These graphs might add to the thread:
http://www.auroradude.com/auroralheight ... ution.html
The first plot is of several thousands of auroral height observations and the second is a graph based on these observations. I borrowed these from "The Polar Aurora" by Carl Stormer
Possibly the Bragg principle applies since a graph of it closely resembles the graph of height distribution of auroras. The fact that the atmosphere is denser at lower altitudes might only change the profile.
I might also point out the fact that ionized oxygen produces a red color at altitudes above about 100 miles and a green color below 100 miles is directly related to the density of the atoms as well as the energy of the particle stream.
The red is caused by less energetic particles that cannot penetrate the atmosphere below about 100 miles. This first excited state lasts about 110 seconds. Lower in the atmosphere a second excited state produced by higher energy particles produces the green. This process lasts only .74 seconds. Atoms here are also raised to the first excited state but are so dense that they collide within the 110 seconds required to release a red photon so the red emission is "quenched".
That there are red and green present at the same time would indicate a mixture of particles of different energies.
Next stop... the twilight zone...