I have been working with David Arctur from the OGC Interoperability Institute
who has a keen interest in coordinate reference systems (CRS) and how they are
described in metadata and in the OGC specs. When David says “coordinate
reference system,” I think he primarily means “datum” but projection and other
attributes also fit in to the question.

The primary concern is how do we describe these geospatial details in a way that
can ensure transformations between systems and comparisons between data sets.
For example, a data set including a location measured by a GPS needs to map the
location exactly to a data set from an aircraft measurement from years earlier
that used a different CRS.

To be honest, I don’t entirely understand the issue, but apparently there is no
consistent way to completely and adequately describe a CRS in the various
metadata standards and specs. David would like to change that and is interested
in getting an OGC special interest group working on the issue and also figure
out a way to recommend how CRS could be described in the IPY metadata profile.

This could be an opportunity for improving how polar data sets are handled in OGC. Would anyone
be interested in working with David on this and possibly attending a meeting next Month?

David writes:

This is a multi-faceted problem involving gaps in metadata content,
user interface support, and even OGC specs (possibly gaps in GML, which needs
some exploration). I think it would be helpful to bring together someone from
NSIDC (or IPY) experienced in developing polar projections, with others from OGC
experienced with the EPSG database as well as OGC/ISO metadata specs. ... This
seems like the kind of activity which could be productively addressed through an
OGC working group including NASA, your organization, Roger Lott, and others from
industry who may be the ones who actually solve the problem. This topic could be
of interest to more than one current working group; the GML WG, Metadata WG, and
Earth Observation, Natural Resources & Environment (EONRE) WG come to mind. I’d
be glad to talk with these WG chairs to see which would be most interested in
adding this topic to their agenda.

The next OGC Technical Committee meeting will be in St Louis, March 24-27, see
http://www.opengeospatial.org/event/0803tc for meeting details. It’s still a
little early; there’s no schedule of WG sessions posted there yet, but it’s
unlikely your delegate would need to attend for more than 1-2 of the 4 days
unless they happened to want to attend additional WG sessions. I will see that
your delegate receives a complementary registration to make it easier on your
budget.

If you are interested please respond to this post or contact David directly (darctur <AT> ogcii.org).

Thanks,

-m.

This is an important area that I believe many of us have (or will in true interdisciplinary work) experienced the hard way (I had a hard time collocating 50km NWP data and 1 km AVHRR data in our visualisation systems some years ago. I am very interested in this topic, and we actual had a discussion on this topic earlier today at my institute. Do you have a link to a working area or something like that? I cannot however commit resources in this at present (linking IPY and WIS is our major concern at present, along with interfacing metadata from various systems, and data subsetting to provide various NWP data).

Øystein,

David A. is setting up a telecon Friday to help determine how CRS’s are best captured in IPY metadata. You may want to sit in. I will be traveling Friday, so I suggest you contact David directly. Bruce Raup at NSIDC is representing the issue for us.

-m.

I’ll see what I can do or if I can get someone else (among our local OGC-expertise) to participate.