In the technique of radio magneto tellurics (RMT), plane wave EM radiation incident upon Earth's surface is used to measure subsurface properties. I have read a number of different sources on the technique. Essentially, Maxwell's equations couple the electric field amplitude and magnetic field amplitude of perpendicular components of the impinging wave source (I think the coupling is generally more complicated, but the plane wave assumption limits the wave orientation to a plane and thus the z-field component is zero). Now, the ratio of the magnetic field amplitude to electric field amplitude is given as the ratio of the magnetic permeability of the medium to the electric permittivity. Since most materials have a magnetic permeability almost the same as the vacuum permeability and they are static, this ratio is a proxy measurement of the impedance of the medium. Another way to think about this is that the wave induces secondary currents in the medium, which produce their own waves of the same frequency and superimpose on the original wave to reduce the amplitudes of the electric and magnetic field components respectively.
My question is this: The measurement is made at the surface of the medium of interest (the Earth-sky boundary as far as I'm concerned). In all the resources I have read, they say that because the skin depth is frequency dependent, making this measurement for a range of frequencies allows one to infer the spatial distribution of impedance. What I don't understand is how a measurement at the surface can tell you things about the subsurface, unless there are reflections. However, the resources never mention measuring reflected waves. Do I misunderstand something about light? I'm kind of thinking about causality here, because the light travels in time from the atmosphere, then interacts with the boundary and continues propagating through the medium. Is it possible that the wavelength of light determines what the "boundary" is? For example, we could call the boundary the region that extends below Earth's surface up to a tenth of a wavelength or so. Then for very low frequency, the "boundary" could be on the order of km. But it still seems to violate causality because the light is sensing stuff ahead of it which determines how it acts at the surface. I understand that the skin depth determines to what depth the wave can induce secondary currents. Do these secondary currents create waves that propagate back to the surface?
I know there are a lot of resources on RMT, and I've looked pretty hard but none address these confusions.
PS: I didn't post this on Earth Science stack exchange because I feel like the essence of the problem is physics (even though the method is used in geophysics).