[PROJ] Transformations in ITRF2020

[Javier Jimenez Shaw]

Hi

I am now facing transformations with ITRF2020 with time(yes, because
Skylark, but I guess more NTRIPs are going to do (or already doing) the
same)

The point is when the destination CRS does not have a direct transformation
from ITRF2020 (or almost).

For instance, EPSG:4150, CH1903+ (the geographic of LV95, in Switzerland)
is only connected to ETRS89. See that CH1903+ is in the European tectonic
plate, as ETRS89

A direct transformation is completely ballpark

echo 42 6 0 2024.6 | ./cs2cs EPSG:9989 EPSG:4150 -d 9 --3d
42.000000000 6.000000000 0.000000000 2024.6

However, from ITRF2020 to ETRS89 there is some change due to the time (the
European plate moving at 2.5 cm/a since 2015):
echo 42 6 0 2024.6 | ./cs2cs EPSG:9989 EPSG:4937 -d 9 --3d
41.999994981 5.999991317 -0.001861237 2024.6

and from ETRS89 to CH1903+ due to the datum offset:
echo 42 6 0 2024.6 | ./cs2cs EPSG:4937 EPSG:4150 -d 9 --3d
42.000742733 6.000670188 -60.682799297 2024.6

Looking at the pipelines, the steps would be ITRF2020 -> ETRF2020 -> ETRS89
-> CH1903+
(btw, only since PROJ 9.4.1, when ETRF2020 was added)

I understand that the only way to get the proper transformation is chaining
these two manually. Via ETRS89 ... because I know it. Right?

Well, the future does not look so promising to me: in EPSG I only see
transformations from ITRF2020 to ETRF family, NAD83(2011) family and
NAD83(CSRS)v8. That is Europe, USA and Canada. That's all, folks.

Thanks,
.___ ._ ..._ .. . ._.  .___ .. __ . _. . __..  ... .... ._ .__
Javier Jimenez Shaw
Mercator str 42.
Null-Island
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/proj/attachments/20240827/4996adfd/attachment.htm>