Hi Tod....
Something strikes me as odd: GeoX is defined as a dimension of 500.
But then declared as a variable with a shape of "pixels". To follow
CF conventions, I think you want to declare it as:
int GeoX(GeoX);
so that it is a "coordinate variable" (in Sect 5.2 of the CF
conventions - see
<http://cf-pcmdi.llnl.gov/documents/cf-conventions/1.3/ch05s02.html>).
You might also need a "coordinates" attribute for your 2D fields so
that they know to use lat_g and lon_g. It's a little hard to
interpret this automatically since two variables that apparently
contain latitude values are in the file.
I note that netcdf tools does not find any coordinate system or grids
in the file -- probably mostly due to the lack of being able to find
geo-locations.
....but I'm sure Don can help better than I.
tom
On Tue, Aug 12, 2008 at 2:29 PM, Tod Hagan <tod@xxxxxxxxxxxxxxx> wrote:
> All,
>
> Two questions:
>
> First, regarding
> http://eclipse.ncdc.noaa.gov:9090/thredds/nfdownload/iscpp-dx/2006/ISCCP.DX.0.GOE-10.2006.04.02.1800.CSU
>
> Is it possible to open this file with IDV? None of the options seem to
> work. It's a netCDF file; ncview's display of its 2D fields look
> reasonable.
>
> According to the global attributes it's formatted following "CF-1.0"
> conventions. Here's an excerpt from the start of 'ncdump -c' output:
>
> dimensions:
> pixel = 94379 ;
> GeoX = 500 ;
> GeoY = 550 ;
> variables:
> int pixel(pixel) ;
> pixel:long_name = "Pixel Number" ;
> int lat(pixel) ;
> lat:long_name = "Latitude" ;
> lat:units = "degrees_north" ;
> lat:scale_factor = 0.1f ;
> lat:add_offset = -90.f ;
> int lon(pixel) ;
> lon:long_name = "Longitude" ;
> lon:units = "degrees_east" ;
> lon:scale_factor = 0.1f ;
> int GeoX(pixel) ;
> GeoX:long_name = "X Position" ;
> GeoX:_CoordinateAxisType = "GeoX" ;
> int GeoY(pixel) ;
> GeoY:long_name = "Y Position" ;
> GeoY:_CoordinateAxisType = "GeoY" ;
> int gxpos(GeoX, GeoY) ;
> gxpos:long_name = "X Position" ;
> int gypos(GeoX, GeoY) ;
> gypos:long_name = "Y Position" ;
> int lat_g(GeoX, GeoY) ;
> lat_g:long_name = "Latitude" ;
> lat_g:units = "degrees_north" ;
> lat_g:scale_factor = 0.1f ;
> lat_g:add_offset = -90.f ;
> int lon_g(GeoX, GeoY) ;
> lon_g:long_name = "Longitude" ;
> lon_g:units = "degrees_east" ;
> lon_g:scale_factor = 0.1f ;
>
> Second, browsing THREDDS catalogs is hampered by the number of parse
> errors that are encountered, using both versions 2.3 and 2.5 of IDV. A
> couple of examples:
>
> Expanding xlink
> node:http://lead.unidata.ucar.edu:8080/thredds/topcatalog.html
> Error parsing xml:
> http://lead.unidata.ucar.edu:8080/thredds/topcatalog.html
> The declaration for the entity "HTML.Version" must
> end with '>'.
> The declaration for the entity "HTML.Version" must
> end with '>'. line:31 column:3
>
> Creating
> ui:http://eclipse.ncdc.noaa.gov:9090/thredds/catalog/iscpp-dx/catalog.html
> Error parsing xml: The declaration for the entity "HTML.Version"
> must end with '>'.
> The declaration for the entity "HTML.Version" must
> end with '>'. line:31 column:3
>
> Has anyone else seen this? Is it a problem with IDV or TDS? Apparently
> the THREDDS catalog validator that used to be available has disappeared.
>
> Thanks.
>
> Tod
>
> --
> Tod Hagan
> Information Technologist
> AIRMAP/Climate Change Research Center
> Institute for the Study of Earth, Oceans, and Space
> University of New Hampshire
> Durham, NH 03824
> Phone: 603-862-3116
>
>
> _______________________________________________
> idvusers mailing list
> idvusers@xxxxxxxxxxxxxxxx
> For list information, to unsubscribe, visit:
> http://www.unidata.ucar.edu/mailing_lists/
>
--
Tom Whittaker
University of Wisconsin-Madison
Space Science & Engineering Center (SSEC)
Cooperative Institute for Meteorological Satellite Studies (CIMSS)
1225 W. Dayton Street
Madison, WI 53706 USA
ph: +1 608 262 2759
