Defining a projection for SetProjection()

[Frank Warmerdam]

David Stajan wrote:
> PROJCS["AGD_1966_AMG_Zone_55",
> 	GEOGCS["GCS_Australian_1966",
> 		DATUM["D_Australian_1966",
> 			SPHEROID["Australian",6378160,298.25]],
> 		PRIMEM["Greenwich",0],
> 		UNIT["Degree",0.017453292519943295]],
> 	PROJECTION["Transverse_Mercator"],
> 	PARAMETER["False_Easting",500000],
> 	PARAMETER["False_Northing",10000000],
> 	PARAMETER["Central_Meridian",147],
> 	PARAMETER["Scale_Factor",0.9996],
> 	PARAMETER["Latitude_Of_Origin",0],
> 	UNIT["Meter",1]]
> 
> My best attempt to define this using the proj.4 syntax is:
> 
> PROJECTION
> "proj=utm"
> "zone=55"
> "units=m"
> "ellps=aust_SA"
> "lat_0=0"
> "lon_0=147"
> "x_0=500000"
> "y_0=10000000"
> "k_0=0.9996"
> "a=6378160"
> "rf=298.25"
> "pm=greenwich"
> END

David,

I would suggest "+proj=utm +zone=55 +south +ellps=aust_SA".

The +pm=greenwich setting is unnecessary since this is the default.

It is imprudent to specify settings like lat_0, lon_0, x_0, y_0
and +k_0 that are implied by proj=utm and a zone.

It is imprudent to doubly define the earth model - using
both +ellps and +a, +rf.

I note you missed the important +south modifier on your utm
projection.

Note that the projection as defined does not know how to switch
to the WGS84 datum (if that is an objective).  PROJ 4.5.0 and
earlier will do a change of ellipsoid which may introduce
unexpected shifts.  PROJ 4.6.0 will assume no datum shift is
needed.

Finally, I hate the old syntax for PROJECTION blocks of putting
each parameter on it's own line.  Furthermore there are some issues
with parsing it in esoteric circumstanes.  I'd encourage:

PROJECTION
   "+proj=utm +zone=55 +south +ellps=aust_SA"
END

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