[OSRS-PROJ] Prime Meridian Support in PROJ.4

[Frank Warmerdam]

Luis Mota wrote:
> Just sth I noticed:
> 
> You cite the srtext definition in the spatial_ref_sys table:
> 
>> PROJCS["Portuguese_National_Grid",GEOGCS["GCS_Lisbon_Lisbon",DATUM["D_Lisbon",SPHEROID["International_1924",6378388,297]],PRIMEM["Lisbon",-9.131906111111112],UNIT["Degree",0.017453292519943295]],PROJECTION["Transverse_Mercator"],PARAMETER["False_Easting",200000],PARAMETER["False_Northing",300000],PARAMETER["Central_Meridian",1],PARAMETER["Scale_Factor",1],PARAMETER["Latitude_Of_Origin",39.66666666666666],UNIT["Meter",1]] 
>>
> 
> 
> However, when I looked at the definition found in 
> http://gdal.velocet.ca/cgi-bin/csv2html/TABLE=p_meridian/CODE=8902/, the 
> value for the prime meridian is -9.0754862. Which one is right? I don't 
> have the EPSG data here to check it out (I have no application to read 
> an Access DB... )
> 

Louis,

The definition for the Lisbon Prime Meridian in EPSG is declared to be
in sexagesimal degrees (DD.MMSSsss) so the value of -9.0754862 should
be interpreted as -9d7m54.862s or -9.0 - 7.0/60 - 54.862/3600 which
works out to -9.13190611111 which closely matches the WKT definition which
is in decimal degrees.  Lots of gotchas in extracting data from the EPSG
tables!

Note that I think Pauls point in quoting the WKT was to show that the
Central Meridian parameter to the Transverse Mercator projection (which was 1)
is likely relative to the Lisbon prime meridian, not to Greenwich, otherwise
the central meridian would be a long ways away from Portugal.  An unlikey
choice for a national coordinate system.

So based on that one example, I would say that the longitude parameters
for projections are usually expressed relative to the prime meridian of
the GCS, not greenwich.  Based on that we could say that the pj_inv() and
pj_fwd() function would not need to be changed as they operate in local
GCS space and are "unaware" of their prime meridian.  Furthermore, all
longitude parameters passed to PROJ.4 would be interpreted as relative to
the prime meridian in effect.  The +pm parameter would only be needed when
desiring to convert between different geographic coordinate systems.

 > What idea is this to publish EPSG data in Access format?

The use of Access makes it easy to implement a nice GUI front end, and to
more clearly express the connections in the data model.   I personally don't
like it because it depends on proprietary software, and puts non-Windows
users a step away from the original source data which is why I provide
a program for dumping the tables to text files for the use of libgeotiff
and GDAL.  However, there are good and practical (if arguable not good
enough) reasons why EPSG did as they did.

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    | Geospatial Programmer for Rent


----------------------------------------
PROJ.4 Discussion List
See http://www.remotesensing.org/proj for subscription, unsubscription
and other information.