Thursday, February 23, 2012

Between Earth and Uranus: Part I

Seekers after alien life tell us to follow the water
For some years now, extrasolar headlines have heavily featured Super Earths – objects not quite like our ordinary Earth, yet much more reminiscent of home than the typical high-mass worlds that dominate the exoplanet census. Over the last several months in particular, as radial velocity and transit searches have attained the requisite degree of sensitivity, we’ve been hearing enthusiastic reports of habitable Super Earths: big sisters of our blue marble that orbit at just the right distance from their host stars to support bodies of liquid water, if only their physical and chemical compositions are just right, too.

Recent candidates for the Goldilocks prize – not too hot, not too cold, not too hard, not too soft – have been announced around a nearby K dwarf (HD 85512 b), a nearby M dwarf (GJ 667 Cc), and a faraway Sun-like star (Kepler-22b). Each announcement has been accompanied by an artist’s view of the exciting new real estate – appealing spheres with frothy white clouds and a hint of blue oceans that seem inspired more by the allegedly Earthly aspect of their subjects than by their demonstrably Super features.

 Artist’s view of Kepler-22b, an object in the telluric to gas dwarf range (“Super Earth”) orbiting a G-type star about 620 light years away. Image credit: NASA/Kepler Mission
Super Earths are typically understood as planets with masses between 2 and 10 times Earth's and radii up to about 3 times Earth's. The term seems to conjure images of big planets with wide oceans and lots of leg room. But the cold, hard reality is that candidate Super Earths are less likely to be scaled-up versions of home than they are to be scaled-down versions of Uranus. To unpack this claim, we need to take a closer look at our distant neighbor in the outer Solar System, as well as at those still more distant extrasolar planets. Only then can we make a sober assessment of the biotic potential of telluric planets orbiting alien suns.

the other ringed planet

Uranus is an odd and interesting place: greenish, very cold, and pitched almost sideways on its axis. Among the Solar System’s eight planets, it is third in diameter, fourth in mass, and seventh in distance from the Sun. It has a deep, stormy hydrogen atmosphere and apparently no solid surface; an astronaut descending through its cloud decks would eventually become enveloped in plasma and slush. Uranus is distinguished by an impressive ring system (second only to the rings of Saturn) and extreme obliquity. While Earth rotates on an axis tilted about 23 degrees with respect to our system’s orbital plane, Uranus is tilted a whopping 97 degrees, so that its north polar region points at the Sun on the northern summer solstice.

Uranus has an extremely tilted axis
Like our system’s other giant planets, Uranus supports an entourage of icy moons. However, given its extreme axial tilt, the Uranian moons circle their host like compartments on a Ferris wheel rather than horses on merry-go-round, as is the case for the other giants’ satellites.

Our understanding of extrasolar planets in other star systems depends heavily on our knowledge of the much more accessible worlds in our Solar System. These fall neatly into three categories on the basis of mass, composition, and distance from the Sun.

toy planetology

The terrestrial or telluric planets have one Earth mass (1 Mea) or less. They are composed almost entirely of elements heavier than helium, with metallic cores, silicate mantles, and relatively thin atmospheres of heavy gases (Venus, Earth, Mars) or no atmosphere at all (Mercury). All travel in the warm, inner regions of the Solar System, where freely orbiting ice particles sublimate in the Sun’s heat.

The gas giant planets have masses in the range of 95 Mea (Saturn; 0.3 Mjup) to 318 Mea (Jupiter; 1 Mjup), with their bulk composition dominated by hydrogen. Although both planets have modest cores of rock, metal, and high-pressure ices, heavy elements constitute only 5%-15% of their overall mass. The remaining 85%-95% is contributed by deep envelopes of hydrogen and helium, in a gaseous phase at the top and metallized at the bottom. Both gas giants orbit outside our system’s ice line, the boundary past which free-floating water molecules stay frozen.

The Sun and the 8 planets at their approximate relative sizes
All these telluric and gas giant planets have naked-eye visibility from the Earth’s surface, so they have been known since prehistory. Well beyond the orbit of Saturn lie the “telescope planets,” Uranus and Neptune, which were discovered only when optical technology achieved the necessary sophistication. They are similar in mass, at 14.5 Mea and 17.2 Mea, respectively, and composed primarily of rock, metal, and ice, with hydrogen envelopes contributing about 15%-20% to their bulk composition. For much of the 20th century, however, astronomers regarded these two objects as smaller versions of Jupiter and Saturn, so that all four outer planets were simply known as “giant planets” or “gas giants.” Only after a series of robotic missions to the outer Solar System returned a better understanding of each planet’s atmosphere and interior structure did it become common to draw a distinction between the gas giants proper, which are mostly hydrogen, and the ice giants (Uranus & Neptune), which are mostly heavy elements.

This neat tripartite schema – rocky planets, icy planets, gas planets – works well for our Solar System. It supports a widely endorsed theory of planet formation, according to which each type of planet in our system formed more or less in the same region where it now orbits. The small rocky planets coalesced along the dry orbits of the inner system; the more massive gas giants emerged in the “sweet spot” of planet formation, where abundant ices supplemented refractory elements, and the resulting cores captured huge quantities of hydrogen; and the middling ice giants grew in the outer regions, where ice is still abundant but slow orbits mean smaller planets with thinner envelopes.

Of course, this toy model of planetology is too simple to explain the substantial gap in mass and radius between Earth, the largest telluric planet, and Uranus, the less massive of the two ice giants, or the equivalent gap between Neptune, the more massive ice giant, and Saturn, the less massive gas giant. Nor can it tell us why the three planetary species have remained near their original homes in the Solar System, while migrating inward to converge on short-period orbits in so many extrasolar systems. For a fuller understanding, we need to confront the accumulated data on that mediagenic new type, the Super Earths, which the past year has brought forth in such abundance.

Extrasolar evidence can now populate the structural void between Earth and Uranus with more than 90 worlds whose estimated or minimum masses fall between 1 Mea and 14.5 Mea, as well as another 80 in the void between Neptune and Saturn, with masses between 17.2 Mea and 95 Mea. In the next installment we’ll take a deeper look into the first of these gaps and investigate the mysteries of the Super Earths.

Sunday, February 5, 2012

Thousands and Thousands of Planets

The three brightest stars in this view form an asterism known as the Summer Triangle. The Kepler field of view occupies the patch of sky between Deneb and Vega. 

Last year saw amazing developments in extrasolar astronomy, and the first weeks of 2012 have continued the trend. Existing techniques for detecting alien planets have dramatically improved, while new search programs – especially the Kepler Mission – have brought a rich harvest of new worlds. As our understanding of the diversity of planetary systems improves, we begin to see past the biases and limitations of previous data. This accumulation of new knowledge makes it clear that our own Solar System is odd, rare, and precious beyond measure.

the first detections

Despite centuries of speculation on the plurality of worlds, as well as a half-century of interstellar adventures in books and movies, the idea of planets orbiting nearby stars remained science fiction until 20 years ago. Late in 1991, a Polish astronomer working at Arecibo Observatory in Puerto Rico announced the detection of at least two planets orbiting a “pulsar” or pulsating neutron star, PSR 1257+12. The orbital motion of these objects had been deduced from minuscule variations in the arrival time of the host star’s radio pulsations. As the discovery team noted a few months later, in a letter to the journal Nature, these two planets cannot be primordial companions of their host. Neutron stars are the remnants of extremely massive stars (spectral types O or B) that have undergone supernova explosions – cataclysms that no planet could survive. Instead, these objects are most likely “second generation planets” that were created when the parent star disintegrated (Wolszczan & Frail 1992).

The pulsar planets were so bizarre, unexpected, and inhospitable to any imaginable form of life that their discovery gained little currency in mass media, which prefer alien worlds that are simply slight variations on Earth (see Krypton, Altair IV, and Tatooine). Similarly scant attention met the announcement, late in 1995 and also in Nature, of a likely gas giant planet orbiting a nearby Sun-like star. This object, 51 Pegasi b, was the first of the Hot Jupiters: exoplanets similar in mass to our own Jupiter, circling stars like our own Sun in incandescent orbits of just a few days. Their precise origin continues to confound theorists.

Nevertheless, as their Swiss discoverers observed, “the search for extrasolar planets can be amazingly rich in surprises” (Mayor & Queloz 1995). And with these two seminal discoveries, planets around alien suns entered our reality.

turn of the millennium

The census of exoplanets grew slowly during the 1990s. Hot Jupiters kept popping up, sometimes under a more anxious nickname – Marauding Jupiters. By the end of 2000, about 50 exoplanets had been announced, the vast majority by observing cyclical variations in the radial velocities of their host stars (sometimes known as the “wiggle” or “wobble” method). No additional pulsar planets had been identified, given the extreme rarity of pulsars in our region of the Milky Way Galaxy.

Two other detection methods were in use by then, with limited success: microlensing and transit surveys. The microlensing technique, which relies on the gravitational magnification produced by chance, transitory alignments of two widely separated stars, had not yet offered a single candidate. The transit technique, which measures the drop in stellar luminosity caused by a planet transiting (i.e., eclipsing) its parent star, had detected one gas giant, HD 209458 b. This discovery was significant because it represented the first time that the mass, radius, and approximate bulk composition of an extrasolar planet could be calculated. On the downside, the target planet was yet another Hot Jupiter.

HD 189733 b, a Hot Jupiter transiting a K-type star only 63 light years away. Image credit: ESA/C. Carreau

None of the exoplanets known in 2000 resembled the Earth, and none of the exoplanetary systems resembled our Solar System. The sensitivity of the radial velocity method was limited to gas giants on orbits of a few days to a few years, ruling out any architectures resembling our home system. Except for the pulsar planets, even the least massive exoplanets known were still substantially heavier than Saturn.

double, double

New discoveries emerged throughout the next decade at an ever-quickening pace. In 2006 the exoplanetary census, as tracked in the Extrasolar Planets Encyclopaedia, reached 200 planets. By late 2009 that number had doubled. A year later it hit 500. By September of 2011 it passed 600, and by mid-November it passed 700. As of today – less than three months later – the count is 758. In other words, more new planets have been identified in the past three months than in the entire decade of the 1990s. At this rate, we can expect the census to reach the magic number of 1000 before the end of 2013.

Along with increasing numbers has come increasing diversity, both of planets and planetary systems and of the techniques available to detect and characterize them. In radial velocity searches, ice giants in the mass range of Neptune and Uranus initially began to appear alongside gas giants. Then in 2005 came the first Super Earth, GJ 876 d, a potentially rocky object only half as massive as Uranus and just 6 times the mass of Earth. In terms of orbital architectures, modest systems with 1 or 2 detectable planets were joined by systems thronging with 5, 6, and 7 planets, most of them orbiting closer to their host stars than Earth is to our Sun.

Microlensing programs finally reported a planet in 2003 (the yield has since reached 14), while transit detections witnessed explosive growth, so that the number of confirmed transiting planets now exceeds 200 and encompasses objects of all types, from brown dwarfs and Super Jupiters to planets smaller than Earth. Even direct imaging – the old-fashioned way of observing, which relies on wavelengths originating from the planet itself rather than on the planet’s effects on its parent star – has seen success, reporting oddly massive planets around hot, nearby stars.

Many different search programs are now engaged in observations around the globe and beyond, contributing to the cascade of new information. Two groups in particular have made multiple headlines over the past few years: the team at Geneva Observatory, using the HARPS spectrograph at La Silla, Chile, to conduct high-precision radial velocity measurements, and the Kepler Mission, using the space-based Kepler Telescope to conduct high-precision transit searches. While HARPS focuses on stars in the immediate Solar neighborhood (especially within a radius of 50 parsecs/160 light years), Kepler observes stars located at a distance of a few dozen to a few thousand parsecs (~100-7000 light years) in a single patch of the sky.

data explosion

The Kepler Telescope was launched on an independent orbit around the Sun in 2009. Its first large dataset was released in 2010. An astonishing number of transiting candidates have been revealed: 706 potential planets in the first report, increasing to 1235 in early 2011, and then to 2000 by the end of 2011. More than 50 of these candidates have already been confirmed by follow-up observations and analyses, meaning that hundreds more transiting exoplanets are likely to augment the census within the next few years.

And that brings us to our present abundance of galactic wonders. These are the extrasolar highlights of 2011, constituting an annus mirabilis for planetary science:

Artist's view of the Kepler-11 system. Image credit: NASA/Kepler Mission
  • Circumbinary planets are real. These are planets that orbit both members of pair of binary stars, and before last year, despite decades of hopeful speculation, they were unknown. Kepler data now demonstrate that “several million” circumbinary planets must exist in our Galaxy (Welsh et al. 2012). In popular media, of course, these objects have been christened Tatooine planets.
  • Kepler-16b, the first circumbinary planet, was announced in Science in September (Doyle et al. 2011). Kepler’s photometer detected a complex pattern of transits indicating that a close pair of stars – one a K dwarf about 69% as massive as the Sun, the other an M dwarf about 20% as massive – is orbited by a cool Saturn-mass planet. (Watch this video.) The two stars (Kepler-16 A and B) share an orbit of 41 days, while the planet (Kepler-16b) orbits the system’s common center of mass in a period of 229 days, a few days longer than the period of Venus around our Sun. On first glance this configuration might suggest a rather steamy giant planet. But since the combined masses and luminosities of the two host stars are less than those of our Sun, Kepler-16b actually orbits in the system’s habitable zone, as Earth does in our system. If perchance Kepler-16b had a massive rocky moon (like Pandora in Avatar), that moon might support life.
  • Low-mass objects in the range of Mars to Neptune vastly outnumber gas giants like Jupiter and Saturn, not simply in our own Solar System but in our general neighborhood of the Galaxy. Kepler data have amply confirmed previous estimates from microlensing searches (Borucki et al. 2011, Sumi et al. 2010). Even better, both Kepler and HARPS demonstrate that Super Earths – objects whose masses and radii fall within a factor of a few of the Earth – are common around stars of a range of spectral classes, from G-type stars like the Sun down to the ubiquitous M dwarfs that comprise at least 75% of the stars in our Galaxy.
Artist's view of the first Earth-size exoplanets. Image credit: NASA/Kepler Mission
  • Super Earths also seem to be relatively common in the habitable zones of their stars – the orbital region where temperatures would support bodies of liquid water on a rocky planet. The Kepler Mission reported one such temperate Super Earth in 2011, known unromantically as Kepler-22b. This object orbits a G-type star like our Sun about 180 parsecs (590 light years) away (Borucki et al. 2012). The HARPS search offered two candidates with even less memorable names: HD 85112 b, orbiting an orange K-type star just 11 parsecs (36 light years) away, and GJ 667 Cc, orbiting an M dwarf that is literally in the Sun’s back yard, at a distance of only 7 parsecs (23 light years). (See Pepe et al. 2011 for HD 85112 b and Bonfils et al. 2011 for GJ 667 Cc.)
  • Other brand-new discoveries and theoretical studies, however, demand a rethinking of the prevailing picture of Super Earths. Although astronomers have often used this nickname to describe any object with a minimum mass smaller than 10 times that of Earth (Uranus, by comparison, has 14.5 times Earth’s mass), it now appears that this mass range includes planets with hydrogen atmospheres. Such objects are better described as Mini Neptunes rather than Super Earths. The statistics of the Kepler-11 system, announced in February 2011, set off the first alarms. This compact system contains several transiting planets with small masses but unexpectedly large radii, implying very deep atmospheres. Then came a study by Leslie Rogers and colleagues, demonstrating that planets with as little as 3 times the Earth’s mass can retain significant envelopes of gaseous hydrogen, provided they maintain an appropriate distance from their host stars. As a result, some or all of the “habitable Super Earths” so breathlessly reported in the mass media are very likely to be uninhabitable siblings of Uranus.
  • Not to worry, though, because Kepler has demonstrated that it can find planets even smaller than Earth, which would necessarily be composed of heavy elements like our home. The first sub-Earth candidate, Kepler-20e (announced in December), is much too hot for liquid water, but perhaps cooler and more congenial worlds will emerge from future analyses.
  • Finally, planetary systems consisting of two or more Super Earth to Neptune-mass planets in a highly compact orbital configuration are very common. HARPS and Kepler reported several such systems in just one year, augmenting the count that has been growing since 2006, when the Neptunian triplets orbiting HD 69830 were announced. By raw numbers, systems containing a single Hot Jupiter still account for more than 25% of all exoplanetary systems, vastly outnumbering the confirmed compact systems with multiple low-mass planets. But numerous studies have found that, once detection bias is removed, far fewer than 1% of all Sun-like stars harbor Hot Jupiters. Kepler data, meanwhile, suggest that compact multiplanet systems are much more common than Hot Jupiters in the Kepler field of view. Unfortunately, we have no guarantee that such compact, low-mass systems are any more likely than Hot Jupiter systems to support habitable planets.

And that's it for last year's marvels in exoplanet exploration. Future postings will examine some of these in more detail and consider new discoveries as they appear. (Life circumstances and general state of sanity permitting, as always.)


Bonfils X, Delfosse X, Udry S, Forveille T, Mayor M, Perrier C, et al. (2011) The HARPS search for southern extra-solar planets XXXI. The M-dwarf sample. Astronomy & Astrophysics, in press.

Borucki WJ, Koch DG, Basri G, Batalha N, Boss A, Brown TM, et al. (2011) Characteristics of Kepler planetary candidates based on the first data set. Astrophysical Journal 728, 117.

Borucki WJ, Koch DG, Batalha N, Bryson ST, Rowe J, Fressin F, et al. (2012) Kepler-22b: A 2.4 Earth-radius Planet in the Habitable Zone of a Sun-like Star. Astrophysical Journal 745, 120.

Doyle LR, Carter JA, Fabrycky DC, Slawson RW, Howell SB, Winn JN, Orosz JA, Prsa A, Welsh WF, Quinn SN, Latham D, Torres G, Buchhave LA, Marcy GW, Fortney JJ, Shporer A, Ford EB, Lissauer JJ, Ragozzine D, Rucker M, Batalha N, Jenkins JM, Borucki WJ, et al. (2011) Kepler-16: A Transiting Circumbinary Planet. Science 333, 1602-1606. 

Mayor M, Queloz D. (1995) A Jupiter-mass companion to a solar-type star. Nature 378, 355-359.

Pepe F, Lovis C, Segransan D, Benz W, Bouchy F, Dumusque X, Mayor M, Queloz D, Santos NC, Udry S. (2011) The HARPS search for Earth-like planets in the habitable zone I. Very low-mass planets around HD 20794, HD 85512, and HD 192310. Astronomy & Astrophysics 534, A58.

Rogers L, Bodenheimer P, Lissauer JJ, Seager S. (2011) Formation and structure of low-density exo-Neptunes. Astrophysical Journal 738, 59.

Sumi T, Bennett DP, Bond IA, Udalski A, Batista V, Dominik M, Fouque P, et al. (2010) A cold Neptune-mass planet OGLE-2007-BLG-368LB: cold Neptunes are common. Astrophysical Journal 710, 1641-1653.

Welsh WF, Orosz JA, Carter JA, Fabrycky DC, Ford EB, Lissauer JJ, et al. (2012) Transiting circumbinary planets Kepler-34 b and Kepler-35 b. Nature 481, 475-479. With online supplementary material.

Wolszczan A, Frail DA. (1992) A planetary system around the millisecond pulsar PSR 1257+12. Nature 355, 145-147.