The Planet Hunters

Milky way

The Milky Way as seen from a telescope in the Namibian desert. (Photo courtesy of Gáspár Bakos)

From Gáspár Bakos’ desk at Princeton, he can see everything that happens at his telescopes on three continents. He can see wild burros nuzzle at the cables in Chile, warthogs wander by in Namibia, and kangaroos come a bit too close for comfort in Australia.

Despite the risks to his equipment, managing telescopes in three time zones across the Southern Hemisphere has a major advantage: it is always nighttime somewhere, so Bakos’ telescope network can search around the clock for planets outside our solar system, or exoplanets.

Bakos, an assistant professor of astrophysical sciences, is one of several Princeton faculty members involved in finding and studying exoplanets. Some researchers, like Bakos, are searching for the faint dimming of starlight that happens when a planet transits in front of a star. Others are trying to achieve what some have called the Holy Grail of planet hunting: direct imaging of an exoplanet. But detecting a planet is just the beginning. Researchers hope that by studying other solar systems they can confirm theories about how planets form and perhaps even learn whether life exists on these other worlds.

“Princeton is making contributions to the search for exoplanets in a number of areas,” said David Spergel, the Charles A. Young Professor of Astronomy on the Class of 1897 Foundation and chair of the Department of Astrophysical Sciences.

Discover and characterize

Just 20 years ago, finding planets around stars other than our sun was thought to be impossible — they were too far away, too dim and too close to the glare of their star. But today, due to creative strategies and new telescopes, about 1,000 planets have been detected and confirmed, with thousands of candidates awaiting confirmation.

The vast majority of the confirmed planets are quite unlike our own, however. Some are gas giants larger than Jupiter that orbit very close to their star — so close, in fact, that they can make an entire circuit around the star in a few days. Contrast this to our solar system, where Mercury, the closest planet to the sun, has an orbital period of 88 days — that is, it takes 88 days to orbit the sun. Earth makes the journey in 365 days.

The reason we’ve found so many of these large planets, which astronomers have nicknamed hot-Jupiters, is because they are relatively easy to detect with current methods. The two most successful ways of finding planets to date are to look for the periodic dimming of starlight when the planet crosses in front of the star, as Bakos’ telescopes do, or to look for the episodic wobbling of the star in response to the planet’s gravitational pull.

Professor Gáspár Bakos

Professor Gáspár Bakos (Photo by Pal Sari)

Through findings from a space-based NASA telescope known as Kepler, which hunted planets from 2009 until earlier this year, we now know that hot-Jupiters are the exception rather than the norm. “We can detect these hot-Jupiters because they pass in front of their stars fairly often and they block a significant fraction of the starlight,” said Bakos, “not because there are more of them than there are other kinds of planets.”

These other kinds of planets could include ones with conditions capable of supporting life. These planets lie in the “habitable zone” not too close and not too far from their star, and are capable of having liquid water on their surfaces.

One of Bakos’ telescope networks, HATSouth, is looking for exoplanets, including possibly habitable ones. HATSouth can detect planets with orbital periods of 15 to 20 days, which may not seem like much, but for certain classes of stars, namely the mid- to late-M dwarf stars, planets with 15-day periods lie in the habitable zone.       Bakos started building his first HAT — Hungarian made Automated Telescope — in 1999 while a student at Eötvös Loránd University in Budapest. The automated telescopes are relatively small — close in size to amateur models — but the lower costs allow more of them to be deployed. An earlier network Bakos built, HATNet, which came online in 2003 and consists of telescopes at sites belonging to the Smithsonian Astrophysical Observatory (SAO) in Arizona and Hawaii, has discovered 43 candidate planets.

HATSouth became operational in 2009 and is a collaboration between Princeton University, the Max Planck Institute for Astronomy, Australian National University and Pontificia Universidad Católica de Chile. Originally funded by the National Science Foundation and SAO, the network consists of six robotic instruments located at Las Campanas Observatory in Chile, the High Energy Stereoscopic System site in Namibia and Siding Springs Observatory in Australia. To date, HATSouth has detected a handful of planets, and has a dozen candidates awaiting confirmation. So far, all of the planets have fairly short orbital periods — they orbit their stars in one to three days — but Bakos is optimistic about HATSouth and his new project, HATPI, for which he has received funding from the David and Lucile Packard Foundation. HATPI is a wide-field camera system that will continuously image the entire night sky at high resolution and precision for five years, with the goal of identifying planets with longer orbital periods. Bakos’ team includes Joel Hartman and Kaloyan Penev, both associate research scholars; Zoltan Csubry, an astronomical software specialist; Waqas Bhatti and Miguel de Val-Borro, both postdoctoral research associates; and Xu (Chelsea) Huang, a graduate student.

Direct imaging

While HATSouth looks for dips in starlight that indicate the presence of a planet, other researchers at Princeton are aiming to directly image exoplanets. Such imaging is possible, for example, when the planet appears on the right or left side of the star rather than directly in front of it. To date, only a handful of exoplanets have been observed this way, because the star’s light is so bright that seeing a nearby planet is like trying to see a speck of dust in the glare of a headlight.


Professor Jeremy Kasdin (Photo by Alexandra Kasdin)

Jeremy Kasdin’s group is working to develop instrumentation for direct imaging. “The idea is to block out the star’s light so that it is possible to see the planet,” said Kasdin, a professor of mechanical and aerospace engineering.

Astronomers have used this concept to study the sun since the 1930s: they place a black disc, called a coronagraph, at the center of the telescope’s image to block light from the sun so they can study solar flares on its surface.

For planet-watching however, this light must be blocked with great precision. Because light acts as a wave, it diffracts around the edge of the telescope and, without a coronagraph, creates concentric patterns on the resulting image (see illustration, page 27), just as water makes ripples in a pond when it flows past an obstacle. These patterns obscure the planet.

To eliminate or change these patterns, researchers at Princeton’s High Contrast Imaging Laboratory, led by Kasdin, are developing a coronagraph with a distinctive shape and arrangement of slits that alter the patterns in ways that can permit detection of planets. This “shaped-pupil coronagraph” is being developed by Kasdin and his collaborators Spergel, Edwin Turner, a professor of astrophysical sciences, Michael Littman, a professor of mechanical and aerospace engineering, and Robert Vanderbei, a professor of operations research and financial engineering, along with postdoctoral research associates Tyler Groff and Alexis Carlotti and graduate students Elizabeth Jensen and A.J. Eldorado Riggs. The coronagraph could be sent up in a space-based telescope mission under consideration for later in the decade.

Pupil diagramThe Kasdin lab is also working on another light-blocking idea called an occulter. This is a giant sail that could fly in space, ahead of a space-based telescope, to block out light from a star. “It is sort of like holding your hand up to block the sun while you watch a bird in the sky,” said Kasdin. The occulter would be launched folded-up, like a flower bud with petals that would unfold in space to create a shield with a diameter of about 40 meters (131 feet) that would fly about 11,000 kilometers — or roughly 6,800 miles — ahead of the telescope to block the star’s light.

Occulter diagramAt Princeton’s Forrestal Campus three miles from the main campus, graduate student Daniel Sirbu is testing four-inch high models of occulters. The team collaborates closely with NASA’s Jet Propulsion Laboratory at the California Institute of Technology where an occulter is being built and tested with the help of engineers at Northrop Grumman and Lockheed Martin.

Video: How an occulter would unfold in space:

In addition to blocking light, Kasdin’s group is working to improve the technology for correcting faulty imaging caused by the Earth’s atmosphere, as well as heat, vibrations and imperfections in the telescopes themselves. All ground-based telescopes suffer from poor imaging quality due to atmospheric water vapor that is present even on cloudless nights, causing turbulence that makes stars appear to twinkle. Astronomers can correct these distortions using a technology known as adaptive optics which involves bendable mirrors. The Kasdin lab is working on improvements to these systems.

Coronagraphs and adaptive optics already are in use in a handful of telescopes, including the Subaru Telescope, operated by the National Astronomical Observatory of Japan (NAOJ), in Hawaii. Princeton researchers, including Kasdin and his colleagues, as well as Turner; astrophysical sciences professor Gillian Knapp; Timothy Brandt, a 2013 Ph.D. in astrophysical sciences; and others, are part of an international collaboration led by NAOJ scientist Motohide Tamura that is known as SEEDS (Strategic Explorations of Exoplanets and Disks with Subaru).

Kasdin’s group is working on designing an instrument, the Coronagraphic High Angular Resolution Imaging Spectrograph (CHARIS), to add to the Subaru Telescope to look at the different kinds of light, or spectra, emitted by a planet. Just as a prism splits white light into its rainbow of colors, CHARIS contains prisms and special filters that allow researchers to see the different wavelengths of light. These wavelengths provide signatures that can reveal the planet’s temperature and hint at which atoms and molecules are present around the planet. Graduate student Mary Anne Peters and Groff are working on CHARIS.

“The instrument makes it possible to look at the spectrum at each point in the image,” said Groff, “so you can distinguish the planet light from the star and see whether the planet’s atmosphere is uniform or cloudy, and you can get an idea of age because as the planet gets older, it cools.”

Beyond detection

Detecting exoplanets, whether by watching for their transits or by direct imaging, is just the first step in developing an understanding of these objects, said Adam Burrows, a professor of astrophysics who uses data gathered from these campaigns to construct theories about the characteristics of exoplanets and planetary systems. “Once we’ve detected planets, how do we figure out their makeup, their atmospheric compositions and temperatures, and their climates? These are the kinds of questions we are interested in answering,” Burrows said.

These questions can be addressed only by detecting and interpreting the spectral emissions that will be detected by instruments such as CHARIS, but so far these are only available for large exoplanets. “As larger ground-based telescopes and space-based missions like the James Webb Space Telescope come online,” Burrows said, “we will have data from planets that are closer in size to the Earth. The objects being studied now are but stepping stones toward the broader characterization of the planets in general in the galaxy and in the universe.”

This broader characterization includes ongoing studies by a number of other Princeton researchers, including Markus Janson, a NASA Hubble Postdoctoral Research Fellow in astrophysical sciences, who studies how planets are formed from dust and debris that orbits the star. Other researchers studying planet formation include Roman Rafikov, assistant professor in astrophysical sciences, and Ruobing Dong, who earned his Ph.D. in 2013 while working on the SEEDS project and is now a NASA fellow at the University of California-Berkeley. Emily Rauscher, a NASA Sagan Postdoctoral Fellow in astrophysical sciences, is studying the climate on these faraway worlds.

Many researchers hope that studying exoplanets will help us learn more not only about planetary formation and solar systems but also about whether other planets exist that could support life. The instruments and telescope networks being developed at Princeton could lead the way. And if a wild burro chews on a cable now and then, well, it is part of the cost of learning what lies outside our solar system.

Box: Data mining for planets

Xu (Chelsea) Huang

Xu (Chelsea) Huang (Photo by Keren Fedida)

Data mining for planets Xu (Chelsea) Huang remembers the thrill of finding her first planet. “It was exciting,” said the graduate student in astrophysical sciences. Huang found that planet and many more in 2012 while looking through a publicly available data set from NASA’s space-based Kepler mission, which scans for dips in starlight as the planet crosses in front of the star. Using techniques developed for analyzing HATNet findings under the guidance of Associate Research Scholar Joel Hartman and Assistant Professor Gáspár Bakos, Huang found 150 potential planets — many of which were hot “super-Earths” that are slightly large than Earth but orbiting their host stars much more closely — that the Kepler team and others had missed. The paper was published earlier this year in the journal Monthly Notices of the Royal Astronomy Society. When the Kepler mission later released an updated list of possible planets, about half the ones that Huang had found were on it.

Box: Forecasting the climate on other worlds

Emily Rauscher

Emily Rauscher, a NASA Sagan Postdoctoral Fellow at Princeton’s Department of Astrophysical Sciences, is modeling the climate on exoplanets. (Photo by Andrew Howard)

Emily Rauscher, a NASA Sagan Postdoctoral Fellow in the Department of Astrophysical Sciences, said that new acquaintances don’t believe her when she says she does climate modeling for exoplanets. Rauscher uses what is known about our solar system, plus the laws of fluid dynamics and the exoplanets’ orbital period and mass, to try to understand the climate on these faraway worlds. “If you watch a planet throughout its orbit, you can see the change in the amount of light emitted from the planet’s night side versus from its day side,” Rauscher said. “Because it is very hot in the day and very cold at night, we expect winds to blow around the planet, and by measuring the difference in brightness coming from the planet, we can detect how the wind affects the planet’s temperature.” Rauscher is fascinated by the idea of life on other planets but said that there is plenty to discover even on uninhabitable hot-Jupiters. “There is a big push to discover Earth-like planets,” she said, “but there is a lot we can learn from studying the planets we know about already.”

Box: Exploring how planets are formed

Studying exoplanets also could help researchers learn more about how planets are formed from dust and debris that orbits the star, said Markus Janson, a NASA Hubble Postdoctoral Research Fellow in the Department of Astrophysical Sciences. Some of the material that doesn’t end up in planets is collected in rings called debris disks, he explained. Our solar system has two such debris disks: an asteroid belt between Mars and Jupiter and the Kuiper Belt beyond Neptune.

“Our conventional theory of how planets form— that dust sticks together and forms into planets like Earth, and that sometimes large amounts of gas accumulate onto a rocky core to form gas giants like Jupiter — is based on what we’ve observed in our solar system.” Janson said. “Now we can study other solar systems, so we can test this theory.” A recent study by Janson and colleagues, accepted for publication by The Astrophysical Journal, indicates that the majority of exoplanetary systems probably did form in this manner.

-By Catherine Zandonella