The hills of West Texas rise in waves around the Hobby-Eberly Telescope, a powerful instrument encased in a dome that looks like the Epcot ball. Soon, it will become more powerful still: Scientists recently primed the telescope to find evidence of dark energy in the early universe, prying open its eye so it can see and process a wide swath of sky. On April 8, scientists will dedicate the new telescope, capping off the $40 million upgrade and beginning the real work.
The dark energy experiment, called Hetdex, isn’t how astronomy has traditionally been done. In the classical model, a lone astronomer goes to a mountaintop and solemnly points a telescope at one predetermined object. But Hetdex won't look for any objects in particular; it will just scan the sky and churn petabytes of the resulting data through a silicon visual cortex. That's only possible because of today's steroidal computers, which let scientists analyze, store, and send such massive quantities of data.
The hope is so-called blind surveys like this one will find stuff astronomers never even knew to look for. In this realm, computers take over curation of the sky, telling astronomers what is interesting and worthy of further study, rather than the other way around. These wide-eyed projects are becoming a standard part of astronomers' arsenal, and the greatest part about them is that their best discoveries are still totally TBD.
To understand dark energy---that mysterious stuff that pulls the taffy of spacetime---the Hetdex team needed Hobby-Eberly to study one million galaxies 9-11 billion light-years away as they fly away from Earth. To get that many galaxies in a reasonable amount of time, they broadened the view of its 91 tessellated stop-sign-shaped mirrors by 100. They also created an instrument called Virus, with 35,000 optical fibers that send the light from the universe to a spectrograph, which splits it up into constituent wavelengths. All that data can determine both how far away a galaxy is and how fast it's traveling away from Earth.
But when a telescope takes a ton of data down from the sky, scientists can also uncover the unexpected. Hetdex's astronomers will find more than just the stretch marks of dark energy. They’ll discover things about supermassive black holes, star formation, dark matter, and the ages of stars in nearby galaxies.
The classical method still has advantages; if you know exactly what you want to look at, you write up a nice proposal to Hubble and explain why a fixed gaze at the Whirlpool Galaxy would yield significant results. “But what you see is what you get,” says astronomer Douglas Hudgins. “This is an object, and the science of that object is what you’re stuck with.”
Hobby-Eberly is only the most recent telescope to open its eyes to this new method. When NASA's Wide-Field Infrared Survey Telescope launches in the 2020s, its view will be 100 times wider than Hubble's---and boast 288 megapixels. It'll check out the effects of dark energy and find new planets, but some of its work is similarly object-agnostic. “The survey science will enable everything from soup to nuts,” says Hudgins, who's working on the program.
The Large Synoptic Survey Telescope---the LSST, not pronounced “Liszt”---will similarly capture the whole sky every few nights, for 10 years, making a decade-long, high-def movie. Its gaze, at any given time, see a space the size of 40 full moons. Electronics and algorithms will auto-check the images to see what has changed or moved or surged or dimmed from night to night.
“The most exciting thing about the LSST will be discovering things that are rare or truly novel or unusual,” says Lucianne Walkowicz, head of the telescope’s eight science collaborations.
Scientists will always need science-of-that-specific-object observations---to follow up on weirdness, to look deeper, to zoom, to see through a different filter. But until pretty recently, they didn't have the option of big, blind surveys. The silicon boxes of the past couldn’t handle them. And---most importantly---they didn't have the torque to twist useful information from the background noise.
Back in the Lone Star State, where Hobby-Eberly will study more than lone stars, the Hetdex team has turned to the Texas Advanced Computing Center. The data will go straight from telescope to computing center, where the scientists hope to auto-process it in near-real-time and store the resulting petabytes. “All I do on a daily basis is just write computer code. That’s what I am now: a computer coder,” says Gebhardt, who is an astronomer.
He believes their processing pipeline will be ready when the full survey starts in August or September. And like Walkowicz and Hudgins, he looks forward to the "strange oddballs," the who-knows-whats.
“You don’t know how to find them, because you don’t know what to look for,” he says.
But because Hetdex isn’t exactly looking for anything, it doesn’t need to know how. And that, really, is why object-agnostic research excites scientists. “I could sit here and describe five things to you," says Hudgins, "and I guarantee that once we do the survey, the most exciting thing is something I didn’t think of."
