I guess this doesn't tell us where the error really is coming from, just (as usual) that error will be present in any navigation.
"The datum (WGS84 vs. NAD27, for example) usually only makes a couple hundred feet difference, and the electronic chart "informs" the chart plotter of its datum." I've personally been in well-surveyed areas where large items (anampitheatre) were over 1/2 mile out of position in NAD27, where they were last surveyed on a USGS topo, versus their actual position under
GPS. The difference depends on many things, among them the location and the quality of the old survey--some of the old baselines just weren't good but pre-GPS simply accepted as 'good enough' for many years.
"Also, the land was where the
GPS claimed it was before and after visiting the Marietas....If you compare the
charts (electronic and otherwise) with Rains's Mexico Guide, you can see that there is a difference in the lat/lon between the two." So, the
GPS and
radar and guide book agree, but the marine
charts (from what source?) were off?
That wouldn't surprise me either. I've seen most of the commercial private mapping (Navteq, etc.) out by 1/3 to 1/4 mile in too many places. Daylight, depth sounders, local knowledge, and the occassional slam tack and about aren't dead yet.