[aprssig] APRS LAT/LONG standards (Geodesy Part 2)

[Gerry Creager N5JXS]

Continuation... Hopefully I've evaded the angry gods of the listserver 
software...

VE7GDH wrote:
> Andre PE1RDW schreef:
> 
> You are probably familiar with the UTM that I mentioned. Universal
> Transverse Mercator is based on 60 zones that are 6º (latitude) wide and
> with the "easting" and "northing" in metres for the coordinates. It is a
> very easy one to use with paper maps that have the UTM grid printed on 
> them.

Universal Transverse Mercator is a PROJECTION and grid system.  It is, 
of necessity, referenced to a DATUM.  The DATUM is not referenced to a 
given projection.  Projections are used to reduce the distortion of 
depictnig a map of a curvilinear surface on a flat, planar surface.  UTM 
happens to be convenient for a number of uses.  It makes math across its 
zone almost trivial, but zone crossings are less simplistic.

> At the speed that helicopters and other aircraft can fly at, and the
> distances they can cover, it makes more sense to use lat / long, but there
> was a recent message from someone that mentioned that the helicopters they
> work with for SAR are now using UTM.

Depending on the GPS receiver you're using in the aircraft, you can 
display almost any coordinate system.  Most I've worked with in the 
civilian world prefer Lat/Lon in decimal degrees.  That is possibly 
because that's how they started and it became a standard by default. 
However, in the military world (at least in the USA and NATO countries) 
the coordinate system looks a lot like our Maidenhead Grid System, save 
it's based on UTM and is called the Military Grid Reference System 
(MGRS).  Given enough characters, one can define a 1m x 1m grid 
coordinate using this system.

> For APRS, the "standard" is WGS84. In spite of that, if you are referencing
> maps calibrated with another datum, I would think you would be better off
> using that same datum in the GPS. I wasn't sure if a GPS always puts out
> WGS84 in NMEA mode or if it puts out a position based on the datum that the
> GPS is set to. I just did a search and came across a page that said "older
> GPS Receiver firmwares always output WGS84, independent of the chosen map
> datum..." so I assume that means that newer receivers might put out
> coordinates based on whatever datum the receiver is set for. If this is the
> case, setting your GPS for APRS to your local datum would mean that there
> would be errors in your position on the APRS-IS & findu.com etc. if that is
> the case. When it comes down to it, the errors will be small if you are
> looking at a map the size of a city. However, if you were zoomed down to
> look at an area of just a few city blocks, it could be significant.

And, just to set the record straight, WGS84 was not always the 
"standard" because so many maps are still published in NAD27.  There was 
a lot of discussion, and a fair bit of vitriole, when this came up the 
first several times.  And there was some good discussion, bereft of 
religious fervor, on all sides.  The final reason for settling on WGS84 
was simply that it WAS defined for GPS use.

If one is working with maps created using another datum, getting the GPS 
to do the datum transformation makes good sense.  However, different 
manufacturers have different plans as to what is spewed out the NMEA 
string.  Some accept the datum selection you've made.  Some retain WGS84 
and solely give your LDC display the selected datum.  You have to 
determine that based on your GPS and its manufacturer.

NOW: For datum error.  NAD27 vs NAD83 can result in errors as large as 
230 meters in horizontal circular error (point-to-point, unadjusted). 
So, thinking the error might be trivial isn't necessarily correct.  In 
my part of Texas, the error is on the order of 110 meters overall with 
different shifts for Northing and Easting.

The method of tranforming data from NAD83 to NAD27 (and vice versa) was 
developed at the US National Geodetic Survey as a tool for determining 
errors in datum shift.  Because of the rather subjective nature of the 
NAD27 datum (please see the Sig archives and look for 'Geodesy' and my 
posts for reference to some of this lore, of contact me off-list), the 
continental US was subdivided into a number of zones and each zone was 
gradated and the correction for North and East offsets was calculated at 
various points.  This subjective correction is as good as it gets. 
There's no rigid mathematical transformation between NAD83 and NAD27 
such as exists for other geocentric datums and NAD83 or WGS84.

I'll entertain questions on-list or off, regarding geodesy and 
cartography.

Be prepared for a pop quiz next class period.

73, Gerry N5JXS
-- 
Gerry Creager -- gerry.creager at tamu.edu
Texas Mesonet -- AATLT, Texas A&M University	
Cell: 979.229.5301 Office: 979.458.4020 FAX: 979.847.8578
Page: 979.228.0173
Office: 903A Eller Bldg, TAMU, College Station, TX 77843