[R-sig-Geo] spTransform: wrong results??

[Roger Bivand]

On Sun, 12 Jul 2009, Agustin Lobo wrote:

> Here you have some results with
> different software for the transformation
> of one single  UTM 31N ED50 coordinate to
> lon,lat ED50 and lon,lat WGS84.
> My understanding of this problem seems to
> be decreasing as I do more testing.
>
> QGIS UTM 31N ED50

This isn't helpful, I'm afraid. You are starting from an unknown, not a 
known. It isn't imposible that QGIS doesn't give a +towgs84 for so-called 
ED50, because ED50 is *not* one datum but many.

Start with an unequivocally known point in a known datum, such as a GPS 
reading in geographical coordinates and WGS84. Work from there, and it 
should get clearer.

Datum definitions are not easy at all, because organisations typically 
always worked within a single datum, and never needed to transform (until 
oil was found across two different jurisdictions using different datum 
standards, hence EPSG).

Roger

>
> POINT(449919.586269 4631874.109627)
>
> TRANSFORMS TO LONGITUDE,LATITUDE:
>
> QGIS ED50
> POINT(2.396864 41.836439)
> QGIS WGS84
> POINT(2.396864 41.836439)
>
> GlobalMapper  ED50  2.39686371° E	41.83643863° N
> GlobalMapper  WGS84 2.39572870° E	41.83535625° N
>
> TNT WGS84
> POINT(2.396864 41.836439)
> TNT ED50
> POINT(2.396864 41.836439)
>
> rgdal ED50 no towgs84 field specified
> WP1	2.39686370617169	41.8364386271064
>
> rgdal WGS84 no towgs84 field specified
> WP1	2.39686370617169	41.8364386271064
>
> rgdal WGS84 with towgs84=-87, -98,-121
> WP1	2.39690750837097	41.8361433350555
>
>
>
> Agus
>
>
>

-- 
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Helleveien 30, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43
e-mail: Roger.Bivand at nhh.no