How Citizen Scientists Discovered the Strangest Star in the Galaxy
It’s a science-geek fantasy: With hard work, sharp wits, and keen eyes, an amateur discovers something all the experts missed. Something big, something that leaves the PhDs confused, perplexed, and delighted.
For the dozen citizen scientists who helped discover a peculiar star called KIC 8462852, this fantasy came true, though the reality is less lone-genius-eureka-moment and more slow-building-ensemble-drama. And they did it with the relentless, unprejudiced curiosity that comes naturally to amateurs.
KIC 8462852 dims suddenly and erratically, unlike any other star known to science. Astronomers have even floated the idea that it is surrounded by a colossal solar-energy-harvesting “megastructure” built by advanced aliens. It has been called “the most mysterious star in the galaxy” and the “WTF star” (that’s “Where’s The Flux,” naturally), but mostly it’s better known as “Tabby’s star” for the astronomer who has taken the lead on investigating it. Thanks largely to the ET dazzle, the star has become a buzzworthy, boldface name in the news and on social media.
After its initial burst of weirdness, Tabby’s star behaved like a normal star for more than two years. But then this past May, Tabby’s star again went haywire. The internet lit up with urgent calls to action.
Now, thanks to observations mobilized by the latest series of dips, a clearer picture of the “WTF star” is finally starting to emerge, as detailed in a new paper in the Astrophysical Journal. Until now, every explanation astronomers had been able to offer up fell short. Starspots? Too small, wrong timing. Orbiting asteroids or comets? Telescopes don’t pick up the signature of radiating heat. A cloud of gas or dust closer to Earth? Maybe, but why aren’t other stars affected in the same way? It’s the kind of stumper that invited improbable explanations, that pit Occam’s razor—which favors the simplest explanation—against Sherlock Holmes’ famous dictum: When you have eliminated the impossible, whatever remains must be the truth…however improbable.
Regardless of what’s actually driving the curious dimming of Tabby’s star, without citizen scientists, it would still be just another undistinguished pinprick on the sky. Here’s the story of how amateurs helped discover the strangest star in the galaxy—and why citizen scientists might be the x-factor in many more discoveries to come.
Thousands of Open Minds
In northern Ontario, when the sky is clear and there’s no moon, Adam Szewczyk can see the Milky Way.
For years, Szewczyk, who lives outside Toronto, had nursed an interest in astronomy. As a kid, he watched Carl Sagan’s TV miniseries “Cosmos” and borrowed books about planets from the library. His parents bought him a small telescope from Walmart as a graduation gift, but he was never able to see much through it. Then, in 2011, he read an article about a web site called Planet Hunters where amateur volunteers could help scientists comb through mountains of data collected by the Kepler Space Telescope. Szewczyk signed up— a “spur of the moment” thing, he recalls—and soon he was scouring Kepler data a few times a week.
Kepler may be a marvel of science, but its operation is dead simple: stare, stare, and stare some more. Twice an hour, almost every hour, for four years, Kepler recorded the brightness of some 150,000 presumably ordinary stars in its field of view. All that staring generated about 2.5 billion data points per year, too much for scientists to comb through by hand. Luckily, they didn’t have to. Kepler scientists fine-tuned their computer programs to pick out the signatures of orbiting planets. They knew, though, that some things would slip through the cracks—planets orbiting stars orbiting other stars, planets being tugged off course by neighboring planets, even things that weren’t planets at all.
“Because of the enormous data volume involved, our planet-finding algorithms were designed to throw away anything that didn’t look like a planet so professional astronomers could focus their efforts on the mission goals,” says Jason Wright, an astronomer at Penn State. To catch the needle-in-a-haystack oddities that would elude the computers, a team led by Yale astronomer Debra Fischer set up Planet Hunters, where they could enlist an army of human eyes to sort through the computers’ discards.
Anyone can sign up to be a Planet Hunter. After a short online training, Planet Hunters are served a series graphs, called light curves, that show the brightness of a target star over a period of 30 days. A typical light curve—that is, a boring one—looks like a scruffy straight line. An interesting light curve, to a planet hunter, is one that dips somewhere along its length. This dip is the signature of a transit, an event in which a planet orbits in front of the star, temporarily obscuring a small fraction of the incoming starlight.
Szewczyk had been on Planet Hunters for a few months when he spotted a particularly strange light curve. It had a dip—that much was obvious. But the dip didn’t conform to the standard profile. “I found its peaks and periods a bit unusual and fascinating, so I decided to mention it in ‘Talk,’ ” the volunteer discussion board, Szewczyk recalls. He dashed off a comment (“bizarre peak—a giant transit”) that caught the attention of other volunteers, including Daryll LaCourse.
LaCourse, a veteran Planet Hunter who estimates that he has classified tens of thousands of light curves on the site, was also intrigued. “[The dip] was not nicely symmetric, as any well-mannered planet transit should be,” he says. A planetary transit usually leaves a U-shaped notch on the light curve, like a cartoon smile, but this dip was straight on one side and droopy on the other. Wondering what they were seeing, LaCourse and other “Talk” regulars emailed astronomer Tabetha “Tabby” Boyajian, then a postdoc at Yale, who served as the liaison between the scientists and the Planet Hunters volunteers. (Boyajian is now an assistant professor at Louisiana State University.)
“When they showed the star to me, we all thought it was a data glitch,” Boyajian says. But the Planet Hunters had already run the light curve through tests that root out faulty data, she says. It had passed: the dip was real.
“Every trained astronomer’s first reaction to seeing the light curve is ‘bad data,’ because we’ve seen bad data so many times that it is almost always the reason we see something totally outside of our range of expectations,” Wright says. “It would not have been spotted without the amateur eyeballs on the data.”
“Professionals know how to look for specific patterns, specific behaviors,” says astronomer Stella Kafka. As director of the American Association of Variable Star Observers, a community of “citizen-astronomers” who collect science data with their backyard telescopes, Kafka is a champion of the discovery power of amateurs. “Non-professional astronomers are not looking for one thing: They can actually see the forest. For that reason, they are more alert to little irregularities that are real, but unusual,” she says. “That’s where discoveries happen.”
“You could call it thousands of untrained eyes, but I call it thousands of open minds,” Kafka says.
Amateurs may be unbiased, but they are not naïve, and the most experienced citizen scientists have developed the deft data-smarts of autodidacts. “The amateurs ended up learning a lot about Kepler and its data set,” Wright says. “They were able to learn what sorts of anomalies are just ordinary data artifacts or rare but boring astrophysical situations and to really focus on the oddballs that suggest something new.”
The combination of clean-slate thinking and in-the-weeds experience is a powerful vehicle for discovery. In fact, LaCourse points out, many of the most active Planet Hunters have left the original online classifier behind, moving on to custom desktop tools of their own design. “Ninety percent of the published discoveries from the project to date are the result of citizen scientists working outside of the web-based viewer and collaborating about their findings on the PH [Planet Hunters] Talk discussion forums,” LaCourse says.
The Waiting Game
Szewczyk’s original oddball transit in 2011 was considered a fluke at first, not a discovery. “[The 2011 light curve] was simply noted with some mild interest and then filed away by the citizen scientists as one curiosity amongst many,” LaCourse says. Fresh Kepler light curves hit the Planet Hunters database every three months, but it took two years for KIC 8462852 to “dip” again—this time, much deeper. And then, two years later, something even stranger: “a series of deep dips quite unlike anything we had seen before,” LaCourse says.
The latest light curves were a cliffhanger, with the measured light levels rising and falling all the way to the tail end of the data set. To see how the story ended, the Planet Hunters would have to wait for the next quarter’s upload.
But the data never came. On May 12, 2013, a pointing system failure ended Kepler’s planned mission. With its once-steady stare now wandering, Kepler was no longer fit for long-term monitoring. (Today, Kepler operates in an alternative observing mode, but cannot hold aim on Tabby’s star.) That left Boyajian and her colleagues, professional and amateur alike, with a problem. They had seen 10 dips, each one different. Some were tiny, barely measurable things. Others plunged as much as 20%. Some lasted for days, others for weeks. Boyajian and her team couldn’t predict when another one might happen, but they didn’t want to miss it. It was like a phone conversation dropped mid-sentence: they were left waiting for KIC 8462852 to call back, and they didn’t want to miss the call when it did.
Meanwhile, by calling in favors from astronomer friends around the world, Boyajian was cobbling together a fuller picture of the star. At the 10-meter Keck telescope on Hawaii’s Mauna Kea, she and her colleagues snapped high-resolution images of KIC 8462852. At the Roque de los Muchachos Observatory in La Palma, Spain, they used a spectrograph to search the star’s light for chemical signatures that might explain its behavior. Her team also combined data from space- and ground-based telescopes to see how much ultraviolet and infrared energy the star was giving off.
Boyajian published a draft of the results in 2015. Of the 49 authors, 11 were amateurs, including LaCourse, who got the second-author spot after Boyajian. The paper asked more questions than it answered, but the authors had little choice. The project was at an impasse: “We’d pretty much exhausted all of our resources,” Boyajian said.
Wright, the Penn State astronomer, entered the fray with a flourish when he hypothesized that the dips could be caused by a “megastructure”—specifically, a swarm of alien-built structures circling around the star. It was not exactly a mainstream explanation, but it was worth considering, Wright thought.
To test that hypothesis or any other, astronomers would have to get detailed measurements of the star during a dip, ideally using multiple telescopes to cover a wide swath of the electromagnetic spectrum. But chances of the planned observations just happening to match the timing of a dip were exceedingly slim.
“We needed a small amount of time every night on a range of small-ish telescopes around the world so we could get something like round-the-clock monitoring for the next dip,” Wright says. But that isn’t how time on professional telescopes is doled out. “Telescope time is normally awarded in units of night or half-nights many months in advance,” Wright explains. Plus, getting time on a research telescope is a competitive process. “Telescope time with professional telescopes is becoming extremely valuable and also rare,” Kafka says. “No one is going to give you a week of time to monitor a star in the 20-meter class” of telescope.
Kafka knew that legions of backyard astronomers would jump at the chance to contribute to observe the star, so she quickly mobilized AAVSO, the amateur astronomer organization, and light curves started pouring in from observers around the world.
Merging light curves from so many different telescopes and observers is like trying to align fingerprints, and it proved nearly impossible to do in real-time. So Boyajian tapped into a network called the Las Cumbres Observatory, a private, non-profit meta-telescope made up of 18 different telescopes. There, astronomers can buy time by the hour. With a “bulk discount,” an hour on the network’s smallest telescopes costs about $125.
The Las Cumbres telescopes, which range from 0.4 meters to two meters across, are split among six different observatories located around the world. Through a series of hand-offs, Las Cumbres can track an object day and night. The entire network is controlled by an artificially-intelligent scheduler that allots time on the telescopes to as many as 50 different astronomy experiments at once. If clouds roll in at one site, the scheduler can even reassign telescopes on the fly. It is the first and, so far, only network of its kind.
“The Las Cumbres Observatory Global Telescope network was practically designed for this sort of thing,” Wright says. “All we needed to use it was funds.”
Wright and Boyajian estimated that it would cost about $100,000 to keep tabs on KIC 8462852 for about two hours a day for a year. Boyajian considered applying for a grant from the National Science Foundation, but she didn’t like her odds: of every ten astrophysics proposals sent in to the NSF, only one or two earns funding. Plus, the winners tend to be proposals with “pretty much guaranteed results,” Boyajian says—and hers was not. Even if she did get a grant, there would be a yearlong lag between writing up the application and cashing the check.
Unwilling to wait on a grant award that might never materialize, Boyajian went back to the crowd: this time, the Kickstarter crowd-funding platform. In May 2016, buoyed by the “megastructure” buzz and a TED talk she had delivered the previous February, she launched “The most mysterious star in the galaxy” on Kickstarter. Las Cumbres gifted the project 200 hours to stay on the star while the Kickstarter commitments came in. In June 2016, the campaign nosed past its goal, and in March, Kickstarter dollars started paying for time on the telescope network.
All along, Boyajian waited for a dip to come. “I remained optimistic, but prepared myself for nothing to happen,” she says. Then, on May 18, the brightness measurements from the Las Cumbres Observatory started dropping. The dip—if it was a dip—was still shy of the “three sigma” benchmark that divides meaningful signals from natural uncertainty. But when astronomers at other telescopes reported seeing it, too, Boyajian decided to “pull the trigger,” and on May 19, word of the dip was rippling across the internet.
Within hours, dozens of research telescopes swung toward KIC 8462852, measuring its brightness across the electromagnetic spectrum. From space, NASA’s Swift satellite scanned for X-ray and ultraviolet radiation, and the WISE spacecraft, serendipitously pointed in the right direction, took infrared readings. Citizen-astronomers joined in, too. Between May and the end of September, KIC 8462852 dipped three times—Kickstarter backers named the dips “Elsie,” “Celeste,” “Skara Brae,” and “Angkor.”
Boyajian and her colleagues are still working to understand their data, though Boyajian says that so far the measurements all point to obscuring dust, which blocks more blue light than red light, as the cause of the dips. But where did the dust come from, where is it located, and why don’t telescopes pick up its heat signature? Boyajian and her colleagues are still trying to figure that out. They don’t know when it might dip again, or for how long, so the key, says Boyajian, is to simply keep watching.
The new observations also all but rule out the possibility of a “megastructure” or any other solid object, which would be opaque at every wavelength. That may be a relief for Szewczyk, who muses about the possibility of aliens around Tabby’s star: “If we somehow get their attention and if they come to ‘consume’ us, I will be guilty as charged for starting all this just by making a ‘blip’ in a ‘Talk’ page.”
Comments
Post a Comment