QUESTION:
How do you determine the position of Pathfinder on the planet and how 
precise is that position? Is it as precise as differential GPS?
     
ANSWER from Mark Adler on July 21, 1997:
No, it's not as precise as differential GPS, but is about on par with
regular GPS.  Though it takes a lot longer to get a fix on Pathfinder, i.e.
days, compared to less than a minute on a handheld GPS receiver.

We can determine the location of Pathfinder to about 100 meters relative to
the center of the planet.  To do that we measure the distance by sending a
signal to the spacecraft, having it reflect that back, and using the time it
takes and the speed of light to get the distance.  We also measure the
velocity of Pathfinder relative to the Earth antenna in one direction by
having it lock onto a frequency from Earth and reflecting that back in
phase.  Then we can measure the doppler shift of the frequency which changes
with the velocity (like a train whistle going by).  Those two types of data
are combined with knowledge about the orbits of Mars and Earth to get the
location relative to the center of Mars.

All of that still doesn't tell us where Pathfinder is on a map of Mars,
since any map needs to be tied to a sphere (or ovoid) spinning about the
center of Mars, and our measurement of that orientation and spin rate has
limited accuracy.  However, we got lucky with Pathfinder and can see several
large features in the distance, like Twin Peaks, that we can unambiguously
identify on Mars maps and with those directly triangulate Pathfinder's
position on the map to a few hundred meters.  That will solve our map-tie
problem and will help greatly in locating future landers.

Also, if we're really lucky, we may be able to actually see Pathfinder in
the very high resolution images of Mars that will be taken by the Mars
Global Surveyor arriving at Mars this Fall.  The whole Pathfinder lander
will take up no more than a pixel or two at the 1.4 meter resolution, but we
may be able to tell that it's Pathfinder by the color of the pixel.

Mark