Think back to the last #nofilter picture you uploaded to your Instagram feed. Was it the horizon of Miami Beach at sunset? The view of NYC from the 17th floor? How about a supermassive black hole?
MIT computer science graduate student Katie Bouman, the brains behind a new algorithm that could make it possible to capture the first-ever image of a black hole, might make that a reality some day. She’ll just need about a dozen unbelievably expensive telescopes.
If you’re thinking that you’ve seen enough black hole photos to suck you in, you’re partially right. Stranger than (science) fiction, telescope images showing vast chasms of nothingness surrounded by new-agey-looking rings of light look almost unreal. There’s just one issue. Luminous discs, the rather stunning result of particle chaos surrounding the void, are the only thing telescope lenses can capture. That impenetrable darkness in the middle shows the aftereffects of the black hole rather than the event horizon. Bouman is looking to change all that. Instead of just an eerie silhouette, she wants to capture the astral remnants in the vortex itself.
Another thing that’s pretty hard to ignore gets in the way of photographing billion-year-dead stars. Reaching one isn’t exactly a day trip. The closest black hole in the Milky Way, Sagittarius A* (read “A-star”), is located in the center of our galaxy over — here’s where “closest” gets ironic — 25,000 light-years away.
This is why we need radio waves if we’re ever going to capture a photo of the mysterious stuff swirling inside a black hole. Radio waves have the almost supernatural power of being able to pass through solids, not to mention that their wavelengths go on (almost) forever. Astronomers use them to explore the far reaches of space to which even several human lifetimes couldn’t buy enough time for a round trip. Using an algorithm that sounds like the latest social media app, CHIRP (Continuous High-resolution Image Reconstruction using Patch priors), Bouman aspires to use Earth as a massive satellite dish that will capture high enough concentrations of radio waves to create images of a black holes.
The Event Horizon Telescope project has already been trying to get these elusive star ghosts to cooperate with a radio telescope lens. The only issue with that is that the world doesn’t have enough radio telescopes. We will need many more than the six that currently exist to observe at a high enough frequency to reach anything as impossibly far away as Sagittarius A-Star. These telescopes are also notoriously picky, needing high elevations whose atmospheres have little to no water vapor.
Frustrated astronomers will finally be able to combine the signals from multiple telescopes using CHIRP, introducing a third telescope into a process that usually involves only two. The extra telescope makes up any differences caused by variables such as elevation and water vapor for a chance to break even.
So, how is the picture actually taken? Bouman acknowledges that “there’s an infinite number of images that will perfectly describe the data [collected by the telescopes].” The fragments of radio telescope data, or “patches,” have to be pieced together into an image that actually looks like something. It gets even more complicated when patches repeat themselves, which is why she built CHIRP to begin with. It pieces these duplicates, called “patch models,” into something recognizable to the human eye. Now, here’s where the Insta-magic happens. While black hole selfies might not be a thing anytime soon, Bouman found out that because patch models have a tendency to mirror each other, her algorithm can be used to take images of just about anything—even on your smartphone.
So, what is a star that’s been decomposing for billions of years supposed to look like? We’ll have to wait until mid-2017 for observatories to start collecting data that could possibly give us the first glamour shot of a black hole. Whatever it looks like, it will probably go viral on NASA’s Instagram.