[Qgis-user] QGIS GNSS NMEA Transformation Problem WGS84 to MGI Austria

[Greg Troxel]

Peter Hartl via QGIS-User <qgis-user at lists.osgeo.org> writes:

> GNSS NMEA interpreted as WGS84 in QGIS can do transformation with
> gridshift from ETRS89 to MGI, that is alredy working very good.

What do you mean by "GNSS NMEA interpreted as WGS84"?  What reference
frame is your GNSS data actually in?   It could be

  WGS84(G2159) if you are using a non-differential solution

  ITRF? if you are using SBAS

  ETRF? if you are using some RTK network

  something else, if your receiver  is combining 4 constellations, each
  of which have their own frames, and doing something that they don't
  document (hint: I have never seen any receiver document this.)


You should be aware that WGS84 is an ensemble, which means that when you
say data is in WGS84 that it is in one of several datums, and you don't
know which one.

The inherent accuracy limitation of the ensemble is about 2m.   So
therefore a shift of 30 cm is not wrong.


Probably, it would be good to figure out how to not use WGS84,
especially if you are tryig to work at sub-meter accuracy.    I have
done this by gathering data in NAD83(2011) epoch 2010.0, which is more
plate fixed, more or less analagous to ETRS89, using RTK with a
reference network operating in NAD83.