When the exoplanet Alpha Centauri Bb was announced in Nature in 2012, it was hailed a watershed moment in the search for Earth-like worlds beyond our solar system. But as eerie as it sounds, it now appears that Alpha Centauri Bb never existed.
According to new research, the closest exoplanet to our own star was probably a ghost — an artefact of an incomplete dataset.
Alpha Centauri Bb was originally spotted using the radial velocity method, a technique which looks at how the gravitational tug of a planet slightly bends its parent star. As Harvard astronomer Xavier Dumusque and his colleagues reported in Nature, the tiny wobble in Alpha Centauri B’s light signature was consistent with a rocky, Earth-sized world, hugging its parent star at an orbital distance a tenth that of Mercury’s. The planet would have been a roiling mass of lava.
But just a year later, the existence of Alpha Centauri Bb was called into question by another team of researchers, who re-imaged the star using different filters and found only weak evidence for a planet. For the past two years, the planet’s existence has remained hotly contested.
The new study, which will be published in the Monthly Notices of the Royal Astronomical Society, may put the final nail in the coffin for Alpha Centauri Bb. As University of Oxford graduate student Vinesh Rajpaul shows, it can be incredibly difficult to distinguish the faint footprints of Earth-sized planets from other sources of background noise, especially if those planets aren’t being monitored continuously. Which Alpha Centauri Bb was not.
To demonstrate how the fraudulent planet could have gotten there, Rajpaul built a computer simulation of a star. He gave his model star no planets, and only observed it sporadically. Running his model to generate synthetic data, Rajpaul found that “planetary” signatures did crop up occasionally. But these patterns were caused by things that had nothing to do with planets at all, like sunspots, or the gravitational tug of another star.
“This is really good work,” Dumusque told National Geographic. “We are not 100 per cent sure, but probably the planet is not there.”
The new findings offer a timely reminder of how critical it is to have long and continuous observations when trying to find really faint, really distant objects. Fortunately, most of the thousands of exoplanets we’ve identified through NASA’s Kepler mission won’t be subject to this particular type of false positive. Kepler was built to stare at one patch of the sky continuously, detecting the faint, periodic shadows of planets as they transit across their stars. It did so for over four years.
On the other hand, we still have to safeguard ourselves against other types of false discoveries, for instance, alien megastructures. It’s all too tempting to gravitate toward the splashy, paradigm-shattering explanation when we find anomalies in our data. It’s usually much harder to learn the truth.
[Read a pre-print of the study at arXiv h/t National Geographic]
Top: Artist’s impression of Alpha Centauri, the closest star to the Solar System. Image Credit: ESO/L. Calçada