[PROJ] Calculating DTK25 from WGS84

[Even Rouault]

Dominik,

first, I'm not familiar at all with DTK25 but
seeing
http://www.adv-online.de/AdV-Produkte/Standards-und-Produktblaetter/Produktblaetter/binarywriterservlet?imgUid=3ab60d30-b338-9eff-de33-50376a112976&uBasVariant=11111111-1111-1111-1111-111111111111
it seems that DTK25 products on Bessel ellipsoid use the Potsdam datum

The corresponding geographic CRS is likely EPSG:4746 "PD/83"

Assuming you have the BETA2007 grid installed,

echo 49.12345 8.55555 | cs2cs -f "%.8f" EPSG:4326  EPSG:4746

49.12450370	8.55651912 0.00000000

However given the formula you give to compute DTK25 sheet numbers, the influence
of the datum shift is completely neglectable, unless you would be
just at the boundary between 2 sheets (but I assume they have some overlap ?)

(56 - 49.12450370) * 10 = 68.75496300000002
(56 - 49.1245) * 10 = 68.75500000000002
(8.55555 -  5.66666666) * 6 = 17.333300040000005
(8.55651912 -  5.66666666) * 6 = 17.339114760000008

> echo 49.12345 8.55555 |>  cs2cs +proj=latlon +ellps=WGS84 +to +proj=latlon +ellps=bessel -f "%.12f"> But it actually does nothing; output coordinates are unchanged.

Yes, to get changes in coordinates, there must be a datum qualifier on each side.
The Bessel ellipsoid itself doesn't define a datum, so you need either to
add a +towgs84 or +nadgrids qualifier, when using this PROJ.4 deprecated syntax
Note also that when using it, you must specify coordinates in long, lat order, whereas
when using EPSG geographic CRS codes, this must be in lat, long order.

Even

-- 
Spatialys - Geospatial professional services
http://www.spatialys.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/proj/attachments/20200408/76b96edd/attachment.html>