'Gravity Spies' Are Hunting for Gravitational Waves

In 1916, Albert Einstein predicted the existence of gravitational waves when he proposed his theory of general relativity. Einstein figured that any object with mass creates a kind of signal, rippling through spacetime, whooshing through the universe at the speed of light with its sound arriving later. This would theoretically mean that the crash of two black holes colliding into each other billions of years ago wouldn’t be detected by humans — and then even, barely so — for millennia.

Einstein’s theory was spectacularly confirmed in September 2015 with a “chirp”. Then in June, scientists confirmed Einstein’s foresight yet again when gravitational waves were detected for a third time, flutters of spacetime when two black holes crashed into into each other between 1.6 and 4.3 billion years ago.

While the detection of gravitational waves was done by scientists using high-precision equipment, the future of this crucial branch of science depends on flocks of unpaid, civilian nerds who want to make a difference.

Gravity Spy, is a volunteer network of 9,000 normals who comb through data collected by the Laser Interferometer Gravitational-Wave Observatory (LIGO), an extremely sensitive detection system that uses laser beams to detect physical distortions caused by passing gravitational waves split between Livingston, Louisiana and Hanford, Washington.

In 2016, a year after the first gravitational wave was detected, the Gravity Spy project was launched with funding from the National Science Foundation. Its goal: Recruit more bodies to sift through the huge swaths of data LIGO generates and identify glitches — instrumental and environmental sources of noises that make LIGO less sensitive to gravitational waves and make it difficult for computers to create models.

“I like to think of it like a microphone,” Joshua Smith, an associate professor of physics at California State University, Fullerton, working on the project, tells Inverse. “You have a microphone and you’re trying to listen to a beautiful quiet song — but your microphone is an electrical device that has all of this crinkling and crackling. We’re trying to figure out how to make those crackling sounds go away so we can hear the beautiful songs.”

Michael Zevin, a Ph.D. student at Northwestern University who helped designed and build Gravity Spy, says that while they’ve been able to confirm three detections of gravitational waves, there have, in turn, been “hundreds of thousands of these triggers” — out-noise features (the technical term for glitches) that make their way into the data. Physicists sometimes know where the noise is coming from, but sometimes they don’t: Strong wind blowing can show up as a glitch or ringing frequencies from LIGO’s mirrors, which are held by a pendulum system, can emerge like a “pluck of a violin string,” Zevin says.

But scientists don’t want to just toss out the glitches that cross their paths; they want to figure out what they are and further improve LIGO’s sensitivity moving forward, too.

“Without good knowledge of the noise in the detectors, it would be near impossible to make discovery of these gravitational waves,” says Zevin. “Even though the Citizens aren’t really searching for gravitational waves themselves, without their help and without the help of the teams that actually do the analysis of all the noise, we would have no way of detecting gravitational waves in the first place.”

On the Gravity Spy website, users are presented with an image of a glitch and asked to identify its morphology. Each shape is considered a different type of glitch — a symmetric teardrop, for example, is called a “blip,” while a thin half-curve is called a “whisper.” These glitches can mimic the detection of astrophysical signals, which is why scientists need humans to identify and characterize the most frequent glitches so they can recognize the subtle differences. As more people identify glitch outliers, the machine learning that underlies Gravity Spy improves as well.

“We’re trying to build a model that combines what computers are good at and what humans are good at to create something that is sustainable and will allow projects to classify even bigger data sets in the future,” says Zevin. “When people identify a new glitch class that appears in the data, they can help us build data sets that computer algorithms can learn off.”

Besides identifying glitches, users are also encouraged to point out new types of glitches, create candidate classes, and find relations between them. But many participants go above and beyond putting in the time to understand the physical causes of glitches, what causes a specific morphology, and learning about the complex rules of gravitation.

“This is a very cooperative community,” Barbara Téglás tells Inverse. “The participants help each other [via an online forum], teach each other, and solve problems together. Many of us read scientific publications and technical papers about the detectors. When the scientific team confirms something we’ve found out, it’s a success for the whole community.”

Téglás has been volunteering as a Gravity Spy since April 2016, and has spent a few hours almost every day since participating. A biotechnology engineer who has had a long, established passion for the sciences, Téglás, who was born in Hungary, logs on from her home in France to participate in multiple citizen scientists projects hosted on Zooinverse. With Gravity Spy, she gets to indulge her curiosity about astronomy and the galaxy, while interacting with a group of people she describes as “helpful, friendly, and intelligent.”

“I chose to participate because I realized it was a great opportunity to be a part of one of the most prominent scientific projects in the world,” says Téglás. “I thought that if I could do something for this, even the tiniest thing, it was worth it.”

And it is worth it: Téglás and her peers are helping play a role in further unlocking the mysteries of the universe. That’s not some empty goal Téglás has outlined: It’s why the discovery of gravitational waves is so important, because their existence can lead to the discovery of thus-far unseen astrophysical objects like neutron stars, supernovas, and even other forms of gravity. Gravitational waves create a ripple through the space they travel through, causing distances to shrink and stretch, and their emergence from entities like black holes confirms those entities exist.

“Even if we were to detect a new event that had the exact same properties of one of the three that we’ve already detected, it would still be a huge leap forward because we’re accessing black holes that are remnants of stars that have long been dead,” says Zevin. “We’re accessing information about them that we’ve never been able to access before — glimpses of the last seconds of these collisions, which allows us to extrapolate tons of information about how stars evolved and the environments in which these black holes really formed.”

Smith, who says he is “completely sold” on a citizen scientists project, is also quick to point out that not only is it that the volunteers are providing scientists with crucial help, but that what they’re doing has a historical precedent. Everyone should feel like they can participate in science, he argues, and if they make that decision then everyone can benefit. People throughout history have made contributions to science where they weren’t necessarily trained experts, Smith reasons, and there shouldn’t be some kind of bar that only allows certain types of people to do science.

Besides, if there was one, they wouldn’t have be able to have the sort of wondrous moments Smith has been lucky enough to encounter as a scientist.

“We expected that the first waves we would see would be tiny, but instead the very first thing we saw was beautiful and gigantic,” says Smith. “This waveform was so clear that we could put the model waveform from Einstein’s theory of general relativity on top of it, and they were a perfect match. It seemed too good to be true.”