How Do Scientists Discover Planets Light Years Away? There's An Art To It, MIT Researchers Say

By now you’ve likely heard about the discovery of K2-12b, a planet orbiting a red dwarf star 110 light years from Earth. It’s the first time scientists have found a planet in the "habitable zone" of its star that also has water vapor in its atmosphere.

It’s also just the latest of some 4,000 exoplanets to be discovered in recent decades, some of them as many as tens of thousands of light years away.

So how exactly do scientists do it?

Well, imagine you are in an empty room, save for a single light bulb at the far end. A single gnat flies in front of the bulb. You might not see the gnat from where you are, but you’ll be able to tell that something disrupted the light. This essentially is how scientists on Earth spot planets in other solar systems.

Only, of course, they’re looking at millions of dim light bulbs from hundreds or thousands of light years away — and nobody knows where the gnats are.

But make no mistake. They are they looking — especially at the Massachusetts Institute of Technology.

"We’re swimming in data, and we’re planet finding central right now," said MIT Professor Sara Seager, an astrophysics and planetary scientist.

So what exactly is this data, and how do planet hunters like Seager use it to identify new planets?

First, Seager said, they send images down from space telescopes in orbit around the earth, like the Transiting Exoplanet Survey Satellite (TESS). The images are of specific, big swaths of the sky and contain thousands of visible stars, which are imaged over and over.

"It’s like a burst shot," said Seager. "If you have a camera that can do a burst of a sports scene, but we’re literally taking images of the sky."

These bursts, though, are taken over the course of a month. For the second step, Seager said, computers go to work identifying each star, and then measuring and plotting precisely how bright the star's light is across all of the images in that month-long burst.

"We then have light curves," explained Seager. "Points as a function of time."

When an orbiting planet passes in front of its star, the amount of light collected drops ever so slightly — just like when that gnat flies past the light bulb. That dimming creates a signature “box shape” in the light curve.

"We have a very simple algorithm that was written down or invented about 20-25 years ago," said Seager.

And that algorithm combs each light curve, hunting for that box shape.

"Once the computer finds that, the computer will flag it," said Seager. "And it becomes what we call a threshold crossing event."

But not every threshold crossing event is a planet passing in front of its star. So, more computer programs are employed to analyze the light curves in greater detail to separate the wheat from the chaff.

"And finally, at the end of the day, we may have gone from 20,000 light curves down to maybe 1,000 that have a threshold crossing event to maybe 50 that we have to look at in more detail," said Seager.

Those planet candidates get farmed out to astronomers around the world for closer individual observation. And the most promising of those get kicked over to five separate follow-up teams. After all that, every once in a while, scientists hit paydirt — and a new planet is discovered.

But why stop there? K2-b12 made headlines not simply because it was found, but because water vapor was also detected in its atmosphere. So how do they figure that out?

"The basic idea is that different chemicals that might be present in the atmosphere tend to block out light at very specific wavelengths," said MIT's Thomas Evans, a specialist in exoplanet atmospheres.

Telescopes like the Hubble have spectrometers that can measure light at a variety of wavelengths, which allows scientists like Evans to determine some of the chemicals present in the exoplanet's atmosphere.

"Currently with Hubble, this instrument only gives us access to certain wavelengths where there are very few molecules we expect to absorb," explained Evans. "Water is the main one, and perhaps methane."

So, scientists like Seager take what is measured — and what’s known about how our atmosphere works, plus the laws of chemistry and physics — to make educated guesses about what else might be in a planet's atmosphere.

"I make a bunch of models and I have to keep on trying until I find the right match that will fit the data," said Seager. "I like to call it the art and science of exoplanet atmospheres."

Despite a number of headlines describing as K2-18b as Earth-like, that’s not entirely true. Seager and Evans say it’s more likely a mini-Neptune, not a super Earth. But make no mistake, says Seager: Finding another Earth is the goal.

"The hope one day is to find an Earth twin around a sun-like star and to be able to recognize it as such and to know that we’re not alone," she said.

And if we do find Earth 2.0 — even if it's just, say, five light years away — that’s about 30 trillion miles. So then we just have to figure out how to get there.