Earlier this year, scientists revealed that they had finally found evidence of theorized primordial gravitational waves, providing the best proof to-date that the Big Bang Theory is correct. Now, however, the researchers have walked back the claim, saying that what they identified may have been clouded by polarized dust. 

Back in March, a team of astronomers from the Harvard-Smithsonian Center for Astrophysics revealed that they had been able to find the signal using the Bicep2 telescope. The announcement was praised as a great scientific achievement, and the team of researchers seemed sure to be tapped for a Nobel Prize. But even as their discovery was being dubbed a huge success, others in the scientific community were questioning the finding. The Boston Globe reports:

Astrophysicist Matias Zaldarriaga of the Institute for Advanced Study in New Jersey said that, like most other people in the field, he was extremely excited when he first heard about the BICEP2 results.... But over the subsequent month, he said, he and a colleague began to wonder whether dust in our own galaxy that absorbs the light of stars and then re-emits it might be an alternative explanation... Other physicists reached similar conclusions. A paper posted by physicists from the University of California Berkeley on Arxiv, a prepublication website, cast further doubt.

The Harvard team had considered dust as a possible problem, but concluded — based on raw data collected by the Planck satellite — that the swathes of sky they were studying were clean enough of dust to corroborate their findings. Now, however, new data has emerged that is forcing the researchers to take a second look. Per the New York Times

[Scientist Raphael] Flauger, however, said he suspected the Bicep group had misread the Planck map, which is based on unpublished data. He did his own analysis under different assumptions and came to the opposite conclusion: The dust had been underestimated and could indeed be a real problem.

When the Harvard scientists finally published their study in the Physical Review Letters on Thursday, they included a caveat that significantly weakened their initial conclusions. They write in the paper's abstract:

We also examine a number of available models of polarized dust emission and find that at their default parameter values they predict power (510)× smaller than the observed excess signal (with no significant cross-correlation with our maps). However, these models are not sufficiently constrained by external public data to exclude the possibility of dust emission bright enough to explain the entire excess signal. 

The issue should be resolved once new Planck data analysis emerges, hopefully within a few weeks.