[gdal-dev] Transformation methods & parameters (OGRCoordinateTransformation)

[César Martínez]

Thanks André, datum_shift.csv is (partially) what I was looking for.

I see however that EPSG:15933 transformation operation is missing on
that list (which in this Spanish example is usually the most accurate
one), maybe because it uses the NTv2 method.

In any case, how can I use the datum_shift.csv parameters to construct
a OGRCoordinateTransformation instance?

César


2016-03-18 18:14 GMT+01:00 Andre Joost <andre+joost at nurfuerspam.de>:
> Am 18.03.2016 um 17:34 schrieb César Martínez:
>
>>
>> This is relevant because sometimes there are several transformation
>> operations available between 2 CRSs, and we'd like our users being
>> able to choose which operation they'd like to use. For instance,
>> between EPSG:23030 and EPSG:4326, at least the following
>> transformations are suitable for Spain: EPSG:15933, EPSG:1633,
>> EPSG:1133, EPSG:1145 [2], each using different methods & parameters
>> and thus providing different accuracy.
>>
>> Is there a simple way to get this information?
>>
>
> EPSG:23030 is ED50 / UTM zone 30N.
>
> In a first transformation step, the data is transformed from the projected
> CRS to the geographic CRS. In your case a transverse mercator projection on
> EPSG:4230 ED50.
>
> The datum shift (with different accuracy depending on the parameters) is
> from EPSG:4230 to EPSG:4326. Look out for the datum_shift.csv file, e.g.
> under bin/gdal-data in the Gisinternal builds for Windows. There you find
> all datum shifts from 4230 to 4326 in the lines labelled SEQ_KEY 306 to 340.
> Remarks, area bounds and parameters follow in the next columns, and the last
> column PREFERRED has just one line with entry 1 (SEQ_KEY=308). This one has
> been chosen as default for all projections based on ED50.
>
> You are free to use any of the other ones. For 7-parms trnasformations, you
> might need to change the sign of the rotation parameters RX, RY and RZ.
>
> HTH,
> André Joost
>
>
> _______________________________________________
> gdal-dev mailing list
> gdal-dev at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/gdal-dev



-- 
SCOLAB
http://scolab.es