Please look up "NaN, in netCDF files" in the Ferret Users Guide Index, ../../documentation/users-guide/full-index/index.htmlx
Here is the information:
If there are values of NaN in the file, then NaN must be listed as either the "missing_value" OR "_FillValue" attribute and then NaN is the missing value. Or, the user may specify SET VARIABLE/BAD=NAN (case insensitive) to designate the Fortran value NaN (not a number) as the bad value flag for a given variable in a netCDF dataset.
Ansley
Ryo Furue wrote:
Hi Ferret users, I'm wondering how NaNs in a netCDF file are handled. In the Ferret mailing list archive, I found a comment that NaNs in a netCDF file are OK. But, I get the attached plot from yes? list/k=1 gvel_x_0 [ . . . The field contains only "nan" and "0.0000" . . . ] yes? fill/k=1/lev=(-10,10,1) gvel_x_0 I expected NaN is treated as a missing_value. (The variable gvel_x_0 doesn't have a missing_value or _FillValue attribute, by the way.) The Ferret I use is FERRET v6.2 Linux(g77) 2.4.21-32 - 05/19/09 7-Aug-09 16:20 Regards, Ryo
