re: But the sodium transition is FAR narrower
A 60 Hz sine wave has a wavelength of 5000 kilometers. There would be very little phase difference between lights even 100 KM apart (roughly 7.2 degrees). It's vitally important that the power grid maintain proper synchronization even across different suppliers, so the phase differences for a particular geographic area would be small. You wouldn't necessarily want to limit light collection to the nulls, just chop off the peaks.
Like I said, I don't know if the idea would work or if it would be practical. I'd love for someone to test it. I just find that filtering out any part of the visible spectrum is unfortunate.
