Ooops.
Don Murray wrote:
Briefly looking at the data, it looks like it might adhere to the
reduced horizontal grid structure for CF:
http://www.cgd.ucar.edu/cms/eaton/cf-metadata/CF-1.0.html#grid_ex3
Unfortunately, this part of the convention does is not handleds
in CfView. If this is the type of data you have, we could try
to add support for it in VisAD as time permits.
Actually, while the CfView does not support the compress attribute,
id does support the "coordinate". So, if the structure is modified
to support the CF conventions as follows:
netcdf anal2 {
dimensions:
Time = 1 ;
Latitude_Longitude_ndx = 4402 ;
variables:
float Time(Time) ;
Time:units = "s since 1970-01-01 00:00:00.000 UTC" ;
float Latitude(Latitude_Longitude_ndx) ;
Latitude:units = "deg" ;
float Longitude(Latitude_Longitude_ndx) ;
Longitude:units = "deg" ;
int Latitude_Longitude_ndx(Latitude_Longitude_ndx) ;
Latitude_Longitude_ndx:units = "" ;
double anal0(Time, Latitude_Longitude_ndx) ;
anal0:_FillValue = 9.96920996838687e+036 ;
anal0:units = "hectoPascal" ;
anal0:coordinates = "Longitude Latitude" ;
// global attributes:
:Conventions = "CF-1.0" ;
}
The netCDF adapter handles this. You still have the problem that
the domain is a 1D manifold, but at least VisAD will create
a Field of type:
(Time -> ((Latitude, Longitude) -> anal0))
Don
*************************************************************
Don Murray UCAR Unidata Program
dmurray@xxxxxxxxxxxxxxxx P.O. Box 3000
(303) 497-8628 Boulder, CO 80307
http://www.unidata.ucar.edu/staff/donm
"There's someone in my head, but it's not me" Roger Waters
*************************************************************
