[pct-l] Follow-up to the earlier GPS discussion

[Gary Schenk]

The GPS system calculates the position of the receiver relative to the 
center of the earth. Then a coordinate position is derived based on a 
mathematical model of the Earth.

The vertical coordinate, NAVD 88 elevation, is derived on the position 
of the receiver relative to the equipotential surface, instead of 
relative to sea level as the NGVD 29 datum. That is a surface where 
gravity is the same everywhere. It is a very uneven surface. For 
elevations to be fairly accurate, it has to be well defined.

In, say, Los Angeles, it is very well defined and surveyors can obtain 
sub-centimeter accuracy with GPS. In the heart of the Sierra Nevada, it 
will not be as accurate because the equipotential surface has not been 
as densely mapped as LA.

In southern California the elevation difference between the old datum 
and the new one is around 3 feet in most places.

I hope you are all totally confused now. :-)

They have done amazing things with the software on cheap handheld GPS 
receivers and they can obtain results that instruments costing tens of 
thousands could not obtain in the not too distant past.

Still there are limits, but a hiker isn't going to really notice much in 
the way of error in her GPS. Those errors are glaring to a land 
surveyor, but of no consequence to hikers.

Gary

On 5/19/2010 7:39 PM, Carl Siechert wrote:
> I don't get this. Putting aside my question about how gravity is
> mapped...Where is this elevation mapping data stored? In the receiver?
> And with all the other mapping minutiae stored in most receivers, why
> can't the elevation be more accurate? Btw, in limited observations with
> automotive receivers, I haven't noticed that elevations on a (presumably
> economically justifiable) interstate highway are any more accurate than
> those seen far afield.
> I always assumed that vertical position was calculated on the fly by
> triangulation, just like lat/long. And I further assumed that the reason
> for the usual error is simply that the high angle of the satellite
> signals doesn't allow for as much precision when triangulating on the
> Z-axis. Am I lost?
>
> On Wed, May 19, 2010 at 2:44 PM, <gwschenk at socal.rr.com
> <mailto:gwschenk at socal.rr.com>> wrote:
>
>
>     Vertical position is based on mapping of the earth's gravity. This
>     mapping is densified in more urban areas where it is economically
>     justified. In remote mountain regions it is not considered as
>     important. And so, every so often you will read where someone will
>     complain that the elevation of Whitney is wrong, because his
>     Magellan says so!
>