2
The Hubble space telescope observed the area in the sky known as the Hubble Deep Field for ten consecutive days, taking multiple images in four different filter passbands: near-ultraviolet (300nm), blue-yellow (450nm), red (606nm), and near-infrared (814nm), for a total of 342 individual exposures.
3
There is an extensive literature on direct imaging for planet detection for both ground-based and space-based telescopes. There are two general types of solutions. One tries to minimize the light of the star by directly blocking it inside the telescope, while the other tries to minimize the light of the star by combining it in at least two telescopes and eliminating it through interference. The latter device is known as an interferometer, the former as a coronograph. The most ambitious interferometer proposed is a flotilla of telescopes orbiting around the Sun and maintaining a precise formation. The design is often associated with NASA's Terrestrial Planet Finder project and the European Space Agency's (ESA) Darwin project. Webster Cash, in “Detection of Earth-like Planets Around Nearby Stars Using a Petal-shaped Occulter,”
Nature,
July 6, 2006, has put forward a similarly ambitious proposal for an enormous coronograph telescope in space.
4
In special cases, when the planets are young, large, and orbit far from their stars, it is possible to discover them directly, as in the spectacular infrared images of star HR 8799 with its coterie of four planets found by Christian Marois et al., “Direct Imaging of Multiple Planets Orbiting the Star HR 8799,”
Science
322 (2008): 1348; and Marois et al., “Images of a Fourth Planet Orbiting HR 8799,”
Nature,
December 23, 2010.
5
SIM PlanetQuest was a NASA mission that was in a detailed design phase a few years ago. S. Unwin et al., “Taking the Measure of the Universe: Precision Astrometry with SIM PlanetQuest” (Astronomical Society of the Pacific, January 2008).
6
A. Wolszczan and D. Frail, “A Planetary System Around the Millisecond Pulsar PSR1257 + 12,”
Nature
355 (1992): 145.
7
Pulsars are the remnants of supernova explosionsâthe end product of the development of a star about ten times more massive than our Sun. Even if the original star had planets, the planets around the remnant pulsar today are not those. We do not have a good idea how the pulsar planets formed after the explosion of the star, and what these planets are made of is not clear.
8
The technique was proposed by M. Holman and N. Murray, “The Use of Transit Timing to Detect Terrestrial-Mass Extrasolar Planets,”
Science
307 (2005): 1288, and by E. Agol et al., “On Detecting Terrestrial Planets with Timing of Giant Planet Transits,”
Monthly Notices of the Royal Astronomical Society
359 (2005): 567, with the practical use of the transit method in mindâtransit timing variations. However, discovering an unseen planet by watching its effect on the orbit of a known planet has a venerable history. This is how the planet Neptune was discovered.
9
J. Lissauer et al., “A Closely Packed System of Low-Mass, Low-Density Planets Transiting Kepler-11,”
Nature
470 (2011): 53. In the case of Kepler-11, all planets were discovered by the transiting method, but transit timing variations allowed for the candidate planets to be confirmed and their masses measured.
10
When applied to stars, the effect is technically referred to as gravitational microlensing, in order to distinguish it from lensing between galaxies.
11
A. Einstein, “Lens-like Action of a Star by the Deviation of Light in the Gravitational Field,”
Science
84 (1936): 506; S. Mao and B. Paczynski, “Gravitational Microlensing by Double Stars and Planetary Systems,”
Astrophysical Journal
374 (1991): L37.
12
J. P. Beaulieu et al., “Discovery of a Cool Planet of 5.5 Earth Mass via Microlensing,”
Nature,
January 26, 2006; D. Overbye, “Astronomers Briefly Glimpse an Earth-like Planet,”
New York Times,
January 25, 2006.
13
OGLE stands for the Optical Gravitational Lensing Experiment, a US-Polish project that uses a telescope in Chile to detect stellar gravitational lensing events.
14
As stars orbit the center of the Milky Way Galaxy and we observe them from our own orbit in the Galaxy, they all appear to shift, albeit very slowly, with respect to each other. Occasionally they will literally pass in front of each other from our point of view. This is the moment when for a brief period of time we can see the gravitational bending of lightâthe effect of gravitational lensing. The smaller the mass of the lens, the briefer the event. With typical orbital speeds of stars in our Galaxy (200â300 km/ sec) and our own motion in the same general direction, the typical duration of a stellar lensing event is several weeks. The lensed star appears to brighten, peak, and then fade back to its original brightness; the peaks typically last just a few days. The signature of a planet is a separate peakâa blip that is superposed on the brightening of the star and lasts for less than a day. Under rare favorable circumstances the orbital motion of the planet may be discernable.
15
A. Gould et al., “Microlens OGLE-2005-BLG-169 Implies That Cool Neptune-like Planets Are Common,”
Astrophysical Journal
644 (2006): L37.
CHAPTER FOUR
1
The march, called “The Transit of Venus March,” was written by John Philip Sousa in 1882 for the nineteenth-century transit of Venus and to honor the first secretary of the Smithsonian Institution.
2
The transits of Venus occur either in pairs separated by an eight-year interval or as a single transit every 121 years; three transits in a short sequence never occur. We live in an era when the transits of Venus come in pairs. The present era started with the transit in 1631 and will end with the transit in 2984, followed by a cycle of single transits. The mechanics of this are nicely described by Eli Maor in
Venus in Transit
(Princeton: Princeton University Press, 2004).
3
W. Sheehan and J. Westfall,
The Transits of Venus
(New York: Prometheus, 2004).
4
See the amazingly successful community of amateur astronomers observing extrasolar planet transits on
www.transitsearch.org
andthe American Association of Variable Star Observers. In 2007 the former succeeded in discovering the transits of an extrasolar planet that had been discovered by the Doppler shift methodâHD 17156b.
5
Stars orbiting each other, as well as spinning on their axis, are randomly oriented in the Galaxy, as studied for many decades in orbits of binary stars.
6
The argument is purely geometrical: if the distribution of inclinations is random, then the probability of transit is (R
s
/2a).
7
The Doppler shift method discovery was described in a paper by T. Mazeh et al., “The Spectroscopic Orbit of the Planetary Companion Transiting HD 209458,”
Astrophysical Journal
352 (2000): L55, while the photometric detection of the transit was announced in the circulars of the International Astronomical Union by David Charbonneau et al. (IAU Circular 7315, 1999) and G. Henry et al. (IAU Circular 7307, 1999).
8
Based on the single known transiting planet HD 209458b, with its fairly deep transits and very “quiet” star, and a simple extrapolation ignoring many subtleties of the simultaneous photometric measurement of many stars (rather than a single one, moreoverâwith a known phase of the planet's orbit), estimates of the number of transiting planets that would be detected within a year run into the hundreds!
9
A. Udalski et al., “The Optical Gravitational Lensing Experiment. Search for Planetary and Low-Luminosity Object Transits in the Galactic Disk. Results of 2001 Campaign,”
Acta Astronomica
52 (2002): 1 and supplement on page 115.
10
The star is the bright star Beta Persei, an eclipsing binary star named by the Arab astronomer Al Gul (a.k.a. Algol), meaning “The Ghul's Head” and also referred to as the “Devil's Star,” most likely because Arab astronomers noticed its regular “blinks.” Italian astronomer Geminiano Montanari (1633â1687) noted its variability in 1670. John Goodricke (1764â1786), a celebrated English astronomer in the field of variable stars, rediscovered its variability, determined that it is strictly regular, and understood the nature of the dimming as eclipses.
11
We described the procedure, as it is now generally applied to all transit searches, in a series of papers. G. Torres, M. Konacki, D. Sasselov, and S. Jha, “Testing Blend Scenarios for Extrasolar Transiting Planet Candidates. I. OGLE-TR-33: A False Positive,”
Astrophysical Journal
614 (2004): 979; and “New Data and Improved Parameters for the Extrasolar Transiting Planet OGLE-TR-56b,”
Astrophysical Journal
609 (2004): 1071.
12
G. Torres et al., “Testing Blend Scenarios,” 979.
13
M. Konacki, G. Torres, S. Jha, and D. Sasselov, “An Extrasolar Planet That Transits the Disk of Its Parent Star,”
Nature
421 (2003): 507.
14
An excellent account of this and many other stories can be found in
The Taste of Conquest
by Michael Krondl (New York:
Ballantine, 2007), a well-researched and entertaining account of one of the main motivations behind the European age of explorationâthe spice trade.
15
“CCD” stands for “charged coupled device,” a silicon chip with 10-micron-size pixels arranged in rows and columns, detecting light and registering its brightness as an electrical charge at each pixel.
16
One station is the Fred Lawrence Whipple Observatory (FLWO) of the Smithsonian Astrophysical Observatory (SAO) on Mount Hopkins in Arizona with four telescopes, and the other is the rooftop of the Submillimeter Array Hangar (SMA) of SAO atop Mauna Kea, Hawaii. These telescopes are modest 0.11m diameter f/1.8 focal ratio telephoto lenses that use front-illuminated CCDs at five-minute integration times.
17
A transit lasts a couple of hours and recurs every few days (for a hot Jupiter), so it is easy to miss them if you have gaps in coverage. Even worse, you often end up with many partial transits (starts or ends) that can be completely useless because the photometry during start and end of night often has systematic errors.
18
Our proposal to NASA did not go through, but the drive to discover and study super-Earths grew stronger. The term “super-Earth” (as well as “super-Venus”) seems to have appeared first in print in our NASA proposalâwe called the imaging/spectroscopic mission ESPI; Melnick et al., “The Extra-Solar Planet Imager (ESPI): A Proposed MIDEX Mission,”
Bulletin of the American Astronomical Society
34 (2001). The context was spectral differences, namely, that we could distinguish between a gas giant, an ice giant, and a super-Earth spectrophotometrically in reflected light. Our ESPI team did not pay attention to refining the criteria for what we called a super-Earth, as the case was marginal for any detection anyway. It was a convenient shorthand and sounded better than “fat-Earth,” a short-lived suggestion in an email written by Tim Brown. In the Kepler proposal, which was written at about the same time as
ESPI, we never gave a name to the 2 and 10 Earth-mass planets. I joined the NASA Kepler team in 2000, a NASA mission that was powerful enough to deliver super-Earths and real analogs to Earth.
The term was used again in Valencia, O'Connell, and Sasselov, “Internal Structure of Massive Terrestrial Planets,”
Icarus
181 (2006): 545, a paper we wrote in 2004, though this time I made an effort in defining itâin terms of mass (1â10 Earth-mass). For the first super-Earth to be discovered (GJ 876d) by Rivera et al. (“A 7.5 M Planet Orbiting the Nearby Star, GJ 876,”
Astrophysical Journal
634 [2005]: 625), the authors did not use a specific term. The discoveries to follow were all based on the Doppler method and hence mass became the defining parameter as the term “super-Earth” was adopted by the observers.
We had an open discussion on the topic at a workshop in Nantes in June 2008. O. Grasset and I pushed for my viewâto call all RV-DETECTED planets below 10 Earth-mass “super-Earths” for the time being, since we'd be unable to distinguish between the subclasses of rocky super-Earths, ocean planets, and mini-Neptunes, and sort things out later. There were all kinds of opinions. For example, Michel Mayor and S. Udry suggested we limit the lower mass bound at 2 Earth-mass. Some were suggesting higher upper mass limits (up to 20â30), and some French colleagues were suggesting alternative names. In the end, there was no particular consensus.
19
Sunspots are temporary perturbations of the solar photosphere, the gaseous shiny surface of the sun, caused by the complex tangles of the solar magnetic field near the sun's surface. A sunspot is a region of the photosphere that has a lower temperature than its surroundings and hence appears black against the shiny surface of the sun. Sunspots are often the size of the projected circle of Mercury or bigger and can change little over days and even weeks. They appear to move slowly as the sun rotates, which is in the same direction as the planets, and in almost the same plane.
20
Eli Maor,
Venus in Transit
(Princeton: Princeton University Press, 2004), gives a detailed account of Gassendi's life and transit observations.
21
See Maor,
Venus in Transit.
22
Maor,
Venus in Transit,
27.
23
K. Chang, “Puzzling Puffy Planet, Less Dense Than Cork, Is Discovered,”
New York Times,
September 15, 2006.
24
See Maor,
Venus in Transit.