vertical wind (w), potential temperature, output on pressure level

[bnepal22]

I am completely new to the UFS model. I am trying to get the vertical wind (w) and potential temperature on a pressure level, similar to w100, w200, w500 etc. Is it possible to get it?

[laurenfrederick]

Hi @bnepal22, @natalie-perlin and @RatkoVasic-NOAA are taking a look and will get back to you with an answer.

[RatkoVasic-NOAA]

@bnepal22 In file diag_table in run directory, you have a list of 3D and 2D fields for output. For your case, there should be already field:
"gfs_dyn", "w", "dzdt", "fv3_history", "all", .false., "none", 2
I looked in control_c192_intel case, and in file atmf012.nc I can find:

        float dzdt(time, pfull, grid_yt, grid_xt) ;
                dzdt:_FillValue = 9.99e+20f ;
                dzdt:cell_methods = "time: point" ;
                dzdt:long_name = "vertical wind" ;
                dzdt:missing_value = 9.99e+20f ;
                dzdt:output_file = "atm" ;
                dzdt:units = "m/sec" ;

That one is on model levels, here is example of interpolated to standard pressure levels, case control_p8_intel, file GFSPRS.GrbF12:

Hera:> wgrib2 -v GFSPRS.GrbF12  | grep DZDT | grep 500
489:31610864:d=2021032206:DZDT Vertical Velocity (Geometric) [m/s]:500 mb:12 hour fcst:

[bnepal22]

@RatkoVasic-NOAA Thank you for your response. Yes, I was able to get the 'w' as a 3d variable as suggested by you, however, I was not able to interpolate it to the regular grid. Currently, I am trying to do it for a stretched grid and using freegrid_parallel tool for to transform the stretched grid to a regular longitude latitude grid for further analysis. Therefore, I wanted to get the w at each pressure level individually as a 2d variable just like zonal (u) and meridional wind (v) by specifying w500, w100 and so on. Currently I am using GFS_v15p2.

[bnepal22]

I think I got the solution. I am now creating a output at model level and then converting hybrid sigma level to pressure level for further analysis using Python. based package PyNgl.

[RatkoVasic-NOAA]

@bnepal22 I'm glad that you found solution that is working for you!

[bnepal22]

But still, I am confused about, how to identify the list of parameters/ variables that can be output and their corresponding field in which they are registered. In my diag_table I have only a few variables. Where to find the list of variables that can be output? I am currently using ccpp_suite = "FV3_GFS_v15p2". And running it for S2S scale with grid stretching.

#Current variables I have in my diag_table
"gfs_sfc", "t2m", "tmp2m", "xapp_ins_6h", "all", .false., "none", 2
"gfs_phys", "totprcp_ave", "prate_ave", "xapp_ins_6h", "all", .false., "none", 2
"gfs_phys", "u10m", "ugrd10m", "xapp_ins_6h", "all", .false., "none", 2
"gfs_phys", "v10m", "vgrd10m", "xapp_ins_6h", "all", .false., "none", 2
"gfs_phys", "tmpmax2m", "tmax_max2m","xapp_ins_1d","all", .false., "none", 2
"gfs_phys", "tmpmin2m", "tmin_min2m","xapp_ins_1d","all", .false., "none", 2

[RatkoVasic-NOAA]

Sure. in this file you have all available diagnostics which you can put in diag_table.
That is place for developers to add new diagnostics for their work.
ufs-weather-model/FV3/ccpp/driver/GFS_diagnostics.F90

[bnepal22]

Thank you very much for this information. I will go through it and will do accordingly.