The ncdump program is often the first command used to look at a new netCDF dataset. Given a netCDF file, ncdump produces on standard output a text representation of specified information in the netCDF file, either CDL or NcML.

• To look at just the "header" information (also known as the file schema or metadata) of a netCDF file, use the "-h" option, for example:

      ncdump -h co2.nc
    

More information is available by using the "-c" option, which specifies all the header information plus values of the coordinate variables.

    ncdump -h co2.nc
  

• To see the same information in NcML, a dialect of XML, use the "-x" option:

      ncdump -x -h co2.nc
    

• To request values for only particular variables, use the "-v variable-list" option.
• To see brief or full annotation of indices of variable values in C notation (0-based) or Fortran notation (1-based), use "-b C", "-b F", "-f C" or "-f F". For example, to get the 200th value of the co2 variable from co2.nc:

      ncdump -v co2 -f F co2.nc | grep "co2(200)"
    

• The Java netCDF package has another implementation of ncdump that can also output CDL or NcML. For example, here's how you could output NcML for co2.nc:

      java -cp toolsUI.jar ucar.nc2.NCdump co2.nc -ncml
  

2007 Unidata NetCDF Workshop for Developers and Data Providers > NetCDF Utilities