The traditional method of orbit determination for a geosynchronous spacecraft is to collect occasional range measurements from a tracking station on the ground. More than one station might be used, but never more than one at a time. Scientists have known this approach would be improved by adding more tracking stations and positioning them as far apart as possible along the circumference of Earth (as seen by the satellite). Unfortunately, many factors make this arrangement impossible. GPS offers a simpler solution.
Viewed from a geosynchronous spacecraft, GPS satellites (on the opposite side of Earth) are positioned somewhat beyond the edge of the planet—providing better tracking geometry than would be possible from any location on the ground. Moreover, the orbital configuration of the GPS constellation ensures that various satellites will be seen at comparatively wide distances from each other, further improving the tracking geometry. Also, in comparison to ground-based stations, GPS satellites exhibit greater relative motion with respect to the geosynchronous satellite.
Of course, a geosynchronous satellite will rarely, if ever, have four GPS satellites in view. Thus, it cannot employ GPS signals the way a user on Earth would. But geosynchronous satellites only use one station in their traditional method of orbit determination anyway, so an absence of more GPS signals is not a deficit.