[Qgis-user] Trimble GeoXT 2005 Accuracy

[Greg Troxel]

Springfield Harrison <stellargps at gmail.com> writes:

> My assumption (now it seems likely wrong) was that SBAS (WAAS in
> southern BC?) would "adapt" to the CRS designated in the receiver
> (NAD83, UTN 10N) and was also localized to the field location (as in
> Wide Area).  Thus QGIS would see the resulting NAD83, UTM 10N SHP file
> as compatible with the project CRS in QGIS (EPSG:26910, NAD83 UTM
> 10N).

You might be right and this is certainly complicated.

GPS navigation solutions without differential are in WGS84(G1762)
because that's the frame the orbits are expressed in.   When you are in
SBAS mode, control stations receive signals and compare pseudoranges to
the expected ones *given the known ITRF2008 coordinates of the reference
stations*, and then encode deltas so that a receiver can calculate the
error to back out as a function of the current time and the current
position.   After doing that, the receiver's solution is in ITRF2008,
not WGS84(G1762), but they are effectively the same thing.

In Garmin navigation-type receivers, data is logged in the solution
frame (labeled as WGS84) regardless of the display.

I really wonder what bits are recorded when you do this, what bits are
transferred, and then what transformation happens before you get a shapefile.

> I'm not sure why I would need to know or use the SBAS Reference Frame
> although the receiver certainly would.

It really depends on what happens in the receiver.

If you were using a local differential reference that was in
NAD83(2011), then the navigation solution would be in NAD83 but the
receiver might not know that.   This is what happens with RTK.

> Changing the project CRS to ITRF2014 (there are 3!, WTF, I used 7912)
> didn't change anything.

One of them is 2D, and one is XYZ.  7912 is the right call.

> I've since deferentially corrected the data with mixed results, some
> better, some worse, relatively.  I need to look at that further.

?

> Anyway, if Pathfinder Office produces a NAD83 UTM 10N SHP file from my
> SSF file (NAD83 UTM 10N), should that not project happily in QGIS set
> to EPSG:26910?  I think the datum transformations would take place in
> the receiver (in concert with SBAS) and Pathfinder Office.

Yes, if the data is properly transformed to NAD83(2011) before QGIS gets
it and QGIS is using some NAD83(2011) all should be ok.

> Perhaps there is more to this, I certainly appreciate your thoughts . . . .

You said consistently 1.2m ish.   I would go measure the same point
maybe 5 times on different days and look at the spread.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 194 bytes
Desc: not available
URL: <http://lists.osgeo.org/pipermail/qgis-user/attachments/20210306/8dd18c78/attachment.sig>