WRF-able data sets

RDA data specialists has been working with wrfhelp to help Weather Research and Forecasting Model (WRF) users more easily get set up and running with RDA-supplied data sets.

Gridded data in GRIB format is used in the WRF Preprocessing System (WPS) to create both initial conditions inside the WRF domains and lateral boundary conditions outside of them.  In order to read in the GRIB data and write it out into a WRF input file, users need to supply a Vtable--a Rosetta stone of sorts that tells WPS which variables to pull-out from the GRIB file.

(You can use more than one input GRIB file.  For instance some users pull SST out of one file and the atmospheric parameters from another one.  You'll need the appropriate Vtable for each input file you use.)

RDA-generated Vtable for FNL, ds083.2
User-contributed Vtables can be found in the WRF WPS directory under ungrib/Variable_Tables.  However, the mapping between a particular GRIB file and the correct Vtable to apply is not simple and clear-cut.

How do you know that the version of a file you have contains the same parameters as an older user-contributed Vtable?  As I explained in What is GRIB2?, it is easy to edit GRIB files to remove or add parameters.

The publicly-distributed version of an analysis file you have may not be the one used internally at some operational centers for which the Vtable was created.  If you tried to apply their Vtable to the data file you have, ungrib.exe may crash.

RDA hosts 110 GRIB1 and 17 GRIB2 datasets.  Of those, our data specialists estimate 13 are suitable candidates for initializing WRF with RDA generated Vtables.  We noticed that most WRF users rely on only a couple of RDA data sets (FNL and GFS)--even when their stated research goals (long time series) would caution against it.  (Read why in Analysis, forecast, reanalysis--what's the difference?)

We wondered if users would branch out and use other GRIB data sets if they could obtain a Vtable that works for those data files.

RDA data specialist (and metadata specialist), Bob Dattore, wrote the software that generates Vtables from the metadata we automatically collect on our data holdings as we archive them.  This will ensure that the Vtables match the data files we serve to you.

Sound good?

How do you find these "WRF-able" datasets?

In the data search matrix on our home page, http://rda.ucar.edu/, select "Supports Project" = "WRF".

Select "Supports Project" = "WRF".  You may need to scroll down the pull-down menu.
That gives you a list of data sets suitable for WRF initialization (using ungrib.exe) as well as WRFDA and FDDA observation and analysis nudging.

For WRF initialization, further narrow down the selection by selecting "Type of Data" = "Grid".
Further narrow down the selection by selecting "Type of Data" = "Grid".
On the data set home page, scroll down to the "WRF Preprocessing System (WPS)" section. Click the Vtable link to download the appropriate Vtable.

Follow the steps in the "Vtable Discussion" of the WRF_NCEP2 help document.

Please let us know about your experiences using our Vtables by sending email to both rdahelp@ucar.edu and the data specialist assigned to the data set you tried.

Go forth and WRF.

We maintain a web page with the full list of data sets with RDA-supplied Vtables.

Datasets with RDA-supplied Vtables:

We have removed ds335.0 Historical Unidata Internet Data Distribution (IDD) Gridded Model Data from the list. ds335.0 is a compound data set made up of files available in the Unidata real-time data distribution stream.

Some component files in ds335.0 are WRF-able.  Please use the Vtables for the individual component data sets listed above.


  1. Could you please provide me with a Vtable for ERA-40 dataset? It seems that both Vtable.ECMWF and Vtable.ERA-Interim are not the right ones.

    1. Hi Stavros.
      When you are referring to "ERA-40", are you referring to ERA-40 data obtained from ECMWF, or ERA-40 Data obtained from the RDA? In either case, the Vtables "Vtable.ECMWF" and "Vtable.ERA-Interim" are unlikely to work in a generic sense for ERA-40, most likely because all required parameters (variables) are not present in the source GRIB files, and/or, are distributed in such a way as to prevent automated generation of a Vtable (as in the case of RDA ERA-40 in which the datafilee are distributed over multiple datasets). Therefore, we have not generated an automated Vtable for RDA ERA-40. We apologize for this inconvenience,


This section is for people who want to discuss using our data holdings effectively. Moderators will delete irrelevant comments.