[MetaCRS] Motion Vote: CS-Map RFC 2 - Redesign Datum Transformation Engine

Norm Olsen norm.olsen at autodesk.com
Mon Jul 12 10:25:20 EDT 2010

There are several downsides to relying on WGS84 as a pivot datum.  Many of these are not current problems; just beginning to become problems.

1> There are now three distinct WGS 84 realizations.  All only slightly different, not different enough yet  to concern most ordinary GIS users.  But nonetheless, they are different.  For example, EPSG now considers what we have always referred to as HPGN/HARN/CSRS as WGS84. 

2> There are datum definitions out there (admittedly very few) which do not reference WGS 84.  That is, the definition goes directly from one non-WGS84 datum to another non-WGS84 datum.  Definitions which convert to/from WGS84 are not available.

3> There are now several grid data file conversion techniques which do not convert from/to a WGS84 equivalent (NAD83, GDS94, ETRF89).  For example, the recently added Swiss .gsb file converts from CH1903 to CH1903+.  An additional transformation (7 Parameter) is required to get to WGS84.  Thus, without special kludges to the code, this transformation file could not be used (does not involve WGS84).

4> We have survived for two decades now assuming the NAD83, GDA94, ETRS89, NZGD2K, and a whole flock of other geocentric datums are all identical with WGS84.  In addition to number one above (multiple realizations of WGS84), the differences between WGS84 and these geocentric datums will become a problem pretty soon.


-----Original Message-----
From: metacrs-bounces at lists.osgeo.org [mailto:metacrs-bounces at lists.osgeo.org] On Behalf Of Frank Warmerdam (External)
Sent: Wednesday, July 07, 2010 9:40 AM
To: Martin Desruisseaux
Cc: metacrs at lists.osgeo.org
Subject: Re: [MetaCRS] Motion Vote: CS-Map RFC 2 - Redesign Datum Transformation Engine

Martin Desruisseaux wrote:
> In other words, the result of two consecutive Datum Shift is usually not 
> expressible in a single set of Bursa-Wolf parameters, except if the 
> datum shifts are only geocentric translations (no scale or rotation).


OK,you are making me think a bit more about what I mean when I talk about
using WGS84 as a pivot datum.

I wasn't thinking the multiple shifts could be reduced to a single Bursa-Wolf
transformation, but rather that each datum shift would still occur by passing
through WGS84.

So for a rather made up example lets assume there is a translation from
Estonia 1992 to ETRS89 and a transformation from ETRS89 to WGS84, I would
like to be able to do the datum transformation from Estonia 1992 to ETRS89
by going from Estonia to ETRS89 to WGS84 and then back to ETRS89.

Thus, I would keep track of the preferred path to WGS84 for Estonia 1992
as being Estonia 1992 to ETRS89 to WGS84 and always apply them as a chain.

That is, I am willing to support chaining, but would still like to have
every transformation go through the grand-central station of WGS84 for

What are the downsides to using WGS84 as a pivot but with chained

Best regards,
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | Geospatial Programmer for Rent

MetaCRS mailing list
MetaCRS at lists.osgeo.org

More information about the MetaCRS mailing list