[OSGeoLive] Mis-projection in OSgeo-Live 16' QGIS only

Even Rouault even.rouault at spatialys.com
Tue Sep 24 18:07:56 PDT 2024 

Jochen,

I don't have a ready-made theory to explain that. I've checked that 
recent and older PROJ versions (the library that powers QGIS for 
coordinate transformation) have identical output for EPSG:31287 both in 
proj.4 strings, WKT2 and WKT:ESRI (the later is the variant use in the 
.prj file of shapefiles). Looking at the EPSG database for 
https://epsg.org/crs_31287/MGI-Austria-Lambert.html also shows no 
changes in ages in its definition. The definition in the open ESRI CRS 
database hasn't changed either in years.

The fact you are observing the same issue with ArcGIS Pro is really 
intriguing. Whereas ESRI software uses part of the open source GIS stack 
in their software, to the best of my knowledge, they don't use PROJ but 
their own Projection Engine. Sorry for asking a question that may sound 
like the one a typical operator of a help center would ask you :-), but 
are you really sure you are testing the exact same files between QGIS 
3.28 and 3.36/ArcGIS Pro ...?

Does the issue if on-the-fly reprojection occurs (and if so to which CRS 
are you reprojecting to?), or also if you keep your project in EPSG:31287

Otherwise please share all the files consisting of his shapefile.

For reference, the expected content for a .prj file using EPSG:31287 
should be:

PROJCS["MGI_Austria_Lambert",GEOGCS["GCS_MGI",DATUM["D_MGI",SPHEROID["Bessel_1841",6377397.155,299.1528128]],PRIMEM["Greenwich",0.0],UNIT["Degree",0.0174532925199433]],PROJECTION["Lambert_Conformal_Conic"],PARAMETER["False_Easting",400000.0],PARAMETER["False_Northing",400000.0],PARAMETER["Central_Meridian",13.3333333333333],PARAMETER["Standard_Parallel_1",49.0],PARAMETER["Standard_Parallel_2",46.0],PARAMETER["Latitude_Of_Origin",47.5],UNIT["Meter",1.0]]

In QGIS, both in the "wrong" and "good" version, please go to the 
Project Properties dialog, CRS tab, and filter on 31287 and paste the 
output of the bottom left pane with the WKT and proj.4 strings. It 
should be:

WKT PROJCRS["MGI / Austria Lambert",     BASEGEOGCRS["MGI",         
DATUM["Militar-Geographische Institut",             ELLIPSOID["Bessel 
1841",6377397.155,299.1528128,                 LENGTHUNIT["metre",1]]], 
         PRIMEM["Greenwich",0,             
ANGLEUNIT["degree",0.0174532925199433]],         ID["EPSG",4312]],     
CONVERSION["Austria Lambert",         METHOD["Lambert Conic Conformal 
(2SP)",             ID["EPSG",9802]],         PARAMETER["Latitude of 
false origin",47.5,             ANGLEUNIT["degree",0.0174532925199433], 
             ID["EPSG",8821]],         PARAMETER["Longitude of false 
origin",13.3333333333333,             
ANGLEUNIT["degree",0.0174532925199433],             ID["EPSG",8822]], 
         PARAMETER["Latitude of 1st standard parallel",49,             
ANGLEUNIT["degree",0.0174532925199433],             ID["EPSG",8823]], 
         PARAMETER["Latitude of 2nd standard parallel",46,             
ANGLEUNIT["degree",0.0174532925199433],             ID["EPSG",8824]], 
         PARAMETER["Easting at false origin",400000,             
LENGTHUNIT["metre",1],             ID["EPSG",8826]],         
PARAMETER["Northing at false origin",400000,             
LENGTHUNIT["metre",1],             ID["EPSG",8827]]],     
CS[Cartesian,2],         AXIS["northing (X)",north,             
ORDER[1],             LENGTHUNIT["metre",1]],         AXIS["easting 
(Y)",east,             ORDER[2],             LENGTHUNIT["metre",1]],     
USAGE[         SCOPE["Topographic mapping (medium and small scale)."], 
         AREA["Austria."],         BBOX[46.4,9.53,49.02,17.17]],     
ID["EPSG",31287]]

Proj4 +proj=lcc +lat_0=47.5 +lon_0=13.3333333333333 +lat_1=49 +lat_2=46 
+x_0=400000 +y_0=400000 +ellps=bessel 
+towgs84=577.326,90.129,463.919,5.137,1.474,5.297,2.4232 +units=m +no_defs

If you are doing a reprojection in QGIS, please also go to the 
Transformations tab in the Project Properties dialog, and show the 
details of the Datum transformation which is used, in particular the 
string that starts with "+proj=pipeline"

(not sure what you referred to with the "bottom of this email". an 
attachment? if so, it didn't get through)

Even

Le 24/09/2024 à 20:13, Jochen Albrecht via osgeolive a écrit :
> Hi
>
> I am teaching with the latest version of OSgeo-Live 16, which runs 
> QGIS 3.28.
> I am using for my class a small demo shapefile, which happens to be in 
> EPSG:31287 (Austria, Styria). QGIS 3.28 places this file into the 
> Mediterranean Sea, off the coast of Italy/France.
> Earlier versions of QGIS (in previous years of teaching), as well as 
> my current stand-alone version 3.36 are placing the data correctly 
> into the province of Styria in Austria. The problem occurs only in 
> version 3.28.
>
> The problem is aggravated by the fact that any attempt at transforming 
> the data is rendered futile in this particular version of QGIS because 
> none of the reprojection dialogues allow me to specify a 
> transformation. I even tried to define a new coordinate system using 
> the transformation parameters given by QGIS 3.38 (see the bottom of 
> this email). Applying these, places the data again into the Mediterranean.
>
> I have been teaching with QGIS for over 20 years, and this is a new 
> one for me. The easy solution would be to just not use version QGIS 
> 3.28. But it is the one that has been burned into the current 
> OSgeo-Live distribution. The whole purpose of OSgeo-Live is to work 
> with versions of the individual software packages that have proven to 
> play nice with each other and I do not want to experiment and have 
> students upgrade to a newer version of GIS.
>
> At this point, I am at a loss to advise my students what to do, or to 
> even explain what is happening here. It is as if the particular 
> coordinate system library for this version has a bug that nobody has 
> come across before. Am I missing something?
> Last minute update: the same error occurs in the latest version of 
> ArcGIS Pro 3.3. I have no idea what those two have in common, but it 
> suggests to me that it has something to do with a flawed coordinate 
> system library rather than the application software itself.
> Again, any suggestion for how to address this (or even explain it to 
> my students) would be appreciated.
>
> Cheers,
>       Jochen
>
> Dr. Jochen Albrecht, GISP (he/him/his)
>
> Professor for Computational and Theoretical Geography
>
> Department of Geography and Environmental Science 
> <http://www.geo.hunter.cuny.edu/>
>
> Hunter College CUNY
>
> 695 Park Avenue
>
> New York, NY 10065
>
> Immediate Past-President, GIS Certification Institute 
> <https://www.gisci.org/>
>
>
> 	
>
>
> Recently published: GIS and Housing: Principles and Practices 
> <https://sites.google.com/rojasap.com/gisandhousing/>*
> *
>
> _______________________________________________
> osgeolive mailing list
> osgeolive at lists.osgeo.org
> https://lists.osgeo.org/mailman/listinfo/osgeolive

-- 
http://www.spatialys.com
My software is free, but my time generally not.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/osgeolive/attachments/20240925/8225a284/attachment-0001.htm>

More information about the osgeolive mailing list