Gravitational waves are ripples in the fabric of spacetime predicted by Albert Einstein's general theory of relativity, first detected in 2015. But an expected corresponding low-frequency gravitational wave background—a kind of "hum" comprised of a chorus of gravitational waves, most likely emanating from binary pairs of supermassive black holes—has proven more elusive. Now the North American Nanohertz Observatory for Gravitational Waves (NANOGrav) has announced the first evidence of this gravitational wave background. The results and related analyses are described in several new papers published in the The Astrophysical Journal Letters.

The collaboration stopped short of claiming outright detection, opting to describe their results instead as strong evidence of the expected gravitational wave background. That said, "In our statistical analyses, there's a less than 1-in-1,000 chance of nature giving our results without gravitational waves being present," NANOGrav chair Stephen Taylor of Vanderbilt University said during a press briefing.

As previously reported, LIGO detects gravitational waves via laser interferometry, using high-powered lasers to measure tiny changes in the distance between two objects positioned kilometers apart. LIGO has detectors in Hanford, Washington, and in Livingston, Louisiana. (A third detector in Italy, Advanced VIRGO, came online in 2016.) On September 14, 2015, at 5:51 am ET, both detectors picked up signals within milliseconds of each other for the very first time—direct evidence for two black holes spiraling inward toward each other and merging in a massive collision event that sent powerful shockwaves across spacetime. That first direct detection was announced on February 11, 2016, spawned headlines worldwide, snagged the 2017 Nobel Prize in Physics, and officially launched a new era of so-called "multi-messenger" astronomy.

If we've already detected gravitational waves, why is NANOGrav necessary? It's because the collaboration is looking for different frequencies of gravitational waves. LIGO’s range centers on stellar remnant black holes and other celestial objects of similar mass. NANOGrav, in contrast, is a galaxy-sized gravitational wave antenna that is sensitive to much lower frequencies (by roughly 10 orders of magnitude) of gravitational waves than LIGO. "LIGO could never, in its wildest dreams, measure these low-frequency gravitational waves because the detector's just not big enough," NANOGRav scientist Chiara Mingerelli of Yale University told Ars. "Our gravitational waves have wavelengths of light years." So the two approaches are complementary.

The roots of NANOGrav date back to the 1970s when scientists thought it might be possible to take advantage of the Voyager mission to look for changes in the arrival times of the spacecraft's signals as they left the solar system. In the 1980s, scientists proposed using exotic stars called pulsars for a similar purpose. While many pulsars aren't sufficiently stable to do so, the discovery of millisecond pulsars saved the day, since they are ideal "cosmic clocks." As I wrote back in 2014:

Pulsars form when stars more massive than our Sun explode and collapse into neutron stars. As they shrink, they spin faster and faster, because angular momentum is conserved. (Think of what happens when you swing an object around your head on a string: the more you shorten the tether, the faster it goes.) Pulsars also blast out radiation that can be picked up on Earth whenever that beam sweeps into our direction, like the rotating beam of a lighthouse. The fastest pulsars, spinning hundreds of times per second, make excellent clocks—on par with the best atomic clocks.

The idea behind NANOGrav is that as gravitational waves stretch and shrink spacetime, this will disrupt the pulsars' ultra-precise "ticking." There should be a telltale signature in the form of a kind of “shimmering” effect, produced because pulses affected by gravitational waves should arrive slightly earlier or later in response to those ripples in spacetime. By studying the timing of the regular signals produced by many individual millisecond pulsars scattered over the sky at the time—called a "pulsar timing array"—NANOGrav tries to detect minute changes in the Earth's position due to the effects of gravitational waves. It just takes many years to do so.

NANOGrav's first data set covered five years of observational data for 17 millisecond pulsars, and the collaboration has been adding more pulsars to their array ever since. The 2014 data release covered nine years of observation. Next, there was the release of the 11-year data set, which initially generated considerable excitement among NANOGrav scientists because a signal popped out that looked a lot like the gravitational wave background. Alas, "It turned out to be Jupiter," said Mingerelli. "It just so happens that Jupiter has an orbital period of almost 11 years."