[Met_help] [rt.rap.ucar.edu #89106] History for Evaluation of PM2.5 on MET

[John Halley Gotway via RT]

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

I chatted with you a little bit last week.
Our team (Georg Grell and Ravan Ahmadov) would like to use met to do  evaluation for FV3 and RAP/HRRR model for all of our chemical variables.
For now , we have the model data and station observation data for the surface PM2.5.
I am new to met and I'm wondering grid_stat is what we are looking for?

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

MET reads gridded forecast data in GRIB 1, GRIB 2, CF-compliant NetCDF, or the NetCDF output of the wrf_interp utility.  May be easiest to just have you come by my office or point me to a sample file you’re trying to use.

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

Thank you very much. I will come over later.
FYI, my HRRR data is on theia (variable smoke):
/scratch3/BMC/wrf-chem/KaYee/met/HRRR_output/out_3D_2018-08-11_09_00_00.nc
My observation data is here (variable PM):
/scratch3/BMC/wrf-chem/KaYee/met/PMobs_905sites_JulyAugust2018.nc
I'm not able to read both netcdf data.

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

Subject: Evaluation of PM2.5 on MET
From: John Halley Gotway
Time: Wed Feb 27 11:11:35 2019

Hi Ka Yee,

I moved this question over to met-help...

I took a look at your data on theia in /scratch3/BMC/wrf-
chem/KaYee/met/HRRR_output.  The NetCDF files in there look a lot like
WRF output files... but with a big caveat.  All of the global
attributes which define the projection information are missing from
this file.  Presumably, you ran some NCO commands to subset your WRF
output file.  I'd recommend that you look at those steps and carry
forward WRF's global attributes to the output file.  If those were
present then MET would be able to read this data.

Whenever getting started with a new dataset in MET, I always recommend
running the plot_data_plane tool.  That demonstrates that MET is
actually able to successfully read your gridded data.  Here are the
commands I was trying to run:

module use /contrib/modulefiles
module load met/8.0
cd /scratch3/BMC/wrf-chem/KaYee/met/HRRR_output
plot_data_plane out_3D_2018-08-11_11_00_00.nc ~/test.ps 'name="smoke";
level="(0,0,*,*)"; file_type=NETCDF_PINT;'

The file_type = NETCDF_PINT; option is telling MET to interpret this
data as if it were output from the WRF p_interp or wrf_interp utility.
Since the global attributes are missing, it isn't able to read this
data.

FYI, MET can easily read the GRIB2 data from that directory.  I ran
plot_data_plane to display 2-meter temperature from one of these
files:

plot_data_plane wrfprs_hrconus_06.grib2 ~/TMP_Z2.ps 'name="TMP";
level="Z2";'

If the "smoke" data is included in these GRIB2 files, we should just
read it from there!

Next, looking at the point observations:
/scratch3/BMC/wrf-chem/KaYee/met/PMobs_905sites_JulyAugust2018.nc

MET includes some pre-processing utilities to put point observations
into a common format.  Unfortunately, this data is not already in that
common format.  So we'd need to reformat it to make it look like what
MET is expecting.


Thanks,
John

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