[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: 19990916: Perl, Coordinate Variables

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: Re: 19990916: Perl, Coordinate Variables
  • Date: Fri, 17 Sep 1999 09:54:43 -0600
>To: address@hidden
>From: "Ze'ev Gedalof" <address@hidden>
>Subject: Perl, Coordinate Variables
>Organization: University of Victoria Tree-Ring Laboratory
>Keywords: 199909162334.RAA27466 netCDF Perl coordinate variable conventions

Hi Ze'ev,

> I am a new NetCDF user (using the Perl extension), and am a little
> perplexed by the definition of coordinate variables given in the version 2
> and 3 guides.  I am working with gridded climate data (SLP and SST) which I
> would like to coordinate by lat, long, and time.  I think that I have
> managed to create these variables as coordinate variables (ncdump -c lists
> them as such), but I can't figure out how to use them now.  Is there a more
> extensive definition somewhere of what coordinate variables are and how to
> use them?  

In terms of the netCDF library, coordinate variables are merely a
convention that applications are free to use or ignore.  Some
applications use coordinate variables in accessing and subsetting
data; others require that you use indices, so you need to write a
layer to translate coordinate variable values into indices.  For
climate data, probably the most elaborate conventions for coordinate
variables are documented in "GDT netCDF conventions for climate data,
version 1.3" available from 

  http://www-pcmdi.llnl.gov/drach/GDT_convention.html

There has been extensive discussion on the netcdfgroup mailing list of
how to strengthen the coordinate variable conventions to make them
more useful, and of whether multidimensional coordinate variables
should be supported.  To browse some of these discussions and the
proposals that resulted from them, see

 http://www.unidata.ucar.edu/packages/netcdf/coords/proposals.html

and

 http://www.unidata.ucar.edu/packages/netcdf/coords/

While there were several good proposals, no consensus was achieved by
the discussions, so you are free to pick your own extended coordinate
variable conventions.  Unfortunately, none of these conventions are
supported by all netCDF software, though there is good support for use
of coordinates in the FAN package and a few others, accessible from
the "Software for Manipulating or Displaying NetCDF Data" document at

 http://www.unidata.ucar.edu/packages/netcdf/software.html

--Russ

_____________________________________________________________________

Russ Rew                                         UCAR Unidata Program
address@hidden                     http://www.unidata.ucar.edu