I had an odd thought earlier, whilst typing an e-mail.
Just what is the average metallicity of the galactic disk?
I’m not sure, but I have a suspicion that it might be at least a little bit subsolar. Why is this bothering me? Well, I have a suspicion this could impact on the Fermi Paradox. My thought goes something like this…
Consider a somewhat metal-poor star, say [Fe/H] = -0.3 – that is, about 50% the heavy-element abundance of the Sun . Let’s assume its protostellar disk shares the same composition – not wholly unreasonable, since they formed in the same place! Presumably, the star’s relative lack of heavy elements will be reflected in the planets that it forms.
Now, first off, you won’t get as many planets around metal-poor stars; planet-formation seems to correlate with stellar metallicity. That said, you do get at least some planets around low-Z stars. But, until relatively recently in galactic history, most stars will have been ‘low’ metallicity. The interstellar medium didn’t originally contain metals; heavy atoms got there as a result of pollution by supernovae and red giants, and that process takes time. And the galactic disk still contains plenty of metal-poor regions even now. (A star having a subsolar metallicity is not a guarantee that it’s older than our Sun – something that’s irritatingly inconvenient for my own research!)
Anyway, I can imagine terrestrial planets existing around our [Fe/H] = -0.3 star. I can even imagine them having a reasonable amount of carbon, oxygen, nitrogen and hydrogen, so the absolute-minimum basis for CHON life is there. Probably there’s at least a dash of other stuff too. But what I imagine it particularly lacking is big, heavy atoms – like rare-earth elements and so on.
Supposing life gets started on such a planet. Suppose even some sort of civilisation gets going. But presumably you’re not going to be able to have much of an electronics industry without metals or rare-earth elements? And if building even just a radio is difficult – let alone TVs, radar etc. – then you’re not going to be leaking much EM radiation into space. And without much leakage, you’re certainly not going to be detectable at interstellar distances.
And if the galaxy as a whole is, on average, metal-poor relative to our Sun, does this also mean that advanced electronics industries are also rare? Food for thought…
 In astronomy, a ‘metal’ is anything heavier than helium – i.e., something that couldn’t have been formed during the Big Bang. The [Fe/H] scale is a logarithmic scale, based on the relative levels of iron as compared to hydrogen; [Fe/H] = 0.0 is 100% solar, i.e. 10^0 = 1. [Fe/H] = -1.0 is 10% of solar and [Fe/H] = +1.0 would be 10x the solar content of metals. In practise, the highest values of [Fe/H] seem to be about +0.5ish. The record object for lowest metallicity, however, is HE 0107-5240, with a breathtakingly-low metallicity of [Fe/H] = -5.2.
 In stars with so-called ‘solar’ abundances, you can use the amount of iron as a sort of standard calibrator for other elements. In less well-behaved stars, those with non-solar abundances, this breaks down, and you have to be a bit careful. In these cases it can be more useful to use [M/H], which is exactly the same logic as Fe, except that M is the total amount of metals against the total amount of hydrogen.