GPS query

[Ian Pearse]

Apologies if this has been looked at before. I've just been skimming through "To Rise From Earth" by Wayne Lee, borrowed from my local library. The section about GPS satellites and how GPS receivers fix their position on Earth mentions "clock skew", the positional error produced by the difference in accuracy between the satellite clocks and the GPS receiver clock. This necessitates a fourth satellite lock to correct it. It doesn't mention any differences due to relativistic effect - moving clocks go slow. Is it safe to assume the relativistic difference would be too small to matter?

[gwiz]

At the clock accuracy required by GPS, relativistic effects do have to be considered, not just the special relativity velocity effect but also the general relativity effect of being higher in the earth's gravity well.

[Ian Pearse]

I'd forgotten about the general relativity effects... out of curiosity, can someone who is up on this sort of thing estimate what sort of margins we're talking about here?

[gwiz]

From a little googling, I found this link which gives a few numbers:
www.astronomy.ohio-state.edu/~pogge/Ast162/Unit5/gps.html
It turns out that the general relativity effect is the larger one.