Elsevier

Icarus

Volume 250, April 2015, Pages 215-221
Icarus

Crater-ray formation by impact-induced ejecta particles

https://doi.org/10.1016/j.icarus.2014.11.030Get rights and content

Highlights

  • We carry out impact experiments using granular targets.

  • We observed the pattern of impact-induced ejecta using a high-speed camera.

  • Mesh pattern, loosely woven with spaces like a net, was observed.

  • The features of the mesh pattern are similar to some rays on the Moon.

  • Numerical simulation shows that inelastic mutual collisions lead to clear pattern.

Abstract

We performed impact experiments with granular targets to reveal the formation process of crater “rays”, the non-uniform ejecta distributions around some fresh craters on the Moon and planets. We found mesh patterns, loosely woven with spaces like a net, as ejecta. A characteristic length of spaces between meshes was evaluated, and an angle, defined as the ratio of the characteristic length to the distance from the ejection point, was obtained as ∼a few degrees. These features are similar to the results of the analyses of the ray patterns around two lunar craters, Glushko and Kepler. Numerical simulations of granular material showed that clear mesh pattern appeared at lower coefficients of restitution between particles but was less clear at larger one, suggesting that the inelastic collisions between particles cause the clear mesh-pattern formation of impact ejecta.

Introduction

Around many craters on the solid surfaces of celestial bodies such as the Moon, “rays”, which extend from craters in sub-radially to radially oriented filaments and diffuse patches (Melosh, 1989), are observed (Fig. 1). The nature of rays has been studied long before the first spacecraft missions to the Moon, and the understanding of the lunar rays has been furthered by high-resolution images taken by recent lunar and planetary spacecraft. On the brightness of the rays, Hawke et al. (2004) summarized that lunar rays are bright due to the compositional contrast with the surrounding terrain (e.g., ray material containing highlands-rich primary ejecta and the adjacent dark mare surface) and/or the presence of immature (fresh, high-albedo) material. Various models on the formation process of such contrast have been proposed such as the emplacement of primary ejecta, the deposition of local material from secondary craters, high-speed debris ejected from the spall zone, and surface scouring by ejecta (e.g., Oberbeck, 1971, Allen, 1977, Schultz and Gault, 1985, Melosh, 1989, Hawke et al., 2004). However, another remarkable feature of crater rays, the spatial non-uniformity, seems to have not been much discussed; the number of studies on the generation mechanism of the non-uniformities of rays is relatively small (e.g., Andrews (1977) found the ray formation in explosion cratering experiments, where spatial patterns may correlate with the non-uniformities of high-velocity detonation products, and Shuvalov (2012) recently numerically simulated the ray formation process based on the interaction of shock waves with old craters, where the ray pattern should relate with the non-uniformities of existing craters).

The purpose of this paper is to provide an alternative model on the generation mechanism of the spatial non-uniformities of crater rays. We focus on the granular materials such as regolith and small fragments caused by impact as ejecta. Granular materials that begin in initially uniform density with some velocity distribution dissipate their initial kinetic energy through the inelastic collisions between granular particles. In the field of physics, it has been shown that inelastically colliding particles spontaneously form “clusters”, locally overdense (the high degrees of the concentration of particles) regions, and that the density of the granular system becomes nonuniform (e.g., Goldhirsch and Zanetti, 1993, McNamara and Young, 1994, Kudrolli et al., 1997, Goldhirsch, 2003). Recently, Möbius (2006) and Royer et al. (2009) carried out experiments using freely falling granular streams in a vacuum and indicated that cluster formation also occurred even in the gravity field. Based on these phenomena, in this paper, we consider the formation process of rays such that spatial non-uniform patterns are formed from the collisions between ejecta particles. First, we report the results of impact experiments with granular targets; we show the spatial distributions of granular materials during its flight after impacts, taking consecutive images of ejecta by a high-speed camera. Then, we compare the results of the experiments and some lunar rays based on the Fourier transformation analysis. Moreover, numerical simulations show that inelastic properties of ejected particles are important for the formation of clear patterns in ejecta curtains.

It is noted that, though the presence of air certainly affects ejecta emplacement (e.g., Schultz, 1992, Barnouin-Jha and Schultz, 1996, Barnouin-Jha and Schultz, 1998, Barnouin-Jha et al., 1999, Suzuki et al., 2013), the effects of air are not considered in this work.

Section snippets

Experimental methods

We carried out impact experiments with an air gun at University of Occupational and Environmental Health. Cylindrical aluminum projectiles with a diameter of 10 mm and a height of 10 mm (2.1 g in mass) were accelerated and impacted on the surface of granular targets with an angle of 90° at ∼30–110 m s−1. We used three kinds of granular materials as targets shown in Fig. 2: (a) spherical glass-beads with a size of 90–110 μm and a bulk density ρbulk of 1.44 g cm−3, (b) “garden” sands with a size of

Results

Fig. 3 shows consecutive images of a glass-beads target. A projectile from the top impacted perpendicularly onto the surface of the target with an impact velocity of 94.7 m s−1. It is clearly seen in each frame that the spatial distribution of ejecta is a “mesh” pattern loosely woven with spaces like a net.

In Fig. 4a, we show three ejecta distributions, which are taken along three black horizontal lines, A, B, and C, in the lower-left panel in Fig. 3 with a length of 512 pixels (87 mm). The

Crater rays on the Moon

We analyzed lunar rays around two similar sized craters, Glushko (Fig. 1a) and Kepler (Fig. 1b). We select these craters because some rays around these craters lay on lunar mare and are relatively clear due to high contrast. The distributions of the rays along three lines, 1, 2, and 3, for Glushko crater (Fig. 1a) with a length of 128 pixels (97.3 km) are shown in Fig. 5a. The distributions are oscillatory curves. We carried out Fourier transformation analysis and the power spectra were

Summary

We performed impact experiments with granular targets to reveal the pattern formation process of impact ejecta and observed the mesh pattern of ejecta, loosely woven with spaces like a net (Fig. 3). The obtained characteristic angle θc of spaces between meshes was similar to that obtained in the analyses of the rays around two craters on the Moon. Two-dimensional numerical simulations with granular media qualitatively showed that inelastic collision between particles was a key process for the

Acknowledgments

The authors would like to thank T. Tanigawa, Y. Cho, T. Hamura, Y. Sekine, and K. Kurosawa. We are also grateful to two anonymous reviewers for helpful comments.

References (25)

  • I. Goldhirsch

    Rapid granular flows

    Annu. Rev. Fluid Mech.

    (2003)
  • I. Goldhirsch et al.

    Clustering instability in dissipative gases

    Phys. Rev. Lett.

    (1993)
  • Cited by (24)

    • Crater-ray formation through mutual collisions of hypervelocity-impact induced ejecta particles

      2020, Icarus
      Citation Excerpt :

      Results show that θc remains almost constant ~0.06 rad across time after the impact (though the result with GB300 largely scatters at the later stages (>~0.5) due to the sparseness of ejecta distribution and the low contrast of images, the average during the stages is similar to that at the earlier stages). A constant angle implies that the pattern expands geometrically and does not evolve; the formation process in the ejecta curtain has finished at the observation time, which is consistent with a previous study (Kadono et al., 2015). The fact that no difference in θc is found among the results of GB100_100, GB100_2, and GB100_6 indicates that no dependence of θc is expected on impact velocity.

    • The mixing of lunar regolith: Vital updates to a canonical model

      2018, Icarus
      Citation Excerpt :

      The flux of secondaries onto the Moon is not uniformly distributed; rather, the likelihood a point is overturned by a secondary impact is largely dependent on proximity to the primary impact crater. Even proximity is no guarantee when secondary-forming ejecta fall in clusters, spokes, and rays (Schultz and Gault, 1985; Vickery, 1986; Melosh, 1989; Kadono et al., 2015) and shallow angle primary impacts display an asymmetric ejecta distribution (Gault and Wedekind, 1978). In addition to the relationship between size, velocity, and impact angle, secondaries are also controlled by the location of their parent primary impact.

    • The length of lunar crater rays explained using secondary crater scaling

      2018, Icarus
      Citation Excerpt :

      Secondary craters outside of the continuous ejecta blanket are not randomly distributed. Some mechanism, possibly inelastic collisions between ejecta particles (Kadono et al., 2015), organizes ejecta to form high spatial density clusters of secondary craters. The model contains four poorly constrained variables describing ejecta behavior and one arbitrarily chosen variable, P.

    View all citing articles on Scopus
    View full text