There's a question at the heart of SETI that doesn't get nearly enough attention. It isn't whether aliens exist and it isn't whether we have the technology to detect them. It's a far more practical problem that, with a billion stars in our Galaxy and finite telescope time, how do you decide which ones to actually listen to?

Most SETI programmes answer that question with proximity. Point the dish at the nearest, brightest stars and hope for the best. It's a reasonable strategy, the closer the source, the stronger the signal but it does ignores something fundamental. Not all stars are equally good candidates for hosting complex life. Some burn too hot and die too quickly, some are too young while some are chemically impoverished in ways that make planet formation, let alone evolution, deeply unlikely. Listening to those stars isn't just a long shot, it might be a complete waste of time.

There are up to 400 billion stars in the Milky Way. As this image of the star-forming region LH 95 suggests, searching for life among them makes looking for a needle in a haystack seem straightforward (Credit : ESA/Hubble)

That's the problem a Turkish high school student named Sahin Torlakcik decided to tackle. The result, just published in a peer reviewed journal, is the Torlakcik Catalog, a rule based filtering system applied to nearly 1.75 million stars from the Gaia space telescope's data archive. The goal wasn't to find the best candidates. It was to systematically identify the worst ones and set them aside.

Using seven stellar parameters, the model highlights stars unlikely to host complex life and removes them from the search list. Stars more than one and a half times the mass of our Sun burn through their fuel too fast and since you need at least a few billion years for complexity to evolve, the heavyweight stars simply don't live long enough. Stars younger than three billion years face the same problem from the other direction, life on Earth took roughly that long just to get past single celled organisms. Stars with too little iron and other heavy metals are poor candidates for forming rocky planets in the first place. Binary systems (two stars orbiting each other) create gravitational environments that can destabilise planetary orbits entirely. Throw in photometric variability and chromospherically active (flares and other such active regions in the chromosphere) red dwarfs blasting their planets with radiation, and you have seven scientifically grounded reasons to say - not that one.

Artist impression of a red dwarf (M dwarf star) that, according to a study by high school student Sahin Torlakcik, are a good candidate for the search for life (Credit : NASA/Walt Feimer)

Applied to the full sample, the model excludes roughly 55 per cent of stars, leaving 777,835 high priority candidates. Age and metallicity do the heaviest lifting, each responsible for cutting around 29 per cent of the total. The retained population is dominated by K dwarfs and quiet M dwarfs which are older, stable, chemically rich stars that give complex life a fighting chance.

Stellar ages from Gaia carry substantial error bars, sometimes spanning billions of years, and how Torlakcik handles that uncertainty is one of the most interesting choices in the study. Rather than apply a hard cut at three billion years and discard anything that looks younger, Torlakcik applies the threshold to the upper age bound instead. The effect is dramatic and a whopping 355,086 stars that a blunter approach would have eliminated are saved. It's a small methodological decision with a large practical consequence.

The catalogue was also cross matched against the Breakthrough Listen programme's primary target list which is the most prominent radio SETI survey currently operating. More than half of matched Breakthrough Listen targets would be flagged for exclusion under these habitability criteria, mostly due to low metallicity. That isn't a criticism of Breakthrough Listen, instead it reflects two genuinely different philosophies, one optimising for detectability, the other for plausibility. The two approaches are complementary, and using both together produces a shortlist of stars that are both close enough to hear and promising enough to be worth hearing from.

The catalogue and tool are all freely available online so that any telescope programme in the world can now apply this filter to any stellar dataset it chooses. Not bad for a high school project.

Source : Where Not to Look: A Parametric Avoidance Model for SETI Target Selection