Centauri Dreams

Strong evidence for water in the atmosphere of the hot Jupiter Tau Boötis b has turned up, thanks to work by Geoffrey Blake (Caltech) and graduate student Alexandra Lockwood. But what’s intriguing about the find isn’t the water — we’ve found water vapor on other planets — but the method of detection. Lockwood and Blake used a modified radial velocity technique that has previously been deployed to detect low mass ratio binary stars. A top-flight instrument like the Near Infrared Echelle Spectrograph (NIRSPEC) at the W. M. Keck Observatory in Hawaii can separate the planetary and stellar components spectroscopically to produce this result.

Image: Simulated data showing the method used for detecting water vapor features around the hot Jupiter tau Boötis b. In this example, the planetary signal has been increased in strength by several orders of magnitude relative to the actual signal. The dotted lines show the blue- and red-shifts of the planetary and stellar lines in the data, respectively. Credit: Alexandra Lockwood/Caltech.

Thus the toolkit of exoplanet science continues to grow. We already know how to study planetary atmospheres when a planet transits its primary as seen from Earth, allowing us to separate direct light from the star from light that has passed through the atmosphere. Another way of studying an atmosphere is through direct imaging, but here we rely on planets far enough from their host star that we can separate them from the star’s glare, the classic problem of planet hunters. But a non-transiting world like Tau Boötis b demands a different approach, and the spectroscopic separation of radial velocity signals appears to be one solution.

Lockwood is lead author of the paper on this work, which appears in The Astrophysical Journal Letters:

“The information we get from the spectrograph is like listening to an orchestra performance; you hear all of the music together, but if you listen carefully, you can pick out a trumpet or a violin or a cello, and you know that those instruments are present. With the telescope, you see all of the light together, but the spectrograph allows you to pick out different pieces; like this wavelength of light means that there is sodium, or this one means that there’s water.”

It was back in 2012 that two independent teams studying Tau Boötis b were able to separate the radial velocity of the planet from that of the star by studying shifts in the spectral lines of carbon dioxide. The technique as expanded into the infrared now gives us the molecular signature of water. I should mention here that a team in 2010 detected carbon monoxide in the atmosphere of another hot Jupiter, the much studied HD 209458 b, a detection that provided what researchers call the ‘spectroscopic orbit’ of the system and thus revealed the true mass of the planet.

So more is at play here than the detection of yet another component of a hot Jupiter’s atmosphere. Separating the two radial velocity components helps to determine a planet’s mass, which in earlier radial velocity studies could only be an estimate of the planet’s minimum mass. The actual mass could be much higher than what astronomers could measure, given their lack of knowledge about the inclination of the planet to the star. By detecting a signature from both planet and star, this method allows us to explore the system in greater detail. From the paper (with reference to the 2012 work on carbon dioxide mentioned above):

…this technique can be applied to non-transiting, RV-detected exoplanets to extract the unknown inclination and true mass. CRIRES [the High-Resolution IR Echelle Spectrometer installed at ESO’s Very Large Telescope] was also used to detect CO on τ Boötis b (Brogi et al. 2012), the first ground-based detection of a short-period non-transiting exoplanet atmosphere, a result confirmed shortly thereafter by Rodler et al. (2012). These studies provide the true mass of the planet and probe the chemical composition of its atmosphere. A combination of high signal-to-noise, high spectral resolution, and coverage of multiple CO overtone lines was required to achieve the sensitivity required for these detections.

So by finding the signature of water in the atmosphere of Tau Boötis b, the Caltech team is also extending work that creates what Lockwood calls the ‘3-D motion of the star and the planet in the system.’ It’s a method that deepens our understanding of the planetary system by giving us a true mass reading, and although at present its applications are limited to close-in gas giants like Tau Boötis b around nearby bright stars, the hope is that improvements in telescopes and infrared spectroscopy will extend the method to smaller planets around dimmer stars.

“While the current state of the technique cannot detect earthlike planets around stars like the Sun, with Keck it should soon be possible to study the atmospheres of the so-called ‘super-Earth’ planets being discovered around nearby low-mass stars, many of which do not transit,” adds Blake. “Future telescopes such as the James Webb Space Telescope and the Thirty Meter Telescope (TMT) will enable us to examine much cooler planets that are more distant from their host stars and where liquid water is more likely to exist.”

The paper is Lockwood et al. “”Near-IR Direct Detection of Water Vapor in tau Boötis b,” The Astrophysical Journal Letters, published online 24 February 2014 (preprint).

CSIRO, the Commonwealth Scientific and Industrial Research Organisation in Australia, is announcing the detection of violent events around the pulsar PSR J0738-4042, some 37,000 light years from Earth in the constellation Puppis. This southern hemisphere constellation was originally part of a larger constellation called Argo Navis, depicting the ship made famous by the journey of Jason and the Argonauts. But Argo Navis was divided into three smaller constellations, leaving Puppis (The Stern) as something of a mythological fragment.

Image: An artist’s impression of an asteroid breaking up. Credit: NASA/JPL-Caltech.

Whatever its origins, Puppis is also home, from our Earthly perspective, to a pulsar around which radiation and sleeting high energy particles are common. Ryan Shannon, a member of the CSIRO research team, has previously examined how an infalling asteroid from a violent disk around a pulsar might affect it, slowing the pulsar’s spin rate and affecting the shape of the radio pulse that we see on Earth. That’s a useful signal, because pulsars flash a radio beam with great regularity, so that any disruptions call attention to themselves. Says Shannon:

“That is exactly what we see in this case. We think the pulsar’s radio beam zaps the asteroid, vaporising it. But the vaporised particles are electrically charged and they slightly alter the process that creates the pulsar’s beam.”

Long term monitoring has shown that the pulse shape of PSR J0738-4042 changes multiple times between 1988 and 2012, with one event that implies a rock of about a billion tonnes interacting with the pulsar. The abstract of the paper on this work summarizes the finding:

The torque, inferred via the derivative of the rotational period, changes abruptly from 2005 September. This change is accompanied by an emergent radio component that drifts with respect to the rest of the pulse. No known intrinsic pulsar processes can explain these timing and radio emission signatures.

And so we get an interesting mechanism for the formation of large objects around a pulsar. Dust and debris formed in the original stellar explosion that gave birth to the pulsar itself could fall back toward the spinning remnant to create a debris disk that, in turn, gives birth to larger objects. Recall that two planet-sized objects were detected in 1992 around the pulsar PSR 1257+12, though not necessarily formed by the same mechanisms (a third planet was detected five years later). We have much to learn about the various ways pulsar planets may form.

The environment that spawns these results has to give us pause. “If a large rocky object can form here,” Ryan Shannon adds, “planets could form around any star.” True enough, though we should add that pulsar planets still seem to be the rarest of the breed. In any case, this detection is remarkable. We’re reaching out to a distance far larger than that between the Earth and the galactic core to track conditions in this system. With so many ongoing projects, we can all too easily become blasé about just what is happening here, but let’s keep that sense of wonder at full throttle. Twenty-five years ago we had no confirmed exoplanets, and now we’re investigating asteroids in a system 37,000 light years from Earth? A golden era of discovery indeed…

The paper is Brook et al., “Evidence of an Asteroid Encountering a Pulsar,” Astrophysical Journal Letters Vol. 780, No. 2 (2014), L31 (abstract).

Following up on yesterday’s post on Gaia, it seems a good time to discuss PLATO, the European Space Agency’s planet hunting mission, which has just been selected for launch by ESA’s Science Policy Committee. The agency’s Cosmic Vision program has already selected the Euclid mission to study dark energy (launch in 2020) and Solar Orbiter, an interesting attempt to study the solar wind from less than fifty million kilometers. Solar Orbiter will surely return data we’ll want to discuss here in terms of magsails, electric sails and other ways to harness a solar wind about which we have much to learn.

Solar Orbiter’s launch is the closest of the three, scheduled for 2017, with PLATO pegged for 2024, the launch to be from the European spaceport in Kourou (French Guiana) aboard a Soyuz booster. Note that date, because it’s expected, as this BBC story notes, that the ground-based European Extremely Large Telescope (E-ELT) will be operational in Chile by 2024, a reminder that it should be powerful enough, with its 39-meter primary mirror, to carry out studies of planetary atmospheres as determined by targets PLATO can provide.

The mission will operate from the L2 Lagrangian point 1.5 million kilometers from the Earth, from which vantage PLATO will turn 34 individual telescopes and cameras on to a field of view that encompasses half the sky. Up to a million stars will be under investigation, looking for the characteristic lightcurve of a planet transiting the star as seen from Earth. PLATO will also have a strong asteroseismology component, allowing even more precision in characterizing its planetary finds, especially since these will be followed up with ground-based radial velocity observations that will benefit from having data on surface pulsations on target stars.

Thus PLATO’s unravelled acronym: PLAnetary Transits and Oscillations of stars. The ultimate goals are exactly those we’d expect: To discover and characterize relatively nearby planetary systems, detecting Earth-sized planets and ‘super-Earths’ in the habitable zone around solar-type stars while measuring solar oscillations in the host stars. Working at optical wavelengths, PLATO should be able to determine planetary mass with a precision of 10 percent, planetary radius with a precision of 2 percent and stellar age up to 10 percent.

The BBC quotes Don Pollacco (University of Warwick, leader of the PLATO Science Consortium), on how the mission differs from those that have gone before:

“PLATO will be our first attempt to find nearby habitable planets around Sun-like stars that we can actually examine in sufficient detail to look for life. Nearly all the small transiting planets discovered so far have been beyond our technology to characterise. PLATO will be a game-changer, allowing many Earth-like planets to be detected and confirmed and their atmospheres examined for signs of life.”

The contrast is, of course, with Kepler, whose stars — in a magnitude range between 7 and 17 — were faint enough that follow-up studies for the majority of candidates are difficult if not impossible. The interesting Lost in Transits blog, written by a PhD student at the University of Warwick and thus likely connected with Don Pollacco at the same school, points out that Kepler’s wide field and large camera array will be turned to brighter stars (magnitude 4-16) with equipment sufficient to follow up even small Earth-class planets around these stars. The blog also points to the effectiveness of asteroseismology in this work:

This ability to survey bright stars also allows astronomers to perform extremely sensitive measurements of the stars themselves. By using variations in starlight caused by ripples on the star’s surface, astronomers can accurately pin down not only the size of the star but also the age of the star system. This means, not only can Plato find exoplanets around bright stars, but it can also determine the size and age of many of these planets to a precision only previously dreamed of.

Ahead for PLATO are further refinements and finalization of the design, along with selection of an industrial prime contractor and, within the next two years, the final adoption of the mission, but the support given by the ESA Science Policy Committee, unanimous, strengthens our hopes that PLATO will fly. The fudge factor in that statement simply reflects the disappointments we’ve seen on both sides of the Atlantic with missions like SIM (Space Interferometry Mission) and the Darwin astronomical interferometer all showing promise only to face cancellation. Budgetary realities are always to be reckoned with, but PLATO’s selection is welcome news indeed.

For more detail on PLATO, see Rauer et al., “The PLATO 2.0 Mission” (abstract).

The young star known as L1527 offers a spectacular view at infrared wavelengths, a result of the configuration of gas and dust around it. Have a look at the image below, taken by the Spitzer Space Telescope, where light from the star escapes through the opening provided by a bipolar gas flow, illuminating the gas to highlight a nebula in the shape of a butterfly. Earlier radio studies of this star have shown that L1527 is surrounded by a gas disk that, from our perspective, is seen edge-on. Now new radio observations are helping us characterize the gas itself.

Image: An infrared image of the protostar L1527 taken by the Spitzer Space Telescope. Credit: J. Tobin/NASA/JPL-Caltech.

It’s an interesting investigation because the chemical changes inside a disk as it forms are little understood. Intense observational effort has gone into studying the physical structure of protoplanetary disks, but separating the young disk and the infalling envelope of gas and dust that gives rise to it is difficult. This is where the powerful sensitivity of ALMA (Atacama Large Millimeter/submillimeter Array, a radio interferometer array in Chile’s Atacama desert) may come to our aid. The new work, described in this news release from the University of Tokyo, is built around ALMA’s high spatial resolution observations to study the chemistry of disk formation.

Invisible in infrared light, the gas around L1527 can be readily examined at ALMA’s wavelengths, its molecules characterized in terms of their density, temperature and chemical composition. Researchers under Nami Sakai (University of Tokyo) have been studying the radio emission from cyclic-C3H2 (three carbon atoms in a loop-like structure with two hydrogen atoms attached) and sulfur monoxide (SO) molecules — these are emissions weak enough to be undetectable by other instruments, but within range of the ALMA array.

Because a new stellar system is formed through the gravitational collapse of interstellar materials, you would think that the interstellar gas and dust would simply be incorporated into the new disk structure as is. But Sakai and team have found something different: There is a chemical change associated with the formation of the disk. L1527’s disk has a radius of about 500 AU. Inside 100 AU, the emission from cyclic-C3H2 is weak enough to suggest chemical differentiation between the inner and outer disk. Meanwhile, SO seems to be concentrated in a ring-like structure with a radius of about 100 AU. The image below shows the result.

Image: L1527 observed by Spitzer (Left) and the distributions of cyclic-C3H2 (center) and SO (right) observed by ALMA. ALMA reveals the gas distribution just close to the protostar. Emission from cyclic-C3H2 is weak toward the protostar but strong at the northern and southern parts. Meanwhile, SO has its emission peak near the protostar. Credit: J. Tobin/NASA/JPL-Caltech, N. Sakai/The University of Tokyo.

So we have a striking change in chemical composition about 100 AU out. Sakai’s simulations show that infalling gas from the outer parts of the disk is piling up at what he calls the ‘centrifugal barrier.’ Local heating in this region as the infalling gas accumulates causes distinct chemical changes. Here I’m going to quote Masaaki Hiramatsu (NAOJ Chile Observatory), who put together a backgrounder on Sakai’s work, describing the lively action around this ‘barrier’:

The infalling gas collides with the barrier and is warmed up. SO molecules frozen on the surface of cold dust grains are liberated into the gas phase. The temperature decreases inside the barrier and the SO molecules are frozen again. This is the formation process of the SO ring at 100 AU. Rotating motion dominates inside the centrifugal barrier. Hence, the barrier is the edge of the disk formation region in which eventually a planetary system will be formed.

I’m not aware of earlier work on the chemical differences between protoplanetary disks and the interstellar clouds out of which they form, or of any earlier evidence for major changes in the chemistry of the disk as it emerges. It’s worth asking whether similar situations will be traced in other protoplanetary disks, something that only future observation will tell us, with inevitable implications for how our own Solar System formed if this turns out to be a common process. The paper suggests that micro-analyses of meteorites, spectroscopy of comets, and sample return from the asteroids will help us extend this study to our own system’s formation.

The paper is Sakai et al., “Change in the chemical composition of infalling gas forming a disk around a protostar,” published online in Nature 12 February 2014 (abstract).