Gerald Sylvester <
[email protected]> wrote in message news:<
[email protected]>...
> >>Also, GPS does a terrible job of measuring altitude on its own.
> > Urban myth, repeated often but nonetheless not good information.
>
> Read this as it will provide the mathematical and geographical
> theory why GPS altitudes are not accurate.
>
> http://www.esri.com/news/arcuser/0703/geoid1of3.html
>
> I do admit though that for cycling, who cares if it is off by
> 100 feet from the mean sea level.
That article is interesting but is focused on why the MODEL used for
GPS elevation is inaccurate not on the accuracy (or consistency) is
the instruments themselves. The "urban myth" I was referring to stems
from the oft repeated claim that because of geometric constraints, GPS
is more accurate horizontally than vertically. This assertion is then
commonly extrapolated into a claim that, therefore, GPS is not-at-all
accurate vertically.
Even the cited article refers to variations of "a few meters" citing
an extreme exapmale of 32 meters at one location. Of course even that
is not an outrageous error. In my experience GPS elevation seems to be
very accurate (+/- 30 feet), especially when viewed over time and
averaged at a specific location (typically +/- 5 feet) or when a
viewed continuously such as in looking at the elevation profile
created from a "track log" of a ride (see Topofusion.com) Such
profiles are surprisingly smooth even when trackpoints are recorded at
high resolution such as every 1 or 2 seconds.
If the global elevation model were actually far off then the measured
elevation would always tend toward an inaccurate reading, which it
does not. (And, of course, when I refer to "accuracy" I am comparing
to measurements assumed to be good, such as USGS topo maps)
> >>Barometric pressure is far more accurate.
>
> yes but does need to have the local barometric pressure setting.
> But in a local environment it is more accurate. If you go out
> on a 50 or 100 mile ride straight out, it will be less accurate
> as the local pressures do change and are different.
>
> > So if you don't know where you are and don't know what the barometric
> > trend has been, a GPSR actually gives far MORE reliable elevation
> > information.
>
> possibly. If you know the phone number of a local airport with
> a ATIS/AWOS you can easily get it though (see AIRNAV.COM). Or you can
> just log on to say WUNDERGROUND.COM and get it that way.
I find this comment to be very funny if only for pointing out the
difference in perspectives from which we view this. I use GPS most
commonly for biking and hiking in mountainous areas where there are no
local airports and one does not generally have the option of simply
logging on to WUNDERGROUND.COM to get barometric figures. For that
matter, even assuming the existence of a suitably close airport, due
to the varying terrain it is very common for even a cell phone to be
out of service range or simply blocked (no I can't "hear you now").
But even under such circumstances a GPSR will provide a
more-than-acceptably accurate elevation reading without all that extra
rigmarole.
>
> For flying, I'll take a Garmin 430 or 530. For riding, I'll
> just base it on what others have measured and posted on the internet.
> If I can't find it, it probably isn't a big enough climb anyways.
Flying - now that's a whole different ball game and "accuracy" can
take on a whole new, perhaps life-or-death, significance.
DR
. The short hills are ~100 ft and
