[Proj] Portuguese Modified Grid - Datum 73

[Roger Oberholtzer]

On Thu, 2010-02-18 at 13:24 +0000, Jose Gonçalves wrote:
> Hello
> 
> Some history about that projection: that central point was chosen at a
> mid latitude of the country and 1d00'00"E of the Lisbon Meridian
> (9d07'54.862"W). The small false easting/northing of a few hundred
> meters, which is quite unusual, was chosen in order to create
> projected coordinates similar to projected coordinates in an older
> datum (datum Lisboa).

I have finally obtained test data for conversion from WGS84 lat/longs
into D73 northings/eastings. After earlier discussions on this list
(between Feb 16 and 18, 2010), it was suggested that the appropriate
specification for proj would be:

 "+proj=tmerc "                  // Projection
 "+ellps=intl "                  // Spheroid
 "+k=1.0 "                       // Scale factor at central meridian
 "+x_0=180.598 "                 // False easting
 "+y_0=-86.98999999999999 "      // False northing
 "+lon_0=-8.131906111111112 "    // Longitude of central meridian
 "+lat_0=39.66666666666666 "     // Latitude of origin
 "+towgs84=0,0,0,0,0,0,0 "
 "+no_defs"

With test locations, I have a rather large error.

Reference point (Santa Cruz):

   Latitude: 41.6743830556 Longitude: -8.3312663889 (WGS84)
   Easting:     -16493.683 Northing:     222799.692

Using the above definition, I get this:

   Easting:     -16421.042 Northing:     222981.401

Which is 195.691 meters from the reference calculation. I have points
all around Portugal, and the all seem to be around 195 meters from the
expected location.

Either I am missing something, or the test data is in error. Or both.

Does anyone know what they would expect for the D73 northing/easting at
the WGS84 lat/long test point above?

-- 
Roger Oberholtzer

OPQ Systems / Ramböll RST

Ramböll Sverige AB
Krukmakargatan 21
P.O. Box 17009
SE-104 62 Stockholm, Sweden

Office: Int +46 10-615 60 20
Mobile: Int +46 70-815 1696