Notice: Wiley Online Library will be unavailable on Saturday 27th February from 09:00-14:00 GMT / 04:00-09:00 EST / 17:00-22:00 SGT for essential maintenance. Apologies for the inconvenience.
 Close observations by the Chang'E-2 spacecraft reveal that the surface of asteroid (4179) Toutatis is characterized by abundant impact craters with most of them being degraded by surface resetting. The less degraded large crater with a diameter of ~800 m at the south pole is estimated to be produced by an impactor with a diameter of ~50 m from strength crater scaling relations. From the analysis of large-impact events on highly porous targets, we argue that Toutatis is likely a rubble-pile body and its two lobes are contact binaries. The fact that Toutatis suffered plenty of impacts with seismic shaking resetting the initial surface features but not resulting in catastrophic disruption is probably because of the material's high attenuation of shock wave.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
 The S-type asteroid (4179) Toutatis has moved into a close-approaching orbit to the Earth since 1988 [Whipple and Shelus, 1993]. It is currently in a 3:1 mean-motion resonance with Jupiter, 1:4 resonance with Earth, and has a tumbling rotation [Milani et al., 1989]. It has been observed by using ground-based photometric, spectroscopic, and radar techniques. From these observations, its orbit [Ostro et al., 1995, 1999; Hudson and Ostro, 1995; Scheeres et al., 1998], rotation [Ostro et al., 1999; Spencer et al., 1995; Mueller et al., 2002], shape [Ostro et al., 1995, 1999; Hudson and Ostro, 1995; Hudson et al., 2003], density [Whipple and Shelus, 1993; Hudson and Ostro, 1995], and composition [Howell et al., 1994; Davies et al., 2007; Reddy et al., 2012] have been studied. However, due to the limitation of ground-based observations, its detailed morphology remains unknown. The recent flyby of Chang'E-2 spacecraft should be the only opportunity to observe its detailed morphology at an unprecedentedly close distance for a long time to come.
 Chang'E-2, the second lunar probe of China, was launched on 1 October 2010. After six months of observations in lunar orbit and one year of observations at the Sun-Earth L2 point, it departed on 15 April 2012 to visit Toutatis with the aim to understand its formation and evolutional history. The flyby happened on 13 December 2012 with a relative velocity of 10.73 km/s. Before the encounter, the attitude of the spacecraft was adjusted with one of the solar-wing panels pointing to the target. Toutatis was imaged by the solar-wing panel's monitoring camera in the opposite side of the spacecraft with a closest approach distance of 3.2 km at UTC 08:30:09. This engineering complementary metal oxide semiconductor camera of 1024 × 1024 pixels has a focal length of 9 mm and a field of view of 7.2° × 7.2°, corresponding to a spatial resolution of 0.007°/pixel. The wavelength coverage is 430–780 nm, and the snapshot cadence is 0.2 s. The close approach and imaging campaign lasted about 50 s during which more than 300 optical images with spatial resolutions from 4.5 m to 80 m were obtained. However, due to Chang'E-2 spacecraft's rapid flyby and Toutatis' exceptionally slow spin of 5.4 days [Hudson and Ostro, 1995], only ~45% of Toutatis' surface was imaged, and the rest was in shadow. For convenience, we note the imaged side as the nearside and the other as the farside. Because Toutatis has an irregular two-lobe shape [Hudson and Ostro, 1995; Ostro et al., 1995], like a bowling pin, we use “head” (small lobe), “body” (large lobe), and “neck” (junction between head and body) to denote position on the asteroid for their morphological characteristics in the following part (see Figure 1a).
3 Results and Discussions
 Similar to other asteroids previously explored by spacecraft (e.g., Gaspra, Ida, Mathilde, Eros, Itokawa, Steins, and Lutetia), Toutatis is covered by craters, implying that collision is the principal geologic process shaping its surface. More than 70 craters are identified unambiguously (see Figure S1 of the supporting information) in the observed images with diameters ranging from tens of meters up to 800 m (Figure 1b). Large craters with diameters from 100 m to ~800 m are distributed uniformly across the whole nearside. All these large craters are shallow, often with subdued rims and smooth floors. Their obscure shapes and shallow depths might be attributed to resetting processes such as seismic shaking from subsequent large impacts [Richardson et al., 2004; Thomas and Robinson, 2005]. The ejecta blanketing and downslope motions of materials on a steep slope, as believed to have happened on the surface of Steins [Keller et al., 2010], could cause shallow depression as well. The smooth floors in the large craters indicate that the surface materials are apparently composed of fine-grained particles. This is in agreement with the observation by Ostro et al. , who suggested that about one third of Toutatis' surface was covered by a smooth component.
 Craters with diameter smaller than 50 m have sharpest, least degraded bowl shapes (see Figure 2). Their upheaval rims without rays are consistent with the expectation of impact events on loose material [Melosh, 1989]. The smooth blankets around the rims reveal the presence of a regolith layer on the surface of Toutatis. Smaller craters in unconsolidated regolith are believed to be formed in the gravity regime [Asphaug et al., 1996]. Based on the bowl-shaped craters with diameters less than ~30 m, typical regolith thickness is therefore estimated to be at least several meters on Toutatis' surface [Quaide and Oberbeck, 1968]. However, the clear drop in the number of medium-size craters on the surface of Toutatis with diameters ranging from ~50 to ~120 m is puzzling and unexpected when compared with other observed asteroids (see Figure 3). The depressed numbers of small craters with diameter smaller than ~100 m (Figure 3b) is most probably caused by illumination bias between the terminator and the rest of the nearside. If the number of small crater is normalized by area over which they were counted, their number is still low compared to that of larger craters (see Figure S1 of the supporting information). We therefore argue that it might be attributed to young large-impact events from which the surface of Toutatis was reset.
 Many distinctive positive features with sizes larger than a few tens of meters were found at the north body, the south head, and the neck of Toutatis (Figure 2a). Based on their irregular shape and localized topographic high appearances (red arrows in Figure 2a), they are most probably boulders. Lack of boulders over the places with a featureless smooth surface may imply that the smooth terrain is covered by regolith. Boulders in the north body and the south head are mainly located at the floor of large degraded craters. Several large boulders with sizes of 30–40 m are found around large craters. Their distributions are consistent with the impact experiments on porous targets in which large fragments and boulders mainly locate at the crater floors and rims [Housen and Holsapple, 2003]. However, the boulders in the neck are large (about 50 m in size) and seem to be partially buried. The partially buried boulders on the surface of asteroid have been attributed to seismic shaking during the resurfacing process [Asphaug et al., 2001; Richardson et al., 2004]. However, the low-velocity contact process of the two lobes, during which the target's surface is first damaged and then large fragments are excavated slowly, might offer another explanation for large boulders in the lobe junction, which indirectly implies the contact binaries of Toutatis' two lobes [Hudson and Ostro, 1995].
 Two parallel structures are observed in the middle of the body. Both structures are oriented nearly perpendicular to the asteroid's long dimension and span almost the full width of the body (see Figure 2a). The elongated depressions with several tens of meters in width suggest that both structures might be grooves. The superposed small craters on the depressions indicate a considerable age and the collapse of loose materials. The origin of grooves on small bodies like Phobos and Eros is thought to be related to fractures of their interiors, from which the surface was disturbed and partially restructured [Thomas and Veverka, 1979; Veverka et al., 1994; Cheng et al., 2009]. Fracture energy arguments however suggest that asteroid fractures may be cracks in unconsolidated regolith, rather than solid rock. According to the spatial distribution of grooves on the surface of Toutatis, these two grooves are most probably related to the large-impact basin at the south pole of Toutatis (planes orthogonal to the basin center). The large-impact event produced strong stress wave that was able to deform the surface by the distal effect of stress wave escaping to the surface at considerable distance from the basin. However, large ridges and striations found by radar images as described by Hudson et al.  cannot be identified on the nearside according of the Chang'E-2 observations. They should be on the farside if they exist.
 The large crater located at the south pole is one of the most intriguing characteristics of Toutatis that was not revealed in previous studies. Its diameter of ~800 m is about one third of the mean diameter of Toutatis (diameter of a sphere of equivalent volume, D~2.45 km [Hudson et al., 2003]). The complicated ridge structures and large boulders in the inner part of the basin reveal that this crater is relatively young. The mean density of Toutatis is ~2.1 g/cm3 [Whipple and Shelus, 1993]. It is much smaller than the value of ordinary chondrites (~3.4 g/cm3) that is the meteoritic counterpart of Toutatis according to the spectroscopic observations [Reddy et al., 2012]. This indicates that the porosity of Toutatis could be as high as 35–45%. Large craters on such a highly porous target may result from the process of material compression [Housen et al., 1999] in which case most of the ejecta material never escapes the crater [Housen and Holsapple, 2003, 2012]. This is in agreement with our observations of the absence of an ejecta blanket surrounding this crater.
 The ratio of crater diameter to the equal-mass-sphere diameter of Toutatis is about 0.33, which is not particularly high for porous targets. If we assume a solid impactor and appropriate scaling relation [Melosh, 1989], the energy of the impactor striking a component target for such a crater reaches about 5 × 1011 J, which is considerably larger than the energy required for shattering a solid rock with the same size of Toutatis [Holsapple, 2009]. Therefore, the existence of this large crater may imply that Toutatis is not a monolith, but rather already strongly fractured at the time of the crater formation. Furthermore, the mean density of Toutatis is too low to be made from solid rocks. All of these indicate that Toutatis might be a rubble pile formed from fragments that have adhered over time. This is also consistent with the observations of fresh ridges and boulders at the floor of the large crater.
 Craters on loose particulate targets may result from a material compression effect [Housen et al., 1999], and the corresponding cratering efficiency may therefore be smaller than those formed on other targets [Schultz et al., 2005]. Since the gravity of Toutatis is so low (e.g., ~0.072 cm/s2), strength rather than gravity provides the predominant response for such a large crater on a highly porous small body [Housen and Holsapple, 2003]. If strength scaling applies, the large crater on the south pole was formed by an impactor ~50 m in diameter traveling at 5 km/s with a typical impact angle of 45° [Holsapple and Housen, 2007]. The huge kinetic energy of the impactor (~1.5 × 1015 J) was transferred to compress the materials to form a crater and absorbed by Toutatis because of its very high porosity [Housen and Holsapple, 2003; Holsapple and Housen, 2012]. This consideration also gives us some hints on how Toutatis might have formed from the collision of two separate entities. If the interiors of both small asteroids (e.g., the head and body) were highly fractured and porous because of previous impacts, their mutual collision might absorb their relative kinetic energy, forming a contact binary without breaking apart. Note that, within the uncertainty range, the similar crater cumulative frequencies in both lobes by the rough estimation imply the similar epoch for their formations (see Figure 3).
 The large-impact events on asteroids would cause global shaking to reset the surface morphologies [Sullivan et al., 2002] and cratering record [Greenberg et al., 1996; Thomas and Robinson, 2005]. However, there are at least four craters on the nearside of Toutatis with a similar size as the large crater at the south pole. Did these impact events reset the initial surface features? Or is the current topography the same as prior to these impacts? To know this, we calculated the critical crater diameter of Toutatis, Dcrit, which is the threshold diameter caused by an impact event capable of degrading all prior topography of a scale Dcrit or smaller [Asphaug, 2008]. In the calculation, the crater scaling relations and power-law approximations for stress wave particle velocity attenuation with radial distance are used. Figure 4 shows the seismic attenuation factor α as a function of normalized critical crater diameter χ = Dcrit/D. Here α is the deduced attenuation exponent at which peak particle velocity decays with radial distance. The dark green square plotted along the curve represents the large crater at the south pole of Toutatis. In general, α is a constant with a value of 1.2–1.3 for most asteroids (the dark region) and 1.3–1.4 for porous asteroids (the gray region) when the observed crater is near the critical crater diameter. If α is fixed within the range of 1.25–1.4, predictions show that Toutatis should have a critical diameter of 150–340 m. Nevertheless, there are at least five craters with diameters exceeding this value, analogous to asteroids like Mathilde [Housen et al., 1999], which is crowded with large craters. This means that the surface of Toutatis was reset, and most preexisting small craters were destroyed by surface seismic waves caused by these large-impact events. This might be the reason that most large craters are degraded and few medium-size craters exist on the surface of Toutatis. Additionally, for the large crater at the south pole, α = 1.43. This value is higher than those of common porous asteroids and therefore would attenuate the heavy shock wave. This could explain the crowding of giant craters that did not result in the global disruption of Toutatis.
 Chang'E-2 did not acquire spectroscopic data from Toutatis, but the absence of large variations of surface color in the filtered images (see Figure 2) suggests that the surface composition of Toutatis might be homogenous. This is consistent with the near-infrared spectroscopy (NIR) spectra observations by the UK infrared telescope [Davies et al., 2007], from which no surface heterogeneities were detected [Reddy et al., 2012]. The large crater at the south pole of Toutatis is young and is not contaminated by other impact events, and the materials excavated by the crater-forming process therefore might represent the composition in the interior of the Toutatis. However, no heterogenetic NIR spectra observed at the south pole [Reddy et al., 2012] implies that the interior composition might be the same as that of the surface.
 Even though Toutatis was observed in detail by ground-based optical and radar observations and its shape and bulk density have been determined to high accuracy, the close flyby observations by Chang'E-2 brought us many new insights and several surprises. The presence of boulders, grooves, and craters (with five larger than 200 m) on Toutatis' surface in combination with its small bulk density and large porosity imply that Toutatis might be a rubble-pile body. That Toutatis suffered plenty of large-impact events that did not result in catastrophic disruption and can be attributed to the strong attenuation of shock waves due to its high porosity. The relatively low number of medium-sized craters with diameters ~50–120 m may be related to the global seismic shaking by larger impact events. Such surface resetting is further assisted by the existence of a thick regolith layer. The partially buried boulders in the lobe junction were generated from the two lobes' low-velocity contact. It indirectly implies that Toutatis is a contact binary. All these features provide valuable information on the formation mechanism and evolutional process of Toutatis.
 M.Z. drafted the manuscript and contributed scientific interpretations of the observations with the discussions of W.F. and W.I.. J.H., L.M., X.W., and D.Q. were responsible for the observations of Toutatis by Chang'E-2 spacecraft. T.L. and W.F. counted the craters on the surface of Toutatis. All of the authors, including J.Y., A.X., and Z.T., discussed and provided significant comments on the results and manuscript.
 We thank Chang'E-2 project team members for their work. We thank Andrew Dombard and two anonymous reviewers for their thoughtful reviews and constructive comments on the manuscript. This research was supported by the Science and Technology Development Fund of Macau (019/2010/A2; 004/2011/A1; 048/2012/A2) and the National Natural Science Foundation of China (11203002). This is PKU PRSL contribution 5.
 The Editor thanks Clark Chapman and two anonymous reviewers for their assistance in evaluating this paper.