A WIYN-win partnership

This month, NASA's Transiting Exoplanet Survey Satellite, TESS, will be fully operational. TESS launched on April 18 to begin a 2-year sky survey that is expected to find 50 new Earth-sized planets and as many as 500 planets smaller than twice the size of Earth. The spacecraft is already in its final, 13.7-day orbit around Earth, and sent home this fantastic starfield image as part of its commissioning process:

How many worlds are in that field of view, waiting to be discovered? We'll get a better idea soon. TESS measures starlight over long periods of time. When planets pass in front of their host stars, there's a slight drop in starlight. In the field of exoplanet science, this technique is called the transit method, and it tells us planet diameter, but not much else. In order to verify there's actually a planet present, we need follow-up measurements. That's why the Kepler space telescope differentiates between candidate exoplanets and confirmed exoplanets; the "candidates" are waiting on additional verification.

Additional verification is usually done with ground-based telescopes, using a technique called the radial velocity method. Orbiting exoplanets makes their stars wobble ever so slightly, and tracking those wobbles not only confirms there's a planet present, it also reveals the mass of the planet. When you combine this with the size information from the transit method, you can calculate density. And then, you can figure out whether your exoplanet is rocky like Earth, or gassy like Jupiter.

NASA's astrophysics division has a policy when it comes to ground-based observations: it doesn't do them. Traditionally, it's been the job of the worldwide science community to follow-up on discoveries made by space telescopes like Kepler and TESS.

That's about to change. In a first-of-its-kind partnership, NASA is teaming up with the National Science Foundation to work around its own rule. NASA still won't be directly funding observations; instead, it is paying for a new science instrument to be installed on a 24-year-old telescope at Kitt Peak National Observatory in Southern Arizona. The instrument will be one of the most sensitive of its kind, giving scientists a dedicated facility where they can take a closer look at the thousands of new worlds discovered by Kepler and TESS. 

"To see"

The NASA-funded instrument is called NEID, which comes from the Tohono O'odham word meaning "to see." (Kitt Peak is located on Tohono O'odham land.) NEID is also a nested acronym for "NN-EXPLORE Exoplanet Investigations with Doppler spectroscopy." (I'll unpack NN-EXPLORE in a bit.) The instrument is being built by Penn State University.

NEID is a super-high-precision spectrograph that splits starlight into its component wavelengths the way a prism creates a rainbow. When an exoplanet wobbles its host star, the star spectrum wobbles, too. Scientists watch how that spectrum shifts over time to determine an exoplanet's mass. 

NEID is being installed on Kitt Peak's 3.5-meter WIYN telescope. Jayadev Rajagopal, WIYN's telescope scientist and head of operations, told me NASA's original goal was for the spectrograph to detect star wobbles as small as 10 centimeters per second.

"That's like the speed of a tortoise," he said. No spectrograph has yet been able to reach such extreme precision. In recent years, instruments like European Southern Observatory's HARPS have crossed the one-meter threshold, after years of refinements. Rajagopal said the current record is about 75 centimeters. The final baseline for NEID will be 50 centimeters per second, with a goal to improve that over time.

Such extreme precision requires a very stable operating environment, year after year. A single point of starlight hitting the WIYN mirror gets shunted into a fiber-optic cable, which carries the light to NEID in an adjacent building. NEID sits in a box within a box within a box. First, there's an outer room cooled slightly below room temperature. In that room sits a "meat locker" operating at the same temperature. Inside the meat locker is a vacuum chamber suspended from cables on a cushion of air. And in that vacuum chamber, the NEID optical bench uses liquid nitrogen-cooled optical detectors that remain temperature-stable within a millionth of a Kelvin.

It's an extreme setup for an extreme instrument, located on an observatory mountaintop in the middle of the desert.

"The mountain goes from negative 5 to 40 degrees Celsius in a year—none of that can get through to the bench," Rajagopal said.

NEID measurements must be repeatable over long periods of time, enough for exoplanets to complete full orbits around their host stars. During that time, operators will inevitably have to take the telescope offline for routine maintenance activities such as cleaning and polishing the mirror. To stay calibrated, the instrument uses a laser frequency comb, which creates a known, artificial spectrum against which to compare results. 

If NEID does get to 10 centimeters per second, scientists will face a new challenge: "stellar jitter." At such small scales, star surfaces are not uniform. They balloon, shift, and belch out flares, all of which can throw off radial velocity measurements. Rajagopal said the scientific community is already figuring out ways to filter out these anomalies. For instance, certain star spectral lines represent certain chemical compositions, and stellar jitter acts on those lines in different ways.

"There are certain lines indicative of the photosphere doing something, as opposed to the entire star doing something," said Rajagopal.

A boon for Kitt Peak

NASA's decision to invest in NEID came at a particularly good time for Kitt Peak, which faced a possible loss of National Science Foundation (NSF) funds for some telescopes starting in 2012. Among the telescopes threatened was WIYN, which saw first light in 1994.

"The NSF wasn't pulling out of Kitt Peak because the telescopes weren't scientifically productive," said Lori Allen, the director of Kitt Peak, in an interview. "The issue was that they had new projects to support, and they couldn't support both the old facilities and all the new facilities."

The NSF pays for 40 percent of WIYN operating costs. The other 60 percent comes from the WIYN Consortium, which gets its name from three universities—Wisconsin, Indiana, and Yale—and the National Optical Astronomy Observatory, or NOAO. (NOAO is funded by the NSF.) Losing 40 percent of the telescope's operating funds could be a devastating financial blow.

Fortunately, Douglas Hudgins, who was at the time the program scientist for Kepler, needed a telescope. Hudgins is now NASA's exoplanet exploration program scientist, and the deputy program scientist for Kepler and TESS. He told me that as Kepler continued to rack up exoplanet candidates, the science community was not keeping pace on follow-ups.

"One of the problems I was keenly aware of was that you really need precision radial velocity measurements," he said.

The current best option for U.S.-based radial velocity measurements is the HIRES instrument at Keck Observatory in Hawaii, but Keck time is precious and competitive. When Hudgins heard the NSF was thinking about divesting from WIYN, he thought a NASA-NSF partnership might be a perfect fit.  

"They had the telescope, the facility, and a good chunk of observing time," he said. "It was just one of those moments where I had been thinking about it, but there hadn't been a good opportunity to do it."

Unlike Kepler, TESS is an all-sky survey that will find planets among nearby, bright stars. This means a large telescope like Keck is not necessary for follow-ups; the 3.5-meter WIYN mirror is sufficient if paired with a high-precision spectrometer.

"When NASA expressed an interest in WIYN, we were of course very interested," said Allen. "It certainly looked like a great opportunity to do some really great science."

Thus was born an experimental partnership called NN-EXPLORE, the NASA-NSF Exoplanet Observational Research program. (This also makes NEID a double-nested acronym.) NASA and NOAO put out a call for proposals to build a new spectrometer. Several universities applied, and the finalists were MIT and Penn State, which faced off in a six-month instrument design competition. In February 2016, NASA and NOAO selected Penn State. The spectrometer is expected to be delivered to Kitt Peak at the end of this year, and be ready for science operations in the last quarter of 2019.

A survey shift

Hudgins said NASA has spent a total of $11.1 million building the NEID spectrometer and installing new facilities at the WIYN telescope. The funds, which come from the astrophysics division's exoplanet research and technology budget, also include an instrument scientist that will oversee NEID operations. 

In return, the NSF will continue paying their 40 percent share of WIYN operations. Penn State will receive a block of guaranteed time on the telescope, but other than that, anyone can apply to use NEID for exoplanet research. Hudgins said there are no requirements that observers look strictly at Kepler and TESS findings, though he expects those to be the targets by default.

"Frankly, the most exciting stuff to do in exoplanet science is following up on these missions," he said, while also stressing that all exoplanet science is valuable to NASA because it shapes future mission planning. NEID is expected to be in use for at least 5 years, and likely longer. Radial velocity measurements benefit from longer data collection periods, allowing scientists to uncover longer-orbit exoplanets and filter out noise. 

All NEID data will be processed within 24 hours and hosted by NExScI, the NASA Exoplanet Science Institute. The instrument is expected to produce huge datasets that can be mined by the entire astronomical community. Allen said this transition of the WIYN telescope from general-purpose to survey-oriented work is indicative of a larger trend.

"We used to provide telescopes to the community," she said. "Now we're providing telescopes and data. We're doing more and more big survey projects that are producing very large databases that are beneficial to everybody." Other Kitt Peak telescopes have been similarly repurposed in recent years, including the 4-meter Mayall telescope, which is doing a dark energy survey sponsored by the U.S. Department of Energy, and the 2.1-meter telescope, which is doing a robotic sky survey using adaptive optics.

"Kitt Peak isn't the only mid-scale observatory that has found its funding in jeopardy recently," Allen said. "I think our move to survey-based projects, utilizing other funding streams, is a good model for other observatories."

NASA's Hudgins said he's very happy with how the partnership unfolded, and the larger impact it will have on the astronomical community.

"I'm a firm believer that we always make the science return from our missions larger than what the mission science team will do," he said, referring to Kepler and TESS. "The more science we get out of them, the more return on investment for NASA."