The G-HAT papers have definitely been dense reading. I think Part IV (the paper I’ll be discussing in this blog post) has been the most accessible one to me so far.
My favourite table/figure in this work is simply Table 1, so I’m going to talk about it a lot.
Firstly, I think it encapsulates the fundamental challenges in artifact SETI: 1) the artifact has to exist and 2) we have to be able to tell that there’s an artifact. This sounds simple, but I think is a useful comparison to the fundamental challenges in communication SETI: 1) the (intentional!) signal has to exist and 2) we have to be able to tell that it’s a signal.
Secondly, it’s good to keep perspective in SETI research: not every anomaly is ETI. In fact, all of them so far are not. The table illustrates that for every megastructure-y looking object, there are reasonable (and plausible) “natural confounders”. This makes sense; though we’re trying to wring as much information out of a single lightcurve as we can (and we’re quite good at it – we can even tell the full 3D stellar rotation and planetary orbit geometry) it’s still just a lightcurve, and there are many inputs that produce the same output. Perspective is vital, especially in a paper that is ~hunting for alien megastructures~
Thirdly, I just love lists, as you can probably tell by the way I write my blog posts. The section following the table goes through the six physical “Distinguishing Features of Megastructures” (ex. anomalous masses, aspects, or orbits). The section after that talks about the nine physical “Confounding Natural Sources of Megastructure Signatures” (ex. starspots, ring systems, or non-transited stars in the field of view aka. “blends”). I’m italicizing the word physical to illustrate exactly what it is that I like about the structure of this section: it shows what distinguishing properties of the systems being observed would be visible to an observer within the system. But we are not within the system – we’re working from lightcurves. And that’s where Table 1 comes in: showing exactly which of the physical properties discussed in (perhaps agonizing) detail would cause which of the 10 lightcurve anomalies in the Table.
I will now briefly summarize the rest of the paper, which I found generally less interesting to me. The next section talks about a few objects in particular that show some of the transit anomalies discussed in the previous section. The section after that discusses how to distinguish a signal beacon from a constant source from an information-rich signal by doing statistical analyses in both the frequency and time domains. The authors quantify it with a “normalized information content statistic”. I’ll admit that the methods in this section were mostly over my head, but I think (hypothetically) that the uniform application of them to future SETI studies would be a fruitful pursuit.