Barometric altimeter and elevation accuracy a bit of data

I recently got a GPSMap60CSx which has a barometric altimeter which the unit
uses to aid in altitude determination. From the best I have been able to
tell, in the absence of manually calibrating the altimeter with known
elevation, the receiver compares the barometric data with the gps elevation
data over some period of time and thereby derives a barometric calibration.
The result is supposed to be improved elevation accuracy and reduced scatter
in elevation data.

On a recent trip, the jeep trail I was on came within a few feet of a USGS
vertical control station, so I did a little comparison between my Geko 201
which I placed on the benchmark and the 60CSx a few feet away on the
windshield of my Jeep. The benchmark was at 1685 feet (NAVD88). Using
tracklog data from the two gps units I found the following:

The Geko gave an average height of 1668 ft; an error of -17 ft with a
standard deviation of 8 feet using 7 datapoints.
The 60CSx recorded 14 datapoints that gave an average elevation of 1678 feet
after correcting for the height above the benchmark for a -7 ft error and a
standard deviation of 2 ft.

WAAS was off on both receivers. I attempted to use WAAS on the Geko, but
was unable to receive the correction signal.

This is only one very simplistic experiment, but it may give some idea of
the usefulness of having the barometric altimeter. I only thought to do
this analysis after the fact, or I might have taken more data.

--
Tom
http://home.att.net/~tbharvey/