So I read through this paper and I would encourage everyone else to do so as well. It's very well-written and should be readily understandable by a non-technical audience. For those who don't want to, I give a brief summary and some comments below.
First,
the author. Definitely not a crackpot. You should take him
seriously seriously.
The short version is that there are photographic plates of the "alien megastructures star" (hereafter AMS) dating back over a century. The brightness variation of the star had not been previously measured from this data set since measurements using photographic plates are extremely tedious and time-consuming compared to modern methods from CCD imaging, and there was no good reason to select this particular star for analysis. In any case, these weren't taken often enough to discover the huge rapid dips in the star's brightness, which occur on a timescale of days.
But given Kepler's discovery, it's worthwhile to look at the historical data to see if the AMS was also doing interesting things on a longer timescale. Hardly anyone looks at photographic plates any more, or even knows how to analyse them correctly. Fortunately Prof. Schaefer is one such person.
Schaefer went and looked at the original photographic plates and measured the brightness of the star over time. To check his procedure and for any possible systematic problems with the data he also looked as several other stars. He found that the check stars stayed essentially constant in brightness whereas the AMS brightness decreases by far more than can be explained by errors in the analysis or problems with the data - by an amount which has never been seen for a star of this type before.
He also looked at the brightness estimated from digital scans of the plates, which previous authors have dismissed as showing nothing of interest. His findings are different, claiming that actually the scans support his visual estimates from the original plates. He does not comment on why the previous researchers did not find this.
According to Schaefer, if this dimming was to be explained by comets it would require a ludicrous number - something like half a million or more each over 200 km in diameter. Even explaining a single Kepler dip requires 36 such giant comets, whereas the largest comet known in our own Solar System is just 60 km in diameter.
However, I disagree with Schaefer on one point. He argues that since the AMS is showing two extraordinary behaviours - dimming on two very different timescales - we should, by Occam's Razor, search for a single common cause. Thus since it's patently ridiculous to invoke comets as the cause of the century-long dimming, they can't be the cause of the more rapid, random dimming either. I would argue that if we see two behaviours occurring on such very different timescales, it doesn't make much sense to argue for a common cause - there's no good reason to think that a process which works over a century should be connected to one that works over a few days. Which is
not to say that comets look like a very likely explanation, mind you.
The paper has been submitted to the
Astrophysical Journal where someone who knows a lot more about stars than I do will be able to do a proper analysis.
http://arxiv.org/abs/1601.03256?utm_content=buffer78842&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer
