RE: Handling multiple projections in CF

• To: "Rich Signell" <rsignell@xxxxxxxx>, <galeon@xxxxxxxxxxxxxxxx>
• Subject: RE: Handling multiple projections in CF
• From: "Woolf, A (Andrew)" <A.Woolf@xxxxxxxx>
• Date: Tue, 5 Dec 2006 07:48:14 -0000

I think from an OGC perspective, there's not a lot to pick between them
- they all offer slightly different CF-mechanisms for specifying the
actual gridpoint positions under different CRS. Any OGC geometry is
associated with (and specified in terms of) just one CRS - through the
gml:srsName attribute of gml:AbstractGeometryType. Transformations
between CRS may in general be handled through algorithms (e.g. ref
proj4) using the geodetic parameters of the CRS.

However the important point is that long-standing netCDF tradition (and
HDF, and others) specify in one way or another actual gridpoint
locations through additional variables providing separately ordinate
values over axes of the grid on which they vary; in addition, an
efficiency may be realised for cases where the coordinate system axes
align with the grid axes - allowing one-d coordinate variables in
CF-speak. OGC's implementation of ISO 19123 (the ISO 'grids' spec)
provides only gml:Grid (non geo-referenced Grid) and gml:RectifiedGrid
(regularly spaced grid), but not gml:ReferenceableGrid (irregular
spacing). A discussion at the forthcoming OGC TC (joint coverages/GML
working group) in San Diego will look at OGC document 06-160 - a change
request to incorporate CV_ReferenceableGrid (the irregular grid case).
That document proposes a mechanism in line with netCDF convention, and
any GALEON members present are encouraged to attend.

-----Original Message-----

From: owner-galeon@xxxxxxxxxxxxxxxx [mailto:owner-galeon@xxxxxxxxxxxxxxxx] On Behalf Of Rich Signell

Sent: 05 December 2006 03:09
To: galeon@xxxxxxxxxxxxxxxx
Subject: Fwd: Handling multiple projections in CF

Galeon Folk,

There has been a lot of discussion lately on the CF group about how to handle multiple projections in CF.

Basically it's been boiled down to three different possible approaches, summarized on this wiki page:

http://cf-pcmdi.llnl.gov/trac/wiki/MultipleProjections

Since you folks have experience with both CF and the OGC conventions, do you see any good reason to pick one over the other (easier to implement or more similar syntax to OGC conventions, etc)?

Thanks,
Rich

• References:
  • Fwd: Handling multiple projections in CF
    • From: Rich Signell