[OSRS-PROJ] Re: Transverse Mercator revisited

[Strebe at aol.com]

I don't know anything about ESRI's implementation. Wallis claims Dozier's 
implementation is inefficient and the derivation not well motivated. I haven't 
looked at Dozier. I had never checked my own implementation against UTM because 
I don't much care about large-scale mapping, but your note induced me to. I 
find that my calculations of the complete ellipsoidal transverse Mercator agree 
with Snyder's UTM formulation [Snyder: Map Projections - A Working Manual] to 
sub-millimeter accuracy from +/-90.0 degrees latitude and +/- 3.5 degrees 
longitude from central meridian, at half degree intervals. Of course, it extends 
to 90 degrees away from the central meridian, and numerical checks demonstrate 
that it is conformal to within one part per billion throughout the spheroid.

Regards,
daan Strebe


In a message dated 10/13/03 14:58:12, gerald.evenden at verizon.net writes:

> Maybe a month or so ago there was a claim about ESRI developing a 
> Transverse
> Mercator projection from an "obscure" NOAA publication.  Part of the claim 
> was
> that the projection was good to 90 degrees from the CM.  I more or less said
> "bovine excrement" and off list got into a discussion with someone who
> supported the claim.
> 
> Subsequently I managed to get a reference to the NOAA article and locate
> the author who was kind enough to send me a copy.  After messing around with
> the article I wrote the following critque and, as you will note, have not 
> got
> a projection that will extend to 90 degrees longitude---nor 90 NS for that
> matter.
> 
> The reason I post it here is that if anyone has a copy of ESRI's software 
> and
> could puch in a few bench mark conversions I would be most grateful.
> Does ESRI's software go to 90 degree longitude??
> 
> A copy of this critique has also been sent to Jeff Dozier.  Sorry, but some
> comments relate to material of the article and are not meaningful without
> having a copy.
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/proj/attachments/20031014/ee90f963/attachment.html>