This paper discusses the value of conducting a search for extraterrestrial intelligence by searching for seemingly impossible astrophysical phenomena. The advantage of this approach is two-fold: in the absence of discovering an extraterrestrial civilization, it may be possible to uncover unknown exotic physics or astrophysical processes.
The specific goal of this study is to query the sky for “disappearing stars” – the idea being that advanced civilizations around these stars could either be harnessing the stellar energy (which would then be re-radiated as waste heat) or intentionally cloaking themselves. This task is accomplished by cross-matching the United State Naval Observatory B1.0 catalog (USNO-B1.0) and the Sloan Digital Sky Survey (SDSS). Objects of interest include those which appear in USNO-B1.0 but not in SDSS, and, if confirmed, would identify stars that have disappeared over the decade lag between these two surveys.
An initial query yields several thousand candidates – stars appearing in USNO-B1.o but not in SDSS. However, this could occur for a variety of reasons significantly more mundane than an advanced extraterrestrial civilization, e.g., diffraction spikes or image artifacts in the SDSS frames which cause the SDSS pipeline to overlook the star. A visual inspection of the initial list of candidates reduces the target sample to 148 objects.
These 148 candidates must be combed for false positives (appears in the USNO-B1.0 catalog when it should not) and false negatives (appears in SDSS, but is not detected.) Most of the candidates fall into the former case, where no object appears in the images that were used to generate the USNO-B1.o catalog. A large fraction also fall into the latter case, where the stars are present in the SDSS frames but the coordinates between the two systems do not agree. In the end, one suspicious candidate is identified. It appears in two separate images used for generating the USNO-B1.o catalog, but cannot be found in the SDSS images.
This study is very good in the sense that it reports the null result as an upper limit. It starts with a clearly stated problem, addresses the problem in a clear way, and translates the result into an upper limit. As acknowledged by the authors, the study critically suffers by probing only a small fraction of the sky, and by probing only a short period of time (~10 years) during which the star could have disappeared. The authors look forward with great anticipation to larger surveys (e.g., GAIA and LSST) that could extend this search to much larger scales. They also suggest that much of this work could be efficiently accomplished by employing citizen science, which has proven itself extremely beneficial to other astronomical projects.