[Proj] Difference between same projections types

Martin Desruisseaux martin.desruisseaux at geomatys.fr
Tue Feb 14 10:47:37 PST 2012


Hello Mustafa

Le 14/02/12 18:50, Mustafa646 a écrit :
> 1.  (+proj=longlat +ellps=WGS84 +datum=WGS84 +no_defs )
This syntax is Proj.4 specific, and may not be understood by other software 
(unless those software provide a Proj.4 compatibility layer). Note however that 
since many projects use Proj.4 for their internal map projections, they 
indirectly use that syntax. For example the PostGIS extension to the PostgreSQL 
database use Proj.4 internally, so its "spatial_ref_sys" table expect that 
syntax. This is non-standard however.


> 2. GEOGCS["WGS 84",DATUM["WGS_1984",SPHEROID["WGS
> 84",6378137,298.257223563,AUTHORITY["EPSG","7030"]],TOWGS84[0,0,0,0,0,0,0],AUTHORITY["EPSG","6326"]],PRIMEM["Greenwich",0,AUTHORITY["EPSG","8901"]],UNIT["degree",0.0174532925199433,AUTHORITY["EPSG","9108"]],AUTHORITY["EPSG","4326"]]
This syntax, named "Well Known Text" (WKT), has been standardized by the OGC 
01-009 specification and is (in theory) an international standard. The syntax is 
described here:

     http://www.geoapi.org/3.0/javadoc/org/opengis/referencing/doc-files/WKT.html

Alternatively you can go straight to the source: 
http://www.opengeospatial.org/standards/ct

This syntax is used in ".prj" file that sometime come with "TIFF + World File" 
images for example. In theory, this syntax should be understood by many 
software. In practice, different software understand it in different ways, 
sometime because the standard had some room for interpretation, and sometime 
because not every developers have implemented the standard correctly.

As a side note, the above WKT is note quite right. It is missing the axis order 
declaration - so it is not a real definition for "EPSG:4326". It is rather a 
definition for "CRS:84" (defined in WMS specification).

By the way, if there is any PostGIS developer reading this list... last time I 
looked, the WKT defined in the PostGIS "spatial_ref_sys" table for France 
Lambert projections declare a wrong "prime meridian" value (more specifically, 
the units are wrong. It should be declared in gradiant - not degrees - because 
the prime meridian shall inherits the unit of the enclosing GEOGCS. While some 
other parts of the WKT specification are vague, this one is clear). I filled a 
bug report to PostGIS years ago, but it doesn't seem to have been addressed...


> 3. EPSG: 4326
This one is only a reference to a definition which shall be provided elsewhere. 
This is often a primary key in a database. Understanding this value require that 
the software had a connection to an appropriate database. In this particular 
case, the database is (or is derived from) the EPSG database: http://www.epsg.org

When using software that can understand such codes, providing the EPSG codes is 
often both the easiest and the most accurate way to define the CRS (assuming 
that the database used by the software doesn't contain errors).

Regards,

     Martin




More information about the Proj mailing list