[postgis-users] Weird behavior transforming from srid 4269 to 2163

[Paragon Corporation]

Disregard this last question if its not a trivial one.  We just decided not
to use this projection for this project since it renders differently under
proj 4.5 than 4.6 and found one that seems to work for us that will render
the same under both.

Our main concern was not so much that we keep data in 2163 projection, but
that a lot of the tools we use are compiled with the older proj library and
we didn't want to change that at this moment. So this little difference in
behavior would cause potential problems when reprojecting non-postgis data.

Thanks,
Regina 

-----Original Message-----
From: postgis-users-bounces at postgis.refractions.net
[mailto:postgis-users-bounces at postgis.refractions.net] On Behalf Of Paragon
Corporation
Sent: Tuesday, January 29, 2008 9:12 AM
To: 'PostGIS Users Discussion'
Subject: RE: [postgis-users] Weird behavior transforming from srid 4269 to
2163

Anyway to make the new bahave like the old? 

-----Original Message-----
From: postgis-users-bounces at postgis.refractions.net
[mailto:postgis-users-bounces at postgis.refractions.net] On Behalf Of Frank
Warmerdam
Sent: Tuesday, January 29, 2008 8:40 AM
To: PostGIS Users Discussion
Subject: Re: [postgis-users] Weird behavior transforming from srid 4269 to
2163

Paragon Corporation wrote:
> I'm having some strange discrepancies between two postgis servers in 
> how they are transforming the same data from 4269 to 2163.  I'm hoping 
> someone could shade some light on what could be the problem.
> 
> The problematic server is running Red Hat 4.1.2-14 and postgis version 
> "POSTGIS="1.3.2" GEOS="2.2.3-CAPI-1.1.1" PROJ="Rel. 4.6.0, 21 Dec 2007"
> USE_STATS"
> 
> The one that I think is right is running Windows 2003 server 
> "POSTGIS="1.3.1" GEOS="3.0.0rc4-CAPI-1.3.3" PROJ="Rel. 4.5.0, 22 Oct 2006"
> USE_STATS"
> 
> So the bad one is running a newer version of the PROJ library, but I 
> wasn't the one who compiled and installed, so not sure if there was a 
> step
missing.

Regina,

I suspect your EPSG:2163 is evaluating to:

  +proj=laea +lat_0=45 +lon_0=-100 +x_0=0 +y_0=0 +a=6370997 +b=6370997
+units=m
+no_defs

In PROJ 4.5 the transformation from EPSG:4269 (NAD83) to EPSG:2163 would be
accomplished by transforming NAD83 coordinates into geocentric coordinates
and then back into lat/long on the EPSG:2163 ellipsoid resulting in a
substantial shift in latitude.

However, after a half a decade of complaints about this "default ellipsoid
conversion" which is usually the wrong thing to do, PROJ 4.6 finally changed
the behavior to doing no datum shift if either the source or the destination
lacks apparent datum shift information.  That means, in this case, that the
input coordinates are being used as if they were already on the spherical
ellipsoid and no datum/spheroid shift operation is done.

I really think this is the appropriate change in your case, but it is
certainly a change in behavior.  It may also not agree with data previously
converted to EPSG:2163.

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

_______________________________________________
postgis-users mailing list
postgis-users at postgis.refractions.net
http://postgis.refractions.net/mailman/listinfo/postgis-users



_______________________________________________
postgis-users mailing list
postgis-users at postgis.refractions.net
http://postgis.refractions.net/mailman/listinfo/postgis-users