[Met_help] [rt.rap.ucar.edu #86852] History for METviewer: comparing FSS from two different systems?
John Halley Gotway via RT
met_help at ucar.edu
Mon Sep 24 16:25:10 MDT 2018
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Hello. I'm wondering if it's possible to compare FSS generated under
MET with that from the VSDB in METviewer - by plotting them in the same
frame and possibly plot the difference. Attached are xmls and pngs for
the FSS vs. neighborhood sizes plots generated from 1) MET grid_stat
(conusnest2, at 20180905_200949) and 2) VSDB (conusnestfvs, at
20180828_171213).
The set up for the two individual plots are different: for the MET plot,
the dependent variable is APCP_24, fixed value has OBS_THRESH > 10,
independent variables are INTERP_PNTS (neighborhood sizes in grid
points). For the VSDB/FVS plot, the dependent variable is APCP/24 >
010.0, and independent variable is OBS_THRESH (< 005, < 014, < 024 ...
etc, these correspond to the MET plot's INTERP_PNTS - e.g. '014'
correspond to the 3x3 neighborhood size). Is there a way to display
both on the same plot, presumably by using Y1/Y2 series variables? I
have not been able to do this (failed attempt:
plot_20180905_204709.xml). I suppose there is no way to choose
different independent variables (because it's Y1/Y2 series, not
Y1/X1/Y2/X2 series?). I've been experimenting with varies settings and
the printouts of the two curves do seem to match (looking at the
overlay, with a light source in the back), but I'm wondering if there is
currently a way to do this in METviewer.
Thanks,
Ying
--
Ying Lin
NCEP/EMC/Verification, Post-processing and Product Generation Branch
NCWCP Cubicle No. 2015
Ying.Lin at noaa.gov
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: METviewer: comparing FSS from two different systems?
From: John Halley Gotway
Time: Mon Sep 10 16:09:38 2018
Ying,
Thanks for sending such a detailed example. I went through the same
steps
you did and unfortunately came to the same result you did.
Since the MET plot defines the independent variable as INTERP_PNTS and
the
VSDB plot defines the independent variable as OBS_THRESH, I'm not able
to
get these two lines on the same plot. And they can't go on the Y1 and
Y2... since even there you need the same independent variable.
Likely the only way to get them on the same plot would be to make them
use
the independent variable. That would likely mean changing how
METviewer
load FSS data into it's database. Perhaps it should be made to load
what's
currently in OBS_THRESH into INTERP_PNTS instead. And we should parse
off
the threshold from the forecast variable name and place that in
OBS_THRESH.
For example, when the variable is "APCP/24>010.0", set OBS_THRESH =
">010.0".
Do you agree that the problem seems to lie in the METviewer load
logic?
Thanks,
John
On Wed, Sep 5, 2018 at 2:55 PM Ying Lin via RT <met_help at ucar.edu>
wrote:
>
> Wed Sep 05 14:55:27 2018: Request 86852 was acted upon.
> Transaction: Ticket created by ying.lin at noaa.gov
> Queue: met_help
> Subject: METviewer: comparing FSS from two different systems?
> Owner: Nobody
> Requestors: ying.lin at noaa.gov
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=86852 >
>
>
> Hello. I'm wondering if it's possible to compare FSS generated
under
> MET with that from the VSDB in METviewer - by plotting them in the
same
> frame and possibly plot the difference. Attached are xmls and pngs
for
> the FSS vs. neighborhood sizes plots generated from 1) MET grid_stat
> (conusnest2, at 20180905_200949) and 2) VSDB (conusnestfvs, at
> 20180828_171213).
>
> The set up for the two individual plots are different: for the MET
plot,
> the dependent variable is APCP_24, fixed value has OBS_THRESH > 10,
> independent variables are INTERP_PNTS (neighborhood sizes in grid
> points). For the VSDB/FVS plot, the dependent variable is APCP/24 >
> 010.0, and independent variable is OBS_THRESH (< 005, < 014, < 024
...
> etc, these correspond to the MET plot's INTERP_PNTS - e.g. '014'
> correspond to the 3x3 neighborhood size). Is there a way to display
> both on the same plot, presumably by using Y1/Y2 series variables?
I
> have not been able to do this (failed attempt:
> plot_20180905_204709.xml). I suppose there is no way to choose
> different independent variables (because it's Y1/Y2 series, not
> Y1/X1/Y2/X2 series?). I've been experimenting with varies settings
and
> the printouts of the two curves do seem to match (looking at the
> overlay, with a light source in the back), but I'm wondering if
there is
> currently a way to do this in METviewer.
>
> Thanks,
>
> Ying
>
> --
> Ying Lin
> NCEP/EMC/Verification, Post-processing and Product Generation Branch
> NCWCP Cubicle No. 2015
> Ying.Lin at noaa.gov
>
>
>
>
------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #86852] METviewer: comparing FSS from two different systems?
From: Ying Lin
Time: Mon Sep 10 19:10:02 2018
Thanks John. I don't know enough about METviewer say anything about
load logic ... Yes, "APCP/24>010.0" is actually OBS_THRESH =
">010.0".
Would loading the current (VSDB's ) OBS_THRESH (< 005, < 014, < 024
...)
in as INTERP_PNTS allow an overlay of FSS lines from two systems,
since
as you said below even for Y1/Y2 plots one needs the same independent
variable - you and I know that "< 014" is equivalent to
NBRHD/INTERP_PNTS = 9, but how would METviewer know that?
Ying
On 09/10/2018 06:09 PM, John Halley Gotway via RT wrote:
> Ying,
>
> Thanks for sending such a detailed example. I went through the same
steps
> you did and unfortunately came to the same result you did.
>
> Since the MET plot defines the independent variable as INTERP_PNTS
and the
> VSDB plot defines the independent variable as OBS_THRESH, I'm not
able to
> get these two lines on the same plot. And they can't go on the Y1
and
> Y2... since even there you need the same independent variable.
>
> Likely the only way to get them on the same plot would be to make
them use
> the independent variable. That would likely mean changing how
METviewer
> load FSS data into it's database. Perhaps it should be made to load
what's
> currently in OBS_THRESH into INTERP_PNTS instead. And we should
parse off
> the threshold from the forecast variable name and place that in
OBS_THRESH.
>
> For example, when the variable is "APCP/24>010.0", set OBS_THRESH =
> ">010.0".
>
> Do you agree that the problem seems to lie in the METviewer load
logic?
>
> Thanks,
> John
>
>
> On Wed, Sep 5, 2018 at 2:55 PM Ying Lin via RT <met_help at ucar.edu>
wrote:
>
>> Wed Sep 05 14:55:27 2018: Request 86852 was acted upon.
>> Transaction: Ticket created by ying.lin at noaa.gov
>> Queue: met_help
>> Subject: METviewer: comparing FSS from two different systems?
>> Owner: Nobody
>> Requestors: ying.lin at noaa.gov
>> Status: new
>> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=86852 >
>>
>>
>> Hello. I'm wondering if it's possible to compare FSS generated
under
>> MET with that from the VSDB in METviewer - by plotting them in the
same
>> frame and possibly plot the difference. Attached are xmls and pngs
for
>> the FSS vs. neighborhood sizes plots generated from 1) MET
grid_stat
>> (conusnest2, at 20180905_200949) and 2) VSDB (conusnestfvs, at
>> 20180828_171213).
>>
>> The set up for the two individual plots are different: for the MET
plot,
>> the dependent variable is APCP_24, fixed value has OBS_THRESH > 10,
>> independent variables are INTERP_PNTS (neighborhood sizes in grid
>> points). For the VSDB/FVS plot, the dependent variable is APCP/24
>
>> 010.0, and independent variable is OBS_THRESH (< 005, < 014, < 024
...
>> etc, these correspond to the MET plot's INTERP_PNTS - e.g. '014'
>> correspond to the 3x3 neighborhood size). Is there a way to
display
>> both on the same plot, presumably by using Y1/Y2 series variables?
I
>> have not been able to do this (failed attempt:
>> plot_20180905_204709.xml). I suppose there is no way to choose
>> different independent variables (because it's Y1/Y2 series, not
>> Y1/X1/Y2/X2 series?). I've been experimenting with varies settings
and
>> the printouts of the two curves do seem to match (looking at the
>> overlay, with a light source in the back), but I'm wondering if
there is
>> currently a way to do this in METviewer.
>>
>> Thanks,
>>
>> Ying
>>
>> --
>> Ying Lin
>> NCEP/EMC/Verification, Post-processing and Product Generation
Branch
>> NCWCP Cubicle No. 2015
>> Ying.Lin at noaa.gov
>>
>>
>>
>>
--
Ying Lin
NCEP/EMC/Verification, Post-processing and Product Generation Branch
NCWCP Cubicle No. 2015
Ying.Lin at noaa.gov
------------------------------------------------
More information about the Met_help
mailing list