > > > > > > > > I agree. Those data values do look bad. I strongly
suspect
> > that
> > > > > > this
> > > > > > > > > isn't a problem caused by grid_stat. Instead I'm
guessing
> > that
> > > > > > > > observation
> > > > > > > > > data you're passing to grid_stat is bad. I ran the
following
> > > > > > command to
> > > > > > > > > visualize the grid_stat pairs output and the
resulting image
> > is
> > > > > > attached.
> > > > > > > > > The values in this data file range from 0 to 4.5e30!
> > > > > > > > > met-5.0/bin/plot_data_plane
> > > > > > > > grid_stat_270000L_20110423_030000V_pairs.nc
> > > > > > > > > obs.ps 'name="OBS_APCP_A24_FULL"; level="(*,*)";'
> > > > > > > > >
> > > > > > > > > Please take a look at the observation data you're
passing to
> > > > > > grid_stat to
> > > > > > > > > see if it's corrupted in some way.
> > > > > > > > >
> > > > > > > > > Thanks,
> > > > > > > > > John
> > > > > > > > >
> > > > > > > > > On Wed, May 13, 2015 at 11:12 AM, Geeta Geeta via RT
<
> > > > > > met_help at ucar.edu>
> > > > > > > > > wrote:
> > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > Wed May 13 11:12:32 2015: Request 71923 was acted
upon.
> > > > > > > > > > Transaction: Ticket created by
geeta124 at hotmail.com
> > > > > > > > > > Queue: met_help
> > > > > > > > > > Subject: another problem with gridstat
> > > > > > > > > > Owner: Nobody
> > > > > > > > > > Requestors: geeta124 at hotmail.com
> > > > > > > > > > Status: new
> > > > > > > > > > Ticket <URL:
> > > > > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
> > > > > > > > >
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > Hi John. I have another stange problem. I had run
grid
> > stat for
> > > > > > another
> > > > > > > > > > date 22 April2011. I got all outputs nicely. but
when I
> > try to
> > > > > > > > visualise
> > > > > > > > > > the OBS output using the ncview, it looks very
strtange. I
> > am
> > > > > > sending
> > > > > > > > the
> > > > > > > > > > NC file for your reference.
> > > > > > > > > > also when i saw the data in the OBS file, it shows
me some
> > very
> > > > > > strange
> > > > > > > > > > values of precipitation. Absolutely unrealistic.
> > > > > > > > > >
> > > > > > > > > > -rw-rw-r-- 1 aditya aditya 286376 May 13 22:03
> > > > > > > > > > grid_stat_270000L_20110423_030000V.stat-rw-rw-r--
1 aditya
> > > > aditya
> > > > > > > > 2750
> > > > > > > > > > May 13 22:03
> > > > grid_stat_270000L_20110423_030000V_fho.txt-rw-rw-r-- 1
> > > > > > > > aditya
> > > > > > > > > > aditya 2750 May 13 22:03
> > > > > > > > > > grid_stat_270000L_20110423_030000V_ctc.txt-rw-rw-
r-- 1
> > aditya
> > > > > > aditya
> > > > > > > > > > 268 May 13 22:03
> > > > > > > > grid_stat_270000L_20110423_030000V_vl1l2.txt-rw-rw-r--
1
> > > > > > > > > > aditya aditya 882 May 13 22:03
> > > > > > > > > > grid_stat_270000L_20110423_030000V_sl1l2.txt-rw-
rw-r-- 1
> > aditya
> > > > > > aditya
> > > > > > > > > > 324 May 13 22:03
> > > > > > grid_stat_270000L_20110423_030000V_pstd.txt-rw-rw-r--
> > > > > > > > 1
> > > > > > > > > > aditya aditya 229 May 13 22:03
> > > > > > > > > > grid_stat_270000L_20110423_030000V_prc.txt-rw-rw-
r-- 1
> > aditya
> > > > > > aditya
> > > > > > > > > > 229 May 13 22:03
> > > > > > grid_stat_270000L_20110423_030000V_pjc.txt-rw-rw-r-- 1
> > > > > > > > > > aditya aditya 229 May 13 22:03
> > > > > > > > > > grid_stat_270000L_20110423_030000V_pct.txt-rw-rw-
r-- 1
> > aditya
> > > > > > aditya
> > > > > > > > > > 20250 May 13 22:03
> > > > > > > > grid_stat_270000L_20110423_030000V_nbrctc.txt-rw-rw-
r--
> > > > > > > > > > 1 aditya aditya 998 May 13 22:03
> > > > grid_stat_270000L_2011042!
> > > > > > > > > > 3_030000V_mcts.txt-rw-rw-r-- 1 aditya aditya
2310 May
> > 13
> > > > 22:03
> > > > > > > > > > grid_stat_270000L_20110423_030000V_mctc.txt-rw-rw-
r-- 1
> > aditya
> > > > > > aditya
> > > > > > > > > > 11396 May 13 22:03
> > > > > > > > grid_stat_270000L_20110423_030000V_cts.txt-rw-rw-r-- 1
> > > > > > > > > > aditya aditya 3046 May 13 22:03
> > > > > > > > > > grid_stat_270000L_20110423_030000V_cnt.txt-rw-rw-
r-- 1
> > aditya
> > > > > > aditya
> > > > > > > > > > 84240 May 13 22:03
> > > > > > > > grid_stat_270000L_20110423_030000V_nbrcts.txt-rw-rw-
r--
> > > > > > > > > > 1 aditya aditya 21644 May 13 22:03
> > > > > > > > > > grid_stat_270000L_20110423_030000V_pairs.nc-rw-rw-
r-- 1
> > aditya
> > > > > > aditya
> > > > > > > > > > 31428 May 13 22:03
> > > > > > > > > >
> > grid_stat_270000L_20110423_030000V_nbrcnt.txtaditya at agniilap
> > > > > > > > > > :~/geeta/installs/met/RUN_MET-22apr2011/out-
22apr2011$
> > > > > > > > > > pls point me to right direction. geeta
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > >
> > > > > >
> > > > >
> > > >
> > > >
> > >
> >
> >
>
------------------------------------------------
Subject: another problem with gridstat
From: John Halley Gotway
Time: Tue May 19 10:03:19 2015
Geeta,
I can't remember, are both of your forecast and observation fields
that
you're passing to Grid-Stat in NetCDF format? Assuming that's the
case,
I'd suggest the following.
I've attached a simple Rscript that I called "get_percentile.R" which
can
be used to compute the percentile of interest. Here's how you'd call
it:
Rscript get_percentile.R sample.nc APCP_12 90
It takes 3 arguments, the NetCDF file to be used, the variable name to
be
read, and the percentile to be written (0 - 100). It writes out the
requested percentile value. Currently, it's computing the requested
percentile over all valid data values. For precipitation, you might
want
to ignore the 0's. But I'm really not sure on that one. Currently,
it's
including the 0's.
You could do the following right before calling Grid-Stat...
(1) Choose a percentile you want to use. Let's say the 90th as an
example.
(2) Run get_percentile.R on your forecast NetCDF file and store the
result:
setenv fcst_perc_value `Rscript get_percentile.R fcst_file.nc
APCP_24 90`
(3) Run get_percentile.R on your observed NetCDF file and store the
result:
setenv obs_perc_value `Rscript get_percentile.R obs_file.nc APCP_24
90`
(4) Edit your Grid-Stat configuration file to use those variables we
just
set:
In the fcst section...
cat_thresh = [ >=${fcst_perc_value} ];
And in the obs section...
cat_thresh = [ >=${obs_perc_value} ];
(5) And finally, call Grid-Stat using that configuration.
Hope that helps get you going.
Thanks,
John
On Tue, May 19, 2015 at 4:23 AM, Geeta Geeta via RT
<met_help at ucar.edu>
wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
>
> Thanks john. Can U give an idea so as to how to do it???. I can
try.
> Geeta
>
> > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > From: met_help at ucar.edu
> > To: geeta124 at hotmail.com
> > Date: Mon, 18 May 2015 09:44:10 -0600
> >
> > Geeta,
> >
> > MET does not currently support percentile thresholding. However,
we have
> > it defined as a development task and plan to add it in the next
year.
> For
> > the time being, you'd have to compute the percentile value outside
of MET
> > and use that to define your threshold.
> >
> > Thanks,
> > John
> >
> > On Sat, May 16, 2015 at 12:22 PM, Geeta Geeta via RT
<met_help at ucar.edu>
> > wrote:
> >
> > >
> > > <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
> > >
> > > hi John.
> > > sorry for another query.
> > > At present i am taking the thresolds (>=0.1, >=2.5 ) etc for
converting
> > > FCST and OBS fileds to binary.
> > > My question is if I have to use 90 th percentile, 75th
percentile etc
> as
> > > the the thresholds, then how will I go about doing that????
> > >
> > > what I should do. ???
> > > Geeta
> > >
> > > > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
gridstat
> > > > From: met_help at ucar.edu
> > > > To: geeta124 at hotmail.com
> > > > Date: Fri, 15 May 2015 13:27:17 -0600
> > > >
> > > > Geeta,
> > > >
> > > > MET version 5.0 includes some additions to the NBRCNT line
type... it
> > > > contains values for FSS, UFSS, and AFSS which are Fractions
Skill
> Score,
> > > > Uniform FSS, and Asymptotic FSS.
> > > >
> > > > UFSS is computed as 0.5 + base_rate / 2.0. In your email, you
stated
> > > that
> > > > FSS is only considered useful if it's > 0.5. I believe you're
> actually
> > > > supposed to compare it to UFSS, which is always a little bit
greater
> than
> > > > 0.5. AFSS is computed as (2.0 * frate * orate) / (frate*frate
+
> > > > orate*orate), where frate and orate are the fcst and obs rates
> > > > respectively. I believe that AFSS gives you the maximum FSS
value
> > > possible
> > > > given the number for forecast and observed events.
> > > >
> > > > I believe I already answered your question about aggregating.
You
> can
> > > use
> > > > the stat_analysis tool to do this. I'd suggest using the
> -fcst_init_beg
> > > > and -fcst_init_end or -fcst_valid_beg and -fcst_valid_end
options for
> > > > stat_analysis to define the beginning/ending dates for the
Summer and
> > > > Monsoon seasons.
> > > >
> > > > As for interpreting FSS, yes, the higher the better. FSS is
always
> > > between
> > > > 0 and 1, and the closer to 1, the better the performance.
> > > >
> > > > Thanks,
> > > > John
> > > >
> > > >
> > > >
> > > > On Fri, May 15, 2015 at 12:23 PM, Geeta Geeta via RT <
> met_help at ucar.edu>
> > > > wrote:
> > > >
> > > > >
> > > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923 >
> > > > >
> > > > > thank s John.
> > > > > also pls comment about Useful FCST aswell. and a few other
things
> I had
> > > > > asked in my mail.
> > > > >
> > > > >
> > > > > > Subject: Re: [rt.rap.ucar.edu #71923] another problem with
> gridstat
> > > > > > From: met_help at ucar.edu
> > > > > > To: geeta124 at hotmail.com
> > > > > > Date: Fri, 15 May 2015 12:11:40 -0600
> > > > > >
> > > > > > Geeta,
> > > > > >
> > > > > > Sure, you can compute the fractions skill score for
comparing
> > > forecast
> > > > > and
> > > > > > observed thunderstorms. When computing FSS, you first
threshold
> the
> > > > > fields
> > > > > > to define events and non-events. Then you look at
successively
> > > larger
> > > > > and
> > > > > > larger areas around each grid point to see how the
forecast event
> > > > > frequency
> > > > > > compares to the observed event frequency. Seems like
applying
> that
> > > to
> > > > > > thunderstorms would be reasonable.
> > > > > >
> > > > > > Also, applying it to rainfall (and monsoons) would be fine
as
> well.
> > > Keep
> > > > > > in mind that Grid-Stat is the tool that computes FSS. And
you
> run
> > > > > > Grid-Stat once for each evaluation time. If you're
evaluating
> once
> > > per
> > > > > > day, you'd run Grid-Stat 122 times for the 122 days of
monsoon.
> And
> > > > > you'd
> > > > > > end up with 122 FSS values. You could look at those as a
time
> > > series, or
> > > > > > you could use the Stat-Analysis tool to aggregate them
together
> into
> > > a
> > > > > > single FSS value, like this:
> > > > > >
> > > > > > stat_analysis -job aggregate -line_type NBRCNT -lookin
> > > out/grid_stat
> > > > > >
> > > > > > It all sounds like a reasonable approach to me. Just be
sure to
> pick
> > > > > > thresholds (for the thunderstorms and monsoons) that
capture the
> > > "events"
> > > > > > that you're interested in studying.
> > > > > >
> > > > > > Thanks,
> > > > > > John
> > > > > >
> > > > > > On Fri, May 15, 2015 at 10:52 AM, Geeta Geeta via RT <
> > > met_help at ucar.edu>
> > > > > > wrote:
> > > > > >
> > > > > > >
> > > > > > > <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
> >
> > > > > > >
> > > > > > > Hi John.
> > > > > > > Just want to confirm a few points.
> > > > > > > Suppose I calculate FSS for two seasons. say Summer (92
days)
> and
> > > > > monsoon
> > > > > > > (122 days) in India.
> > > > > > > I plot FSS vs spatial Scale for both seasons.
> > > > > > >
> > > > > > > 1. How can aggregate the FSS for entire season . I mean
I will
> > > have one
> > > > > > > UNIQUE plot of FSS vs scale for a day ???. How will I
> aggregate???
> > > > > > > 2. Calculated FSS has to be greater than 0.5 at a
particular
> > > spatial
> > > > > scale
> > > > > > > to considered useful FCST by the model. Is this
statement
> > > correct????
> > > > > > > 3. the season having larger FSS means that MODEL has
more
> skill as
> > > > > > > compared to the other season???
> > > > > > >
> > > > > > > Can U pls clarify.
> > > > > > > Geeta
> > > > > > >
> > > > > > > From: geeta124 at hotmail.com
> > > > > > > To: met_help at ucar.edu
> > > > > > > Subject: RE: [rt.rap.ucar.edu #71923] another problem
with
> > > gridstat
> > > > > > > Date: Fri, 15 May 2015 22:09:07 +0530
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > Hi John.
> > > > > > > I will download the image of 3B42 TRMM and send u. I am
unable
> to
> > > open
> > > > > the
> > > > > > > page now.
> > > > > > >
> > > > > > > I have one more query reg the FSS. I am trying to apply
FSS for
> > > some
> > > > > > > thunderstorm simulations.
> > > > > > > so essentially, it will tell me if the frequency of FCST
and
> OBS
> > > are
> > > > > > > close???
> > > > > > >
> > > > > > > Can I apply this technique for evaluation of the MONSOON
> rainfall
> > > for a
> > > > > > > season (122 days in India) for entire southern
India???. so
> for
> > > this I
> > > > > > > will be taking a very large domain. I expect that in
this case
> FSS
> > > > > will be
> > > > > > > better as compared to that of the thunderstorm cases???
> > > > > > >
> > > > > > > Will this be a viable idea???.
> > > > > > > I need your feedback reg this.
> > > > > > >
> > > > > > > Geeta
> > > > > > >
> > > > > > >
> > > > > > > > Subject: Re: [rt.rap.ucar.edu #71923] another problem
with
> > > gridstat
> > > > > > > > From: met_help at ucar.edu
> > > > > > > > To: geeta124 at hotmail.com
> > > > > > > > Date: Wed, 13 May 2015 15:48:20 -0600
> > > > > > > >
> > > > > > > > Geeta,
> > > > > > > >
> > > > > > > > OK, the NetCDF matched pairs file you sent looks
reasonable
> to
> > > me.
> > > > > You
> > > > > > > are
> > > > > > > > comparing it to some plot from the TRMM website and
say that
> they
> > > > > differ.
> > > > > > > > Can you please send me the plot to which you're
comparing
> this
> > > data
> > > > > as
> > > > > > > well
> > > > > > > > as the link you used to grab it?
> > > > > > > >
> > > > > > > > I can't really comment on this until I understand the
> discrepancy
> > > > > you're
> > > > > > > > seeing.
> > > > > > > >
> > > > > > > > Thanks,
> > > > > > > > John
> > > > > > > >
> > > > > > > >
> > > > > > > > On Wed, May 13, 2015 at 3:23 PM, Geeta Geeta via RT <
> > > > > met_help at ucar.edu>
> > > > > > > > wrote:
> > > > > > > >
> > > > > > > > >
> > > > > > > > > <URL:
> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
> > > >
> > > > > > > > >
> > > > > > > > > hi John. I just now did whole process of making
Observed
> data
> > > > > again.
> > > > > > > Now
> > > > > > > > > sending you new NC file.
> > > > > > > > > But the aggregated data of 8 observations (06UTC of
22
> APR2011
> > > -
> > > > > 03UTC
> > > > > > > of
> > > > > > > > > 23 April 2011) looks very different from the plotted
data
> on
> > > TRMM
> > > > > > > website.
> > > > > > > > >
> > > > > > > > > > Subject: Re: [rt.rap.ucar.edu #71923] another
problem
> with
> > > > > gridstat
> > > > > > > > > > From: met_help at ucar.edu
> > > > > > > > > > To: geeta124 at hotmail.com
> > > > > > > > > > Date: Wed, 13 May 2015 12:04:54 -0600
> > > > > > > > > >
> > > > > > > > > > Geeta,
> > > > > > > > > >
> > > > > > > > > > I agree. Those data values do look bad. I
strongly
> suspect
> > > that
> > > > > > > this
> > > > > > > > > > isn't a problem caused by grid_stat. Instead I'm
> guessing
> > > that
> > > > > > > > > observation
> > > > > > > > > > data you're passing to grid_stat is bad. I ran
the
> following
> > > > > > > command to
> > > > > > > > > > visualize the grid_stat pairs output and the
resulting
> image
> > > is
> > > > > > > attached.
> > > > > > > > > > The values in this data file range from 0 to
4.5e30!
> > > > > > > > > > met-5.0/bin/plot_data_plane
> > > > > > > > > grid_stat_270000L_20110423_030000V_pairs.nc
> > > > > > > > > > obs.ps 'name="OBS_APCP_A24_FULL"; level="(*,*)";'
> > > > > > > > > >
> > > > > > > > > > Please take a look at the observation data you're
> passing to
> > > > > > > grid_stat to
> > > > > > > > > > see if it's corrupted in some way.
> > > > > > > > > >
> > > > > > > > > > Thanks,
> > > > > > > > > > John
> > > > > > > > > >
> > > > > > > > > > On Wed, May 13, 2015 at 11:12 AM, Geeta Geeta via
RT <
> > > > > > > met_help at ucar.edu>
> > > > > > > > > > wrote:
> > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > Wed May 13 11:12:32 2015: Request 71923 was
acted upon.
> > > > > > > > > > > Transaction: Ticket created by
geeta124 at hotmail.com
> > > > > > > > > > > Queue: met_help
> > > > > > > > > > > Subject: another problem with gridstat
> > > > > > > > > > > Owner: Nobody
> > > > > > > > > > > Requestors: geeta124 at hotmail.com
> > > > > > > > > > > Status: new
> > > > > > > > > > > Ticket <URL:
> > > > > > > https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=71923
> > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > > Hi John. I have another stange problem. I had
run grid
> > > stat for
> > > > > > > another
> > > > > > > > > > > date 22 April2011. I got all outputs nicely. but
when I
> > > try to
> > > > > > > > > visualise
> > > > > > > > > > > the OBS output using the ncview, it looks very
> strtange. I
> > > am
> > > > > > > sending
> > > > > > > > > the
> > > > > > > > > > > NC file for your reference.
> > > > > > > > > > > also when i saw the data in the OBS file, it
shows me
> some
> > > very
> > > > > > > strange
> > > > > > > > > > > values of precipitation. Absolutely unrealistic.
> > > > > > > > > > >
> > > > > > > > > > > -rw-rw-r-- 1 aditya aditya 286376 May 13 22:03
> > > > > > > > > > > grid_stat_270000L_20110423_030000V.stat-rw-rw-
r-- 1
> aditya
> > > > > aditya
> > > > > > > > > 2750
> > > > > > > > > > > May 13 22:03
> > > > > grid_stat_270000L_20110423_030000V_fho.txt-rw-rw-r-- 1
> > > > > > > > > aditya
> > > > > > > > > > > aditya 2750 May 13 22:03
> > > > > > > > > > > grid_stat_270000L_20110423_030000V_ctc.txt-rw-
rw-r-- 1
> > > aditya
> > > > > > > aditya
> > > > > > > > > > > 268 May 13 22:03
> > > > > > > > > grid_stat_270000L_20110423_030000V_vl1l2.txt-rw-rw-
r-- 1
> > > > > > > > > > > aditya aditya 882 May 13 22:03
> > > > > > > > > > > grid_stat_270000L_20110423_030000V_sl1l2.txt-rw-
rw-r--
> 1
> > > aditya
> > > > > > > aditya
> > > > > > > > > > > 324 May 13 22:03
> > > > > > > grid_stat_270000L_20110423_030000V_pstd.txt-rw-rw-r--
> > > > > > > > > 1
> > > > > > > > > > > aditya aditya 229 May 13 22:03
> > > > > > > > > > > grid_stat_270000L_20110423_030000V_prc.txt-rw-
rw-r-- 1
> > > aditya
> > > > > > > aditya
> > > > > > > > > > > 229 May 13 22:03
> > > > > > > grid_stat_270000L_20110423_030000V_pjc.txt-rw-rw-r-- 1
> > > > > > > > > > > aditya aditya 229 May 13 22:03
> > > > > > > > > > > grid_stat_270000L_20110423_030000V_pct.txt-rw-
rw-r-- 1
> > > aditya
> > > > > > > aditya
> > > > > > > > > > > 20250 May 13 22:03
> > > > > > > > > grid_stat_270000L_20110423_030000V_nbrctc.txt-rw-rw-
r--
> > > > > > > > > > > 1 aditya aditya 998 May 13 22:03
> > > > > grid_stat_270000L_2011042!
> > > > > > > > > > > 3_030000V_mcts.txt-rw-rw-r-- 1 aditya aditya
2310
> May
> > > 13
> > > > > 22:03
> > > > > > > > > > > grid_stat_270000L_20110423_030000V_mctc.txt-rw-
rw-r-- 1
> > > aditya
> > > > > > > aditya
> > > > > > > > > > > 11396 May 13 22:03
> > > > > > > > > grid_stat_270000L_20110423_030000V_cts.txt-rw-rw-r--
1
> > > > > > > > > > > aditya aditya 3046 May 13 22:03
> > > > > > > > > > > grid_stat_270000L_20110423_030000V_cnt.txt-rw-
rw-r-- 1
> > > aditya
> > > > > > > aditya
> > > > > > > > > > > 84240 May 13 22:03
> > > > > > > > > grid_stat_270000L_20110423_030000V_nbrcts.txt-rw-rw-
r--
> > > > > > > > > > > 1 aditya aditya 21644 May 13 22:03
> > > > > > > > > > > grid_stat_270000L_20110423_030000V_pairs.nc-rw-
rw-r-- 1
> > > aditya
> > > > > > > aditya
> > > > > > > > > > > 31428 May 13 22:03
> > > > > > > > > > >
> > > grid_stat_270000L_20110423_030000V_nbrcnt.txtaditya at agniilap
> > > > > > > > > > > :~/geeta/installs/met/RUN_MET-22apr2011/out-
22apr2011$
> > > > > > > > > > > pls point me to right direction. geeta
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > > >
> > > > > > > > > >
> > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > > >
> > > >
> > >
> > >
> >
>
>
------------------------------------------------
More information about the Met_help
mailing list