[Geotiff] Encoding of "ETRS89 / ETRS-LAEA" (i.e. EPSG:3035)

[Peter Strobl]

Frank,

Thanks for your kind words, they give us some confidence that our
specification is justifiable. For now we will leave it as is.

However, we don't get your point in saying "I still think you would be
better off the ProjectedCSTypeGeoKey to 3035." We would love to do so, but
GeoTiff specs for that Key reserve only the range [20000, 32760] for EPSG
Projection System codes. In no case an EPSG number can be used directly but
the projection has to be identified in a list and the respective number,
which is never the same as the EPSG code has to be used. We don't see the
possibility to set the ProjectedCSTypeGeoKey to any EPSG code. This would
clearly violate the GeoTiff specs and thus jeopardize any attempt of
standardisation. Ideally the GeoTiff specs would be amended, but this seems
a hard thing to do...

Regarding the PROJCS, you are right, it is tweaked. I produced one on
Windows and got the following: 'PROJCS["JRC-EGCS", ...' On Linux
this file was rejected by gdal so to process it further we kicked out all
the '"' stuff. Then it was digested silently. If written by gdal on
Linux we get 'PROJCS["JRC-EGCS", ...' which I guess is what it should look
like.

Best regards & have a nice weekend!

Peter & Pierre


-----Original Message-----
From: Frank Warmerdam [mailto:warmerdam at pobox.com] 
Sent: 05 June 2008 20:03
To: Peter Strobl
Cc: geotiff at lists.maptools.org; 'Pierre Soille'
Subject: Re: [Geotiff] Encoding of "ETRS89 / ETRS-LAEA" (i.e. EPSG:3035)

Peter Strobl wrote:
> Unfortunately most software on earth seems not capable to read such a
> GeoTiff properly (or better: the way we think it should be read).
> Even gdal holds a surprise. If processed by gdal, which usually means a
> transformation of the projection information into wkt format (right?) the
> following result is achieved:   
> 
> PROJCS[JRC-EGCS,
>        GEOGCS[ETRS89,
>               DATUM[European_Terrestrial_Reference_System_1989,
>                     SPHEROID[GRS1980,6378137,298.2572221000027,
>                              AUTHORITY[EPSG,7019]],
>                     AUTHORITY[EPSG,6258]],
>                     PRIMEM[Greenwich,8.190126211110344e-320],
>                     UNIT[degree,0.0174532925199433],
>                     AUTHORITY[EPSG,4258]],
>        PROJECTION[Lambert_Azimuthal_Equal_Area],
>        PARAMETER[latitude_of_center,52],
>        PARAMETER[longitude_of_center,10],
>        PARAMETER[false_easting,4321000],
>        PARAMETER[false_northing,3210000],
>        UNIT[metre,1,AUTHORITY[EPSG,9001]]]
> 
> This means that the PCSCitationGeoKey gets obviously lost. Instead the
> GTCitationGeokey (KeyID 1026) is quoted in the "PROJCS" field, pretending
it
> is describing the projection.
> 
> Why is this? 

Peter,

I agree that the selection of GTCitation instead of PCSCitation
for the PROJCS[] name by GDAL seems imprudent.  However, the citations
are just informational, and the definition otherwise seems fine.

BTW, did GDAL really produce a WKT report without quotes around
the various names?  I'm rather surprised to see that.

If you like, you can file a ticket against GDAL suggesting that it
use PCSCitation in preference to GTCitation for the PROJCS.  But otherwise
your file seems correct.  The fact that many packages can't read it isn't
strictly your fault, though I would not call the GDAL behavior failure, just
slightly suboptimal.

Nevertheless, I still think you would be better off the
ProjectedCSTypeGeoKey to 3035.

Best regards,
-- 
---------------------------------------+------------------------------------
--
I set the clouds in motion - turn up   | Frank Warmerdam,
warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | President OSGeo, http://osgeo.org