One Big Ball of Rock

The planet of HD 149026; artwork by one of the researchers
This illustration, by exoplanet researcher Gregory Laughlin, attempts to be 'at least marginally scientifically accurate,' he says. The planet's night side should be so hot that it glows from dull red to bright orange-hot, depending on how deep we see into its expected refractory cloud layers. 'It's much hotter than, say, a very hot barbeque briquet,' says Laughlin. The side that's lit by the nearby star, by comparison, is dazzlingly brilliant. There is limb-darkening around the planet's night edges.
Artwork by Gregory Laughlin.
Less than a decade ago, each new extrasolar planet discovery was greeted with accolades and headlines. Now, with about 160 exoplanets on the books, it takes something special to attract attention. Today an international consortium of American, Japanese, and Chilean astronomers did just that. This afternoon the group announced perhaps the most bizarre planet yet: an object with a core of heavy elements that may amount to 65 or 70 times the mass of Earth.

The newfound body — not to be confused with another possibly rocky planet announced two weeks ago with a much lower mass — is a whole new animal. It contains as much or more heavy elements (elements heavier than hydrogen and helium) than all the planets and asteroids in our solar system combined. Astronomers have assumed that virtually all the exoplanets found to date are gas giants like Jupiter and Saturn, with heavy elements such as oxygen, silicon, carbon, and iron constituting at most one-fourth of their masses. But the new planet appears to be one-half to two-thirds heavy stuff. "This object is odd, even given the weird zoo of planets found so far," says Alan Boss (Carnegie Institution of Washington).

The planet orbits the type-G0 IV star HD 149026, which is about 260 light-years from Earth. The star, magnitude 8.2, is easily visible in a small telescope (or even binoculars) a couple degrees northwest of the globular cluster M13 in Hercules, now high in the evening sky. The star is a little larger, brighter, and more massive than the Sun, and it is beginning to evolve off the main sequence toward becoming a red giant.

A Very Special Find

The story of the discovery began when team leader Debra A. Fischer (San Francisco State University) and her colleagues called attention to the star for its high content of heavy elements; stars of this sort have proven to be the most likely to host detectable planets. In July 2004 Bun'ei Sato (Kobe University, Japan) repeatedly measured the star's radial velocity with the 8.2-meter Subaru Telescope in Hawaii. These observations revealed a wobble that was likely induced by a planet in orbit. Observations in February and April 2005 at the Keck Observatory in Hawaii, by Geoffrey W. Marcy (University of California, Berkeley) and his colleagues, confirmed that a planet circles the star very rapidly: every 2.877 days. Based on the strength of its tug on the star, the planet, now named HD 149026b, had to have a minimum mass of about 115 Earths, or 0.36 Jupiters.

Artist's rendition of planet in transit
This artist's rendition shows the size of HD 149026b as it crosses the face of its Sunlike star. It blocks 1/330 of the star's light.
Painting by Lynette Cook.

Then came a breakthrough. During the early morning of May 11th,
Gregory W. Henry (Tennessee State University) followed the star with a robotic 0.8-meter telescope at the Fairborn Observatory in Arizona. Henry measured a 0.003-magnitude (0.3 percent) drop in the star's brightness that lasted for three hours at exactly the time when the planet would transit the star if its orbital plane were in our line of sight. Observations three nights later confirmed the transits. This is the eighth exoplanet discovered to cross the face of its host star.

The tight orbit
Although the planet orbits just 0.042 a.u. (6.3 million miles) from the center of its star, there's still plenty of room between them. Everything in this scene is drawn to scale.
Sky & Telescope diagram; data courtesy Gregory Laughlin.

The transits were the key that unlocked the planet's physical nature. The very slight amount of stellar dimming astonished team members. They were expecting a planet about the size of Jupiter, but the minuscule brightness drop means the planet's diameter is only 72 percent that of Jupiter. And due to the transit, astronomers know the planet's orbit is inclined nearly edge-on to our line of sight. This fortuitous alignment tells astronomers that the minimum mass they measured from the wobble is equal to the true mass.

Knowing both the planet's mass and size, astronomers easily calculated its mean density: 1.4 times that of water. The other seven transiting exoplanets all have considerably larger diameters and lower densities. "The other transiting planets are clearly made mostly of hydrogen and helium gas," says team member Gregory P. Laughlin (University of California, Santa Cruz).

What Is It Telling Us?

In the solar system, Saturn is closest in mass to the new find, with 95 Earth masses. But HD 149026b's density is 1.7 times greater — even though, unlike frigid Saturn, it must be roasted to red heat by being so close to its star. Given that Saturn is 25 percent heavy elements and has a core mass of perhaps 20 Earths, theorists judge that the new planet must be one-half to two-thirds heavy elements to account for its small size and high density. The material likely includes rock, metals, and compounds such as water, methane, and ammonia, which planetary astronomers call "ices."

Cutaways compared
Models of Jupiter's interior and the interior of HD 149026b are compared. In reality, the heavy elements may be partly dissolved in the next outer layer rather than sharply segregated into a core as shown here.
Sky & Telescope diagram; data courtesy Gregory Laughlin.

A model of planetary interiors developed by Peter Bodenheimer (UC, Santa Cruz) predicts that HD 149026b has a dense core of 65 to 70 Earth masses, perhaps surrounded by a layer of highly compressed liquid water, a layer of hydrogen and helium under such pressures that they behave like liquid metals, and a deep atmosphere of primarily hydrogen and helium gas. Independent calculations by the team of Mark Marley and Jonathan Fortney (NASA/Ames Research Center) and Didier Saumon (Los Alamos National Laboratory) reach a similar conclusion. But as Bodenheimer notes, "The heavy elements are not necessarily all in the core. Some of the material could be dissolved in the gaseous envelope, as is the case for Jupiter. But in either case, a large total heavy-element fraction is needed to explain the small radius."

HD 149026b is not a place you'd want to visit. The surface gravity at the top of the core could be as high as 10 g's, and the overlying pressure and temperature would simultaneously scrunch and melt a human into an organic puddle. A person would fare little better at the cloudtops, where the close proximity to the star — about 0.042 the average Earth-Sun distance — raises temperatures to a scalding 1,270° C (2,300° F).

Theorists are struggling to understand how any planet could accumulate so much heavy-element material. HD 149026b presumably formed when small planetesimals collided in a disk around the star during the star's infancy, building up a large rocky body that accreted more surrounding material. But as Alan Boss notes, "I have not seen any core-accretion models that predict the formation of such a beast. I suspect that the core-accretion folks will be scratching their heads for awhile over how this thing could have formed." Saumon adds, "I don't think anyone has a plausible formation mechanism for this new planet yet."

Model densities
Two possible models of the new planet's interior are compared to interior models of Saturn and Neptune. One model of the new planet assumes a core of rock only, the other of 'ices' only. Saturn and Neptune contain some of each. In each case, the center of the planet is at left, and its surface is at right.
Sky & Telescope diagram; data courtesy Jonathan Fortney.

The star itself has a concentration of heavy elements 2.3 times that of the Sun, so its protoplanetary nebula must have had plenty of planet- and core-building material. But models predict that planets with such massive cores will quickly sweep up huge amounts of gas too from the disk, forming an overgrown version of Jupiter or Saturn. Yet this planet has comparatively little gas. "This planet is some sort of hybrid between Jupiter/Saturn and Uranus/Neptune," says Fortney. He notes that most of Jupiter's and Saturn's masses are in their hydrogen/helium envelopes, whereas most of Uranus's and Neptune's masses are in cores of rock and ice. Like other close-in massive planets, HD 149026b presumably formed much farther from its star and migrated inward due to the gravitational drag effect expected in a protoplanetary disk.

Laughlin speculates that the high concentration of heavy elements could be the result of two or more planets that collided. If so, the collision probably should have knocked HD 149026b's orbital plane out of the star's equatorial plane. A transit observation made at Keck on June 25th, which is currently being analyzed by Marcy and his colleagues, should determine the star's rotational axis with respect to the planet's orbit.

"Another possibility is that maybe this object migrated in quickly and accumulated solid material sent in from the disk," says Laughlin. Marley suggests, "Perhaps this object formed in a gas-poor region of the nebula and was able to grow a massive core. I'm sure we'll hear a lot of theories."

Tristan Guillot (Cote D'Azur Observatory, France) notes that unlike planets residing far from the powerful gravity of their host stars, close-in planets like HD 149026 cannot eject small rocky bodies out of their planetary systems. Huge numbers of incoming small bodies should eventually impact close-in planets, adding to their inventory of heavy elements. "There are routes to forming these kinds of beasts," says Guillot.

A Target for Amateurs?

HD 149026b presents a tempting target for advanced amateur astronomers armed with CCD cameras and able to do precision photometry. Amateurs have already detected the deeper transits of the two previously known exoplanets orbiting the stars HD 209458 in Pegasus and TrES-1 in Lyra. But as Laughlin points out, the new planet's transits will be harder to record; the amount of stellar dimming is only a fifth as much as in HD 209458 and TrES-1. Making matters worse, there are no good reference stars of comparable brightness in HD 149026's field of view, which makes precision photometry all the harder.

Finder chart
A finder chart for the 8th-magnitude star HD 149026 in Hercules, high overhead these evenings. The starting landmarks are 3.5-magnitude Eta Herculis (one of the corners of the Hercules Keystone) and the 6th-magnitude globular cluster M13. North is up and east is left. For the scale of the chart, note the declination degree scale at right. (The star is at right ascension 16h 30m 29.6s, declination +38° 20' 50'). Click the chart for higher resolution.
Sky & Telescope diagram.

"It will be an exciting challenge for the amateurs to detect this transit, but I think they'll be able to do it," says Laughlin, who has organized the global amateur network to look for transiting exoplanets. Laughlin adds that amateurs will have to improve their techniques to detect HD 149026b's events, but this capability will be highly beneficial when they carry out additional searches.

Amateur transit observations could help refine the exact orbital period, which will enable professionals to schedule the perfect time to observe the star with the Hubble and Spitzer space telescopes to try to detect other properties of the planet, such as its temperature and whether it has rings or an extended atmosphere. Ground-based professional telescopes will continue to monitor the star to see if it harbors more planets yet unknown.

The unusual and unexpected nature of HD 149026b highlights the fact that aside from the eight known transiting exoplanets, astronomers really know very little about the other 150-plus bodies, other than their orbits and minimum masses. "This detection certainly emphasizes that some of these nontransiting planets may look structurally different than expected,"
says Debra Fischer. Her team's discovery paper has been accepted for publication in the Astrophysical Journal, and the team has set up a Web site with much information on the find.

"We still need to verify the data; there's an outside chance the radius estimate is incorrect," warns Adam Burrows (University of Arizona). He points out that only a handful of transits have been observed, and those have a relatively low signal-to-noise ratio. "But if the data are correct, it's one of the most interesting objects of the 160 known exoplanets."