[PROJ] Derived Vertical CRS

[Vautour, André]

I would have replied to the original email, but I wasn’t on the mailing list when it came in.

Using a DerivedCRS with a vertical offset and slope transformation seems wrong to me. In ISO 19111 and EPSG, a conversion is an operation that stays in the same datum, and a transformation is an operation that changes the datum. Vertical offset and slope is clearly going from one vertical datum to a different datum. Are you saying that your Custom Vertical CRS is still using the EGM2008 geoid datum after a vertical offset and slope operation?

Examples of conversions that stay in the same vertical datum would be a Depth vertical CRS, that has a height-depth reversal conversion, or a Feet vertical datum that has a unit conversion to the metres CRS. In all those cases, the source (derived) and target (base) datums are the same:
NAVD88 Height (ftUS): https://epsg.org/crs/wkt/id/6360
NAVD88 Depth: https://epsg.org/crs/wkt/id/6357

Should this not be a BoundCRS that has custom as the source,  EGM2008 height as the target, and the vertical offset as the abridged transformation? Am I right to assume that your Custom Vertical CRS is not using the EGM2008 datum?
André

From: PROJ <proj-bounces at lists.osgeo.org> On Behalf Of Even Rouault
Sent: Saturday, July 30, 2022 14:36
To: Javier Jimenez Shaw <j1 at jimenezshaw.com>; proj <PROJ at lists.osgeo.org>
Subject: Re: [PROJ] Derived Vertical CRS

External Email

Javier,

I've addressed this in https://github.com/OSGeo/PROJ/pull/3270.

Regarding the WKT issue, I've raised this to the OGC CRS SWG list as the grammar of WKT doesn't allow currently a unitless parameter, even if the EPSG Vertical Offset and Slope method requires it. I'd suggest using something like UNIT["none",1] instead of ANGLEUNIT["arc-second",4.8481368111E-06], until that's cleared up

Even
Le 29/07/2022 à 21:16, Javier Jimenez Shaw a écrit :
Hi

Thanks to Even's PR #3200 "Implement Vertical Offset and slope transformation method", I am trying to use a Derived Vertical CRS using the conversion method "Vertical Offset and Slope" (that will be a perfect combination with "Affine parametric transformation" in the horizontal... I hope)

The WKT2 I have created is this

VERTCRS["Custom Vertical",
    BASEVERTCRS["EGM2008 height",
        VDATUM["EGM2008 geoid"]],
    DERIVINGCONVERSION["vertical offs. and slope",
        METHOD["Vertical Offset and Slope",
            ID["EPSG",1046]],
        PARAMETER["Ordinate 1 of evaluation point",46.9166666666669,
            ANGLEUNIT["degree",0.0174532925199433],
            ID["EPSG",8617]],
        PARAMETER["Ordinate 2 of evaluation point",8.18333333333361,
            ANGLEUNIT["degree",0.0174532925199433],
            ID["EPSG",8618]],
        PARAMETER["Vertical Offset",-0.245,
            LENGTHUNIT["metre",1],
            ID["EPSG",8603]],
        PARAMETER["Inclination in latitude",-0.21,
            ANGLEUNIT["arc-second",4.8481368111E-06],
            ID["EPSG",8730]],
        PARAMETER["Inclination in longitude",-0.032,
            ANGLEUNIT["arc-second",4.8481368111E-06],
            ID["EPSG",8731]],
        PARAMETER["EPSG code for Horizontal CRS",4326,
            ANGLEUNIT["arc-second",4.8481368111E-06],
            ID["EPSG",1037]]],
    CS[vertical,1],
        AXIS["gravity-related height (H)",up,
            LENGTHUNIT["metre",1]],
    USAGE[
        SCOPE["unknown"],
        AREA["World"],
        BBOX[-90,-180,90,180]]]

The first strange thing is that without the unit in the parameter "EPSG code for Horizontal CRS", projinfo "$(cat derived_vertical.wkt)" shows a warning:
Warning: Parsing error : syntax error, unexpected ID. Error occurred around:
            ID["EPSG",1037]]],

When I try to see how the conversion would be with the command
projinfo -s EPSG:3855 -t "$(cat derived_vertical.wkt)"
I get
PROJ string:
Error when exporting to PROJ string: Unsupported conversion method: Vertical Offset and Slope

Using it in a compound crs and using cs2cs does not get the proper transformation neither.

Am I creating the WKT2 string properly to create a derived vertical crs (to be used in a compound) with that conversion method?

Thanks

PS. I am using master from today
PS2.  projinfo -s EPSG:5728 -t EPSG:5730 shows a coordinate operation that uses "Vertical Offset and Slope", with a nice PROJ string

.___ ._ ..._ .. . ._.  .___ .. __ . _. . __..  ... .... ._ .__
Entre dos pensamientos racionales
hay infinitos pensamientos irracionales.



_______________________________________________

PROJ mailing list

PROJ at lists.osgeo.org<mailto:PROJ at lists.osgeo.org>

https://lists.osgeo.org/mailman/listinfo/proj

--

http://www.spatialys.com

My software is free, but my time generally not.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/proj/attachments/20220802/a5950dd4/attachment-0001.htm>