[Met_help] [rt.rap.ucar.edu #93298] History for NetCDF input to Grid Stat

[John Halley Gotway via RT]

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

We'd like to stick to verifying gridded forecasts and gridded observations here at CPC. 

Our forecasts and observations are all in NetCDF 4, and are not entirely CF-compliant. So I guess here are my specific questions for now:

(1) Can Grid Stat read NetCDF 4?
(2) Since our NetCDF files are not entirely CF-compliant, is there some way to tell MET (in a config file, for example) what the names of the time and space dimensions are?
(3) If these are not possible, will they be in a future version of MET?

Thanks a lot,
Mike

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

Subject: NetCDF input to Grid Stat
From: John Halley Gotway
Time: Mon Dec 02 11:13:40 2019

Mike,

I see you have a question about NetCDF inputs to the Grid-Stat tool.

Yes, Grid-Stat (and the rest of the MET tools) can read gridded
NetCDF4 files.  If you would like to send a sample NetCDF CPC data
file, I could try it out and see what changes would be needed for MET
to read it.  You can post sample data to our anonymous ftp site:

https://dtcenter.org/community-code/model-evaluation-tools-met/met-
help-desk#ftp

Assuming MET isn't able to read your NetCDF files as-is, your options
are:

(1) Change the format to actually be CF-compliant and then we can
double-check to be sure MET can read it.

(2) Use the existing python embedding capability to tell MET how to
read them.  You'd write a python script to define the metadata and
data values and pass them to MET via memory.  Here's some examples of
python embedding:

https://dtcenter.org/community-code/model-evaluation-tools-met/sample-
analysis-scripts

The existing implementation works well for most tools, but not mtd,
ensemble-stat, or series-analysis, which read multiple input files.

(3) If/when we get the feature listed below implemented, it could
largely replace the python embedding capability for you:

https://github.com/NCAR/MET/issues/1020

Hope that helps.

Thanks,
John

------------------------------------------------
Subject: Re: [rt.rap.ucar.edu #93298] NetCDF input to Grid Stat
From: mike.charles at noaa.gov
Time: Mon Dec 02 11:42:37 2019

Thanks Jon, Python embedding sounds perfect, I will take a look at the
docs and get back to you if I can’t get it working.

--
Mike Charles
Climate Prediction Center

> On Dec 2, 2019, at 1:24 PM, John Halley Gotway via RT
<met_help at ucar.edu> wrote:
>
> Mike,
>
> I see you have a question about NetCDF inputs to the Grid-Stat tool.
>
> Yes, Grid-Stat (and the rest of the MET tools) can read gridded
NetCDF4 files.  If you would like to send a sample NetCDF CPC data
file, I could try it out and see what changes would be needed for MET
to read it.  You can post sample data to our anonymous ftp site:
>
> https://dtcenter.org/community-code/model-evaluation-tools-met/met-
help-desk#ftp
>
> Assuming MET isn't able to read your NetCDF files as-is, your
options are:
>
> (1) Change the format to actually be CF-compliant and then we can
double-check to be sure MET can read it.
>
> (2) Use the existing python embedding capability to tell MET how to
read them.  You'd write a python script to define the metadata and
data values and pass them to MET via memory.  Here's some examples of
python embedding:
>
> https://dtcenter.org/community-code/model-evaluation-tools-
met/sample-analysis-scripts
>
> The existing implementation works well for most tools, but not mtd,
ensemble-stat, or series-analysis, which read multiple input files.
>
> (3) If/when we get the feature listed below implemented, it could
largely replace the python embedding capability for you:
>
> https://github.com/NCAR/MET/issues/1020
>
> Hope that helps.
>
> Thanks,
> John


------------------------------------------------
Subject: NetCDF input to Grid Stat
From: melissa.ou at noaa.gov
Time: Mon Dec 02 12:41:53 2019

ok thanks for letting me know. hopefully you can get it working!

On Mon, Dec 2, 2019 at 1:42 PM mike.charles at noaa.gov via RT <
met_help at ucar.edu> wrote:

> Thanks Jon, Python embedding sounds perfect, I will take a look at
the
> docs and get back to you if I can’t get it working.
>
> --
> Mike Charles
> Climate Prediction Center
>
> > On Dec 2, 2019, at 1:24 PM, John Halley Gotway via RT
<met_help at ucar.edu>
> wrote:
> >
> > Mike,
> >
> > I see you have a question about NetCDF inputs to the Grid-Stat
tool.
> >
> > Yes, Grid-Stat (and the rest of the MET tools) can read gridded
NetCDF4
> files.  If you would like to send a sample NetCDF CPC data file, I
could
> try it out and see what changes would be needed for MET to read it.
You
> can post sample data to our anonymous ftp site:
> >
> >
> https://dtcenter.org/community-code/model-evaluation-tools-met/met-
help-desk#ftp
> >
> > Assuming MET isn't able to read your NetCDF files as-is, your
options
> are:
> >
> > (1) Change the format to actually be CF-compliant and then we can
> double-check to be sure MET can read it.
> >
> > (2) Use the existing python embedding capability to tell MET how
to read
> them.  You'd write a python script to define the metadata and data
values
> and pass them to MET via memory.  Here's some examples of python
embedding:
> >
> >
> https://dtcenter.org/community-code/model-evaluation-tools-
met/sample-analysis-scripts
> >
> > The existing implementation works well for most tools, but not
mtd,
> ensemble-stat, or series-analysis, which read multiple input files.
> >
> > (3) If/when we get the feature listed below implemented, it could
> largely replace the python embedding capability for you:
> >
> > https://github.com/NCAR/MET/issues/1020
> >
> > Hope that helps.
> >
> > Thanks,
> > John
>
>
>

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