<div dir="ltr">Thank you very much for this research. I am surprised that we have not met in that rabbit hole. I first got interested in this when trying to see how close I could match the coordinates obtained with an hour of data collection using my SXblue II L1 GNSS WAAS unit to a USGS monument on a nearby mountain top. With guidance from a USGS specialist on datum and epoch conversion, the result was about 5.5 inches which I was very happy with for my purposes. That was in 2017. In 2020, having plenty of time to obsess over unimportant things, I fell back in that same rabbit hole as I had more reason to correctly set the CRS on WAAS coordinates.<div><br></div><div>I will read your references to try to figure out whether WAAS is ITRF2008 or 2014. But since the difference is millimeters, it is not really important. My hope is to have a good CRS defined for SBAS in QGIS and that null transforms are not used between CRS which are "just" a meter different. When I save WAAS data as ITRF2008 (I can't remember which EPSG), QGIS 3.16 thinks the datum is invalid. 2014 seems to be OK.</div><div><br></div><div>Stewart</div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Wed, Mar 17, 2021 at 6:14 PM Nicolas Cadieux <<a href="mailto:njacadieux.gitlab@gmail.com">njacadieux.gitlab@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<div>
<p>Hi,</p>
<p>This last question brought me down a rabbit hole that took me a
while to find (at least partially). Of course, this is far away
from the original question of "Do GPX files contain CRS
information?". The answer to that was no. The datum is WGS84 (in
it's most current iteration, or WGS84(G1762). Transformation
parameters to and from ITRF have been published and can be found</p>
<ul>
<li><a href="https://confluence.qps.nl/qinsy/latest/en/world-geodetic-system-1984-wgs84-182618391.html#id-.WorldGeodeticSystem1984(WGS84)v9.1-WGS84realizations" target="_blank">https://confluence.qps.nl/qinsy/latest/en/world-geodetic-system-1984-wgs84-182618391.html#id-.WorldGeodeticSystem1984(WGS84)v9.1-WGS84realizations</a></li>
<li>This file contain multiple transformations with sources
<a href="https://confluence.qps.nl/qinsy/files/latest/en/182618383/182618384/1/1579182881000/ITRF_Transformation_Parameters.xlsx" target="_blank">https://confluence.qps.nl/qinsy/files/latest/en/182618383/182618384/1/1579182881000/ITRF_Transformation_Parameters.xlsx</a></li>
<li>This contains info on WGS84 and NAD83
<a href="https://mcraymer.github.io/geodesy/pubs/nad83_agu2007spr.pdf" target="_blank">https://mcraymer.github.io/geodesy/pubs/nad83_agu2007spr.pdf</a></li>
<li><a href="https://www.unoosa.org/documents/pdf/icg/2018/icg13/wgd/wgd_12.pdf" target="_blank">https://www.unoosa.org/documents/pdf/icg/2018/icg13/wgd/wgd_12.pdf</a><br>
</li>
</ul>
<p>Other question that were raised: What are the other reference
frames used by the various GNSS services:</p>
<ul>
<li><a href="https://www.sciencedirect.com/science/article/pii/S0273117720308292#" target="_blank">https://www.sciencedirect.com/science/article/pii/S0273117720308292#</a>!</li>
<ul>
<li>"The TRFs realized by the GPS, GLONASS, Galileo, and
BeiDou-2 and BeiDou-3 broadcast ephemerides are the orbital
realizations of WGS 84 (G1762′), PZ90.11, GTRF19v01, and BDCS
respectively."</li>
<li>More info here.
<a href="https://gssc.esa.int/navipedia/index.php/Reference_Frames_in_GNSS" target="_blank">https://gssc.esa.int/navipedia/index.php/Reference_Frames_in_GNSS</a><br>
</li>
</ul>
</ul>
<p>This document made a comparison comparison between broadcast
reference frames and ITRF:</p>
<ul>
<li><a href="http://www.epncb.eu/_newseventslinks/workshops/EPNLACWS_2017/pdf/06_Open_Session/04_Broadcast-Precise.pdf" target="_blank">http://www.epncb.eu/_newseventslinks/workshops/EPNLACWS_2017/pdf/06_Open_Session/04_Broadcast-Precise.pdf</a></li>
<ul>
<li>"Reference frames:<br>
•GPS and Galileo broadcast reference frames are aligned with
ITRF: translations are less than 0.10 m and rotations are less
than 2 milli-second of arc<br>
•GLONASS M broadcast reference frame is offset to ITRF by at
most 0.27 ±0.04 m in Y and maximum rotation is 4 ±2
milli-second of arc about Y.<br>
•GLONASS K broadcast reference frame is offset to ITRF by at
most 1.06 ±0.17 m in Y and maximum rotation is 19 ±2
milli-second of arc about X.<br>
•BeiDoubroadcast reference frame is offset to ITRF by at most
0.26 ±0.18 in X and Y, and maximum rotation is 2 ±1
milli-second of arc about Z."</li>
</ul>
</ul>
<p>Then there are questions pertaining to SBAS around the world:</p>
<ul>
<li><a href="https://www.faa.gov/about/office_org/headquarters_offices/ato/service_units/techops/navservices/gnss/library/factsheets/media/SBAS_Worldwide_QFact.pdf" target="_blank">https://www.faa.gov/about/office_org/headquarters_offices/ato/service_units/techops/navservices/gnss/library/factsheets/media/SBAS_Worldwide_QFact.pdf</a></li>
<ul>
<li>It state the various services</li>
<ul>
<li>Wide Area Augmentation System (WAAS)</li>
<li>European Geostationary Navigation Overlay Service (EGNOS)</li>
<li>Multi-functional Transport Satellite (MTSAT) Satellite
Based Augmentation System (MSAS) (Japan)</li>
<li>GPS Aided Geostationary Earth Orbit (GEO) Augmented
Navigation (GAGAN) (India)</li>
<li>System of Differential Correction and Monitoring (SDCM)
Russia</li>
<li>Korean Augmentation Satellite System (KASS)</li>
<li>BeiDou Satellite Based Augmentation System (BDSBAS)</li>
</ul>
</ul>
<li>Also found this
<a href="https://www.gps.gov/technical/ps/2008-WAAS-performance-standard.pdf" target="_blank">https://www.gps.gov/technical/ps/2008-WAAS-performance-standard.pdf</a></li>
<li>And this <a href="https://www.nstb.tc.faa.gov/" target="_blank">https://www.nstb.tc.faa.gov/</a></li>
</ul>
<p>Finally, there was the question of the introduction of new
reference frames when a SBAS services was used. I was skeptical
of this but you guys are right. Using a WAAS service does
introduce a new reference frame as the "... GEO satellites do not
belong to any satellite positioning service (e.g. GPS, GLONASS),
ephemeris for those satellites are not externally available.
Therefore, it is the SBAS that is in charge of providing the user
with the GEO ephemeris. Keep in mind that all components are
expressed in ECEF reference coordinates and the time offset is
with respect to SBAS Network time (SNT)."
(<a href="https://gssc.esa.int/navipedia/index.php/The_EGNOS_SBAS_Message_Format_Explained#SBAS_broadcast_data" target="_blank">https://gssc.esa.int/navipedia/index.php/The_EGNOS_SBAS_Message_Format_Explained#SBAS_broadcast_data</a>)</p>
<p>Same information is stated here
(<a href="https://www.sciencedirect.com/science/article/pii/S0273117720308292#" target="_blank">https://www.sciencedirect.com/science/article/pii/S0273117720308292#</a>!)</p>
<ul>
<li>"Also, note that the methodology and results reported here are
only valid for direct, real-time unaugmented GNSS applications.
Any form of augmentation including all types of differential
positioning, Assisted GPS, or specialized augmentation, such as
the US Federal Aviation Administration (FAA’s) Wide Area
Augmentation System, immediately introduce an alternate TRF with
its own relationship to ITRF2014."</li>
</ul>
<p>And here
(<a href="https://gssc.esa.int/navipedia/index.php/The_EGNOS_SBAS_Message_Format_Explained#SBAS_broadcast_data" target="_blank">https://gssc.esa.int/navipedia/index.php/The_EGNOS_SBAS_Message_Format_Explained#SBAS_broadcast_data</a>)</p>
<ul>
<li>
<h4>"Message type 9</h4>
<p>Message type 9 contains the information about the GEO
navigation.
As GEO satellites do not belong to any satellite positioning
service (e.g. GPS, GLONASS), ephemeris for those satellites
are not externally available. Therefore, it is the SBAS that
is in charge of providing the user with the GEO ephemeris.
Keep in mind that all components are expressed in ECEF
reference coordinates and the time offset is with respect to
SBAS Network time (SNT)."</p>
</li>
</ul>
<p>Finally, I tried to "“If you can find a clear statement of what
frame any SBAS uses, I'd love to see a URL/pointer.” I found these
documents.</p>
<ul>
<li><a href="https://www.gsa.europa.eu/sites/default/files/brochure_os_2017_v6.pdf" target="_blank">https://www.gsa.europa.eu/sites/default/files/brochure_os_2017_v6.pdf</a></li>
<ul>
<li>This one talks about the EGNOS reference frame<br>
</li>
<li>On page 20, "Strictly speaking, the time and position
information that are derived by an SBAS receiver that applies
the EGNOS corrections are not referenced to the GPS Time and
the WGS84 reference systems as defined in the GPS Interface
Specification. Specifically, the position coordinates and time
information are referenced to separate reference systems
established by the EGNOS system, namely the EGNOS Network Time
(ENT) timescale and the EGNOS Terrestrial Reference Frame
(ETRF). However, these specific EGNOS reference systems are
maintained closely aligned to their GPS counterparts and, for
the vast majority of users, the differences between these two
time/terrestrial reference frames are negligible."</li>
<li>P 22 "<br>
The ETRF is periodically aligned to the ITRF2000 in order to
maintain the difference between the positions respectively
computed in both frames below a few centimetres. The same can
be said about the WGS84 (WGS84 (G1150) aligned to ITRF2000).
Conversion of ETRF data into WGS84 (G1150) is obtained by
applying the offset that exists at a certain epoch between the
ETRF and the ITRF2000 to the ITRF2000 to WGS84 (G1150) frame.
Note that currently these last two reference frames are almost
equivalent (offsets minor than 2cm). This means that, for the
vast majority of applications, it can be considered that the
positions computed by an EGNOS receiver are referenced to
WGS84 and can be used with maps or geographical databases in
WGS84."</li>
</ul>
</ul>
<p>For the other SBAS services, I found this documents<br>
</p>
<ul>
<li><a href="https://www.mdpi.com/2072-4292/11/4/411" target="_blank">https://www.mdpi.com/2072-4292/11/4/411</a> : Evaluation of Orbit,
Clock and Ionospheric Corrections from Five Currently Available
SBAS L1Services: Methodology and Analysis</li>
<ul>
<li>They state: "In addition, SBAS-derived and IGS precise
orbits are formally referred to different reference frames.
The current IGS precise orbits are referred to IGS14, which is
aligned with the latest International Terrestrial Reference
Frame (ITRF) called ITRF2014 [20]. <u>WAAS adopts WGS84 as
the coordinate reference system to broadcast satellite orbit
corrections </u>[21], and the most recent WGS84
realization(G1762) agrees with ITRF2008 at the centimeter
level [22]. As presented in the International Terrestrial
Reference Service (ITRS) website
(<a href="http://itrf.ensg.ign.fr/trans_para.php" target="_blank">http://itrf.ensg.ign.fr/trans_para.php</a>), the differences
between <span style="font-size:16.6043px;font-family:sans-serif">ITRF2008 and ITRF2014 are at the
millimeter-level. <u>EGNOS reference frame, named EGNOS
Terrestrial </u></span><span style="font-size:16.6043px;font-family:sans-serif"><u>Reference Frame (ETRF)</u>,
is periodically aligned on ITRF within a consistency of a
few centimeters [</span><span style="font-size:16.6043px;font-family:sans-serif">23</span><span style="font-size:16.6043px;font-family:sans-serif">].</span><span style="font-size:16.6043px;font-family:sans-serif"> We cannot find any official documentation
or information about the reference frames adopted for the </span><span style="font-size:16.6043px;font-family:sans-serif">MSAS,
GAGAN, and SDCM. However, an alignment to WGS84 can be
expected since the requirements </span><span style="font-size:16.6043px;font-family:sans-serif">of
international air navigation must be fulfilled [</span><span style="font-size:16.6043px;font-family:sans-serif">5</span><span style="font-size:16.6043px;font-family:sans-serif">].
Therefore, the differences between reference</span><span style="font-size:16.6043px;font-family:sans-serif">
frames mentioned above would not exceed a few centimeters,
which can be neglected in the accuracy </span><span style="font-size:16.6043px;font-family:sans-serif">assessment of SBAS-derived orbits."</span></li>
</ul>
</ul>
<p>Hope this helps if you are crazy enough to read it!</p>
<p>Cheers.</p>
<p>Nicolas</p>
<p><br>
<span style="font-size:12.3892px;font-family:sans-serif"><span style="font-size:12.3892px;font-family:sans-serif"></span></span></p>
<p><span style="font-size:12.3892px;font-family:sans-serif"><span style="font-size:12.3892px;font-family:sans-serif"></span></span></p>
<div>On 2021-03-06 10:19 p.m., Nicolas
Cadieux wrote:<br>
</div>
<blockquote type="cite">
<div>
<blockquote type="cite">
<div dir="ltr">“If you can find a clear statement of what
frame any SBAS uses, I'd love<br>
to see a URL/pointer.”</div>
</blockquote>
</div>
<div><br>
</div>
<a href="https://gssc.esa.int/navipedia/index.php/Reference_Frames_in_GNSS" target="_blank">https://gssc.esa.int/navipedia/index.php/Reference_Frames_in_GNSS</a>
<div><br>
</div>
<div>My guess is that each constellation will use there own
reference frame for the WAAS system they choose to implement.
Each can be tied to a specific ITRF version using published
transformation parameters. My guess is that a gps unit will use
these parameters and use a specific ITFR only when mixing
satellites from various constellations as a way to bridge the
information from one frame to another. Error are under a few
centimetres. Most consumer GPS have a errors that are much
larger related to stuff like clocks, broadcast éphémérides and
atmospheric conditions. The specific ITFR chosen is probably
decided by each manufacturer depending on the published data.
Since WGS84 and PZ-90 (Russians CRS) can both be easily tied to
ITRF2008, I would expect that RF would be the one used in
current GPS units. I will ask the sxblue guys to confirm this.
<br>
<br>
<div dir="ltr">Nicolas Cadieux
<div><a href="https://gitlab.com/njacadieux" target="_blank">https://gitlab.com/njacadieux</a></div>
</div>
<div dir="ltr"><br>
<blockquote type="cite">Le 5 mars 2021 à 16:48, Greg Troxel
<a href="mailto:gdt@lexort.com" target="_blank"><gdt@lexort.com></a> a écrit :<br>
<br>
</blockquote>
</div>
<blockquote type="cite">
<div dir="ltr"><span></span><br>
<span>Nicolas Cadieux <a href="mailto:njacadieux.gitlab@gmail.com" target="_blank"><njacadieux.gitlab@gmail.com></a>
writes:</span><br>
<span></span><br>
<blockquote type="cite">
<blockquote type="cite"><span>For elevation, I read the
spec as saying that the datum is "WGS84</span><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><span>orthometric height", meaning
that one takes WGS84 ellipsoidal height and</span><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><span>uses EGM2008 to get a height
that is sort of "above sea level". The</span><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><span>notion that the height is
ellipsoidal height is to me unreasonable. </span><br>
</blockquote>
</blockquote>
<blockquote type="cite"><span></span><br>
</blockquote>
<blockquote type="cite"><span>If the standard says
orthometric height, it means that it takes the</span><br>
</blockquote>
<blockquote type="cite"><span>ellipsoïdal height and then
applies the geiod model (in this case</span><br>
</blockquote>
<blockquote type="cite"><span>EGM2008 or Earth gravitational
model 2008). This is the height where</span><br>
</blockquote>
<blockquote type="cite"><span>the average sea level would be
given the local gravity on land.</span><br>
</blockquote>
<blockquote type="cite"><span>Orthometric height is the
geiod height or the height above the average</span><br>
</blockquote>
<blockquote type="cite"><span>sea level.</span><br>
</blockquote>
<span></span><br>
<span>Agreed but what it says is:</span><br>
<span></span><br>
<span> <xsd:element name="ele" type="xsd:decimal"
minOccurs="0"></span><br>
<span> <xsd:annotation></span><br>
<span> <xsd:documentation> Elevation (in meters)
of the point. </xsd:documentation></span><br>
<span> </xsd:annotation></span><br>
<span> </xsd:element></span><br>
<span></span><br>
<span>To me, "elevation" always means some kind of
orthometric height. I have</span><br>
<span>never heard anyone call an ellipsoidal height
elevation. Given the</span><br>
<span>notion of WGS84 in GPX, and that WGS84 defines
orthometric height, I</span><br>
<span>find this unambiguous -- but not comfortably so.</span><br>
<span></span><br>
<blockquote type="cite">
<blockquote type="cite"><span>I would suggest to the
Jeremy to understand the delta from "WGS84" to</span><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><span>GDA94. I'm not a
<a href="http://geodesy.expert.au" target="_blank">geodesy.expert.au</a>, but my impresssion is that it's</span><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><span>only a few meters and that
it is therefore unlikely that points from a</span><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><span>Garmin unit have errors that
are small enough to notice that. I have</span><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><span>not been able to notice the
NAD83(2011)/WGS84(G1762) shift (about a</span><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><span>meter) with L1-only
navigation solution GPS. I can resolve it very</span><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><span>clearly with dual-frequency
multi-constellation RTK.</span><br>
</blockquote>
</blockquote>
<blockquote type="cite"><span></span><br>
</blockquote>
<blockquote type="cite"><span>In North American, most
devices do not make a difference between Nad83</span><br>
</blockquote>
<blockquote type="cite"><span>(revised models) and WGS84
(revised models). I imagine this is</span><br>
</blockquote>
<blockquote type="cite"><span>probably the case with GDA94,
specially if GDA94 was identical to</span><br>
</blockquote>
<blockquote type="cite"><span>WGS84 original in the
beginning (i’am not sure this is the case, I</span><br>
</blockquote>
<blockquote type="cite"><span>really don’t know here).</span><br>
</blockquote>
<span></span><br>
<span>Agreed. I think what you are saying is that when one
asks a device to</span><br>
<span>datum transform from WGS84 to NAD83 it will use null
transform.</span><br>
<span></span><br>
<blockquote type="cite">
<blockquote type="cite"><span>Despite "GPX is WGS84", if
the GPS receiver was receiving differential</span><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><span>corrections, either locally
or via SBAS such as WAAS, then the output</span><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><span>coordinates are no longer in
WGS84 and are instead in the differential</span><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><span>system's frame. WAAS is I
believe in something like ITRF2005, but it's</span><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><span>very hard to figure that out
precisely. (My understanding is that at</span><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><span>least most of Australia
currently has no available SBAS, but almost all</span><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><span>measurements made in the US
with navigation-grade equipment are with</span><br>
</blockquote>
</blockquote>
<blockquote type="cite">
<blockquote type="cite"><span>WAAS.)</span><br>
</blockquote>
</blockquote>
<blockquote type="cite"><span></span><br>
</blockquote>
<blockquote type="cite"><span>Weird... I would expect the
coordinates to be a simple corrections of</span><br>
</blockquote>
<blockquote type="cite"><span>whatever version of WGS84 is
currently in use...</span><br>
</blockquote>
<span></span><br>
<span>I expected that too. It seems not to be though.</span><br>
<span></span><br>
<span>The reference stations that generate the coordinates
don't have a way to</span><br>
<span>get precise WGS84(G1762) coordinates. And,
GLONASS/Galileo/BeiDou don't</span><br>
<span>use WGS84. It all amounts to a bunch of frames which
are for practical</span><br>
<span>purposes equivalent (ITRF2008 is a good overall
description today, I</span><br>
<span>think, but that and ITRF2014 are really close).</span><br>
<span></span><br>
<span>My theory is that until you get to RTK, you just
aren't going to get</span><br>
<span>sub-meter. So worrying about which modern (>=
2005) flavor of</span><br>
<span>WGSF84/ITRF/IGS is academic.</span><br>
<span></span><br>
<span>If you can find a clear statement of what frame any
SBAS uses, I'd love</span><br>
<span>to see a URL/pointer.</span><br>
</div>
</blockquote>
</div>
</blockquote>
<pre cols="72">--
Nicolas Cadieux
<a href="https://gitlab.com/njacadieux" target="_blank">https://gitlab.com/njacadieux</a></pre>
</div>
</blockquote></div>