How to use WRF outputs (in NetCDF Format) as boundary conditions

Questions/topics related to the first-guess input meteorological data, and Vtables
Post Reply
katierobinson
Posts: 3
Joined: Wed May 27, 2020 9:25 pm

How to use WRF outputs (in NetCDF Format) as boundary conditions

Post by katierobinson » Thu May 28, 2020 2:30 pm

Hi,
I want to use the outputs from a previously run WRF simulation as the boundary conditions for another, higher resolution simulation.

I converted the wrfout* netcdf file to GRIB format using: cdo -v -f grb -copy -selgrid,1,5,6 wrfout_d01_1979-01-01_00:00:00 wrfout_d01_1979-01-01_00:00:00.grb

I then ran ungrib successfully, but when running metgrid I get the error: ERROR: The mandatory field TT was not found in any input data

I am using Vtable.ARW, but it does not appear to be able to find the T variable in the converted file.
cdo sinfo of the original file gives this:

File format: netCDF2
-1 : Institut Source Ttype Levels Num Gridsize Num Dtype : Parameter name
1 : unknown unknown instant 1 1 98441 1 F32 : LU_INDEX
2 : unknown unknown instant 1 1 26 2 F32 : ZNU
3 : unknown unknown instant 1 1 27 3 F32 : ZNW
4 : unknown unknown instant 1 1 4 4 F32 : ZS
5 : unknown unknown instant 1 1 4 4 F32 : DZS
6 : unknown unknown instant 1 1 98441 1 F32 : VAR_SSO
7 : unknown unknown instant 26 2 98728 5 F32 : U
8 : unknown unknown instant 26 2 98784 6 F32 : V
9 : unknown unknown instant 27 3 98441 1 F32 : W
10 : unknown unknown instant 27 3 98441 1 F32 : PH
11 : unknown unknown instant 27 3 98441 1 F32 : PHB
12 : unknown unknown instant 26 2 98441 1 F32 : T

But cdo sinfo of the converted grib file gives this:

File format: GRIB
-1 : Institut Source Ttype Levels Num Gridsize Num Dtype : Parameter name
1 : unknown unknown instant 1 1 98441 1 P16 : var1
2 : unknown unknown instant 1 1 98441 1 P16 : var6
3 : unknown unknown instant 26 2 98728 2 P16 : var7
4 : unknown unknown instant 26 2 98784 3 P16 : var8
5 : unknown unknown instant 27 3 98441 1 P16 : var9
6 : unknown unknown instant 27 3 98441 1 P16 : var10
7 : unknown unknown instant 27 3 98441 1 P16 : var11
8 : unknown unknown instant 26 2 98441 1 P16 : var12
9 : unknown unknown instant 1 1 98441 1 P16 : var19

How can I fix this so that the parameter names are not lost when converting from netcdf to grib? Or is there another way to use WRF outputs as boundary conditions for running WPS?

Thanks!

Ming Chen
Posts: 1013
Joined: Mon Apr 23, 2018 9:42 pm

Re: How to use WRF outputs (in NetCDF Format) as boundary conditions

Post by Ming Chen » Sat May 30, 2020 3:54 pm

I wonder whether you can run ndown? I suppose ndown can do exactly what you want to. Please see the details at
https://www2.mmm.ucar.edu/wrf/users/doc ... way_Nested

if for some reasons you don't want to run ndown, then I would suggest you extract required variables from wrfout files, write them in intermediate format, and then run metgrid.exe. In this case you don't need to go through the trouble to transfer wrfout to grib, and ungrib the files.
WRF Help Desk

katierobinson
Posts: 3
Joined: Wed May 27, 2020 9:25 pm

Re: How to use WRF outputs (in NetCDF Format) as boundary conditions

Post by katierobinson » Sat May 30, 2020 5:23 pm

Thanks for your reply!
Is there a method you would recommend for extracting the required variables from wrfout files to write them in an intermediate format??

Ming Chen
Posts: 1013
Joined: Mon Apr 23, 2018 9:42 pm

Re: How to use WRF outputs (in NetCDF Format) as boundary conditions

Post by Ming Chen » Sat May 30, 2020 7:56 pm

Below are the variables required for WPS (and later for running WRF):
Mandatory:
3d upper air: horizontal winds, temperature, relative humidity, geopotential height
3d soil: soil temperature and soil moisture
2d fields: surface pressure, sea-level pressure, land mask, Skin temperature, 2-m T & RH,, 10-m U and V
Optional (but desirable):
2d fields: topography elevation of input data, SST, sea-ice
Fortran codes, or NCL, or other post process tools like RIP and UPP can read wrfout files, put variables on pressure levels, and remap to different grids, etc. After that, you can write the data in intermediate format. Detailed description can be found at:
https://www2.mmm.ucar.edu/wrf/users/doc ... gical_Data
WRF Help Desk

Post Reply

Return to “Meteorological Input Data”