Matt-
Matthew Foster wrote:
I think a big part of the problem with AWIPS NetCDF is that AWIPS
predates the CF convention by quite a bit (it looks like CF 1.0 came out
in 2003) and even predates COARDS (1995?). There was probably some
other convention around when AWIPS (WFO-Advanced) was under development,
but I don't know what that would be.
Someone can correct me, but I believe that the AWIPS convention grew out
of the NetCDF Users Working Group (NUWG) convention
(http://www.unidata.ucar.edu/software/netcdf/NUWG/) and as Matt points
out, predated the CF conventions. However, FSL (now GSD), modified that
to work in their environment (AWIPS) and I think the merging of the
various vertical dimensions was one of the modifications.
I could be mistaken in my timeline. Feel free to correct me if I have
it wrong.
I think the take away message from this discussion should be to use a
convention when writing your netCDF files and make sure that convention
is documented so that data consumers (e.g. the IDV) can write readers
for it. Unidata is actively involved in the development of the CF
conventions and we suggest using that as a starting point. As Tom
points out, we are in the process of working with SSEC and NCDC to
develop a CF standard for storing satellite imagery. There is also a
proposal for storing point observations.
The IDV/netCDF-Java teams have done a lot of work to support various
conventions (and non-conventions) in these packages. We also are
willing to work with data providers to help them conform to existing
conventions.
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
*************************************************************