WASHINGTON — The discovery four years ago of a rocky, not-too-distant planet that seemed a prime candidate for alien life was a thrilling development in the search for other Earths – proof, it seemed, that our planet was not the only one with just the right mix of life-sustaining conditions.

But two scientists at Penn State University say they have debunked the possibility of a “Goldilocks” planet once believed to revolve around Gliese 581, a faint dwarf star that is 20 light-years from Earth.

Using a new technique they are developing to confirm the existence of small, hard-to-detect planets, the researchers say they have determined that physical changes within the star itself created the illusion of an orbiting Goldilocks planet and a second one that also does not exist.

Three other planets do exist, they said, in the Gliese 581 planetary system, but none within the star’s “habitable zone” – an area just the right distance from the star to allow the presence of liquid water.

“It’s bittersweet,” said Suvrath Mahadevan, one of the researchers who conducted the study, published Thursday afternoon in the journal Science. “We are pleased the technique works, but on the other hand, we have disproved these two planets. It would have been nice if they had existed.”

Steven Vogt of the UCO/Lick Observatory at the University of California at Santa Cruz, one of the astronomers who first reported the existence of the “Goldilocks” planet, did not respond to emails this week. Paul Butler of the Carnegie Institution of Washington, his co-researcher, declined to comment for this story.

This is not the first time the planet’s existence has been questioned.

Vogt and Butler reported their initial discovery in the Astrophysical Journal in September 2010, reporting that they had found, for the first time, something long predicted by astronomers: a planet, orbiting a distant star, with the right balance of temperature and mass to host liquid water and an atmosphere – two conditions presumably necessary to support life as we know it.

The planet, Gliese 581G, was about three or four times the mass of Earth, they said. Unlike Earth, which rotates on its axis, Gliese 581G was fixed in relation to its sun, with only one side perpetually bathed in light, they said. The most likely area for life, they said, was in a band around the planet where the dark side met the illuminated one.

“The logic now says there are lots of planets like this out there,” Vogt told The Washington Post at the time. In later interviews, he said he believed that the planet probably hosted alien life.

Since then, however, a number of astronomers have cast doubt on the existence of Gliese 581G. A Swiss team reported that it was unable to find evidence of the planet. Vogt and Butler published a follow-up paper defending their research.

“We stand by our data and results and are hard at work obtaining more of our own data on this system,” Vogt told Space.com in February 2011.

The Penn State researchers used a different measure of the star’s activity to reach their conclusions. Using their new technique to process existing data, they said, they were able to correct for activity on the star’s surface – a sun spot, for example – that could be misinterpreted as evidence of a planet around the star. In the new analysis, evidence that the three true planets in the system exist was strengthened, while the suspect planets disappeared, they said.

Paul Robertson, the lead researcher on the new paper, said his intent was not to critique Vogt and Butler’s methods but to provide a new tool for scientists hunting for new planets.

“I think the astronomers did more than due diligence,” he said of Butler, Vogt and other researchers who have looked at this system.