[Proj] Finnish KKJ in proj 4.4.9

[Martin Vermeer]

On Tue, 2005-12-13 at 09:15 +0100, Roger Oberholtzer wrote:

...

> I have a Windows program that claims to do the correct transformations:
> 
> 	Program name EurefMuunnos
> 
> 	Author and date Sakari Haljala, NLS 4.9.2003
> 
> 	Copyright National Land Survey of Finland, Helsinki 2003
> 
> This is what we are trying to match. These are some sample data points
> they provided:

I try the following:

./cs2cs +proj=latlong +ellps=WGS84 +to +init=epsg:2392

> EUREF-FIN- geographic coordinates in format: aamm.mmm = degrees+minutes
> (latitude, longitude):
> 
> 6011.048456 2456.961511

24d56.9615'E 60d11.0485'N
2552856.87      6675107.38 -24.57 (see below... matches within a metre)

> 6011.917399 2456.117327

24d56.117327'E 60d11.917399
2552053.36      6676709.62 -24.57 (again a match... the -24m is
approximately minus the geoid height, as the KKJ datum is approximately
defined on a non-geocentric ellipsoid matching the geoid surface over
Finland)

> 6012.283731 2455.387040

(Try for yourself)

> 6013.139141 2457.086345 
> 
> 6010.570229 2455.930532 
> 
> 6012.467191 2456.127325
> 
> 6013.045471 2458.825683 
> 
> KKJ- plane coordinates in Gauss-Krüger zone no. 2 (northing, easting):
> 
> 6675107.237 2552856.284 
> 
> 6676709.563 2552052.761
> 
> 6677380.277 2551368.203 
> 
> 6678990.902 2552915.797 
> 
> 6674205.682 2551915.400 
> 
> 6677730.555 2552047.542 
> 
> 6678840.480 2554525.154 
> 
> How would I set that up in cs2cs? The input is EUREF-FIN, not WGS84. In
> fact, my real data (not these test points) will be WGS84 from a GPS
> receiver. But I would like to verify these points as well.

How to do it from inside C, I don't know, but you should be getting the
same results. Sakari's program is OK (if not, thousands of customers
here would have screamed :-)

BTW EUREF-FIN is in practice the same as WGS84 on this accuracy level.
The differences are caused by

1) The Defense Dept's WGS84 and the IGS's ITRFxx (xx = year number)
differ nowadays on the few-centrimetre level, due to being different
datums based on different observational material.
2) WGS84/ITRFxx are not only geocentric, but defined to represent a
system that does not include plate tectonics. EUREF-FIN, a Finnish
realization of ETRF89, is rigidly connected to the Eurasian plate and
coincided with ITRF89 at 1989.0, _but not anymore_ due to the plate
motion.

But as said, if a few dm will do, forget about this. 

We could probably improve the precision of the KKJ transformation from
the present +/- 1 m (and up to several metres even on Åland) down to +/-
1 dm, by calculating and including a datum grid representing the
regional deformations of KKJ. That would be quite a job though.

- Martin

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://lists.osgeo.org/pipermail/proj/attachments/20051213/032af4ea/attachment.sig>