Monday, August 6, 2012

The Study of Light: Transiting Planets

Distances on an astronomical scale are somewhat incomprehensible to us as humans. Currently, there’s no feasible way we can personally go anywhere farther than perhaps Mars, all we can send as a human representative is some kind of unmanned robot. (As a side note, Curiosity has landed!) In order to study the vast depths of space, we’re relatively limited to one of our senses, our eyes. Studying astronomy is essentially detecting and analyzing photons: for now, all we can do is measure light.

But what we can infer and learn from that light is astronomical on its own scale. From stellar light, we can learn about stars’ chemistry, size, distances, and the list goes on. One of the most recent and exciting developments in our knowledge of outer space is something that has been theorized for centuries--the existence of extrasolar planets--and comes simply from observing light.

Just like when our moon eclipses the Sun, when exoplanets eclipse, or transit, their host star, the star grows visibly dimmer. The method of observing planet transits is centuries old, and originated in our own Solar System. The early transits of the inner planets in our Solar System were predicted by Johannes Kepler, which were generally very accurate. This isn’t surprising, as the transit method hinges directly upon Kepler’s Third Law.

But planets are tiny compared to stars. Observing an exoplanet transit requires very precise instrumentation to be able to detect such tiny fluctuations in stellar flux. However, modern technology has evolved to the point where we can do this, measuring changes up to a part in ten thousand, or better. Thus, when a star has a planetary companion and its orbit is oriented just right, we observe a dip in the star’s light curve. Occultations, sometimes called secondary transits, occur when a planet passes behind its host star and there is a consequent dip in light, due to the planet’s light being lost. However, this is a much smaller effect, and is currently limited to only the biggest and hottest planets.
From transits, we observe three quantities: , where P is period, tau is transit duration, and delta is how “deep” the light curve is in relation to the star’s non-transit stellar flux, a quantity equal to .

The transit detection method has evolved exponentially, simply beginning with Mercury and Venus passing in front of the Sun.

1631: Mercury’s transit was predicted and observed.

1639: Transit of Venus was observed by Horrocks and Crabtree. As a result, the estimated size of the solar system increased ten times, but the estimate was still smaller than the actual size.

A mural at Manchester Town Hall by Ford Madox Brown. From wikimedia commons.

During the mid-1900s, we hadn’t observed an exoplanet transit yet, but this did not mean that scientists didn’t speculate about them.

1952: Otto Struve asserts that during a transit, the observed stellar flux would decrease in magnitude by .02. This corresponds to a fractional change of 1/50. This would have been observable by photometric technology of the time, but none were detected.
1984: Borucki and Summers postulate that in order to detect terrestrial planets, more precision was needed to lessen atmospheric influence. Their proposed program required monitoring hundreds or thousands of stars and expected to detect one planet per year.
1992: Tutukov estimates that 30% of all stars have planetary systems. He said the search for planets around sun-like stars was “unpromising” using the radial velocity detection method, and we should approach finding exoplanets with the transit method. He estimates that for 4000 stars with , ten should have planetary systems detectable by transit. In addition, he postulated that brown dwarfs cannot be detected except by transit.

The first extrasolar transits observed were from exoplanets that had already been detected with radial velocity measurements.

1995: Mayor et al. discover a Jupiter-sized companion to 51 Pegasi from radial velocity measurements. The planet has a semimajor axis smaller than Mercury’s, and is either a gas giant or a brown dwarf. This is the first exoplanet discovered around a sun-like star.
1999: Henry, Marcy, Butler, and Vogt find a transiting 51 Peg-like planet. First radial velocity was used to detect the planet, then the transit was observed--the first extrasolar transit. The planet’s radius, mass, and density were derived to imply a gas giant in a circular orbit.
1999: Charbonneau, Brown, Latham, and Mayor publishes a paper simultaneously with the Henry group, also using radial velocity to find an exoplanet and then acquiring its light curve. It was also a gas giant, with a semimajor axis of .047 AU and signs of water vapor. It was determined to have an atmosphere, and was one of the first to be observed spectroscopically. The team found the planet’s mass, radius, surface gravity, escape velocity, and average density. This planet turned out to be the exact same planet that the Henry group was observing, and therein lies controversy on who actually recorded the first extrasolar transit.

Finally, a planet was actually discovered via transit, giving rise to the mass transit surveys we see today. Simply analyzing photons from a distant star pointed to the existence of a planetary companion.

2003: Konacki et al. discovered the first planet using the transit method. After the radial velocity method was combined with the transit spectra, the planet was found to have .9 Jupiter mass, a semimajor axis of .023 AU, and a temperature of 1900 K (hot!).
2004: Alonso et al. conducted the first transit survey.
2006: CoRoT, which stands for Convection, Rotation, and planetary Transits; is launched in December by the French. Its primary purpose was to use stellar seismology to discern the inner structure of stars, but it also used transits to detect companion planets. It aims to find telluric rocky exoplanets similar to Earth. It does not publish planet candidates, only fully characterized planets.
The COROT Space Telescope. (Copyright 2006 CNES)

2009: The Kepler mission is launched in March, aiming to find potentially the largest sample of exoplanets to date through transit photometry. It is designed to discover Earth-like planets in the habitable zone of their host stars by focusing on one small (but deep) portion of the sky that includes about 150,000 target stars.
Kepler's field of vision in the Milky Way. From wikimedia commons.

Scientific discovery, as seen with the transit detection method, is a lot like a fire. Once the first spark is ignited, it grows, expanding and encompassing. The method of detecting planets via transit originally began as a prediction, but has developed into a reality which has enabled us to learn so much more about planets beyond our solar system, planets around distant specks of light. And the beauty of discovery is that there is no limit; our knowledge of exoplanets continues to broaden, and there will always be fascinating things to uncover.


  1. Your prediction was right: I did like this one! Very nice post. The historical summary is concise yet clear. One request: I'd like to see more about how tau and T come into play and help us interpret the transit light curve.

  2. Cool timeline! It's amazing how much progress has been made in the last 50 years...

    It's crazy to think of someone in 1631 predicting and observing a transit. How did they (they being... who, Kepler?) observe that first transit of Mercury? Did they just stare at the sun constantly and wait for something to happen? How did Kepler predict when the transit would happen, and how did he know it would be visible from wherever they were observing (I assume Europe)?