[PROJ] Syntax to change projection from UTM to TM

[Greg Troxel]

Javier Jimenez Shaw via PROJ <proj at lists.osgeo.org> writes:

> I think you are misunderstanding some concepts.
> This presentation about the American reference systems may help you
> https://geodesy.noaa.gov/web/science_edu/webinar_series/changes-afoot-after-2022.shtml

I hope Ulises is working in meters.  But, regardless of claimed
deprecations, sometimes people use UTM in survey feet, sometimes in
international feet, and these days more in meters.

> or this paper
> https://www.fig.net/resources/proceedings/fig_proceedings/fig2023/papers/cinema03/CINEMA03_dennis_12044.pdf

I think that is still in the future, because it is about NATRF2022 which
is not yet released.    So far I am unaware of any actual use of
SPCS2022.

But this is an awesome reference for understanding.

> You cannot change it for UTM (actually it does not affect the results. It
> is ignored in PROJ if my checks were correct).

Interesting.  probably proj should throw an error.

> When you center your transverse Mercator in your (small) area of work
> (setting lon_0 and lat_0), you want to use something near k=1. Because you
> do not want to cover the wide range of 6 degrees of the UTM zone.

Agreed.  Arguably you want something so that the average distortion over
your area is zero, but for 10s of km and probably  even  100km, 0 is
good enough and simpler.

> There is an effect due to the elevation (factor de escala de altura). The
> higher you are, the more apart are points with same lat-lon, just because
> the radius at this point is not the radius of the ellipsoid, but something
> a bit bigger.
> You are in Mexico City (that is high ;), so the custom TM should have a k >
> 1 to compensate the elevation respect to the ellipsoid.

I have long wondered about this.  My impression is that in surveying,
the actual horizonal distance is measured and then this distance is
reduced to the geoid/ellipsoid.  That is, one is intending to measure
the distance not between two points, but the distance between their
projections along the plumb line to the geoid.  I may be confused about
this; the SPCS2022 paper makes me lean that way.  But maybe they want
the SPCS distance to match EDM even if you reduce both?

But it really depends on what  you want, and if you want the property
that distances in UTM are one thing or the other.

> What always make me nuts is the false easting and northing. They are just
> to have positive numbers in your coordinates (it is just adding an offset
> to the final numbers). I am never sure if they are "real" meters or
> "distorted" meters (I think the latter).

I'm pretty sure they are just added after projecting.   With UTM this
lets easting be positive even W of the central meridian, and northing be
positive in the southern hemisphere.

Really, the only thing that's meaningful with UTM is

  - transform back to geodetic

  - do cartesian distance from E1,N1 to E2,N2 (same zone only!) and
    treat that as a linear horizontal distance, generally not worrying
    about the scale factor