I'm reading through the book Introduction To Laser Technology and just got to the section on Fabry-Perot interferometers. The book talks about the seemingly paradoxical behaviour of the optical power passing through a pair of 95% mirrors, separated by an integer multiple of the half-wavelength, not being 90.25% but instead ~100%.
I understand that this is due to wave interference, and that a resonant Fabry-Perot transmits almost all of the light, whereas a non-resonant Fabry-Perot reflects almost all of the light.
What has thrown me a little is what the book says about the optical energy in a resonant Fabry-Perot:
For a Fabry-Perot to be resonant, the separation between its mirrors must be equal to an integral number of half-wavelengths of the incident light. Such an interferometer is shown in Figure 4.12. Note here that because the mirrors are separated by an integral number of half-wavelengths, the light is exactly in phase with itself after one round trip between the mirrors. Thus, all the waves traveling in one direction (say, left to right) are in phase with each other. And the waves moving right to left will likewise all be in phase.
In this case, all the individual waves between the mirrors add together and result in a substantial amount of power bouncing back and forth between the mirrors. For the interferometer shown in Figure 4.10, about 20W will circulate between the mirrors, even though only 1W is incident on the interferometer.
Those 20 watts are constantly reflecting off the second mirror, which transmits 5%. That is where the 1W of transmitted light, shown in Figure 4.10, comes from.
A recreation of Figure 4.10 is as follows:
A recreation of Figure 4.12 is as follows:
The book explains that Figure 4.12 shows the same wave bouncing left-to-right, right-to-left, and back again.
What I'm not confused about is the transmission of 1W to the right, and I also (mostly) understand why no optical power is reflected back to the source. What I am confused about is how the 20W power figure arises; the book doesn't give much of an explanation for this. I inferred that 20W comes from the 95% reflectivity and 1W input, but the depiction and explanation don't explain how 20W of optical power appear from 1W of input.
I can almost get my head around it by thinking more along the lines of power density than just power, with a kind of energy storage in the space (cavity?) between the two mirrors, but I can't quite fully grasp it. Can someone explain this?

