[PROJ] Syntax to change projection from UTM to TM

Greg Troxel gdt at lexort.com
Fri Dec 6 17:44:32 PST 2024


Ulises Ibarra via PROJ <proj at lists.osgeo.org> writes:

> * I need to specify in detail because I think I have no idea what I'm doing
> :P.

> * So I understand that the k is not necessary to use in the command. Is
> this correct?

Javier said you had the wrong value.  UTM is a particular instance of
TM.  Besides a defined central meridian scale factor, there are 60 zones
each with their central meridian.  So try it with no +k at all.

> * I don't understand what you said about there being people who use both
> NAD83 and WGS84 and I am using  ITRF2008, could you explain it better?
> Maybe I shouldn't specify the ellipsoid?

I guessed from the zone that you are in North America; maybe that was
wrong.

Each datum has an associated ellipsoid.  Many (large majority?) use
GRS80.  WGS84, used in GPS and by the US DoD, but not really of interest
for precise positioning, uses an ellipsoid which is almost GRS80,
arguably because of rounding and a different approach to specifying it.
But it is simply different.  The effect of this is tiny; my memory is
that it's about 1 mm at the earth's surface for coordinates.

When you ask proj to transform from some projected coordinate system
like UTM to geodetic (lat/lon) coordinates, you need to specify an
ellipsoid.  Unless the coordinates are in a CRS which specifies both UTM
and a base datum, in which case it's implicit.

But you have data in ITRF2008, and you aren't telling proj that.  You
are just doing projections.  That's totally ok -- that's how proj used
to be.  But you need to be aware of the datum and ellipsoid yourself.  I
think you are getting it right.


More information about the PROJ mailing list