⇐ ⇒

[CF-metadata] CF-1 map projections export/import from GDAL - update & request for help with Polar stereographic

From: Patrick Sunter <patdevelop>
Date: Fri, 21 Oct 2011 18:53:23 +1100

Hi there,

I posted to this list a while back in the context that a few of us are
working on better import/export capability of projected raster data
into NetCDF CF-1 from the GDAL open source library - notes on our work
are at http://trac.osgeo.org/gdal/wiki/NetCDF_exportUpdateChanges .
These changes should make it through to the trunk of GDAL pretty soon.

So as well as letting people on this list know in case this is of
general interest - we'd also really appreciate any help from people
who are up with, or have examples of, some of the less common
projections that CF-1 allows you to encode. Leaving aside the current
discussions of possibilities of WKT and Datum support - we aimed to
get as good a coverage of current CF-1.5 projections and their
attributes as possible.

This Wiki page gives a summary of status of where the projection support is at:
http://trac.osgeo.org/gdal/wiki/NetCDF_ProjectionTestingStatus

In particular, the Polar Stereographic CF-1 attributes have been a bit
hard to 'map' to the CF-1 equivalents:
If you look at http://www.remotesensing.org/geotiff/proj_list/polar_stereographic.html,
there is a 'latitude of natural origin', but the CF-1.5 version
specified that 'latitude_of_projection_origin' must be either +90 or
-90 (to encode the pole). So I'm wondering if the CF-1
'standard_parallel' for this projection is the equivalent here?

There were also issues with a few of the 'variants' of CF-1
projections, e.g. those that say you can used either a
standard_parallel or scale_factor - listed at the bottom of the Wiki
page linked above.

In general my testing process has involved seeing if the exported
projected rasters passes both the CF-checker, and also can be loaded
and projected in the correct location in IDV, which uses NetCDF Java's
on the backend and thus their implementation of CF-1 projection
handling.

I have attached our main testing sample to that Wiki page in both it's
original UTM SRS, and also a Tarball of files exported in all the CF-1
projections so far.

thanks, Patrick.

BTW: I did look at the FGDC "Content Standard for Digital Geospatial
Metadata" document referenced on the CF-1.5 App F page, but apart from
specifying the attribute names for each projection, it didn't seem to
help much with relating them to other map projection definition
systems.

====
Patrick Sunter
Victorian Partnership for Advanced Computing
Senior Computational Scientist
Received on Fri Oct 21 2011 - 01:53:23 BST

This archive was generated by hypermail 2.3.0 : Tue Sep 13 2022 - 23:02:41 BST

⇐ ⇒