New revisions of chartdatum norway v2023b

[himsve]

New separation model model between ETRS89 ell. height and CD_NORWAY

The GeoTiff no_kv_CD_above_Ell_ETRS89_v2023b.tif is the lastest version of the vertical offset grid (separation model) from ETRS89 ell. heights (EPSG:4937) to CD_NORWAY depths (EPSG:9672).

• no_kv_CD_above_Ell_ETRS89_v2023b is a replacement of the current version no_kv_CD_above_Ell_ETRS89_v2021a
• no_kv_CD_above_Ell_ETRS89_v2023b.tif consists of two bands, vertical offsets (band 1) and vertical offset standard deviations (band 2). Band 2 is marked vertical_offset_accuracy in meters as GDAL Metadata. So far vertical accuracy metadata is missing in PROJ.
• The modified transformation is requested in EPSG, but not yet published.

[jjimenezshaw]

Are you going to delete it, or create a new one?

Actually I see it was already superseded by https://epsg.org/transformation_10130/ETRS89-to-CD-Norway-depth-2.html, but not deleted.

[rouault]

The metadata of the grid is incorrect. It should not be TYPE=VERTICAL_OFFSET_VERTICAL_TO_VERTICAL, but TYPE=VERTICAL_OFFSET_GEOGRAPHIC_TO_VERTICAL

I also think we should let the previous file for reproducability for older transformations. The EPSG supersession mechanism should ensure the new grid is used by default, once it has made its way through EPSG.

[rouault]

@himsve You can drop the changes in the generated files (files.geojson, index.html) frmo the PR if that makes your life easier. I can easily refresh them on my side once this PR is merged. I suspect you might perhaps have some git issues with CRLF if you work on Windows.

[rouault]

@kbevers Is it fine to merge that right now in master, or will that cause trouble in your 1.16 release process?

[kbevers]

Is it fine to merge that right now in master, or will that cause trouble in your 1.16 release process?

I prefer that you wait until 1.16 is done