[Proj] Conversion from WGS84 to UTM31

[Oscar van Vlijmen]

> From: Fabien Brachere <fabien.brachere-eikonex.net>
> Subject: [Proj] Conversion from WGS84 to UTM31

> I'm trying to convert data from my GPS to UTM31 coordinates and I use
> something like that:
...
> 0.69315 42.785283333
...
> transdat ( http://www.killetsoft.de/transdat.htm ) the conversion result
> is:
> 311405.583 4739753.703 which is exactly where I was with my GPS

I calculated the conversion with my stuff; it's up to you or others to get
it into cs2cs.

First a datumshift from WGS to International 1924.
Bundesamt für Kartographie und Geodäsie has good parameters.
Look for Spain - EST99.
http://crs.bkg.bund.de/crseu/crs/eu-national.php
There are parameters available for the region in question, going from
ES_ED50 to ETRS89. The ED50 datum is on an ellipsoid equal to International
1924 and ETRS89 is nearly the same as WGS84.
I understand that you want to go in the opposite direction, from WGS84 to
ED50, so the mentioned parameters should be negated.
The transformation method is ISO (like PROJ.4 does), it est, an approximated
position vector Helmert transform. So the rotation angles don't have to be
changed after the first negation of all parameters.

The parameters from ES_ED50 -> ETRS89 are:
dx=-131.0; dy=-100.3; dz=-163.4;
rx=-1.244; ry=-0.020; rz=-1.144; ds=9.39;
(units: m, m, m, ", ", ", ppm)

With your:
lat=42.785283333; lon=0.69315; ht=0;
I arrive at:
lat=42.786375; lon=0.6943564; (deg)

The (not rounded) coordinates fed to a UTM projector, set to the correct
ellipsoid (ED50 or International 1924) and zone (31), gives the coordinates:
x=311408.377; y=4739755.778; (m)
The differences of 2.8 m and 2.1 m can be attributed to the use of a
slightly different set of parameters. The "CRS-EU" parameters are meant for
applications needing an accuracy of ca. 1.5 m.

The geocentric shift you mentioned:
dx=-84; dy=-97; dz=-117;
gives indeed:
x=311405.5829; y=4739753.7031; (m)
but as mr. Mugnier said, there are better parameters suited for the region
in question. See above!