Geographic Information Systems Stack Exchange is a question and answer site for cartographers, geographers and GIS professionals. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

First off, I am a recent convert to QGIS and am greatly impressed with its ease of use and speed compared with the ESRI products with which I am more familiar - my congratulations and thanks to its developers (including the plug-ins).

I have surveyed (and corrected) DGPS (vector point) data collected to OSGB36 (British National Grid); these positions match acceptably well with the OSGB36 map layers to which I have access when imported as CSV and converted to SHP in QGIS, projected using EPSG:27700 (~1m error, although error could be in map layer which is suspect).

I'm attempting a four-stage export from QGISv1.7.4 to Google SketchUp (SU) via Google Earth (GE), as follows:

stage 1: In QGIS, Save as../KML with CRS set to WGS84 EPSG:4326 (or ETRS89);
stage 2: Open KML in GE;
stage 3: Save as KMZ;
stage 4: Open in SU.

Unfortunately, the exported WGS84 positions in the KML file are in error by ~8m - I've checked them against OSTN02 conversion using tool on OS website , I've also checked GE projection (by eye: converting GE lat/long to OSGB36 and comparing with OSGB36 vector map layer). (SU fails to open the KMZ file as well, but that's not a QGIS issue, I imagine - any hints welcome...).

So the evidence seems to point towards an error in the transformation from OSGB36 to WGS84 in stage 1, i.e. with QGISv1.7.4.

Any suggestions as to where the fault in my process might lie, or to an alternative approach, gratefully received...

thanks - JS

share|improve this question
A question about the coordinates in your original CSV file. Were these read out of the GPS receiver as National Grid coordinates, or as WGS84 lat/lon coordinates? – nhopton May 8 '12 at 14:27
The rover and base station GPS data (presumably WGS84) were post-processed to DGPS using OS RINEX station corrections to produce National Grid eastings and northings with height AODN. – jack_sprat May 8 '12 at 19:15
Ah, you're using professional kit. Taking readings of OS eastings and northings off the screen of a recreational GPS receiver can lead to discrepancies of about eight metres (this grade of instrument doesn't do the transformations very accurately). Perhaps it might help if you posted the data for a few of your points. – nhopton May 8 '12 at 19:27
OSGB36_mE_obs,OSGB36_mN_obs,GPS_ht_obs,ht_mAODN,label,KML_lat,KML_long,OS_lat,OS‌​_long,OS_GPSht 277014.8215,720032.2866,126.2391,72.5446,remed1,56.3566399801,-3.9920204605,56.3‌​5657627,-3.99198303,126.239 277024.1095,720037.1405,126.1231,72.429,remed2,56.3566859775,-3.9918725155,56.35‌​662226,-3.99183509,126.123 277030.6197,720044.4488,126.2305,72.5367,remed3,56.3567532905,-3.9917706459,56.3‌​5668957,-3.99173322,126.231 277033.6013,720056.2183,126.1593,72.4657,remed4,56.3568597428,-3.9917279344,56.3‌​5679602,-3.99169052,126.159 is that legible? – jack_sprat May 8 '12 at 21:22
~7.5m error in KML & OSTN02 lat/long positions – jack_sprat May 8 '12 at 21:30

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.