gps accuracy

[Mark Sylvester <msylvester@gn.apc.org>]

> > From: saba@dge.inpe.br

> > Hi all,

> > Does anybody know how precise would be the GPS in locating a stratospheric ball
> > obaloon at the altitude of 37 km. Could it be less precise? Does anybody know w
> > here could I find this information? 

The satellites for the gps are pretty high up - the number that 
comes to mind is 20 000 km - so there should be no essential 
difference between performance on the surface of the earth or 37 
km up. It is possible that the receiver uses some calculating 
tricks that assume that the receiver is near the surface, but 
then the system apparently works on aircraft at 10km altitude...

Just how far your balloon travels may be relevant if you have to 
use differential correction. This is necessary if you want 
better than the 150m error that SA imposes, and can get you down 
to 1-5m uncertainty. The problem is that if the balloon covers 
100s of km then correction data from a single base station may 
not work. Mind you, if it does travel that far then I guess you 
won't mind a 150m error.

This information is correct to the best of my understanding! I had 
a hard time finding authoritative information about gps myself, 
and wasted a lot of time with misinformation, some of it from 
engineers involved in marketing the products. In the end I have 
found you need to check with someone who is using gps for the 
application you have in mind. 

I would worry about retrieving the expensive hardware when the 
balloon comes down!

Mark.

Mark Sylvester, Duino, Trieste, Italy.