[gdal-dev] showing vertical datum in wgs84 or its variants

Even Rouault even.rouault at spatialys.com
Sat Sep 15 14:32:59 PDT 2018


Michael,

> Coordinate System is:
> COMPD_CS["WGS 84 / UTM zone 51N + WGS 84 (G1674)",
>     PROJCS["WGS 84 / UTM zone 51N",
>         GEOGCS["WGS 84",
>             DATUM["WGS_1984",
>                 SPHEROID["WGS 84",6378137,298.257223563,
>                     AUTHORITY["EPSG","7030"]],
>                 AUTHORITY["EPSG","6326"]],
>             PRIMEM["Greenwich",0,
>                 AUTHORITY["EPSG","8901"]],
>             UNIT["degree",0.0174532925199433,
>                 AUTHORITY["EPSG","9122"]],
>             AUTHORITY["EPSG","4326"]],
>         PROJECTION["Transverse_Mercator"],
>         PARAMETER["latitude_of_origin",0],
>         PARAMETER["central_meridian",123],
>         PARAMETER["scale_factor",0.9996],
>         PARAMETER["false_easting",500000],
>         PARAMETER["false_northing",0],
>         UNIT["metre",1,
>             AUTHORITY["EPSG","9001"]],
>         AXIS["Easting",EAST],
>         AXIS["Northing",NORTH],
>         AUTHORITY["EPSG","32651"]],
>     GEOCCS["WGS 84 (G1674)",
>         DATUM["World_Geodetic_System_1984_G1674",
>             SPHEROID["WGS 84",6378137,298.257223563,
>                 AUTHORITY["EPSG","7030"]],
>             AUTHORITY["EPSG","1155"]],
>         PRIMEM["Greenwich",0,
>             AUTHORITY["EPSG","8901"]],
>         UNIT["metre",1,
>             AUTHORITY["EPSG","9001"]],
>         AXIS["Geocentric X",OTHER],
>         AXIS["Geocentric Y",OTHER],
>         AXIS["Geocentric Z",NORTH],
>         AUTHORITY["EPSG","7662"]]]
> 
> (and the z values do get properly projected). But when I convert from vrt to
> geotiff,

EPSG:32651+7662 is an invalid CRS. The vertical part should be a vertical
CRS, and not a Geocentric CRS as here.
GDAL / VRT is forgiving, but there's no way to encode this in GeoTIFF.

There's no way in WKT 1 to model a ProjectedCRS with a ellipsoidal height
(WKT 2:2018 will allow it with the concept of Projected 3D CRS),
so using plain EPSG:32651 is your best option here.

~~~~~

Actually I lied... There would be a way, but it is definitely not recommended.
The historical GeoTIFF spec allows for "ellipsoid Vertical CS"
( http://geotiff.maptools.org/spec/geotiff6.html#6.3.4 ), but ISO 19111 does
not allow this, and this will be likely no longer available as it in a future
revision of the GeoTIFF spec.

So if using listgeo to dump the TIFF you want to fix, and you insert the following in the
Keyed_Information section of the listgeo dump

      GTCitationGeoKey (Ascii,45): "WGS 84 / UTM zone 31N + WGS84 ellipsoid height"
      VerticalCSTypeGeoKey (Short,1): VertCS_WGS_84_ellipsoid
      VerticalUnitsGeoKey (Short,1): Linear_Meter

and then using geotifcp -g to forge a new GeoTIFF file, you will get

COMPD_CS["WGS 84 / UTM zone 31N + WGS 84 ellipsoid height",
    PROJCS["WGS 84 / UTM zone 51N",
        GEOGCS["WGS 84",
            DATUM["WGS_1984",
                SPHEROID["WGS 84",6378137,298.257223563,
                    AUTHORITY["EPSG","7030"]],
                AUTHORITY["EPSG","6326"]],
            PRIMEM["Greenwich",0,
                AUTHORITY["EPSG","8901"]],
            UNIT["degree",0.0174532925199433,
                AUTHORITY["EPSG","9122"]],
            AUTHORITY["EPSG","4326"]],
        PROJECTION["Transverse_Mercator"],
        PARAMETER["latitude_of_origin",0],
        PARAMETER["central_meridian",123],
        PARAMETER["scale_factor",0.9996],
        PARAMETER["false_easting",500000],
        PARAMETER["false_northing",0],
        UNIT["metre",1,
            AUTHORITY["EPSG","9001"]],
        AXIS["Easting",EAST],
        AXIS["Northing",NORTH],
        AUTHORITY["EPSG","32651"]],
    VERT_CS["unknown",
        VERT_DATUM["Not specified (based on WGS 84 ellipsoid)",2002,
            AUTHORITY["EPSG","6030"]],
        UNIT["metre",1.0,
            AUTHORITY["EPSG","9001"]],
        AXIS["Up",UP]]]

But this would be considered as invalid as well by today's standards.
(EPSG:6030 is not a vertical datum, but a geodetic datum)

Even

-- 
Spatialys - Geospatial professional services
http://www.spatialys.com


More information about the gdal-dev mailing list