Mxsmanic wrote:
Marco Leon writes:

I thought the primary factor in determining the differences between
baro altitude and GPS altitude was the difference between "sea level"
and the WGS84 datum plane at the particular location?

In theory, yes. But in practice there is also a lot of inaccuracy in
altitude readings via GPS. The system is not designed to provide
highly accurate altitude, as the original users (soldiers in the
field, ships) didn't require it.

Altitude has always been very important to field soldiers and "ships"
whose responsibilities include flinging large projectiles long
distances. Among the multitude of variable factors of which indirect
fire requires calculation, the accuracy of the weapon and target
constants x,y and z are most important. It's tough to hit targets
12-20 miles distant if you don't know the difference in altitude
between weapon and target.

GPS, being a military system, was finally derived (IMO, and
suprisingly, opinions vary) by combining the technologies being
developed by the Navy (Project Timation) and the Air Force (System
621B).

Since both services incorporate aviation and one goal of military
avaition is to operate in "all weather" the ultimate goal of GPS/WAAS
altitude accuracy is to permit pilots to land safely in zero-zero
visibility without the need, expense (and redundant technologies) and
limitations of the present ILS.
See: http://gps.faa.gov/programs/waas/que...swers-text.htm

The altimeter/GPS altitude accuracy debate is beyond my expertise but
my experience includes a different altimeter indication after landing
at an airport I had left earlier in the day while my handheld $150 WAAS
capable GPSr indicated its identical reading to 1/10ft.

Barometric altimeters are analog and the finest divisions of the scale
IME are 20', so I would not expect them to be any more accurate than +-
10'. I think WAAS combined with DGPS will soon provide digital
accuracy... far beyond any ordinary need.

"Both horizontal and vertical changes in position can be measured to an
accuracy of a few millimeters (horizontal) to several millimeters
(vertical).
http://volcanoes.usgs.gov/About/What...ation/GPS.html

The GPS altitude accuracy "problem" is mostly attributable to the
limitations and generality of the WGS84 datum geoid height
http://www.colorado.edu/geography/gc...gif/geoid2.gif
which could have easily been overcome by WAAS correction and simply
adding more satellites (and postioning them optimally). Economics have
precluded this rush to solution.

The bottom line is, IMO, WAAS altitude is far more accurate than
altimeter... depending on how many birds are in view. 4 (depending on
positional diversion and attitude relative to the receiver), I'd trust
GPS if my life depended on it.
-----

- gpsman