Between 2005 and 2009, a team of astronomers using the European Southern Observatory in Chile’s Atacama Desert made a discovery: the first ever potentially habitable planet outside our solar system. It was orbiting a star called Gliese 581, in the constellation Libra, 20 light years away. Then, in 2012, another potentially life-supporting planet was found, this time with data from the Keck Observatory, near the top of a dormant volcano in Hawaii.
But these planets may not actually exist, according to a paper published by astronomers at Penn State University and the University of Texas. The study, which came out in the journal Science on Thursday, has added yet more fuel to an intense debate that has raged for the last 10 years.
Astronomers discover new planets by looking for changes in starlight. Sometimes, the gravitational pull of a planet will make a star wobble. As the star moves, the wavelength of its light changes, explained Suvrath Mahadevan and Paul Robertson, two of the study’s authors.
“If the star were moving towards you, that wavelength would get shifted towards the blue, but if it were moving away from you, it would go towards the red,” said Mr. Mahadevan.
Yet that shift in wavelength is not necessarily caused by the pull of a nearby planet. The surface of the star is boiling with activity, the gas rising and then sinking again. It can be easy to mistake this “stellar noise” with the movement of the whole star.
The discoverers of GJ 581d and GJ 581g saw a change in wavelength, and attributed it to the pull of these planets. But in this new study, Mr. Mahadevan’s team found that the flickering caused by stellar noise was almost identical to that, which made the discoverers infer the presence of planets. They were unable to distinguish the effect of these planets from the changes that were going on within the star itself, making them doubt that the planets existed at all.
“It’s bittersweet,” said Mr. Mahadevan. “These planets that were very promising don’t exist.”
His disappointment is echoed by many, because these were “Goldilocks planets,” situated in the swathe of space around a star where it is neither too hot nor too cold for life to exist.
Sara Seager, an astronomer at MIT, explained that the letdown would be greater in the case of GJ 581d. “It was one of our favorite planets. It motivated people to consider planets quite different from Earth as potentially habitable,” she said.
Gj 581g is less of a letdown because astronomers already doubted its existence. “It has a very weak signal,” she said. “It made a big news splash, but the way it was discovered wasn’t completely airtight.“
Skygazers should not mourn the loss of these planets yet. Debra Fischer, an astronomer at Yale, thinks very highly of the new study, but for her it does not prove that these planets don’t exist. ”It’s the interpretation of the data that the Penn State team is calling into question,“ she said. ”People have to build new instruments and find out what’s really going on.“
Both Ms. Fisher and Mr. Mahadevan are doing exactly that. His will try to cancel out stellar noise; hers will try to better understand it. And for Penn State’s Paul Robertson, one of the authors of the new study, the finding is a step in the right direction. By subtracting the stellar noise, the team was able to confirm the existence of other planets.
Eric Boodman: email@example.com or 412-263-3772. @EricBoodman