[Met_help] [rt.rap.ucar.edu #89965] History for grid_stat Variable Name Output Change
John Halley Gotway via RT
met_help at ucar.edu
Tue Apr 30 12:04:40 MDT 2019
----------------------------------------------------------------
Initial Request
----------------------------------------------------------------
Hello there,
I am in the process of upgrading my Excessive Rainfall Outlook (ERO)
verification code from METv6.0 to METv8.0. During this process, I noticed
that the observation variable name after running grid_stat has changed
between these two versions. My methodology is the same, and applies a
circular neighborhood of 9 grid points in the interp definition of the
grid_stat configuration file (old configuration file is config_file_old,
new configuration file is config_file_new).
In more detail if I run:
path_to_metv8.0/grid_stat obs_file fcst_file config_file_old -outdir ./
The name of my observational data within the netcdf file is called
OBS_ST4gFFG_Surface_FULL_MAX_49
However, if I run
path_to_metv6.0/grid_stat obs_file fcst_file config_file_new -outdir ./
The name of my observational data within the netcdf file is called
OBS_ST4gFFG_Surface_FULL_MAX_81
The data inside each file is exactly the same, but I am wondering why the
metadata label is different? For reference, my files are stored here:
https://ftp.wpc.ncep.noaa.gov/erickson/dtc2/. The grid_stat netCDF output
for METv6.0 is grid_stat_out_OLD.nc and for METv8.0 is grid_stat_out_NEW.nc
Thanks again!
Mike
--
Michael J. Erickson
Research Scientist
Cooperative Institute for Research in Environmental Sciences (CIRES)
NOAA/NWS/NCEP/Weather Prediction Center
Phone: 301-683-1546
----------------------------------------------------------------
Complete Ticket History
----------------------------------------------------------------
Subject: grid_stat Variable Name Output Change
From: John Halley Gotway
Time: Mon Apr 29 12:48:09 2019
Hi Mike,
I see that you have a question about differences between met-6.0 and
met-8.0 when smoothing data.
I used the data and config files you sent and ran Grid-Stat for both
versions 6.0 and 8.0. First, I diffed the output in the .stat files
and
see that in both met-6.0 and met-8.0, INTERP_PNTS = 49. However, in
the
NetCDF matched pairs output files, met-6.0 has MAX_81 whereas met-8.0
has
MAX_49, as you describe.
Here's the logic. Earlier versions of MET only supported a square
interpolation region. A square with a width of 9 includes 81 grid
points
(9 * 9). Then in met-6.0, we added support for circular regions as
well.
Since we already had the "width" option defined, we continued to use
it for
circles but interpret it as the diameter of the circle. And we
updated the
logic to report the ACTUAL number of points included in the circular
region. And that's correctly reported in the .stat output as
INTERP_PNTS =
49.
However, we failed to correctly indicate that in the Grid-Stat NetCDF
variable names. So somewhere between met-6.0 and met-8.0 we corrected
the
variable names to indicate the actual number of grid points included
in our
circle with a diameter of 9. And that's the reason for the
discrepancy.
Make sense?
Thanks,
John
On Mon, Apr 29, 2019 at 12:19 PM Michael Erickson - NOAA Affiliate via
RT <
met_help at ucar.edu> wrote:
>
> Mon Apr 29 12:19:25 2019: Request 89965 was acted upon.
> Transaction: Ticket created by michael.j.erickson at noaa.gov
> Queue: met_help
> Subject: grid_stat Variable Name Output Change
> Owner: Nobody
> Requestors: michael.j.erickson at noaa.gov
> Status: new
> Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=89965 >
>
>
> Hello there,
>
> I am in the process of upgrading my Excessive Rainfall Outlook (ERO)
> verification code from METv6.0 to METv8.0. During this process, I
noticed
> that the observation variable name after running grid_stat has
changed
> between these two versions. My methodology is the same, and applies
a
> circular neighborhood of 9 grid points in the interp definition of
the
> grid_stat configuration file (old configuration file is
config_file_old,
> new configuration file is config_file_new).
>
> In more detail if I run:
> path_to_metv8.0/grid_stat obs_file fcst_file config_file_old -outdir
./
>
> The name of my observational data within the netcdf file is called
> OBS_ST4gFFG_Surface_FULL_MAX_49
>
> However, if I run
> path_to_metv6.0/grid_stat obs_file fcst_file config_file_new -outdir
./
>
> The name of my observational data within the netcdf file is called
> OBS_ST4gFFG_Surface_FULL_MAX_81
>
> The data inside each file is exactly the same, but I am wondering
why the
> metadata label is different? For reference, my files are stored
here:
> https://ftp.wpc.ncep.noaa.gov/erickson/dtc2/. The grid_stat netCDF
output
> for METv6.0 is grid_stat_out_OLD.nc and for METv8.0 is
grid_stat_out_NEW.nc
>
> Thanks again!
>
> Mike
> --
> Michael J. Erickson
>
> Research Scientist
> Cooperative Institute for Research in Environmental Sciences (CIRES)
> NOAA/NWS/NCEP/Weather Prediction Center
> Phone: 301-683-1546
>
>
------------------------------------------------
Subject: grid_stat Variable Name Output Change
From: Michael Erickson - NOAA Affiliate
Time: Mon Apr 29 12:55:21 2019
Hi John,
Thank you for your quick response! That makes sense to me.
Later code looks for naming specifically with 81 rather than 49,
although I
can adapt it to be more flexible.
Mike
On Mon, Apr 29, 2019 at 6:48 PM John Halley Gotway via RT
<met_help at ucar.edu>
wrote:
> Hi Mike,
>
> I see that you have a question about differences between met-6.0 and
> met-8.0 when smoothing data.
>
> I used the data and config files you sent and ran Grid-Stat for both
> versions 6.0 and 8.0. First, I diffed the output in the .stat files
and
> see that in both met-6.0 and met-8.0, INTERP_PNTS = 49. However, in
the
> NetCDF matched pairs output files, met-6.0 has MAX_81 whereas met-
8.0 has
> MAX_49, as you describe.
>
> Here's the logic. Earlier versions of MET only supported a square
> interpolation region. A square with a width of 9 includes 81 grid
points
> (9 * 9). Then in met-6.0, we added support for circular regions as
well.
> Since we already had the "width" option defined, we continued to use
it for
> circles but interpret it as the diameter of the circle. And we
updated the
> logic to report the ACTUAL number of points included in the circular
> region. And that's correctly reported in the .stat output as
INTERP_PNTS =
> 49.
>
> However, we failed to correctly indicate that in the Grid-Stat
NetCDF
> variable names. So somewhere between met-6.0 and met-8.0 we
corrected the
> variable names to indicate the actual number of grid points included
in our
> circle with a diameter of 9. And that's the reason for the
discrepancy.
>
> Make sense?
>
> Thanks,
> John
>
>
> On Mon, Apr 29, 2019 at 12:19 PM Michael Erickson - NOAA Affiliate
via RT <
> met_help at ucar.edu> wrote:
>
> >
> > Mon Apr 29 12:19:25 2019: Request 89965 was acted upon.
> > Transaction: Ticket created by michael.j.erickson at noaa.gov
> > Queue: met_help
> > Subject: grid_stat Variable Name Output Change
> > Owner: Nobody
> > Requestors: michael.j.erickson at noaa.gov
> > Status: new
> > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=89965 >
> >
> >
> > Hello there,
> >
> > I am in the process of upgrading my Excessive Rainfall Outlook
(ERO)
> > verification code from METv6.0 to METv8.0. During this process, I
noticed
> > that the observation variable name after running grid_stat has
changed
> > between these two versions. My methodology is the same, and
applies a
> > circular neighborhood of 9 grid points in the interp definition of
the
> > grid_stat configuration file (old configuration file is
config_file_old,
> > new configuration file is config_file_new).
> >
> > In more detail if I run:
> > path_to_metv8.0/grid_stat obs_file fcst_file config_file_old
-outdir ./
> >
> > The name of my observational data within the netcdf file is called
> > OBS_ST4gFFG_Surface_FULL_MAX_49
> >
> > However, if I run
> > path_to_metv6.0/grid_stat obs_file fcst_file config_file_new
-outdir ./
> >
> > The name of my observational data within the netcdf file is called
> > OBS_ST4gFFG_Surface_FULL_MAX_81
> >
> > The data inside each file is exactly the same, but I am wondering
why the
> > metadata label is different? For reference, my files are stored
here:
> > https://ftp.wpc.ncep.noaa.gov/erickson/dtc2/. The grid_stat netCDF
> output
> > for METv6.0 is grid_stat_out_OLD.nc and for METv8.0 is
> grid_stat_out_NEW.nc
> >
> > Thanks again!
> >
> > Mike
> > --
> > Michael J. Erickson
> >
> > Research Scientist
> > Cooperative Institute for Research in Environmental Sciences
(CIRES)
> > NOAA/NWS/NCEP/Weather Prediction Center
> > Phone: 301-683-1546
> >
> >
>
>
--
Michael J. Erickson
Research Scientist
Cooperative Institute for Research in Environmental Sciences (CIRES)
NOAA/NWS/NCEP/Weather Prediction Center
Phone: 301-683-1546
------------------------------------------------
Subject: grid_stat Variable Name Output Change
From: John Halley Gotway
Time: Mon Apr 29 13:20:12 2019
Mike,
I was a bit surprised that a circle with diameter = 9 would only
result 49
grid points. I ran a test using regrid_data_plane and pcp_combine.
(1) Pick a single obs point.
(2) Run regrid_data_plane to get the max in a 9x9 circle and a 9x9
square.
(3) Use pcp_combine to compute the difference (square minus circle).
Attached is an image showing the difference. The red area are points
included in the 9x9 square that are not included in the 9x9 circle. I
suppose that's not an unreasonable result. What do you think?
Thanks,
John
On Mon, Apr 29, 2019 at 12:55 PM Michael Erickson - NOAA Affiliate via
RT <
met_help at ucar.edu> wrote:
>
> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=89965 >
>
> Hi John,
>
> Thank you for your quick response! That makes sense to me.
>
> Later code looks for naming specifically with 81 rather than 49,
although I
> can adapt it to be more flexible.
>
> Mike
>
>
>
> On Mon, Apr 29, 2019 at 6:48 PM John Halley Gotway via RT <
> met_help at ucar.edu>
> wrote:
>
> > Hi Mike,
> >
> > I see that you have a question about differences between met-6.0
and
> > met-8.0 when smoothing data.
> >
> > I used the data and config files you sent and ran Grid-Stat for
both
> > versions 6.0 and 8.0. First, I diffed the output in the .stat
files and
> > see that in both met-6.0 and met-8.0, INTERP_PNTS = 49. However,
in the
> > NetCDF matched pairs output files, met-6.0 has MAX_81 whereas met-
8.0 has
> > MAX_49, as you describe.
> >
> > Here's the logic. Earlier versions of MET only supported a square
> > interpolation region. A square with a width of 9 includes 81 grid
points
> > (9 * 9). Then in met-6.0, we added support for circular regions
as well.
> > Since we already had the "width" option defined, we continued to
use it
> for
> > circles but interpret it as the diameter of the circle. And we
updated
> the
> > logic to report the ACTUAL number of points included in the
circular
> > region. And that's correctly reported in the .stat output as
> INTERP_PNTS =
> > 49.
> >
> > However, we failed to correctly indicate that in the Grid-Stat
NetCDF
> > variable names. So somewhere between met-6.0 and met-8.0 we
corrected
> the
> > variable names to indicate the actual number of grid points
included in
> our
> > circle with a diameter of 9. And that's the reason for the
discrepancy.
> >
> > Make sense?
> >
> > Thanks,
> > John
> >
> >
> > On Mon, Apr 29, 2019 at 12:19 PM Michael Erickson - NOAA Affiliate
via
> RT <
> > met_help at ucar.edu> wrote:
> >
> > >
> > > Mon Apr 29 12:19:25 2019: Request 89965 was acted upon.
> > > Transaction: Ticket created by michael.j.erickson at noaa.gov
> > > Queue: met_help
> > > Subject: grid_stat Variable Name Output Change
> > > Owner: Nobody
> > > Requestors: michael.j.erickson at noaa.gov
> > > Status: new
> > > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=89965
> >
> > >
> > >
> > > Hello there,
> > >
> > > I am in the process of upgrading my Excessive Rainfall Outlook
(ERO)
> > > verification code from METv6.0 to METv8.0. During this process,
I
> noticed
> > > that the observation variable name after running grid_stat has
changed
> > > between these two versions. My methodology is the same, and
applies a
> > > circular neighborhood of 9 grid points in the interp definition
of the
> > > grid_stat configuration file (old configuration file is
> config_file_old,
> > > new configuration file is config_file_new).
> > >
> > > In more detail if I run:
> > > path_to_metv8.0/grid_stat obs_file fcst_file config_file_old
-outdir ./
> > >
> > > The name of my observational data within the netcdf file is
called
> > > OBS_ST4gFFG_Surface_FULL_MAX_49
> > >
> > > However, if I run
> > > path_to_metv6.0/grid_stat obs_file fcst_file config_file_new
-outdir ./
> > >
> > > The name of my observational data within the netcdf file is
called
> > > OBS_ST4gFFG_Surface_FULL_MAX_81
> > >
> > > The data inside each file is exactly the same, but I am
wondering why
> the
> > > metadata label is different? For reference, my files are stored
here:
> > > https://ftp.wpc.ncep.noaa.gov/erickson/dtc2/. The grid_stat
netCDF
> > output
> > > for METv6.0 is grid_stat_out_OLD.nc and for METv8.0 is
> > grid_stat_out_NEW.nc
> > >
> > > Thanks again!
> > >
> > > Mike
> > > --
> > > Michael J. Erickson
> > >
> > > Research Scientist
> > > Cooperative Institute for Research in Environmental Sciences
(CIRES)
> > > NOAA/NWS/NCEP/Weather Prediction Center
> > > Phone: 301-683-1546
> > >
> > >
> >
> >
>
> --
> Michael J. Erickson
>
> Research Scientist
> Cooperative Institute for Research in Environmental Sciences (CIRES)
> NOAA/NWS/NCEP/Weather Prediction Center
> Phone: 301-683-1546
>
>
------------------------------------------------
Subject: grid_stat Variable Name Output Change
From: John Halley Gotway
Time: Mon Apr 29 13:26:10 2019
Actually, I am a bit more concerned. A circle with diameter = 9 is
really
just a matrix of offsets from -4 to +4 in both the x and y directions.
I
used R to compute the distances for each of 81 offsets for those 81
points. Looking at the resulting differences, exactly 49 of them are
<= 4
while 69 of them are <= 4.5.
Based on this, I suspect there's some truncation occurring. I suspect
we're check a radius of 4 instead of 4.5.
I need to ask talk to folks around here some more about this.
Thanks,
John
On Mon, Apr 29, 2019 at 1:19 PM John Halley Gotway <johnhg at ucar.edu>
wrote:
> Mike,
>
> I was a bit surprised that a circle with diameter = 9 would only
result 49
> grid points. I ran a test using regrid_data_plane and pcp_combine.
> (1) Pick a single obs point.
> (2) Run regrid_data_plane to get the max in a 9x9 circle and a 9x9
square.
> (3) Use pcp_combine to compute the difference (square minus circle).
>
> Attached is an image showing the difference. The red area are
points
> included in the 9x9 square that are not included in the 9x9 circle.
I
> suppose that's not an unreasonable result. What do you think?
>
> Thanks,
> John
>
> On Mon, Apr 29, 2019 at 12:55 PM Michael Erickson - NOAA Affiliate
via RT <
> met_help at ucar.edu> wrote:
>
>>
>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=89965 >
>>
>> Hi John,
>>
>> Thank you for your quick response! That makes sense to me.
>>
>> Later code looks for naming specifically with 81 rather than 49,
although
>> I
>> can adapt it to be more flexible.
>>
>> Mike
>>
>>
>>
>> On Mon, Apr 29, 2019 at 6:48 PM John Halley Gotway via RT <
>> met_help at ucar.edu>
>> wrote:
>>
>> > Hi Mike,
>> >
>> > I see that you have a question about differences between met-6.0
and
>> > met-8.0 when smoothing data.
>> >
>> > I used the data and config files you sent and ran Grid-Stat for
both
>> > versions 6.0 and 8.0. First, I diffed the output in the .stat
files and
>> > see that in both met-6.0 and met-8.0, INTERP_PNTS = 49. However,
in the
>> > NetCDF matched pairs output files, met-6.0 has MAX_81 whereas
met-8.0
>> has
>> > MAX_49, as you describe.
>> >
>> > Here's the logic. Earlier versions of MET only supported a
square
>> > interpolation region. A square with a width of 9 includes 81
grid
>> points
>> > (9 * 9). Then in met-6.0, we added support for circular regions
as
>> well.
>> > Since we already had the "width" option defined, we continued to
use it
>> for
>> > circles but interpret it as the diameter of the circle. And we
updated
>> the
>> > logic to report the ACTUAL number of points included in the
circular
>> > region. And that's correctly reported in the .stat output as
>> INTERP_PNTS =
>> > 49.
>> >
>> > However, we failed to correctly indicate that in the Grid-Stat
NetCDF
>> > variable names. So somewhere between met-6.0 and met-8.0 we
corrected
>> the
>> > variable names to indicate the actual number of grid points
included in
>> our
>> > circle with a diameter of 9. And that's the reason for the
discrepancy.
>> >
>> > Make sense?
>> >
>> > Thanks,
>> > John
>> >
>> >
>> > On Mon, Apr 29, 2019 at 12:19 PM Michael Erickson - NOAA
Affiliate via
>> RT <
>> > met_help at ucar.edu> wrote:
>> >
>> > >
>> > > Mon Apr 29 12:19:25 2019: Request 89965 was acted upon.
>> > > Transaction: Ticket created by michael.j.erickson at noaa.gov
>> > > Queue: met_help
>> > > Subject: grid_stat Variable Name Output Change
>> > > Owner: Nobody
>> > > Requestors: michael.j.erickson at noaa.gov
>> > > Status: new
>> > > Ticket <URL:
https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=89965
>> >
>> > >
>> > >
>> > > Hello there,
>> > >
>> > > I am in the process of upgrading my Excessive Rainfall Outlook
(ERO)
>> > > verification code from METv6.0 to METv8.0. During this process,
I
>> noticed
>> > > that the observation variable name after running grid_stat has
changed
>> > > between these two versions. My methodology is the same, and
applies a
>> > > circular neighborhood of 9 grid points in the interp definition
of the
>> > > grid_stat configuration file (old configuration file is
>> config_file_old,
>> > > new configuration file is config_file_new).
>> > >
>> > > In more detail if I run:
>> > > path_to_metv8.0/grid_stat obs_file fcst_file config_file_old
-outdir
>> ./
>> > >
>> > > The name of my observational data within the netcdf file is
called
>> > > OBS_ST4gFFG_Surface_FULL_MAX_49
>> > >
>> > > However, if I run
>> > > path_to_metv6.0/grid_stat obs_file fcst_file config_file_new
-outdir
>> ./
>> > >
>> > > The name of my observational data within the netcdf file is
called
>> > > OBS_ST4gFFG_Surface_FULL_MAX_81
>> > >
>> > > The data inside each file is exactly the same, but I am
wondering why
>> the
>> > > metadata label is different? For reference, my files are stored
here:
>> > > https://ftp.wpc.ncep.noaa.gov/erickson/dtc2/. The grid_stat
netCDF
>> > output
>> > > for METv6.0 is grid_stat_out_OLD.nc and for METv8.0 is
>> > grid_stat_out_NEW.nc
>> > >
>> > > Thanks again!
>> > >
>> > > Mike
>> > > --
>> > > Michael J. Erickson
>> > >
>> > > Research Scientist
>> > > Cooperative Institute for Research in Environmental Sciences
(CIRES)
>> > > NOAA/NWS/NCEP/Weather Prediction Center
>> > > Phone: 301-683-1546
>> > >
>> > >
>> >
>> >
>>
>> --
>> Michael J. Erickson
>>
>> Research Scientist
>> Cooperative Institute for Research in Environmental Sciences
(CIRES)
>> NOAA/NWS/NCEP/Weather Prediction Center
>> Phone: 301-683-1546
>>
>>
------------------------------------------------
Subject: grid_stat Variable Name Output Change
From: John Halley Gotway
Time: Mon Apr 29 13:55:14 2019
OK, nevermind, I talked to Tressa about it and drew some pictures, and
I'm
satisfied that MET is doing something reasonable. A circle with
diameter =
9 includes 9 *grid points* in each direction. But those 9 grid points
represents 8 grid lengths between them. So applying a radius = 4 is a
reasonable thing to do.
I'll go ahead and resolve this ticket now.
Thanks,
John
On Mon, Apr 29, 2019 at 1:25 PM John Halley Gotway <johnhg at ucar.edu>
wrote:
> Actually, I am a bit more concerned. A circle with diameter = 9 is
really
> just a matrix of offsets from -4 to +4 in both the x and y
directions. I
> used R to compute the distances for each of 81 offsets for those 81
> points. Looking at the resulting differences, exactly 49 of them
are <= 4
> while 69 of them are <= 4.5.
>
> Based on this, I suspect there's some truncation occurring. I
suspect
> we're check a radius of 4 instead of 4.5.
>
> I need to ask talk to folks around here some more about this.
>
> Thanks,
> John
>
> On Mon, Apr 29, 2019 at 1:19 PM John Halley Gotway <johnhg at ucar.edu>
> wrote:
>
>> Mike,
>>
>> I was a bit surprised that a circle with diameter = 9 would only
result
>> 49 grid points. I ran a test using regrid_data_plane and
pcp_combine.
>> (1) Pick a single obs point.
>> (2) Run regrid_data_plane to get the max in a 9x9 circle and a 9x9
square.
>> (3) Use pcp_combine to compute the difference (square minus
circle).
>>
>> Attached is an image showing the difference. The red area are
points
>> included in the 9x9 square that are not included in the 9x9 circle.
I
>> suppose that's not an unreasonable result. What do you think?
>>
>> Thanks,
>> John
>>
>> On Mon, Apr 29, 2019 at 12:55 PM Michael Erickson - NOAA Affiliate
via RT
>> <met_help at ucar.edu> wrote:
>>
>>>
>>> <URL: https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=89965 >
>>>
>>> Hi John,
>>>
>>> Thank you for your quick response! That makes sense to me.
>>>
>>> Later code looks for naming specifically with 81 rather than 49,
>>> although I
>>> can adapt it to be more flexible.
>>>
>>> Mike
>>>
>>>
>>>
>>> On Mon, Apr 29, 2019 at 6:48 PM John Halley Gotway via RT <
>>> met_help at ucar.edu>
>>> wrote:
>>>
>>> > Hi Mike,
>>> >
>>> > I see that you have a question about differences between met-6.0
and
>>> > met-8.0 when smoothing data.
>>> >
>>> > I used the data and config files you sent and ran Grid-Stat for
both
>>> > versions 6.0 and 8.0. First, I diffed the output in the .stat
files
>>> and
>>> > see that in both met-6.0 and met-8.0, INTERP_PNTS = 49.
However, in
>>> the
>>> > NetCDF matched pairs output files, met-6.0 has MAX_81 whereas
met-8.0
>>> has
>>> > MAX_49, as you describe.
>>> >
>>> > Here's the logic. Earlier versions of MET only supported a
square
>>> > interpolation region. A square with a width of 9 includes 81
grid
>>> points
>>> > (9 * 9). Then in met-6.0, we added support for circular regions
as
>>> well.
>>> > Since we already had the "width" option defined, we continued to
use
>>> it for
>>> > circles but interpret it as the diameter of the circle. And we
>>> updated the
>>> > logic to report the ACTUAL number of points included in the
circular
>>> > region. And that's correctly reported in the .stat output as
>>> INTERP_PNTS =
>>> > 49.
>>> >
>>> > However, we failed to correctly indicate that in the Grid-Stat
NetCDF
>>> > variable names. So somewhere between met-6.0 and met-8.0 we
corrected
>>> the
>>> > variable names to indicate the actual number of grid points
included
>>> in our
>>> > circle with a diameter of 9. And that's the reason for the
>>> discrepancy.
>>> >
>>> > Make sense?
>>> >
>>> > Thanks,
>>> > John
>>> >
>>> >
>>> > On Mon, Apr 29, 2019 at 12:19 PM Michael Erickson - NOAA
Affiliate via
>>> RT <
>>> > met_help at ucar.edu> wrote:
>>> >
>>> > >
>>> > > Mon Apr 29 12:19:25 2019: Request 89965 was acted upon.
>>> > > Transaction: Ticket created by michael.j.erickson at noaa.gov
>>> > > Queue: met_help
>>> > > Subject: grid_stat Variable Name Output Change
>>> > > Owner: Nobody
>>> > > Requestors: michael.j.erickson at noaa.gov
>>> > > Status: new
>>> > > Ticket <URL:
>>> https://rt.rap.ucar.edu/rt/Ticket/Display.html?id=89965 >
>>> > >
>>> > >
>>> > > Hello there,
>>> > >
>>> > > I am in the process of upgrading my Excessive Rainfall Outlook
(ERO)
>>> > > verification code from METv6.0 to METv8.0. During this
process, I
>>> noticed
>>> > > that the observation variable name after running grid_stat has
>>> changed
>>> > > between these two versions. My methodology is the same, and
applies a
>>> > > circular neighborhood of 9 grid points in the interp
definition of
>>> the
>>> > > grid_stat configuration file (old configuration file is
>>> config_file_old,
>>> > > new configuration file is config_file_new).
>>> > >
>>> > > In more detail if I run:
>>> > > path_to_metv8.0/grid_stat obs_file fcst_file config_file_old
-outdir
>>> ./
>>> > >
>>> > > The name of my observational data within the netcdf file is
called
>>> > > OBS_ST4gFFG_Surface_FULL_MAX_49
>>> > >
>>> > > However, if I run
>>> > > path_to_metv6.0/grid_stat obs_file fcst_file config_file_new
-outdir
>>> ./
>>> > >
>>> > > The name of my observational data within the netcdf file is
called
>>> > > OBS_ST4gFFG_Surface_FULL_MAX_81
>>> > >
>>> > > The data inside each file is exactly the same, but I am
wondering
>>> why the
>>> > > metadata label is different? For reference, my files are
stored here:
>>> > > https://ftp.wpc.ncep.noaa.gov/erickson/dtc2/. The grid_stat
netCDF
>>> > output
>>> > > for METv6.0 is grid_stat_out_OLD.nc and for METv8.0 is
>>> > grid_stat_out_NEW.nc
>>> > >
>>> > > Thanks again!
>>> > >
>>> > > Mike
>>> > > --
>>> > > Michael J. Erickson
>>> > >
>>> > > Research Scientist
>>> > > Cooperative Institute for Research in Environmental Sciences
(CIRES)
>>> > > NOAA/NWS/NCEP/Weather Prediction Center
>>> > > Phone: 301-683-1546
>>> > >
>>> > >
>>> >
>>> >
>>>
>>> --
>>> Michael J. Erickson
>>>
>>> Research Scientist
>>> Cooperative Institute for Research in Environmental Sciences
(CIRES)
>>> NOAA/NWS/NCEP/Weather Prediction Center
>>> Phone: 301-683-1546
>>>
>>>
------------------------------------------------
More information about the Met_help
mailing list