The Lingam and Loeb paper “Natural and artificial spectral edges in exoplanets” contains a lot of elements that I really appreciate in a theoretical SETI paper.
The primary conceit of the paper is that, as vegetation produces a reflective spectral “red edge”, artificial materials could produce similar reflective spectral signatures. These signatures, described in the paper as a distinctive change in reflectivity over a narrow bandwidth, could be detectable if they covered enough of an exoplanet’s surface.
The authors find that silicon produces one of these “artificial edges” in the ultraviolet. Even 10% coverage by silicon could be detectable in certain circumstances (tidally-locked planets around M-dwarfs) with next generation telescopes like WFIRST and JWST. Given an assumption about the composition and reflectivity of the material, you can easily get the coverage fraction. From that fraction, you can guess their power usage (this is assuming that the silicon signature is produced by large-scale photovoltaic arrays).
The element I most appreciated in this paper was the authors’ clear statement of assumptions and acknowledgements of alternatives and difficulties in their methods. They make it clear that their calculations are predicated on the idea that the civilization is getting their power supply from their host star, not from geothermal energy or nuclear fusion (which they state would cause other signatures, but they don’t try to predict them in this paper). They add caveats that would affect the signature’s detectability from Earth: hazy atmosphere on the planet, strong winds, and cloud cover. They give solid reasons for choosing silicon as the element to focus on in the paper, based on nucleosynthetic abundance, but also show some spectral signatures of other plausible bases for solar panels. Finally, they considered false positives that could also cause a similar signature, in particular a natural material called enstatite*, and ways to differentiate the two (eg. looking for energy redistribution on the surface of the planet from dayside to nightside).
All in all, I found this to be a very convincing and self-aware paper, and I’m very excited to see an artifact-search like this be conducted in the next decade or so!
* Some facts about enstatite (because part of me always wanted to be a geologist)! It’s a common mineral found in igneous and metamorphic rocks and is a 5-6 on the Mohs scale. It’s essential in some Earth mantle materials and is commonly found in asteroids. It has, in fact, been found in crystalline form in some planetary nebulae.