Frank's wrote:

Thanks for your comments Tom, and you raise some valid points. Jerry Sevick
"The Short Vertical Antenna and Ground Radial", pp 25, 26, does state that
the procedure is accurate to within 25%, but does not provide any
independant verification of these claims. The method was developed by M. C.
Waltz at Bell Labs, but, again, nothing was ever published.

It would be interesting to develop a more accurate method. While my
measurement of 52 mS/m may not be very realistic it is evident that this
region does have a very high soil conductivity. Ground-wave daylight
reception of AM broadcast stations, with strong signals, at well over 300
miles is possible.

There are more accurate methods to calculate ground conductivity, but
what's the point? The skin depth in soil is on the order of 10 or 20
feet, depending on the frequency and soil quality. This means that
substantial current is flowing down to a few times this depth. Certainly
where I live, and I'd bet that in most locations, the conductivity is
far from uniform. So in order to know the conductivity of the soil which
is carrying current, you'd need to measure it down to several tens of feet.

Once you had that data, what would you do with it? Currently available
modeling programs assume homogeneous ground to an infinite depth. So
you'd have to choose some single value from among your measurements if
your objective is to get better accuracy from a program. But there's no
evidence that a homogeneous ground with any single value of conductivity
will behave the same as a stratified ground.

So having even an extremely accurate measure of surface conductivity at
a particular radio frequency (and it does vary with frequency) still
gives you much too little information to build even a crudely accurate
model of the actual ground in which the current is flowing.

Roy Lewallen, W7EL