NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
I have a netCDF variable that has _FillValue attribute set to NaN (undefined value from, for example, dividing a number by 0). When I tried to set it to a numeric value using ncatted, it changes all zeroes in data to the fill value as well in the process. Is this a bug is there a way to retain the zero values while resetting the NaN fillvalue? The following is the dump of a sample file I used. Both 0 and NaN in data become fill value (-999.) while I was expecting the zeros to remain. The command I used to set fillvalue is: ncatted -a "_FillValue,A,o,d,-999." nan.nc ******** Start of nan.nc dump ******** netcdf nan { dimensions: latitude = 3 ; longitude = 3 ; variables: double A(latitude, longitude) ; A:_FillValue = NaN ; A:long_name = "A Variable" ; A:units = "mm" ; double longitude(longitude) ; longitude:long_name = "longitude" ; longitude:standard_name = "longitude" ; longitude:units = "degrees_east" ; double latitude(latitude) ; latitude:long_name = "latitude" ; latitude:standard_name = "latitude" ; latitude:units = "degrees_north" ; // global attributes: :Conventions = "CF-1.4" ; data: A = 0.02, 0, NaN, 0.3, 0.4, 0.5, 0, 0.6, 0.8 ; longitude = -80, -79, -78 ; latitude = 10, 11, 12 ; } ******** End of nan.nc dump ******** Thanks. Jianfu Pan
netcdfgroup
archives: