A recently commissioned planet-finding instrument has been used to study a young solar system around the star AU Microscopii, leading to the discovery of rapidly moving features in the dust disk around the star. See Letter p.230
In the southern sky hangs the constellation Microscopium, 'the microscope', one of several minor constellations named after scientific instruments in the eighteenth century. Using some much more recent instrumentation, Boccaletti et al.1 (page 230 of this issue) have now observed with fresh clarity a young solar system nestled within that constellation, homing in on a dusty ring of debris around the star AU Microscopii (AU Mic). These observations from the ground match or exceed the resolution with which this debris disk was previously seen by the Hubble Space Telescope. When the authors compared the new images with the older Hubble data, they discovered several localized areas of enhanced brightness, perhaps clouds or clumps of dust, moving outwards from the star surprisingly fast on trajectories suggesting that the clouds are likely to escape into interstellar space.
Twenty years ago this month, astronomers announced the detection of a planet orbiting a Sun-like star2, launching a revolution that has led to detections of thousands of exoplanets with a dizzying diversity of properties. But the vast majority of such discoveries have been made indirectly, primarily by measuring minute variations in the light of the host star from which the presence of a planet can be inferred.
Directly imaging anything orbiting a nearby star is a daunting observational challenge; even large planets such as Jupiter are hundreds of thousands or millions of times fainter than the associated star, and are easily lost in the glare of scattered starlight. But that challenge is worth pursuing, because if one can see something directly, one can measure its spectrum, enabling detailed physical characterization. The same holds true for rings of dusty debris, which can be produced by the evaporation of comets or collisional destruction of asteroids. All-sky infrared surveys have taught us that such debris disks are common, but only a small fraction has been seen directly.
AU Mic hosts one such disk, first imaged more than a decade ago3. As stars go, it is small, nearby and young: half the mass of the Sun, 10 parsecs (32 light years) away and about 25 million years old. Observations4 have established that it is surrounded by a belt of planetesimals at a radius of about 40 astronomical units (1 AU is Earth's distance from the Sun), similar to the Kuiper belt of our Solar System. Occasional collisions of the planetesimals in the AU Mic disk liberate dust particles that are blown outwards by the stellar wind. The total mass of dust is about that of the Moon, ground fine and spread widely. And just as our Solar System's dust is non-uniformly distributed, the disk around AU Mic has clumps and asymmetries, which some have interpreted5,6,7,8 as signs of an unseen planet stirring up the smaller bodies.
This intriguing evidence led to AU Mic becoming one of the first targets for SPHERE (Spectro-Polarimetric High-contrast Exoplanet Research; Fig. 1), a planet-finding instrument, soon after it started operating in 2014. SPHERE is a specialized 'high-contrast' imaging system, designed to allow the Very Large Telescope in Chile to point towards a bright star while blocking almost all of that star's light, and thus allowing a view of the star's close environs. Developed over the past decade, it uses a combination of sophisticated optical, spectroscopic and analytical techniques, and was intended to enable the detection of light up to one million times fainter than the host star.
So, did SPHERE meet expectations? Boccaletti et al. present a convincing affirmative with their observations of AU Mic. Their data are beautiful and impressive, matching the Hubble images in terms of clarity of fine structure, and exceeding Hubble in terms of the ability to peer close to the central star. What was definitely not expected, however, was what they saw around AU Mic: the clumpy dust clouds that extend across much of the disk to the southeast of the star have moved outwards over the past few years, by between 3 and 8 AU each (see Fig. 1 of the paper1).
This unexpected result led the team to revisit some of the earlier Hubble data with more careful analysis. By separately analysing two data sets that had previously been combined, they found that the outward motion could also be seen in the Hubble data between 2010 and 2011. Although it is well understood that individual dust grains would be gradually blown out of the system, this would be expected to be a fairly continuous, steady-state process — not something that would produce a chain of discrete clouds of dust, stretching across an apparent distance equal to the diameter of our Solar System and moving outward as a coherent pattern. Furthermore, the observed speed increases roughly linearly with apparent distance from the star, from 4 to 10 kilometres per second; the motion of the outermost clumps is fast enough for bound orbits around the star to be ruled out. These clouds seem to be blowing out into interstellar space.
The authors readily admit that they do not have a good explanation for what is going on. They present several potential hypotheses, from resonant waves of dust induced by an unseen planet, to debris from massive asteroid collisions, to material ejected from the debris ring as a result of periodic stellar flares. But none of these is fully satisfactory. One highly speculative idea is that stellar flares are interacting with a planetary magnetosphere (the region in which charged particles are affected by a planet's magnetic field) or a circumplanetary ring like that of Saturn, in which case the projected orbital motion of a planet around the star could explain the variation in cloud velocities. Further investigations of AU Mic's debris disk are surely a high priority for SPHERE, even as the instrument moves into full use studying a large sample of targets.
SPHERE is not the only game in town for high-contrast imaging. Over the past few years, instruments of this type have become available at many large telescopes, each with their own strengths and specializations. In fact, the AU Mic system has also recently been observed by the Gemini Planet Imager at the Gemini South Telescope in Chile9, although this instrument's field of view is too small to see the high-speed outer clumps detected by SPHERE. Additional competition comes from several instruments that detect slightly longer infrared wavelengths (SPHERE operates in the visible and near-infrared regions of the spectrum), at which planets can be brighter and the demands for adaptive-optics systems are less stringent.
But even with the latest instrumentation and large surveys, the challenge of imaging planets is so great that only a modest number are likely to be seen over the next few years. Cases such as that of AU Mic, in which disks can be imaged in great detail but any planets present are unseen, are likely to remain more common than directly imaged planets. Lucky for astronomers, then, that circumstellar disks still turn out to have surprises such as the fast-moving dust features of AU Mic.Footnote 1
Boccaletti, A. et al. Nature 526, 230–232 (2015).
Mayor, M. & Queloz, D. Nature 378, 355–359 (1995).
Kalas, P., Liu, M. C. & Matthews, B. C. Science 303, 1990–1992 (2004).
Strubbe, L. E. & Chiang, E. I. Astrophys. J. 648, 652–665 (2006).
Liu, M. C. Science 305, 1442–1444 (2004).
Metchev, S. A., Eisner, J. A., Hillenbrand, L. A. & Wolf, S. Astrophys. J. 622, 451–462 (2005).
Krist, J. E. et al. Astron. J. 129, 1008–1017 (2005).
Fitzgerald, M. P., Kalas, P. G., Duchêne, G., Pinte, C. & Graham, J. R. Astrophys. J. 670, 536–556 (2007).
Wang, J. J. et al. Astrophys. J. Lett. (in the press); preprint at http://arxiv.org/abs/1508.04765 (2015).