Introduction

In the beginning, grib fields were identified by a name, a level and some timing information. Life was simple and the people were happy. This was soon to pass. A hoard of ensemble members appeared in the distance. A dust storm rolled in from the plains. The new power plants started putting sulfar dioxide into the air. All this had to added to the grib format. So names had to have modifiers, the percentile of some particulates of some chemical between size A and B microns. So wgrib2 added "extended names". So you could switch beween the old simpler names and the newer extended names. Why would you want use extended names? Sometimes you want the names to be unique. For example, when creating NetCDF file, you want the TMP from ensemble member 1 to has a different name from ensemble member 2. In another example, you may want to search for a unique field. With the extended name, you only need simple name comparison.

Extended Names

The extended name includes modifiers like the ensemble information, dust information and probabilities. Optionally you can include the level and forecast information (wgrib2 v3.0.2+). More details are given here.

Usage

   -set_ens_name N
      
      N == 0                 default
      N == 1                 add misc terms like ensemble or probability
                               extension for wgrib2 v3.0.2+
      and(N,1) == 1          add misc terms like ensemble or probability
      and(N,2) == 2          add level information
      and(N,4) == 4          add forecast information
      ex N == 3              misc + level information
      ex N == 5              misc + forecast information
      ex N == 6              level information + forecast information
      ex N == 7              misc + level information + forecast information
      N > 7                  for future use

Extended Extended Names

The extended name facility was extended with wgrib2 v3.0.2 in order to fix a problem with converting grib to netcdf with some of the newer NCEP files. When wgrib2 writes netcdf files, the -netcdf concatinates the level information to the extended name to produce the netcdf name for the field. For some of the newer NCEP forecast files, this name wasn't unique. So the extended name needs to optionally include the forecast time information. With this extension, the level information was added as another option field to the extended name.

  $ wgrib2.v3.0.2 FCST.grb -set_ext_num 1 -netcdf FCST.nc
     lost fields because the field names were not unque

  $ wgrib2.v3.0.2 FCST.grb -set_ext_num 5 -netcdf FCST.nc
     extended name include misc-info and forecast-time-info.

The final modification to the extended name, was to make the field and space character a run-time parameter by the option, -set_ext_name_chars. The default values are '.' and '_' for backwards compatibilty. The modification was needed we now are seeing level-info and misc-info with periods in them like "0.5 mb".

-match, -if, and other string matches

The extended names also applies the match inventory. So string matches will have to be rewritten. Fortunately the process is mechanical.

  -match ":RH:"         ->     -match ":RH(:|.)"
  -match ":RH:500 mb:"  ->     -match ":RH(:|.)" -match ":500 mb:"

Examples

-sh-2.05b$ ./wgrib2 chem.grb2 -var
1:0:MASSDEN
-sh-2.05b$ ./wgrib2 chem.grb2 -ext_name
1:0:MASSDEN.hi-res_ctl.Water_Vapour
-sh-2.05b$ ./wgrib2 chem.grb2 -set_ext_name 1 -ext_name
1:0:MASSDEN.hi-res_ctl.Water_Vapour