GFS Analysis and Forecast System has been upgraded wef 1200 UTC, January 14, 2015. Your WRF forecast is likely to fail if you have not yet updated your scripts. Please do the following changes in order for the run to be successful.
Update your Vtable after downloading from this command/link
wget http://www2.mmm.ucar.edu/wrf/src/Vtable.GFS_new
GFS file naming conventions have also been changed, you need to download either of the file instead of gfs.t00z.pgrb2f00
You can download any of the file.
gfs.t00z.pgrb2.0p25.f000 : 0.25 degree
gfs.t00z.pgrb2.0p50.f000 : 0.50 degree
gfs.t00z.pgrb2.1p00.f000 : 1.00 degree
gfs.t00z.pgrb2.2p20.f000 : 2.50 degree
Please note that all gridded pressure GRIB files (*pgrb*) will be up to 20 minutes late. Therefore you will have to adjust your crontab to download the files.
Ungrib and metgrid may be successful but failing to change Vtable may stop real or wrf run or you may have NUM_METGRID_SOIL_LEVELS = 0 error.
Please leave your comments if you have any query
Hi,
ReplyDeleteAs you write I get this "NUM_METGRID_SOIL_LEVELS = 0 error." when I'm trying to run WRF with the new data (gfs.t00z.pgrb2.1p00.f000 : 1.00 degree), even though I have downloaded the new Vtable.
I am linking to it as: ln -sf ungrib/Variable_Tables/Vtable.GFS_new Vtable.
I must be missing something obviously. Do you have any ideas?
Thanks in advance.
Best regards, Müller.
Hi Muller,
DeleteI am not sure how your wps directory has been organised,
Just check your Vtable in your working directory (directory where you have your namelist), it should be linked tVtable.GFS_new
Another thing you can check that you are not downloading the pgrb2b files.
Other than this I don't see any reason why you would get this error.
Regards
Vikas
had same problems; try
Deleteftp//ftp.ncep.noaa.gov.pup.data.nccf.com/gfs/prod/gfs.2015011500/gfs.t00z.pgrb2full.0p50.f000
Hi,
ReplyDeleteI changed the Vtable and now I don't get the NUM_METGRID_SOIL_LEVELS = 0 error any more. But I still have an error due to a different num_metgrid_levels (47 instead of 48) in one gfs file. Did you got an error like this? And in case, have you found a solution?
Thanks in advance,
Best regards, Alberto
Great!!! Thanks from El Salvador!!!
ReplyDeleteHi everybody, I have downloaded the new Vtable, with gfs.t00z.pgrb2full.0p50.f000, and have the same NUM_METGRID_SOIL_LEVELS = 0 error
ReplyDeletePlease check you have linked the new Vtable correctly
DeleteSolved for me.
DeleteHi all,
I have dealt with the same problem (NUM_METGRID_SOIL_LEVELS = 0 error). I couldn't found any concrete solution anywhere, because it is not a clear problem as long as there are people who has found it and people who not.
For those that have it as me try to do this exercise: download the complete GFS files (approx. 65MB each one). Then try to run WPS and WRF again. The problem should desappear. That is because probably you as me are downloading a subset of variables from GFS. As long as these files have change probably we are missing soil moisture, and even though the VTable.GFS_new is correct the complete needed data are not available.
I still have no investigated which variables should I add to my old request to nomads server. If someone could help on these it could be nice.
Sorry for the extension.
Hi,
ReplyDeleteThanks from Czech Republic.
I am new to WRF and also to Linux. I have already had some issues with GFS files so I used another files.
Now, I used this help:
http://lundquistgroup.blogspot.cz/2014/04/using-input-datasets-in-real-wrf-runs.html
and these files:
http://rda.ucar.edu/datasets/ds083.2/index.html#sfol-wl-/data/ds083.2?g=22015
and by running ./real.exe it has error of mismatch of number of soil levels in met_em files and namelist.input or something like that.
but with new GFS table the real.exe runs correctly.