[Proj] Belge 1972 / Belgian Lambert 72 (31370) - towgs84parameters

Thu Jan 21 03:52:03 PST 2010

Jan Hartmann wrote: 

> No, if QGIS uses PROJ, this is just an error. 

Okay, you may be right that QGIS does not use the file gcs.override.csv.

But I see that the nad/epsg file of PROJ.4 contains the same erroneous 
+towgs84 parameters for Belge 1972 as the gcs.override.csv.  
(At least PROJ version 4.6.1). 

> PROJ and EPSG use opposite rotational formulas, and PROJ uses degrees,
EPSG radians. 

I don't agree in the general case.  PROJ uses the Position Vector 
Transform, while EPSG is neutral on the rotation sign convention: 
they use the same sign convention as the original source. 
And PROJ uses arc seconds for rotations, while EPSG is neutral 
on the angle unit: they use the same angle unit as the original source
(usually arc seconds, but sometimes microradians or radians). 
     For the EPSG transforms you quote, EPSG use arc-seconds
for the rotations, but either the Position Vector Transform or the
Coordinate Frame Rotation depending on whether they got the 
transform from Eurogeographics or directly from Belgium. 

Regards,

--
Mikael Rittri
Carmenta AB
SWEDEN
www.carmenta.com 

________________________________

From: Jan Hartmann [mailto:j.l.h.hartmann at uva.nl] 
Sent: Wednesday, January 20, 2010 3:23 PM
To: PROJ.4 and general Projections Discussions
Cc: Mikael Rittri
Subject: Re: [Proj] Belge 1972 / Belgian Lambert 72 (31370) -
towgs84parameters

No, if QGIS uses PROJ, this is just an error. PROJ and EPSG  use
opposite rotational formulas, and PROJ uses degrees,  EPSG radians.
Could you report this bug to the the QGIS team?

Jan 

On 20-1-2010 14:04, Mikael Rittri wrote: 

	Hello Thibaut and Jan,
	The towgs84 parameters you say that QGIS is using, seems to come
from the file 

http://svn.osgeo.org/metacrs/geotiff/trunk/libgeotiff/csv/gcs.override.c
sv

	that is used by libgeotiff and GDAL.  The datum shifts in this
file have been 
	constructed manually.  For Belge 72 they seem to be wrong, as
you noted:

	      The signs of DX, DY and DZ are wrong,
	      The signs of RX, RY and RZ are correct, provided that the
COORD_OP_METHOD_CODE is changed to 9607,
	      The DS is 1.0000012747, but should be -1.2747 (expressed
in unity instead of parts per million, and wrong sign). 

	(This is assuming that the datum shift EPSG:15928 is correct.)

	I think the parameters in gcs.override.csv came from an
information source
	that gave the datum shift in the direction from WGS84 to Belge
72.  

	Best regards,

	--
	Mikael Rittri
	Carmenta AB
	SWEDEN
	www.carmenta.com 

________________________________

	From: proj-bounces at lists.maptools.org
[mailto:proj-bounces at lists.maptools.org] On Behalf Of Jan Hartmann
	Sent: Wednesday, January 20, 2010 11:29 AM
	To: PROJ.4 and general Projections Discussions
	Subject: Re: [Proj] Belge 1972 / Belgian Lambert 72 (31370) -
towgs84parameters

	Hi Thibaut,

	the most recent version of the EPSG database, 7.4,
(www.epsg.org) has two definitions for the datum shift from Belge
Lambert to WGS84 (they call it a coordinate transformation). I'l give
the parameters as a PROJ string, with rotations converted from radians
to degrees and sign-converted

	nr 162 and 164 (accurate to a meter) :  +proj=lcc
+lat_1=51.16666723333333 +lat_2=49.8333339 +lat_0=90
+lon_0=4.367486666666666 +x_0=150000.013 +y_0=5400088.438 +ellps=intl
+towgs84=-99.059,53.322,-112.486,0.419,-0.830,1.885,-1 +units=m +no_defs

	nr 163 and 166: (accurate to 20 cm): +proj=lcc
+lat_1=51.16666723333333 +lat_2=49.8333339 +lat_0=90
+lon_0=4.367486666666666 +x_0=150000.013 +y_0=5400088.438 +ellps=intl
+towgs84=-106.8686,52.2978,-103.7329,-0.3366,0.457,-1.8422,-1.2747
+units=m +no_defs 

	The second one if the official one from the Belgian National
Geographic Institute (http://www.ngi.be/FR/FR4-4.shtm). Both should give
approximately the same results. I have used epsg:31370, which uses the
second definition without errors. The errors in your picture, about 100
meters, look as if no datum transformation has been applied at all.

	You could test this by transforming your data points manually
with cs2cs for both towgs84 strings, and compare the results with the
position of the WGS84 coordinates in QGIS  

	Jan

	On 20-1-2010 10:18, Thibaut Gheysen wrote: 

		Hi,

		I have a problem in QGIS 1.4.0 when I reproject a gpx
file into Belgian lambert 72 coordinate system : the waypoints are not
correctly placed on the map (see this image :
http://www.fsagx.ac.be/gf/outilslogiciels/Garbel/proj4.jpg). I already
have a similar problem when writing a GPS software
(http://lists.maptools.org/pipermail/proj/2006-August/002447.html). To
solve this I have used this towgs84 parameters :
+towgs84=-99.059,53.322,-112.486,0.419,-0.83,1.885,-0.999999. In QGIS,
the towgs84 parameters are different
(+towgs84=106.869,-52.2978,103.724,-0.33657,0.456955,-1.84218,1). When I
use my GPS software to reproject into Belgian Lambert 72 the waypoints
are correctly placed (red points in the image) but not when I use QGIS
(yellow points). I think thus there is a problem in the epsg 31370
definition.

		Best regards,
		Thibaut.

		_______________________________________________
		Proj mailing list
		Proj at lists.maptools.org
		http://lists.maptools.org/mailman/listinfo/proj

	_______________________________________________
	Proj mailing list
	Proj at lists.maptools.org
	http://lists.maptools.org/mailman/listinfo/proj

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/proj/attachments/20100121/6ab94ff2/attachment.html>