Working around lots of insanely brilliant Ph.D. types, I occasionally run into ideas that wouldn't occur to me.
Tonight, there's a total eclipse of the moon, with the "exciting" parts starting in less than 20 minutes. Great! Like many people, I like eclipses. Like many people, I think, "I'll just use my zoom lens to take pictures of the moon at various points during the eclipse. That'll be fun."
Insanely brilliant Ph.D. types, on the other hand, submit proposals for telescope time, with titles like "Refined Measurement of Earth's Transmission Spectrum through a Lunar Eclipse." And the proposals actually get accepted, by a telescope with an 8.2-meter mirror and the highest-resolution visible-light spectrograph on any large telescope in the world.
Yes, kids, if you grow up to be an astronomer, you can actually talk people into letting you point huge telescopes at the moon. Cool, huh?
But the insanely brilliant Ph.D. types aren't taking pictures - they're taking a spectrum of the light being reflected off the moon, at various stages during the eclipse. The're using the 8.2-meter Subaru Telescope, with 11.6 times the light-gathering area of the Hubble Space Telescope, and its High-Dispersion Spectrograph, the highest-resolution visible-light spectrograph on a large telescope anywhere in the world.
And to make it even cooler, they're not taking a spectrum of the moon, or even of the sun (since the moon merely reflects sunlight) - they're taking a transmission spectrum of Earth's atmosphere! When the moon is totally eclipsed, the only light reaching it has been bent through the Earth's atmosphere.
Comparing a spectrum of that light with the known spectrum of ordinary sunlight, and with a spectrum of light coming straight into Earth's atmosphere from a well-known star, will enable them to figure out which features of the spectrum are specifically due to the light passing through Earth's atmosphere. And those spectral features will tell them which chemical elements are in Earth's atmosphere, and how abundant they are. It could even give an estimate of how polluted Earth's atmosphere is, on average.
This technique is applied all the time to distant exoplanets that transit in front of their stars, to get some idea of what their atmospheres are like. But applying it to study Earth's atmosphere, using a telescope on earth, is a pretty neat trick, and definitely an interesting approach to a lunar eclipse.
As for me, I'll just keep taking pictures, and post some when it's over!