Earlier I said a volume of soil between electrodes behaves as a
capacitor in parallel with a resistance.

To conform better to the impedance-frequency response of real soil, a
better simulation is obtained with another resistor in series with the
capacitor. Better still add a 3rd resistor in series with a second
capacitor, both in shunt with R1, C1 and R2.

I have a computer program somewhere which assists in designing a
circuit to simulate a given type of soil. But what use such circuits
might have is a matter for conjecture.

It is more convenient and practical to work in terms of resistance
rather than the scientific term conductivity. Ohm-metres rather than
milli-Siemens. When thinking in terms of conductivity I always feel I
should be standing on my head.

The resistance measured between opposite faces of a 1-metre cube of
the soil is 1000 ohms when the soil has a resistivity of 1000
ohm-meters.

A poor soil is 1000 ohms = 1 milli-S Sea water is 0.22 ohms.

One can visualise a 1 metre cube of the material. The permittivity of
the material being the nunber of times the measured capacitance
exceeds the calculated capacitance between the electrodes when only
air is present. It's about 9 pF.
----
Reg, G4FGQ