[QGIS-Developer] Problem: QGIS Reproject layer Transformation of Lambert North and South Costa Rica to CRTM05

[Eduardo Rojas Rodríguez]

Hi everyone!

Before creating a ticket in QGIS Issue Tracking, I wanted to consult you a
few things to be the most effective for when creating the ticket and to
carry all the necessary information.

I wanted to ask who maintains the algorithm "Reproject layer" of QGIS
provider which transformation method is used?

We have detected that in QGIS the transformation of vector layers from the
Coordinate System Lambert North or South for Costa Rica (obsolete official
projection of Costa Rica) to the current Official Projection that is CRTM05
has a transformation problem.

*the problem:* is when use the "Reproyect Layer" to reproject a vector layer
that is in the coordinate system  "Lambert North or South for Costa Rica" to
the coordinate system "CRTM05" a mismatches occurs from 8 meters up to 100 o
300 meters when the process is performed using the "Reproject layer"
algorithm of the QGIS provider in the processing toolbox.

We believe that this is because there is a problem with the Datum transfer
since Lambert North or South for Costa Rica is Local "Ocotepeque Clarke
1866" while CRTM05 is global "WGS84" for that reason we found necessary to
know how the trasnformation is programmed in QGIS for the Reproject Layer
algorithm. 

for the transformation we need to know if it is using the matrix of
"Molodensky Badekas" of 3 or 7 parameters, the ideal is that it is carried
out using the one of 7 parameters.

We have the necessary parameters For the matrix of 7 parameters which are: 

Parameters of transformation using the Molodensky Badekas matrix of 7
parameters for reproject CRS:
Ocotepeque 1935 / Lambert Costa Rica South EPSG:5457 and Ocotepeque 1935 /
Lambert Costa Rica North EPSG:5456 to CRTM05 EPSG:5367

PAR_X_AXIS_TRANSLATION, (-)213.116
PE_PAR_Y_AXIS_TRANSLATION, (-)9.358
PE_PAR_Z_AXIS_TRANSLATION, (+)74.946
PE_PAR_X_AXIS_ROTATION, (-)2.351418791 (Is enough with 4 decimals)
PE_PAR_Y_AXIS_ROTATION, (+)6.1466912E-02(Is enough with 4 decimals)
PE_PAR_Z_AXIS_ROTATION, (-)6.394208994(Is enough with 4 decimals)
PE_PAR_SCALE_DIFFERENCE, (+)5.22
PE_PAR_X_COORD_OF_ROTATION_ORIGIN (BARICENTRO),617749.7118
PE_PAR_Y_COORD_OF_ROTATION_ORIGIN (BARICENTRO),-6250547.7336
PE_PAR_Z_COORD_OF_ROTATION_ORIGIN (BARICENTRO),1102063.6099

These are the parameters for the transformation.

This will be enough to create the ticket in the QGIS Issue Tracking? Or will
more information be needed? Or explain the problem better?

Thanks for the help!
-

Eduardo :)




-----
Eduardo Rojas Rodríguez
Analista en SIG
Soluciones en Tecnologías
Tel: 506 2505-5453
eduardo.rojas at soltig.net
http://soltig.net


--
View this message in context: http://osgeo-org.1560.x6.nabble.com/Problem-QGIS-Reproject-layer-Transformation-of-Lambert-North-and-South-Costa-Rica-to-CRTM05-tp5319677.html
Sent from the QGIS - Developer mailing list archive at Nabble.com.