Abstract. A brief summary of pre-JWST and new JWST observations of debris disks is presented, with a focus on Fomalhaut. These disks provide the opportunity for a detailed look at processes in exoplanetary systems that are not observable by other means.
Although the planets get most of the publicity, the Asteroid and Kuiper Belts are two additional major components of the Solar System. The planets finished their episodes of orbital migration and collisions long ago (thankfully), but collisions have not completely damped out in these two belts of small bodies.
The occasional collisions in the Asteroid belt break up the participants into smaller scraps that are put onto related orbits that can be traced back to the collision sites. This process creates “families” of asteroids with similar composition as well as related orbits. The collisions are messy, creating a very broad range of fragment sizes, and as these fragments continue to collide (particularly when the orbits bring them back to the original collision site) they yield clouds of dust. Snapshots of the resulting dusty bands around the sun were dramatically captured by the IRAS all-sky infrared survey (Sykes & Greenberg 1986, Sykes 1990, Nesvorny et al. 2003). There must be occasional collisions and generation of dust in the Kuiper Belt also, although it is too far away and the dust emission too faint to be detectable.
A fascinating aspect of these dust belts is that they reflect not only the processes within the Asteroid and Kuiper Belts, but the planets can imprint signatures on them, as shown in Figure 1. In this figure, the particle densities have been weighted by the inverse square of the distance from the sun to simulate a hypothetical brightness distribution as if the Solar System were being observed from outside. Uranus and Neptune as well as Jupiter and Saturn have profound effects on the distribution of dust. In fact, the role of Jupiter in shaping and carving the Asteroid belt is very well known. This suggests an indirect way to study planetary systems around nearby stars, as Liou and Zook (1999) pointed out:
“If an extraterrestrial intelligence were observing our solar system and had the image in Figure 1, it would know (if its knowledge was similar to or better than ours) that at least one giant planet at about 30 AU exists in our solar system [i.e., Neptune].” That is, the structures in circumstellar dusty debris can be used to infer the existence of the planets that are molding them.
It took many years to fulfil this prediction, and we will return to it later. The dusty circumstellar disks are of great interest in themselves also, since they show the presence of belts of smaller bodies analogous to the Asteroid and Kuiper Belts and can reveal processes occurring within them. They are also more readily detected than planets. This is illustrated by the analogy that it is easier to see the paint when it is spread on the wall (i.e., the dust) than when it is in the can (i.e., collected into a planet).
The Solar System belts are very tenuous and would not be detectable around another star with our current technology (Wyatt 2008). However, this pessimistic view was swept away when it was discovered with the IRAS all-sky infrared survey that Vega is about 15 times brighter than expected in the far infrared (Aumann et al. 1984). Like most scientific discoveries, this one was made by accident. A-type stars like Vega had served astronomy well as reliable calibration sources for the 60 years since they were originally suggested for this role by Henrietta Leavitt (1917), and the IRAS observation was planned for routine calibration purposes. The dust responsible lay in the analog to the Kuiper Belt around this star, far enough away that the star heats it only to about 85 Kelvin (that is 85 degrees above absolute zero). This is about 100 times lower temperature than that of the star, resulting in the peak of the emission being at 100 times longer wavelength than that of the star, i.e., in the far infrared near 60 mm. The result is very dramatic in the far infrared because the stellar output has fallen to only a few percent of its value in the visible, making it possible for the cold dust to be dominant. We could say that IRAS discovered the Kuiper Belt around Vega before Jewitt and Luu (1993) found the true one around the Sun (although a footnote might be needed to explain that at the time Pluto was the “ninth planet” but is now classified as the first discovered member of the Kuiper Belt). Originally, this behavior was termed “the Vega Phenomenon,” but it is now known by the more generic term of planetary debris disks, or just “debris disks.”
Several hundred debris disks were found in the IRAS data and many more have been discovered with ISO, Spitzer, and WISE. Their study has become a significant chapter in the astronomy of the past four decades. The results are far too extensive to cover here, but are described in a number of review articles (Wyatt 2008; Matthews et al. 2014, Hughes et al. 2018, Wyatt 2021, Marino 2022). Cold telescopes in space have played the dominant role because sufficient sensitivities are not possible from the ground where the infrared emission from the warm telescope partially blinds detectors.
Rather than describing all of this work, we will track developments in the study of the debris disk around Fomalhaut, a hot star (8600 Kelvin) in the southern hemisphere and at about 25 light years (7.7 parsecs) distance. Among debris disks, Fomalhaut has been kind of a show-off, placed ideally for ALMA and with all the components of other debris disks prominently represented and relatively easily observed. The disk was first well resolved in the submm (Holland et al. 1998, 2003) followed quickly by resolved images in the mid- and far-infrared with Spitzer (Stapelfeldt et al. 2004) and optical (Kalas et al. 2005). As a foundation for the discussion, we will first assemble the highest quality imaging information, proceeding from long wavelengths to short ones.
Figure 2 shows the debris ring as seen by ALMA at a wavelength of 1.3 mm. The ring is at a radius of 143 Astronomical Units (au – the radius of the orbit of the Earth) from the star and only 13.5 au wide. We show Figure 3 for historical reasons – it is the view at 24 and 70 µm with the Spitzer telescope. For the first time it demonstrates that the structure at the shorter wavelength falls inside the outer ring seen with ALMA. Since this image was obtained, the Herschel telescope has obtained higher resolution images in the far infrared, from 70 through 500 µm (Acke et al. (2012) that show the ring in greater detail, i.e. at 70 µm similar to the Spitzer resolution at 24 µm. Figure 4 is the image obtained with JWST (Gaspar et a. 2023). A comparison with the left image in Figure 3 illustrates the huge leap in resolution and understanding. The debris ring was first imaged in the optical with HST by Kalas et al. (2005); Figure 5 combines those data with all the data obtained subsequently with HST into a more complete optical image. The ALMA. JWST/MIRI, and HST images are the only ones available with arcsec or better resolution. They are combined into a single one in Figure 6, which is the basis for a discussion of the processes sculpting the system.
The confinement of the mm-wave flux to such a narrow ring was surprising, leading to the suggestion that two planets, one inside and one outside it, “shepherd” it (Boley et al. 2012), in a way reminiscent of how shepherds over millennia have tended and kept order in their flocks of sheep. This is a realization of the prediction of Liou and Zook (1999) quoted at the beginning of this article, that ring structures carved by the gravity of unseen planets could reveal the presence of those planets. To understand the ring images in general, the other forces on ring articles need to be invoked. The gravitational forces on the particles go as their masses, which are proportional to their volumes. The radiation forces (when they absorb or scatter photons of light there is a resultant push in the direction the photon was going) are proportional to their projected areas. Consequently, there will be a particle size below which radiation is dominant and above which gravitation has the largest influence on their motion; for Fomalhaut and typical particle materials, this division is at a size of about 2 µm (Arnold et al. 2019). The ring seen with ALMA is described as containing “parent-bodies,” meaning particles large enough to have stable orbits under gravity, and that collide destructively with each other to generate smaller debris. Since particles emit and scatter efficiently at wavelengths close to the particle sizes, this indicates the emission is from roughly mm-sized grains, something like beach sand. As they collide and break each other down into fine dust, these sand grains need to be replenished, which occurs when larger bodies collide destructively, breaking each other up into smaller particles. The colliding bodies are in turn replenished through collisions of still larger bodies. For obvious reasons, this process is called a collisional cascade.
In contrast to the mm-wave image, the HST optical image traces scattered light – scattering is efficient for photons similar in wavelength to the grain size, so the grains responsible are of size less than a micron (i.e., a thousand times smaller than the particles dominating the ALMA image). Since the outer ring lights up in the optical, we can conclude that the collisional cascade operating in this ring extends down to these small sizes. The photon pressure force on grains less than about 2 µm exceeds the gravitational force, so the smallest grains are blown away from the star; a careful look at Figure 6 reveals a thin halo outside of the ALMA-imaged ring illustrating this process.
For more than a decade, we had the beautiful HST and ALMA images, and had to content ourselves with the Spitzer image in Figure 3 as our best view at 24 µm and images at similar resolution from the Herschel Telescope at 70 µm and beyond. We resorted to building theoretical models to imagine what was happening with the dusty debris. These models leaned heavily on the structure of the solar system, with an inner ring analogous to the Asteroid belt and an outer disk like the Kuiper Belt, and nothing between (e.g., Su et al. 2013). We pointed JWST/MIRI at Fomalhaut expecting to confirm this picture but, as Figure 4 shows, got quite a shock. Virtually the whole region between the star and outer ring is filled with emitting dust that shows evidence for an inner ring and more. This complex structure is most likely signaling that there are more planets circulating the star.
Inside all this complexity there is another mystery: extremely hot dust confined to a few tenths of an au, revealed through interferometry (Absil et al. 2009) – a technique combining the images from multiple telescopes widely separated, to achieve the kind of sharpness we would get if we could afford a telescope as large as their separation. We are struggling to understand how this dust can be kept in place but does not evaporate so close to the star (e.g., Stamm et al. 2019).
We now have JWST/MIRI images of a number of additional debris disks and the shock in finding emission filling the region between the star and outer disk is no longer even a surprise. A dramatic example is Vega itself, a star very similar in properties to Fomalhaut and at virtually the same distance.
However, the dust in the Vega system is very smoothly distributed without the dramatic structures revealed around Fomalhaut. This disk – it does not resemble a ring – seems to have an inner edge at a few au and a modestly deep groove at about 60 au, just inside Vega’s outer ring at 85 au that started the whole area of study. Vega seems to share the other debris properties of Fomalhaut, i.e., the very hot dust right around the star, the far infrared ring (although more diffuse and perhaps not shepherded by any planets), and the sub-micron-sized grains streaming out under the influence of photon pressure. Yet the differences in the JWST/MIRI images signal that its planetary system may be quite different.
Debris disks are interesting in their own right, as this discussion shows. As we image more, we may get glimpses of the influence of unseen planets as with Fomalhaut. Planets as massive as Jupiter are found around only about 3% of stars (Rowan et al. 2016). The most common “giant” planets have masses less than 10% the mass of Jupiter, more similar to Neptune and Uranus than to Jupiter and Saturn (Ananyeva et al. 2023). The great majority of exoplanets have been discovered through their small effects on the radial velocity of their stars as the orbit and gravity pulls the stars in synchronism, or by the minute dimming of their stars as they transit in front of them. Even Jupiter could be detected directly by JWST only at distances within about 25 light years (e.g., Ygouf et al. 2024). The current examples of directly imaged planets are of objects more massive than Jupiter; specifically, no Jupiter-mass planets have been spotted among the nearby stars. Within 25 light years, there are only eleven stars known to have exoplanets within the habitable zone (Wikipedia 2024), so detecting directly planets in a configuration in any way resembling the Solar System seems to be a remote possibility.
It was therefore exciting when inside the ALMA-imaged ring around Fomalhaut, a faint dot was found that seemed to be a planet orbiting Fomalhaut (Kalas et al. 2008). However, subsequent observations seemed to show it growing a bit in size until finally by 2014 it had blurred out and was undetectable, as can be seen from the bottom set of small cutouts in Figure 5 (Gaspar and Rieke 2020). The most plausible explanation is that it is the result of a devastating collision between two asteroid-sized bodies, with the resulting creation of a copious cloud of dust that masqueraded for a while as a planet (Kenyon et al. 2014, Lawler et al. 2015, Gaspar and Rieke 2020). Perhaps as exciting to the small band of debris disk aficionados as a planet, this object (if this hypothesis is correct) may well represent on a small scale the type of process that creates and sustains debris disks.
In our own asteroid belt, strong episodes of dust generation are associated with asteroid collisions (Sykes & Greenberg 1986, Sykes 1990, Nesvorny 2003). Similarly, in very young debris disks (generally around stars less than 100 Myr old) there are wide excursions in the infrared emission above the stellar photosphere. They indicate that their asteroids are colliding at a high rate and that the dust these collisions create is getting cleared rapidly (e.g., Melis et al.2012, Meng et al. 2012, Su et al. 2019, Su et al. 2022). Figure 7 is an example (from Su et al. 2019). The rate of collisions around the young stars is indicative of the chaotic events as planets and large asteroids are assembled. The infrared emission can help to reveal details of this process. For example, the infrared spectra usually are dominated by either crystalline silicate material (e.g., Olofsson et al. 2012, Su et al. 2019), or by silica (e.g., Rhee et al. 2008, Lisse et al. 2009), a simpler silicon-oxide compound. Both materials are direct indicators of the extreme temperatures reached as kinetic energy is released in a violent collision, and silica in particular is produced in extremely violent hyper-velocity collisions (Johnson et al. 2012). New JWST spectra indicate some variations on these two dominant dust compositions, so once these data have been fully understood we will have some understanding of compositional variations among asteroids orbiting other stars.
The rapidity of the variations contrasts with the behavior of traditional debris disks around older stars, which tend to fade at ages of about 500 Myr (Gaspar et al. 2013, Sierchio et al. 2014). In addition, there can be periodic modulations on top of the general light curve – you can almost see them if you look closely at the 4.5 µm curve for ID8. Explaining the full range of variations has been very challenging. The rapid drops in output require that the emitting dust be very small, so that radiation pressure blowout can expel it quickly. This rules out the conventional concept of generation by collisional cascades, since by definition they operate from a population of larger bodies. Instead, we may be seeing violent collisions from which tiny grains are condensing out of vaporized minerals. The incidence of these collisions around an individual star is extremely high, indicating that the zone where they are occurring is in a state of turmoil. There are a number of possibilities for this condition, but a likely one is that planets in the system are undergoing a scattering event or that a planet is migrating – situations that should be relatively common at the 30 Myr age of the ID8 system. Perturbations from such behavior will deflect planetesimals onto eccentric and inclined orbits, creating a chaotic situation in a formerly peacefully orbiting asteroid family and with an elevated rate of collisions (see Su et al. 2023 for discussion of this process in another highly variable system). Initially some of the resulting dust clouds may be optically thick, slowing their dissipation, but as they clear the fine grains will be ejected by radiation pressure force, yielding a rapid drop in their infrared emission.
In summary, debris disks offer a detailed look at processes in exoplanetary systems that are not observable by other means. This includes the presence and dynamical state of asteroid and Kuiper belts, and even the presence of unseen planets. These disks are a key aspect in our increasing understanding of planetary systems in general.
References
Absil, O., Mennesson, B., Le Bouquin, J.-B., et al. 2009, ApJ, 704.
Acke, B., Min, M., Dominik, C., et al. 2012, A&A, 540, A125.
Ananyeva, Ivanova, Shashkova et al., 2023, Atmosphere, 14, 353.
Arnold, J.A., Weinberger, A.J., Videen, G., et al. 2019, AJ, 157, 157.
Aumann, H.H., Beichman, C.A., Gillett, F.C. et al. 1984. ApJL, 278, L23.
Boley, A.C., Payne, M.J., Corder, S., et al. 2012, ApJL, 750, L21.
Gaspar, A., Rieke, G.H., & Balog, Z. 2013, ApJ, 768, 25.
Gaspar, A. & Rieke, G. 2020, Proceedings of the National Academy of Science, 117, 9712.
Gaspar, A., Wolff, S.G., Rieke, G.H., et al. 2023, Nature Astronomy, 7, 790.
Holland, W.S., Greaves, J.S., Zuckerman, B., et al. 1998, Nature, 392, 788.
Holland, W.S., Greaves, J.S., Dent, W.R.F., et al. 2003, ApJ, 582, 1141.
Hughes, A.M., Duchene, G., & Matthews, B.C. 2018, ARAA, 56, 541.
Jewitt, D. & Luu, J. 1993, Nature, 362, 730.
Johnson, B.C., Lisse, C.M., Chen, C.H., et al. 2012, ApJ, 761, 45.
Kalas, P., Graham, J.R., & Clampin, M. 2005, Nature, 435, 1067.
Kalas, P., Graham, J.R., Chiang, E., et al. 2008, Science, 322, 1345.
Kenyon, S.J., Currie, T., & Bromley, B.C. 2014, ApJ, 786, 70.
Lawler, S.M., Greenstreet, S., & Gladman, B. 2015, ApJ, 802, L20.
Leavitt, H. 1917, Ann. Harvard College Obs., 71, 3.
Liou, J.-C., & Zook, H.A. 1999, AJ, 118, 580.
Lisse, C.M., Chen, C.H., Wyatt, M.C., et al. 2009, ApJ, 701, 2019.
Marino, S. 2022, arXiv:2202.03053.
Matthews, B.C., Krivov, A.V., Wyatt, M.C., et al. 2014, Protostars and Planets VI, 521.
Melis, C., Zuckerman, B., Rhee, J.H., et al. 2012, Nature, 487, 74.
Meng, H.Y.A., Rieke, G.H., Su, K.Y.L., et al. 2012, ApJL, 751, L17.
Nesvorny, D., Bottke, W.F., Levison, H.F., & Dones, L. 2003, ApJ, 591, 486.
Olofsson, J., Juhász, A., Henning, T., et al. 2012, A&A, 542, A90.
Rhee, J.H., Song, I., & Zuckerman, B. 2008, ApJ, 675, 777.
Rowan, D., Meschiari, S., Laughlin, G. et al. 2016, ApJ, 817,104.
Sierchio, J.M., Rieke, G.H., Su, K.Y.L., & Gaspar, A. 2014, ApJ, 785, 33.
Stamm, J., Czechowski, A., Mann, I., et al. 2019, A&A, 626, A107.
Stapelfeldt, K.R., Holmes, E.K., Chen, C., et al. 2004, ApJS, 154, 458.
Su, K.Y.L., Rieke, G.H., Malhotra, R., et al. 2013, ApJ, 763, 118.
Su, K.Y.L., Jackson, A.P., Gaspar, A., et al. 2019, AJ, 157, 202.
Su, K.Y.L., Kennedy, G.M., Schlawin, E., et al. 2022, ApJ, 927, 135.
Su, K.Y.L., Kennedy, G.M., Rieke, G.H., et al. 2023, ApJ, 959, 43.
Sykes, M.V. 1990, Icarus, 85, 267.
Sykes, M.V., & Greenberg, R. 1986, Icarus, 65, 51.
Wyatt, M.C. 2008, ARAA, 46, 339.
Wyatt, M.C. 2021, in Big questions in exoplanetary science, ed.: Nikku Madhusudhan, Bristol: Institute of Physics, Chap 15.
Ygouf, M., Beichman, C.A., Llop-Sayson, J. et al. 2024, AJ, 167, 26.