OBSERVATIONAL EVIDENCE FOR AN IMPACT ON THE MAIN-BELT ASTEROID (596) SCHEILA

Asteroid–asteroid impacts must occur based on numerous spacecraft images of craters on asteroids and the creation of asteroid families via catastrophic fragmentation events (e.g., Themis, Koronis and Eos families) and cratering events (e.g., Juno and Vesta). However, there are few opportunities to observe impacts in progress. We provide unique observational evidence for a natural impact that occurred in late 2010.

The outer main-belt asteroid, (596) Scheila, was discovered in 1906 and was found to exhibit comet-like activity on 2010 December 11 (Larson 2010; Larson et al. 2010). If this activity is cometary (sublimation-driven) in nature, Scheila would be the largest main-belt comet discovered so far with a diameter of 113–120 km (Tedesco & Desert 2002; Usui et al. 2011). It is classified as a T-type asteroid; asteroids of this type have dark and moderately red spectra (DeMeo et al. 2009). No gases were detected by Swift UV–optical observations, suggesting that the outburst was not triggered by the sublimation of ice (Bodewits et al. 2011). Given these findings and other physical evidence (Jewitt et al. 2011; Hsieh et al. 2011; Yang & Hsieh 2011), it is likely that an asteroid collided with (596) Scheila recently and that its activity is likely not produced by sublimation.

However, little is known thus far about the fundamental physical parameters such as grain size and the dust emission date. We made imaging observations of (596) Scheila using ground-based telescopes for three months soon after the discovery of comet-like activity. The obtained data were analyzed to detect faint structures. As a result, we first succeeded in the detection of the linear tail. We examined the position angle of the linear tail based on the dynamics of the dust particles and derived the exact date of the dust emission. We also determined the maximum size of the dust particles.

The observation details are summarized in Table 1. (596) Scheila was observed with three telescopes: the Ishigakijima Astronomical Observatory (IAO) 1.0 m telescope, using a three-channel (the g', Rc, and Ic bands) simultaneous imaging system (Multicolor Imaging Telescopes for Survey and Monstrous Explosions); the University of Hawaii 2.2 m telescope (UH 2.2 m), using a Tektronix 2048 × 2048 pixel CCD; and the Subaru 8.2 m telescope, using the prime-focus camera Suprime-Cam. The fields of view and pixel resolutions of these instruments are 66 × 66 and 039 (IAO), 75 × 75 and 022 (UH 2.2 m), and 34' × 27' and 020 (Subaru), respectively. The data were reduced in the standard manner using the IRAF package for the IAO and UH 2.2 m data, and SDFRED2 for the Subaru data (Ouchi et al. 2004). The original images were bias-subtracted and divided by a flat field. Images from each data set were combined to improve the signal-to-noise ratio and to permit the subtraction of background stars and galaxies. In addition to the above observations, the light curve for (596) Scheila was taken with the 0.4 m reflector telescope at Hamanowa Astronomical Observatory (MPC code D91 Adati) for nine nights between 2010 December 12 and 2011 January 28. The cooled CCD camera SBIG ST-8, with an R-band filter, was employed to determine its rotational period after showing comet-like activity.

3.1. General Appearance

Figure 1 shows time-series images of (596) Scheila observed from 2010 December 12 to 2011 March 2 by our observations. In the December images, three components appear: the northern tail, southern tail, and westward tail. The northern tail was the most prominent structure, and both the northern and southern tails were smoothly curved in the western direction. The westward tail slightly deviated from the anti-solar direction. These components faded rapidly and finally became undetectable after early 2011 February. Instead, a faint linear tail appeared in 2011 February and March. Although triple dust structures were found before 2011 January (Jewitt et al. 2011), our observation provides the first evidence for the linear tail.

Zoom In Zoom Out Reset image size

Figure 2 shows the multi-band images taken at IAO. It is clear that no significant difference appears in the morphology observed in three optical channels (the g', Rc, and Ic bands) at IAO, which implies that the diffuse cloud consisted of dust particles large enough to scatter optical light (i.e., larger than submicron size).

Zoom In Zoom Out Reset image size

3.2. Estimates of the Dust Emission Date and the Grain Size

We examine the identity of the linear tail on the basis of the dust dynamics. The trajectory of a dust particle can be calculated from the ratio of the solar radiation pressure to the solar gravitational attraction, β, and the ejection terminal velocity, v_tml. Assuming spherical particles with a radius a (m) and mass density ρ (kg m⁻³), one can compute the β values. By definition,

$$\begin{equation} \beta = \frac{K Q_{{\rm pr}}}{\rho a}, \end{equation} \tag{ 1 }$$

where K = 5.7 × 10⁻⁴ kg m⁻² and Q_pr is the radiation pressure coefficient averaged over the solar spectrum (Burns et al. 1979). We assumed Q_pr = 1.

Synchrones are the loci of dust particles emitted at the same time but with different β, while syndynes are the loci of dust particles with the same β value but ejected at different times. A synchrone–syndynes analysis is useful only when the terminal velocity of the particles is zero. In general, synchrones can be approximated by radial lines. Because (596) Scheila's linear tail is narrow, suggesting that the particles were ejected with nearly zero terminal velocity, we derived the synchrone position angles and compared them to the observed position angles. The results are shown in Figure 3. It is found that the position of the linear tail after 2011 February coincided with the locus of dust particles ejected with zero velocity on December 3.5 ± 1.0. It is thus likely that dust particles were ejected impulsively rather than continuously. Because the faint linear tail was connected to (596) Scheila on 2011 March 2, large dust particles (≳100 μm) must have been ejected (see Figure 1, bottom panel).

Zoom In Zoom Out Reset image size

Observation with the 0.68 m Catalina Schmidt telescope on 2010 December 3.4 showed a slightly diffuse appearance and integrated magnitude V = 13.2, slightly brighter than the expected asteroid magnitude (Larson 2010; Larson et al. 2010). Our result is consistent with their observation.

3.3. Light Curve

The light curve of (596) Scheila is shown in Figure 4. We derived a rotational period of 15.8480 hr to an accuracy of 1 s. The period we obtained is in complete agreement with that determined in 2006 January, before it showed comet-like activity (Warner 2006). This result suggests that the comet-like activity was not driven by a rotational spin-up (Holsapple & Housen 2007). As we show in the bottom panels in Figure 4, our observational geometry is similar to that of the period before the comet showed activity. Nevertheless, the shape of the light curve changed around the rotational phase 0.9–0.3. It is therefore likely that a fresh surface was excavated by the impact.

Zoom In Zoom Out Reset image size

Based on our analysis, we consider the cause of the comet-like activity on (596) Scheila, although similar discussion was done by different authors (Jewitt et al. 2011; Bodewits et al. 2011). Ice outgassing may cause the activity. However, no spectral line emission that proved the sublimation of ices was reported (Bodewits et al. 2011). In addition, the morphology of the dust cloud is unlike those of typical comets. Rotational breakup could be another explanation of the activity. However, (596) Scheila is not a rapid rotator (15.848 hr); furthermore, the period did not change after the dust emission, as we have shown in the previous section. Accordingly, we can rule out the rotational breakup hypothesis. Electrostatics might launch particles into interplanetary space. As noted above, dust particles up to 100 μm in radius were confirmed by our measurements, which is too large to be launched from a 113 to 120 km body by electrostatics (Lee 1996). It is, therefore, natural to think that (596) Scheila's dust cloud was impact triggered. Assuming this is the case, we examine the impactor mass and the crater size below.

The total magnitude of the cloud was 13.2 ± 0.1 on December 12, which was brighter than the predicted magnitude of 14.3. We assumed that (1) the optical properties of the dust grains were the same as those of (596) Scheila, (2) the ejecta consisted of dust particles ranging from a_min = 0.1–1 μm to a_max = 100 μm with a differential size distribution of q = −3.5 (the size distribution: dn/da∝a^−q, typical for impact debris), and (3) the mass densities of the dust particles and (596) Scheila were 1670 kg m⁻³, which is equivalent to the mass density of the Tagish Lake meteorite (Hiroi & Hasegawa 2003; Zolensky et al. 2002). Under these assumptions, we estimated the ejecta mass to be (1.5–4.9) × 10⁸ kg. We emphasize that a large fraction of the mass fell back to the asteroid's surface because of gravity. Only the small fraction of the ejecta accelerated to a velocity greater than the escape velocity (v_es = 55 m s⁻¹) appeared as a dust cloud from Earth's orbit. To estimate the impactor's mass, we considered two different cases: a cohesive surface with the strength of weak carbonaceous chondrite meteorites and a surface consisting of regolith particles with the size distribution assumed above (Housen & Holsapple 2011). In the first case, the mass fraction of the ejecta with velocities between 55 and 340 m s⁻¹ (the maximum speed is derived in M. Ishiguro et al. 2011) is only 0.7%–1.0% of the total mass excavated from a parabolic crater with a depth–diameter ratio of 1:3. We assumed a tensile strength of 0.3 MPa (Tsuchiyama et al. 2010) and strength scaling of the ejecta. The total mass excavated by the impact is estimated to be (0.2–0.7) × 10¹¹ kg, which is equivalent to the mass of a crater 500–800 m in diameter. We estimated an impactor diameter of 30–50 m, using the scaling parameters for dry soil (Holsapple 1993), the same mass density for the impactor and (596) Scheila, and an impact velocity of 5 km s⁻¹. In the second case, we assumed gravity scaling for the ejecta; as a result, we estimated that approximately 0.6%–0.8% of the total excavated mass had ejection velocities between 55 and 340 m s⁻¹. In this scenario, the blown-out ejecta left a crater 600–800 m in diameter with a depth–diameter ratio of 1:5. We estimated an impactor diameter of 20–30 m in the second case using the scaling parameters of sand with no cohesion as an extreme; however, the fine regolith particles assumed here would have some cohesion that is weaker than that of dry soil. Our estimate for the impactor diameter (20–50 m) is consistent with the impactor size derived by Jewitt et al. (2011) and Bodewits et al. (2011). It is suggested that impact collisions between 100 m class asteroids can occur annually (Jewitt et al. 2010; Bottke et al. 2005). Therefore, it is no surprise that an impact event with a 20–50 m object occurred recently. We assumed a normal impact; however, the real impact probably occurred with an oblique incidence, as discussed below, which requires a larger impactor.

The shape of the light curve changed after the impact event. Since the apparent area of the impact crater (0.2–0.5 km²) accounts for only 0.02%–0.05% of total cross sectional area of (596) Scheila (1 × 10⁴ km²), it is unlikely that the light curve change was caused by the excavation of the fresh material on the impact site. However, we estimated a crater volume of (0.8–3.7) × 10⁸ m², which can cover the entire surface of (596) Scheila if the thickness of the ejecta layer is 2–9 mm. Secondary craters, which were produced by the large boulders with an impact velocity of v_es≲ 55 m s⁻¹, may also contribute to a change in the light curve. Further research on the ejecta is needed to elucidate these points.

From our observations of (596) Scheila using Subaru, IAO, and UH 2.2 m, we find that

• 1.  
  the dust particles ranging from 0.1–1 μm to 100 μm were ejected into the interplanetary space impulsively on December 3.5 ±1.0 day;
• 2.  
  the ejecta mass was estimated to be (1.5–4.9) × 10⁸ kg, suggesting that equivalent mass of a 500–800 m crater was excavated by the event;
• 3.  
  the shape of the light curve changed after the impact event probably because fresh material was excavated around the impact site.

Research at Seoul National University was supported by the National Research Foundation of Korea and a Basic Research Grant. We are grateful to all the staff of the Subaru Telescope and Ishigakijima Astronomical Observatory, which are operated by the National Astronomical Observatory of Japan (NAOJ). Use of the UH 2.2 m telescope for the observations was also supported by NAOJ. We thank Garrett T. Elliott and Marco Micheli at the University of Hawaii for observation support. The instrumentation at Ishigakijima Astronomical Observatory was supported by a Grant-in-Aid for Scientific Research on Priority Areas (19047003). S.H. is supported by the Space Plasma Laboratory, ISAS, JAXA.

Facilities: Subaru - Subaru Telescope, UH:2.2m - University of Hawaii 2.2 meter Telescope