Exoplanets once trumpeted as life-friendly may not exist

Two planets considered among the most promising for life outside the solar system don’t exist, scientists report July 3 in Science. The signals embedded in starlight that were attributed to the planets may instead have been caused by the changing magnetic activity of their star, Gliese 581.

Although the study isn’t the final word on these enticing yet controversial worlds, scientists say it reinforces the need for meticulous analyses to separate planets’ signals from those generated by spots and flares on stars. “This is a big warning concerning the interpretation of [small] signals as being planets,” says Stéphane Udry, an astronomer at the University of Geneva.

Located about 20 light-years away in the constellation Libra, Gliese 581 is a small, dim star similar to tens of billions of other stars in the galaxy. Yet it has grabbed astronomers’ attention due to claims that two very intriguing planets orbit it. Gliese 581d, discovered in 2007, was the first reported exoplanet on which life seemed plausible (SN: 5/23/09, p. 11). Then in 2010, Gliese 581g grabbed headlines when codiscoverer Steven Vogt of the University of California, Santa Cruz said, “The chances of life on this planet are 100 percent” (SN: 10/23/10, p. 5).

A lot of controversy has swirled around the planets of Gliese 581, particularly g, since then. Multiple studies have found no evidence for Gliese 581g, though Vogt’s team stands by its claim. Meanwhile, one study questioned the existence of Gliese 581d. Without planets d, g and a farther away, cooler planet that the new study also questions, Gliese 581 would host just three planets, all of which are too hot to support life.

Penn State astronomer Paul Robertson says he had no horse in the race when he decided to analyze this confounding star. He looked at data from previous studies using the radial velocity method of planet discovery, in which the subtle tug of a planet’s gravity causes the star it orbits to wobble. Astronomers can measure this wobble by studying starlight, which shifts color slightly depending on the direction of the wobble. Robertson set out to determine whether magnetic activity on Gliese 581’s surface, a phenomenon similar to sunspots, could obscure some of that light and cause astronomers to incorrectly measure the wobble and perhaps mistakenly detect planets.

Unlike previous groups that had looked at the data, Robertson and colleagues studied light emitted by energized hydrogen atoms as an indicator of Gliese 581’s magnetic activity. The researchers found a clear link between the amounts of measured wobble and magnetic activity on the stellar surface. When Robertson and his team removed the signals thought to be from the star’s surface, the evidence for three planets became even stronger. But the evidence for three others, including 581d and 581g, almost entirely disappeared.

Robertson says the study is the nail in the coffin for planets d and g. Udry, whose research team discovered 581d, says the evidence is compelling but requires further analysis. Vogt and another scientist who reported 581g’s discovery did not respond to requests for comment.

Greg Laughlin, an astronomer at UC Santa Cruz who has not argued on either side of the Gliese 581 debate, is skeptical of the study, which he says analyzed only a small portion of the available measurements on the star. “This star has resisted airtight analysis for going on 10 years,” he says. “I’m not convinced that the final configuration [of planets] has been published.”

Still, astronomers agree that the study highlights the challenges of finding other Earths. With the newest telescope instruments, measurements of radial velocity have become so precise that astronomers need to agree on rigorous tests to separate the signal from the noise. “If you want to be confident you’ve found an exciting planet,” Robertson says, “then you have to do this type of analysis to make sure you’re not being tricked by the star.”