[Taxacom] Impossible geocodes

[Tom Schweich]

On 1/30/2015 10:27 AM, Curtis Clark wrote:
> On 2015-01-30 1:43 AM, mesibov at southcom.com.au wrote:
>>  DMS is definitely handy for GPS-determined locations.
> Handheld GPS units characteristically give an estimated uncertainty 
> figure. It would be computationally simple for the firmware in the 
> unit to limit the significant figures provided in a DD output to 
> roughly match that uncertainty. Too bad they don't.

I don't know about all GPS receivers, but I know that mine (Garmin 
GS76cx) operates in WGS 1984 and decimal degrees (DD).   Any other 
reading on the dial, whether it's DMS or NAD83/UTM is a mathematical 
transformation from what the instrument is reading from the 
satellites.   Sometimes the transformations are simplified and introduce 
errors. Also, my experience is that converting DMS to something mapable, 
i.e., DD, carries a significant risk of error.

A couple of years ago, we had long discussion on TAXACOM about how to 
record/handle/round/truncate/report coordinates.  I believe the proposed 
number of ways to handle coordinates was the square of the number of 
discussion participants, leaving me feeling that no number was 
trustworthy after someone had fiddled with it. As your mapping guy, my 
advice is to set your GPS receiver to its native geographic system, and 
record the coordinates your GPS receiver said, exactly as your receiver 
said it.  Later, when you want to map your data, you can round, 
truncate, or project your locations.

{As for accuracy, I went out on my bicycle and visited 7 control points 
14 times each, and learned that my GPS receiver was accurate (on that 
day at that place) to between 4-5 meters. The centroid of the points was 
shifted to the northeast about 1.5-2.0 m., something I was unable to 
account for. }

-- 
Tom Schweich KJ6BIT tas27 at schweich.com
http://www.schweich.com
http://twitter.com/schweich