[Proj] How to convert a sphere to ellipsoid with correct datum?
j.l.h.hartmann at uva.nl
Mon Sep 6 08:15:28 EST 2010
On 09/05/10 07:46, Hermann Peifer wrote:
> On 04/09/2010 14:15, Jan Hartmann wrote:
>> Thanks Hermann, that was really great help. To explain why I keep
>> coming back on the issue: I am georeferencing the oldest cadastral
>> maps of the Netherland from 1830. There was no nation-wide
>> triangulation then,..
> You're welcome.
> You wrote in an earlier post that you observed a N-S and a (smaller,
> but noticeable) E-W shift in you transformations. As we all agree (I
> hope), the E-W shift must be unrelated to any change in ellipsoid. In
> this case, I would have immediately checked if the tool's (cs2cs,
> proj) Bonne projection is the culprit. Maybe some rare variant of
> Bonne was used back then in 1830, or a weird projection parameter,
> which is not supported in
> <http://trac.osgeo.org/proj/browser/trunk/proj/src/PJ_bonne.c>, or
> something like that.
> I would have taken a textbook (better: two) and implemented the Dutch
> Bonne in my favourite scripting language in order to verify if this
> possible source of errors can be excluded. It shouldn't take more than
> one sunny Sunday afternoon to do this check.
OK, that's another good suggestion I never thought of. The complete math
behind the projection used in 1850 is described in the "Meetkunstige
Beschrijving van het Koningrijk der Nederlanden" (Geometric Descriptrion
of the Kingdom of the Netherlands), which is even online, I see, now I
am replying this email:
I have a few handbooks on my desk, including Snyder's
I'm not sure I can tackle this on one Sunday afternoon, sunny or rainy,
though. Any bidders from the list :-)? A translation from Dutch to
English would not be beyond my powers.
More information about the Proj