Asteroids and the James Webb Space Telescope

While asteroids were once derided by frustrated astrophysicists as "vermin of the skies," we have come to realize they are invaluable tracers of history long erased from larger bodies. The non-gravitational processes they experience can involve forces so small that they cannot be measured on more massive bodies. Their present-day orbits contain echoes of giant-planet migration, offering constraints that help us understand not only our solar system but others as well. A handful of small and large asteroids have received spacecraft visitors on flyby or rendezvous missions, with one sample return completed and two more in the works. However, spacecraft will never visit the vast majority of asteroids, estimated at roughly 1 million larger than 1 km and millions more smaller than that size. These objects must be observed remotely if the scientific questions they answer are to be addressed. As the flagship space telescope of the coming decade, the James Webb Telescope (JWST) is an obvious place for planetary astronomers to turn.

The small bodies are unusual targets for JWST compared to most solar system objects. Their sizes are often uncertain and typical light curve effects can change brightnesses by a few tenths of magnitudes in an hour or so. However, their centrality to important questions in planetary science was recognized in the Planetary Science Decadal Survey, and JWST observations have the potential to revolutionize their study.

The small bodies have been subject to several classification schemes relating to their physical appearance, orbit, formation location, composition, and other factors. There is evidence that groups of small bodies now in overlapping parts of the solar system may have formed at quite different solar distances (Walsh et al. 2012) and that small body populations that formed together may be scattered over the length of the solar system (Gomes et al. 2005; Morbidelli et al. 2005). In this white paper, we focus on the group of small bodies with semimajor axes from roughly 2 to 5.2 AU, taking in the main asteroid belt and the Hilda and Trojan groups, with a few nuances: regardless of semimajor axis we exclude the near-Earth objects (defined as having perihelia <1.3 AU), which have several unique challenges and are the subject of the white paper of Thomas et al. (2016), and we also include the irregular satellites of Jupiter in this paper under the heading of the Trojan asteroids. Finally, although formally classified by the International Astronomical Union as a "dwarf planet" rather than an asteroid, we include Ceres.

Asteroid science used the Hubble Space Telescope (HST) primarily for improved spatial resolution, as with observations of Vesta (Binzel et al. 1997; Thomas et al. 1997; Zellner et al. 1997), Ceres (Thomas et al. 2005), and Pallas (Schmidt et al. 2009). Deconvolution of Wide Field/Planetary Camera images of even smaller asteroids were used to determine diameters, albedos, and lower limits to axial ratios (Storrs et al. 2005, among others). Observations of main belt comets (MBCs, also called active asteroids) also take advantage of the improved spatial resolution, as well as increased sensitivity of HST, though the MBCs were first detected using smaller, ground-based telescopes. JWST will have a much larger primary mirror than HST (25 m² versus 4.5 m²), providing over five times more light-gathering power. HST's instruments do not currently operate beyond 1.7 μm (Wide Field Camera 3, WFC3) and never operated beyond 2.45 μm (Near Infrared Camera and Multi-Object Spectrometer, NICMOS), while JWST's extended wavelength range allows measurements of hydroxyl and organic fundamental absorptions in the near-IR and thermal emission in the mid-IR. Furthermore, JWST operates at L2, much farther from Earth than HST, enhancing temperature stability and providing longer possible periods of continuous observation. The telescope is optimized for infrared optics, including a cold primary mirror and subsequent optics, and all instruments are cryogenic. These features mean that JWST in general should be an excellent platform for observations of asteroids. With this in mind, the benefits of JWST to asteroid studies can be effectively divided into three categories: improved spatial resolution, better signal-to-noise ratio (S/N) on small targets, and ability to reach wavelengths unreachable from ground-based telescopes. Asteroid-centered projects will most likely make use of JWST for one or more of these reasons.

1.1. Improved Spatial Resolution

1.2. Wider Wavelength Coverage

1.3. Improved Sensitivity

Not only does poor transmission block some wavelengths from measurement from Earth, but it also drastically affects the limiting magnitude of objects. In the 3 μm region, diagnostic for water, hydroxyl, and organic material, targets are typically limited to V ∼ 14 at the NASA Infrared Telescope Facility (IRTF), corresponding to ∼50 km in the mid-belt for carbonaceous bodies. Even with the Keck telescope, 3 μm observations on objects fainter than V ∼ 15.5 are a challenge (M. E. Brown 2015, private communication) and data quality is marginal. This is the case even though objects of V ∼ 18 are commonly observed with the IRTF at 2.5 μm, where transmission is much higher. Figure 2 shows that objects 0.5 km in size and 5% albedo at 2.5 AU can be observed with S/N approaching 10 at 3 μm in spectroscopic mode with NIRSpec in 1000 s. This corresponds to objects with absolute magnitude of 20.3 and an apparent magnitude of 23.1. We note that every numbered asteroid (that is to say, every asteroid with a secure orbit) reaches apparent magnitude brighter than this. The equivalent absolute magnitude for Trojan asteroid sensitivity is 15.3, again fainter than any numbered object in that part of the solar system. This improved sensitivity with NIRSpec on JWST makes discovery and secure orbit determination the limiting factor for observations from the main belt through the Trojan clouds, along with most known irregular satellites of Jupiter as practically every object will be observable with S/N > 10 in 1000 s across the entire 0.6–5 μm range with the R = 100 prism. Because asteroidal spectral features are typically broad, solid-state absorptions, higher spectral resolutions are rarely required. Spectrophotometric observations can be made even more quickly and/or at higher S/N.

Zoom In Zoom Out Reset image size

This situation will not change significantly with future surveys. The Large Synoptic Survey Telescope (LSST) is expected to detect 90% of main belt asteroids to H = 20 and 50% of MBAs to H = 20.7, with 90% of Trojans to H = 17.5 and 50% of Trojans to H = 17.8 detected (LSST 2009). Objects of these sizes are still within a factor of a few of the sensitivity limits mentioned above for 1000 s integrations and are all easily above the sensitivity limits for spectrophotometry with the NIRCam imager in the 1–5 μm region. As a result, almost any arbitrarily chosen known asteroid in the main belt or beyond can be observed by JWST, although due to solar elongation constraints they cannot be observed at any arbitrarily chosen time.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The sensitivity of the Mid-infrared Instrument (MIRI) will allow compositional studies of much smaller objects than currently available. Spitzer studies of asteroids showed some with surprising emissivity patterns interpreted as due to a "fairy castle" structure in the regolith or a high fraction of infrared-transparent materials (Vernazza et al. 2012; Yang et al. 2013). Observations of smaller targets will allow the degeneracy to be resolved. In addition, asteroids typically have emissivity spectra close to blackbodies, with compositional effects playing only a small role. Improved sensitivity will increase the number of targets for which these compositional effects can be seen.

1.4. Improved Contrast

As noted, we anticipate successful asteroid projects will utilize several of the advantages mentioned above. For instance, 10 Hygeia (430 km diameter) reaches 2.77 AU and subtends 034 at perihelion, and 3 Juno (233 km diameter) reaches perihelion inside of 2 AU, subtending 032. Both of these values are very close to the angular size of Pallas when it was observed by Schmidt et al. (2009), shown in Figure 3. NIRSpec will allow image cubes to be obtained of these objects, rather than simple spectrophotometry—this is also shown schematically in Figure 3, where every pixel is associated with a visible–near-IR spectrum. The ability to observe over a continuous period, unlike HST, enables more straightforward coverage over an entire rotation (compared to piecing together observations from different HST orbits).

Zoom In Zoom Out Reset image size

A second example project would be studying MBCs (Figure 4). These objects have cometary tails and comae but are in main belt orbits. Some of these objects are thought to be shedding material after impact events or spin up due to the Yarkovsky–O'Keefe–Radzievskii–Paddack effect, but others appear to have subsurface volatiles driving their activity. Volatiles have never been detected on these bodies, but because almost all of them are 5 km in size or smaller, constraints are quite loose. Observations with NIRCAM and MIRI would easily establish whether ice is exposed on their surfaces as well as provide firmer sizes and albedos, and observations of MBC tails across the NIRCAM and MIRI wavelengths would enable more precise measurements of particle sizes. The rate of MBC discoveries suggests there will be several available during the lifetime of JWST. Searches for gas in MBC comae (as opposed to dust) have placed water outgassing rates of ∼10²⁶ s⁻¹ for several objects (Jewitt 2012). Kelley & Milam (2016) calculated S/N for JWST observations of comets and noted that water outgassing at this rate is detectable to ∼4.5 AU. We can therefore expect to improve upon existing constraints throughout the main asteroid belt.

A third potential project is observations of asteroid collisional family members. For decades, we have expected to see variation in the spectra of family members corresponding to, for instance, crust/mantle/core fragments of differentiated bodies. However, observations in the 0.8–2.5 μm region have found family members to be remarkably similar to one another even when parent bodies are suspected to be differentiated (Bus & Binzel 2002; Bus 2014). This has been interpreted as showing that the collisions creating families disrupt bodies into very small pieces, which then reaccrete. Using JWST, we can reach smaller targets in collisional families and potentially determine the characteristic size of collisional fragments from the size at which spectral diversity appears. Current family studies are typically conducted on 3–4 m class telescopes and are limited to objects of apparent 19th magnitude or brighter, corresponding to absolute magnitudes of H ∼ 15–16.5 for a range of semimajor axes from 2.3 to 3 AU, or diameters of roughly 1–6 km for that range of distances and a range of albedos from 5% to 30%. As noted above and in Figure 2, high-quality NIRSPEC observations can be made for much smaller objects, and NIRCAM observations can probe targets perhaps 10% the size of the current limit.

The extreme sensitivity of MIRI has the consequence that it saturates for surprisingly bright targets. Ceres saturates every filter and grating other than the shortest-wavelength half of the shortest-wavelength grating. Because saturation is related to surface brightness, smaller objects with Ceres' temperature will still saturate MIRI as long as they are larger than 011 in angular diameter, the size of a MIRI pixel. Angular sizes below 011 may or may not saturate MIRI depending on a combination of the fraction of the pixel filled and the temperature. Table 1 shows estimates of the temperatures that are expected at various important distances for asteroids: the inner edge of the asteroid belt at 2.0 AU, the middle of the belt near where Ceres and Pallas orbit at 2.8 AU, the Cybele region at 3.4 AU just beyond what is typically considered the outer edge of the main belt, and 5.2 AU for Trojan asteroids and the irregular satellites of Jupiter.

These temperatures assume a spherical body and are calculated for a position 45° from the subsolar point using a geometric albedo of 0.2, G = 0.15 (resulting in a bond albedo A of 0.08), emissivity () of 0.9, and a beaming parameter (η) of 1). The choice of 45° from the subsolar point represents a balance between the contribution of the hottest part of the surface and the contribution from a larger surface area. The subsolar point will have a temperature 8% higher than the locations 45° from the subsolar point. The subsolar temperature (T_ss) is calculated using the well-known equation

$$\begin{eqnarray*}&&{T}_{\mathrm{ss}}=((1-A)S/(\eta \;\epsilon \;\sigma ))1/4,\end{eqnarray*}$$

where S is the solar constant at the distance of interest and σ is the Stefan–Boltzmann constant (Lebofsky et al. 1986, for example).

Figure 5 overlays the surface brightness of targets of these various temperatures for objects large enough to fill a MIRI pixel. The 3rd–6th columns of Table 1 show the sizes at which MIRI pixels are just filled (at various asteroid distances, assuming JWST is at 1.0 AU and observing at opposition), and the sizes at which various observing modes become possible as a function of distance.

Zoom In Zoom Out Reset image size

The largest Trojan asteroid (624 Hektor) has a diameter of ∼225 km. All Trojans are capable of being observed by MIRI in the 10 μm gratings, though 20 μm spectroscopy cannot be done until smaller sizes are reached. The largest few irregular satellites of Jupiter are also too large for 20 μm spectroscopy or 15 μm imaging, though over 50 of the irregular satellites will not saturate MIRI.

Because of its powerful capabilities, JWST provides unique opportunities for asteroid science. It is likely to be used in solving several important small bodies questions during its lifetime, though naturally it is not the ideal facility for all observations and some observations will be impossible due to saturation. Nevertheless, the small bodies community should take advantage of JWST to the extent possible to obtain data impossible to acquire using other methods.

A.S.R. would like to acknowledge support from NASA Planetary Astronomy Grant NNX14AJ39G and NSF Planetary Astronomy Award 1313144. The authors would like to thank members of the JWST Project at NASA Goddard and staff members at STScI for information and review of this manuscript.