Re: Is that really the best place to build these things?
Yes, I'm afraid you are really fumbling in the dark on this one ;-)
1) The protons in the LHC are sent around the ring in discrete bunches of about 10^10 particles, and actively guided and focused around the circuit by a series of dipole, quadrupole and sextupole magnets. If the ring gets "distorted" over time by geological movement, it doesn't matter - the circle isn't perfect anyway. Heck, it's not even flat - the side nearer the Jura mountains is lower than the other side of the ring. Pretty much all distortions to the ring require real-time magnet adjustments, and even the previous experiment that shared the same tunnel (LEP) could detect the distortions made by the orbit of the moon around the earth, and even the weight of snow + water on the Jura mountains as the seasons passed. The Deutsches Elektronen-Synchrotron (DESY) in Hamburg can even easily detect the "Cultural Noise" of the working week, and how then energy of busy humans in the city above becomes less from Monday to Friday as people get more tired through the week. The relevant plot is here: http://i.imgur.com/WGP661R.png and there is more at http://vibration.desy.de/
2) The effects mentioned in 1) are all important when the particles are circulating for long periods of time (hours) in the storage ring; however these effects are unimportant within the detector volume where particles are only spending a few nanoseconds before either decaying, stopping, or escaping. Gravity really doesn't come into play within the detector volume - if it did, we'd have probably observed the graviton by now, which we haven't. These particles are moving just shy of light speed, and the gravitational differences caused by the odd mountain whilst tracking the particles through the detector volume (about the size of a large 6 story building) are just super-negligible. Errors are more likely to be dominated by the accuracy/calibration/alignment of the various detector components, and the ability to accurately determine the location of the bunch crossing within the detector - figuring out where two bunches of particles moving at light speed actually "meet" is not trivial.
3) Australia, etc, might be great in many ways, but it's *really* inaccessible. Given the thousands of people involved in these experiments, it's not really very practical to put an experiment in the middle of nowhere. Astrophysics suffers this problem with some of their crazier experiments, which really do need square-kilometers of surface detectors.
Anyway, I hope that brief taster helps you to understand just how much physicists think about these things, and how much they have already learnt from the experience of running these incredibly large-scale experiments :-)