Hi Joe, Panoply uses NetCDF-Java underneath. The variables you see are constructed by NetCDF-Java. NetCDF-Java parses StructMetadata.
If it were my choice then I would be using the NetCDF-Java library for this job however my client wants Python. For Python, which is based on NetCDF-C, you need to parse StructMetadata.0 string to reconstruct coordinate variables by yourself. This is good to know. I spent a bucket load of time trying to find, manipulate and figure out why on earth I couldn’t grab the correct fields. I don’t feel so angry with myself now. We provide some Python examples at http://hdfeos.org/zoo for handling the metadata using netcdf4-python. For example, you can parse the StructMetadata using Python regular expression like below and construct lat/lon https://raw.githubusercontent.com/hdfeos/zoo_python/master/zoo/lpdaac/myd/MYD17A2_Gpp_1km.py and visualize it on map. http://hdfeos.org/zoo/LPDAAC/MYD17A2.A2007073.h09v08.005.2007096132046.hdf.py.png We’ll provide the complete Python LandSat 7 example later at hdfeos.org. These are super helpful thank you for sharing them. I don’t know if you guys also develop the netCDF-python driver (documentation) but it would be super helpful if this was made explicit. It is totally cryptic for people to find this out. Once again thank you for the detail above this is super helpful. Lewis
_______________________________________________ Hdf-forum is for HDF software users discussion. [email protected] http://mail.lists.hdfgroup.org/mailman/listinfo/hdf-forum_lists.hdfgroup.org Twitter: https://twitter.com/hdf5
