[Qgis-user] Do GPX files contain CRS information?

[Greg Troxel]

Stewart Holt <stewartbholt at gmail.com> writes:

> "If you can find a clear statement of what frame any SBAS uses, I'd love to
> see a URL/pointer."
>
> I have spent a few hours on many occasions in the last couple of years
> trying to find a definitive answer from an authoritative source. I have
> arrived at the same conclusion that ITRF2008 OR 2014 is likely very close.
> WAAS was not developed to help us with better accuracy. I can find tons of
> information about its application to aviation but only vague references to
> its reference frame. The earliest was ITRF 2000. Next it used 2008 and
> since 2014 exists, I wonder if it is using that. As stated, there is not
> much difference. I have read that the epoch is updated yearly. I would like
> to know the best EPSG to choose for this data. I have been using
> NAD83(2011) EPSG:4369 which is probably fine for most mapping purposes.

<SPECULATION>

  Each new WGS84 realization is very similar to some particular ITRF
  realization.  If I were charged with choosing a reference frame for an
  US-centric SBAS, I'd probably pick the ITRF that corresponds to the
  current WGS84 realization, recognizing that changing my reference
  station coordinates from ITRF2008 to ITRF2014 is unlikely to be
  measurable by anyeno looking at the corrections stream.   I would also
  update the reference station coordinates at the beginning of every
  year, tracking how the GPS control segment is updated.

</>

> Why do I continue to wonder about this? As a matter of personal interest, I
> wanted to know how accurately I could measure the elevation of some of the
> mountains over 4000 feet in the state of GA, USA. Except for a few with
> survey markers, most have elevations estimated from contour lines (40'
> interval). A few years ago, using a SXBlue II GNSS L1 unit, I measured the
> position and elevation of an 1807' mountain with a survey marker near home.
> Converted its USGS NAD83(1994) coordinates to NAD83(HARN), used HTDP to
> adjust for plate drift and landed within one half foot of the measured
> coordinates. Elevation was to the nearest foot. Data was collected for one
> hour and averaged.  I recently repeated the experiment and only came within
> 1.2 feet position and about twice that on elevation. With the new North
> American datum which will replace NAD83 in 2022 or 3 or ? this kind of
> thing will get easier, I hope!

See my hours-ago reply.  If you consider a 2m coordinate difference to
be "different" rather than "exactly matching", you need to be very
careful about NAD83 vs some realizaition of WGS84 and you need to be
careful that qgis is doing the transform that it should among those, and
not picking a null transform because of 1980s history.

Also, NAVD88 height and WGS84 orthometric height are not the same thing.
-------------- 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/20210306/511808e0/attachment.sig>