I wanted to estimate the electric field near the earth's surface. I wanted to use the generalized ohms law. I have an estimated electric conductivity and current density. Please look into the attached file for a full description of the problem.
I want to remark in forward that I am no specialist in atmospheric physics. Nevertheless the problem is an interesting case. We have the following boundary conditions:
-The soil (layer1) is charged negatively.
-The ionossphere (layer2) is charged positively
-The electric field near the surface is relativly stable and amounts to ~150 V/m
The layers conductivity is high comparable to the atmospheric resistance - therefore you have a quick charge exchange within the layers- if charges are generated, they distribute quick on the surfaces so that the spot of generation is unimportant.
The generation of negative charges is obviously the existance of thunderstorms. This generates the separation of charges. Vice verca you have a current of ions to the surface which neutralizes the charges. The constant electric soil field means, that the effective current density must be equal to zero - due to the 2 components of separation and recombination of charges.
This problem is comparable to the problem of a p-n-junction within a diode: You have a "constant" (with respect to time, not with respect to location) electric field. This field is due to a drift current of charges in the neighboring regions and an compensating diffusion current into the opposite direction. This produces a time independent local electric field.
What effects the ionization of solar wind? Of course, it enhances the positive charge and therefore it deflects other positively charged particles coming from sun. So this positively charged layer protects the inner atmosphere from ionization. The generation-ionization process works independently therefore.
If I can help, I want to do this. But I am no specialist in this field. Concerning the extraneous currents - here the influence of the magnetic field must be taken into account. Concerning the electric field you are interested in: I would recommend to solve the Poisson equation. This ought to be possible because the stationary charge distribution is known.