In 2006, scientists at NASA's Goddard Space Flight Center sent a machine called ARCADE into space on a giant balloon, in search of radiation from the universe's earliest stars. ARCADE (Absolute Radiometer for Cosmology, Astrophysics, and Diffuse Emission) carried seven sensors that picked up electromagnetic radiation like radio waves. The plan was to lift it far enough up to prevent the Earth's atmosphere from interfering. Then, the finely-tuned instrument could detect faint radio signals from ancient stars.
Instead, ARCADE detected a huge amount of radio noise—six times louder than scientists had predicted—which has since come to be known as the "space roar." And while there are some theories, we still don't know what's causing it.
Of course, space isn't roaring in any way that our ears could hear. But there are objects in the universe—including some galaxies—which emit radio waves via synchrotron radiation.
According to Dale Fixsen, a University of Maryland research scientist and a member of the ARCADE team, NASA had built devices that detected radio noise before. These worked by looking at one point in the sky, and then at another nearby one for contrast. These instruments were useful for detecting radio-emitting galaxies and supernovas, because they measured the difference between two points. But they couldn't detect the roar.
"If there's a uniform source [of synchrotron radiation], those instruments are blind to it," Fixsen tells mental_floss.
On the other hand, ARCADE used a "large beam" that searched 7 percent of the sky. Because of the large area it searched, and its high-precision sensors, it was the first instrument we've built that could discover the roar.
But it couldn't find out everything. Fixsen says that synchrotron radiation has a characteristic spectrum. And since every source of the radiation displays this same spectrum, ARCADE couldn't discover what was roaring.
Fixsen says that synchrotron radiation usually comes hand in hand with infrared radiation. We've already measured the amount of infrared radiation that the Milky Way emits with the COBE satellite, and according to Fixsen, with our galaxy's level of infrared, it doesn't look like the Milky Way is the source of the synchrotron radiation for the "space roar."
"The relationship is tight for all galaxies we've measured," Fixsen says. "It should hold true for our galaxy as well."
On the other hand, theorists think that we've detected almost all the sources of this radiation outside our galaxy. And we know that none of these sources is causing the "roar."
According to Fixsen, there are a few possible explanations. First, the "roar" could be coming from the earliest stars. The first stars didn't have any dust—because the first dust in the universe was formed within those stars. This could have let those stars create a lot of synchrotron radiation, without a correspondingly high amount of infrared.
Second, the radiation might be coming from gases in large clusters of galaxies—Fixsen says that it would be difficult for the instruments we've used up until now to detect radiation from these.
Third, it could be coming from dim, but extremely plentiful, radio galaxies. Individually, they would be too quiet for us to detect, but en masse they might be loud enough to create the "roar."
But while there are some plausible theories, we still don't have any data to tell us which one is right. Fixsen says that there's been talk about flying ARCADE again (it's currently living in the Goddard Space Flight Center). Or they might use an instrument on the ground next time; Fixsen says they could use the data from the ARCADE mission to calibrate it, and avoid interference from the atmosphere.
But for now, what NASA wrote in its 2009 press release is still true: "The source of this cosmic radio background remains a mystery."