As a theoretical matter, if you have a wire with a high resistance fault to ground:
you would measure the resistance to ground from either end.
This resistance is the sum of the wire resistance, the soil resistance, and the fault resistance.
If you assume that the soil resistance and the fault resistance are constant, then the difference between these two measurements is the wire resistance to the fault from either end.
You know the resistance per foot of wire, so you calculate distance from the wire resistance.
As a practical matter the above theory is pure bovine excrement. The soil resistance values are not constant. The fault resistance is probably not constant. And the wire resistance is so much smaller than the other two values, so the noise on your megohm measurement will swamp the milliohm wire resistance values.
If you have a solid wire to wire bolted fault then you can plausibly locate it with milliohm measurements.
There are better tools for the job.
One approach pulses high voltage into the wire, then you walk along the ground with a probe that detects the leakage.
Another approach uses the fact that signals travel in wire at a known speed, and is essentially radar but in the wire.
See the 'Fault Finder' mentioned by
@ptonsparky or 'Time Domain Reflectrometry'
Jon
