1

I have received a shapefile with a user defined projection (i.e., +proj=krovak +lat_0=-32.923433 +lon_0=116.47485 +alpha=111.423433 +k=1.00004 +x_0=44328.529781 +y_0=1345943.211845 +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +units=m +no_defs).

The site is located in Australia but after automatic re-projection into GDA94 (the default CRS) it places the site in northern China.

I have viewed this particular shape file on the client's GIS system (ArcGIS) and it seems to locate fine.

Any suggestions as to what I need to do or am I missing something?

I have attempted to manually change the CRS (did not work) and I also saved the layer to a new SHP file and manually chose the CRS but this also did not work.

  • Can you add the extent of your data, as reported by properties, metadata tab? – AndreJ May 29 '14 at 10:43
  • 1
    Can you post the Esri .prj file for the data? We use two parameters to set the axis directions (SW or EN). – mkennedy May 29 '14 at 17:16
1

I'm not really sure the data you got is correct. IMHO, the projection provided is Krovak S-JTSK, which was designed to be used in central Europe's Czechia and Slovakia, which is all but Australia.

I would suggest exporting the shapefile into some other format (GeoJSON?) and proceeding from there.

  • To make it more complicated, positive Krovak coordinates are South, West (in this order). QGIS sets both axis to negative to get the orientation right. So I suggest to reproject inside ARCGIS to something useful. – AndreJ May 29 '14 at 15:42
  • Thanks Jan and Andre. The problem lies with the original files. I have requested a new set of files with the correct projection from the client. For all the frustration it was a useful exercise in manipulating projections in QGIS. Thanks to all who have contributed. – user30765 May 30 '14 at 0:02
0

I guess, you might have missed out the -ve sign in Latitude..(i.e., +proj=krovak +lat_0=-32.923433 +lon_0=116.47485 +alpha=111.423433 +k=1.00004 +x_0=44328.529781 +y_0=1345943.211845 +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +units=m +no_defs).

That is the reason why the reprojected data placing over China..

  • Thank you for that suggestion. the -'ve I assumed to be correct for the southern hemisphere. I have created a copy of the CRS and made the latitude positive just to see if that did anything. After applying the new CRS, as expected, nothing changed. In response to Andre the extents of the data are - xMin,yMin 45820.8,48926.6 : xMax,yMax 53683.7,61011.7. – user30765 May 29 '14 at 11:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy