WRF with CFSR Data – Part Two

See Part One for hints on how to fetch CSFR data in an automated way. You only need two of the WRF presets, since SST is included in surface files:

  • pressure level data (pgbh06 files)
  • surface and fluxes (flxf06 files)

I rename the files and strip off the trailing date which refers to how they are archived e.g.

201001011800.pgbh06.gdas.20100101-20100105.grb2  ------>  201001011800.pgbh06.gdas.pressure.grb2
201001020000.flxf06.gdas.20100101-20100105.grb2  ------>  201001020000.flxf06.gdas.sst.grb2

Then you need to run ungrib.exe twice:

  1. in namelist.wps, set prefix=PLEVS and link Vtable.CFSR_press_pgbh06 --> Vtable
  2. in namelist.wps, set prefix=SFLUX and link Vtable.CFSR_sfc_flxf06 --> Vtable

That should create two sets of intermediate files prefixed PRESS and SFLUX. Next run metgrid with the

fg_name = 'PLEVS',  'SFLUX',

Et voila! It should work.

20 thoughts on “WRF with CFSR Data – Part Two

  1. Malleswararao


    Thanks for the WRF with CFSR data for part I and II. It works fine for me. Can you provide necessary information how to use MERRA data as input to WRF and Vtable corresponding to that one.

    Awaiting for your reply,

    Thanks & regards,

  2. Jianfeng Li


    Thanks for your part I and part II. It helps me a lot. But I still have some problems with the ungrib program. I tried to ungrib CFSv2 dataset by using the Vtable from http://drought.geo.msu.edu/data/CFSR4WRF/vtable.php. By using this Vtable, I only need to run ungrib.exe one time. I have done it before and it worked fine. But now there was strange words on the running “name of source model => unknown model from NCEP 197”. I don’t know what happened. The results seems right but I haven’t check them completely. All the following processes can work. Then I tried to ungrib pressure dataset and surface dataset separately according to your suggestions. The same problem! Do you know what’s wrong with that?

    Thank you very much for your time and concern.


  3. Simon

    Hey everyone,

    The WPS process works fine as it is described here. Also wrf.exe works fine as long as I use sf_surface_physics=1. If I want to use any other surface model I do get the error message: Missmatch with input data SF_SURFACE_PHYSICS=1. As anyone used different LSM than the 5 layer diffusion model?



  4. Prasanth

    Dear Simon,
    I followed your procedure but I’m getting this error! please help me!

    [pvk03@brlogin1 WPS]$ ./metgrid.exe
    Processing domain 1 of 1
    Processing 2000-07-01_00
    Oops, something is not right with the Gaussian latitude computation.
    The input data gave the starting latitude as 41.995.
    This routine computed the starting latitude as +- 89.761.
    The difference is larger than 0.01 degrees, which is not expected.
    ERROR: Gaussian_latitude_computation
    [unset]: aborting job:
    application called MPI_Abort(MPI_COMM_WORLD, 0) – process 0

    Thanks in advance

    1. Lydia

      Hi Prasanth,
      How did you solve your problem, the Gaussian_latitude_computation error? I got the same error as you did.

      Thanks a lot,

    2. Daniel


      How did you solve the problem?

      I’m getting the same error (slightly different coordinates, but I am guessing that the underlying the reason is the same).

  5. Felix

    Hi everyone!

    First of all, thanks for the two post about CFSR+WRF! I just have two questions:

    What version of the model are you using?
    I’m using 3.5.1 and there are three Vtables for CFSR: Vtable.CFSR2_web, Vtable.CFSR_press_pgbh06 and Vtable.CFSR_sfc_flxf06. Do you know when it’s appropiate or what are the differences between these three Vtables?

    Kind regards. and again, thanks for the post.

    1. Boreas Post author

      Hi Felix,

      I’m using WRF 3.5. I don’t know what the CFSR_web Vtable refers to. The ones I use are Vtable.CFSR_press_pgbh06 and Vtable.CFSR_sfc_flxf06 for the pressure-level data and surface flux data respectively.

        1. Boreas Post author

          In a New Year’s resolution to resurrect this blog I’m digging through the comments and realised I never replied. There should be no difference if you download CFSR data from NOMADS, in fact that is probably an easier way, thanks!

  6. Giacomo

    Hi, if I download the preset
    ds094.0 “WRF Model Input: Vtable.CFSR – Pressure levels”,
    I get such this file:
    so “pgrbh” instead of your “pgbh06”

    and if I download “WRF Model Input: Vtable.CFSR – Surface”,
    I get then two types of files, such as:

    So those ones instead of your “flxf06”. What should I do? flxf06 files are included instead in the “WRF Model Input: Vtable.SST” preset, but then I would have to download 3 presets: what should I do?


    1. Boreas Post author

      I think the pgrbh are the pressure level files, and the sfluxgrbf are the surface fluxes. From memory, I think that SST is already included in the surface flux, so the SST preset is not needed, but it is a long time since I have used this, so I might be wrong about that.

    2. Andreas

      Giacomo, have you been able to solve this? I mean, could you proceed with the prgbh and sfluxbrbf files? Or did you run into any problems later on?

  7. SOares


    Below is the content of input data for surface.

    1:0:d=2016051318:TMP:surface:6 hour fcst:
    2:1281235:d=2016051318:SOILW:0-0.1 m below ground:6 hour fcst:
    3:1913536:d=2016051318:SOILW:0.1-0.4 m below ground:6 hour fcst:
    4:2548523:d=2016051318:TMP:0-0.1 m below ground:6 hour fcst:
    5:3321462:d=2016051318:TMP:0.1-0.4 m below ground:6 hour fcst:
    6:4087136:d=2016051318:WEASD:surface:6 hour fcst:
    7:4550390:d=2016051318:LAND:surface:6 hour fcst:
    8:4570009:d=2016051318:ICEC:surface:6 hour fcst:
    9.1:4720308:d=2016051318:UGRD:10 m above ground:6 hour fcst:
    9.2:4720308:d=2016051318:VGRD:10 m above ground:6 hour fcst:
    10:6191049:d=2016051318:TMP:2 m above ground:6 hour fcst:
    11:7353550:d=2016051318:SPFH:2 m above ground:6 hour fcst:
    12:8441509:d=2016051318:PRES:surface:6 hour fcst:
    13:9035075:d=2016051318:SOILW:0.4-1 m below ground:6 hour fcst:
    14:9675613:d=2016051318:SOILW:1-2 m below ground:6 hour fcst:
    15:10307092:d=2016051318:TMP:0.4-1 m below ground:6 hour fcst:
    16:11066716:d=2016051318:TMP:1-2 m below ground:6 hour fcst:

    17:11825458:d=2016051318:HGT:surface:6 hour fcst:

    Is it possible if I can edit and modify the content of input data of CFSR because I am interesting only “10:6191049:d=2016051318:TMP:2 m above ground:6 hour fcst:” ?




Leave a Reply

Your email address will not be published. Required fields are marked *