[PROJ] Questions about declaring a CRS
Even Rouault
even.rouault at spatialys.com
Tue Jun 29 08:35:53 PDT 2021
Le 29/06/2021 à 17:27, Floris Vanderhaeghe via PROJ a écrit :
>
> Thank you Even!
>
> To R users, would this be appropriate advice regarding PROJ strings? -
> please comment or edit where needed:
>
> 1. Using a PROJ string to specify a CRS is discouraged
> <https://proj.org/development/reference/functions.html#c.proj_create>
> and currently not much documented by PROJ (PROJ strings are used
> for coordinate operations instead). At least for registered CRSs,
> always use AUTHORITY:CODE or WKT2
> <https://docs.opengeospatial.org/is/18-010r7/18-010r7.html>, not
> PROJ strings.
>
Yes
>
> 1. Also custom CRSs are ideally specified as WKT2
> <https://docs.opengeospatial.org/is/18-010r7/18-010r7.html>, or
> more conveniently as (currently not yet official) PROJJSON
> <https://proj.org/specifications/projjson.html>, which follows the
> WKT2 structure.
>
Yes
>
> 1. When specifying a custom CRS with a PROJ string, the WGS84
> ensemble datum (EPSG:6326
> <https://epsg.org/datum_6326/World-Geodetic-System-1984-ensemble.html>)
> will automatically be adopted if no '+ellps' is specified (and
> with '+ellps', the datum is considered as 'unknown').
>
Yes
>
> 1. Note that EPSG:6326 always refers to the latest WGS84 realization,
> not to a fixed one. If these conditions are fine for your usecase,
> then there's no problem.
>
No. For EPSG, EPSG:6326 is the datum ensemble WGS 84, not the latest
realization. We have indeed discussed a number of times on this list if
we should consider it as the latest realization in practice, but haven't
decided so yet.
>
> 1. For backward compatibility reasons, PROJ strings with '+towgs84'
> or '+datum', although deprecated, will remain supported by PROJ
> and GDAL.
>
Yes
>
> 1.
>
>
> 2. When specifying a custom CRS that doesn't use the WGS84, NAD27 or
> NAD83 datum (which can be specified with '+datum'), it is better
> to use WKT2
> <https://docs.opengeospatial.org/is/18-010r7/18-010r7.html> or
> PROJJSON <https://proj.org/specifications/projjson.html> and
> properly refer the geodetic datum. A typically less accurate
> alternative is defining as a BOUNDCRS by specifying '+towgs84'
> transformation parameters.
>
Yes
>
> With regards
>
> Floris
>
>
>
> mailhandtekening
>
> *Floris Vanderhaeghe*
>
> Flemish Government
> RESEARCH INSTITUTE FOR NATURE AND FOREST
> Team Biometry, Methodology and Quality Assurance
> Havenlaan 88 bus 73, 1000 Brussels
> Belgium
> floris.vanderhaeghe at inbo.be <mailto:floris.vanderhaeghe at inbo.be>
> www.inbo.be <http://www.inbo.be/>
>
> ///////////////////////////////////////////////////////////////////////////////////////////
>
>
>
>
> _______________________________________________
> PROJ mailing list
> 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/20210629/e94edd56/attachment.html>
More information about the PROJ
mailing list