Topband: Modeling "Ground" and losses

[Markku Oksanen]

All
As far as I can understand, wires / radials just under the surface of the ground are simply conductors embedded in a lossy dielectric (with an interface to air near it), similar to carbon filled plastic or other such material.  If the loss and the dielectric constant is low, the wire acts more like free space antenna conductor with standing waves made possible by actually having energy that can be reflected back from the open end. If the loss is high, nothing remains to form a standing wave as the energy in the wire dissipates such that only (mostly) a forward wave is present and no standing wave can be observed.
The reason why this might be difficult to understand is that there are some three factors can be different:  The dielectric constant, the conductivity and the spatial variations of these both in depth and along the surface of the ground and then the geometry of the multiple wires in this medium and the corresponding coupling(s) between the wires.  
The detailed predictions for current profile would seem most beneficial if a symmetric field very large number of very long radials is not possible for space or cost reasons.  As I understand, the primary loss mechanism for ground mounted vertical systems is EM field penetrating the lossy material below.  To lower this loss, one needs to prevent this ground penetration.  Building a "shield", a large, dense radial field is a way to do this.  However, this doesn't change the fact that loss associated with one radial wire changes with current and current profile on it!   I guess this is at the core of the debate here.
I find it remarkable that we have now free tools to solve EM-problems that took computing center size efforts when I got started with this hobby at late 70's.

MarkkuOH2RA/OG2A/WW1C