We investigate the patterns observed in ejecta curtain induced by hypervelocity impact (2-6 km/s) with a variety of the size and shape of target particles. We characterize the patterns by an angle, defined as the ratio of the characteristic length of the pattern obtained by Fourier transformation to the distance from the impact point. This angle is found to be almost the same as that obtained by the reanalysis of the patterns in the previous study at lower impact velocities (Kadono et al., 2015, Icarus 250, 215-221), which are consistent with lunar crater-ray systems. Assuming that the pattern is formed by mutual collision of particles with fluctuation velocity in excavation flow, we evaluate an angle at which the pattern growth stops and show that this angle is the same in the order of magnitude as the ratio of the fluctuation velocity and the radial velocity. This relation is confirmed in the results of experiments and numerical simulations. Finally, we discuss the dependence of the patterns on impact conditions. The experiments show no dependence of the angle on impact velocity. This indicates that the ratio between the fluctuation and radial velocity components in excavation flow does not depend on impact velocity. Moreover, the independences on particle size and particle shape suggest that the angle characterizing the structure of the patterns does not depend on cohesive force. Since cohesive forces should be related with elastic properties of particles, the structure does not depend on elastic properties, though inelastic collisions are important for the persistence and contrast of the patterns.

We performed impact experiments to observe patterns in an ejecta curtain with targets consisting of small sand particles and large inclusions comparable to or smaller than the size of the projectiles. The spatial intensity distributions in the ejecta at early stages of crater formation depend on the size of the inclusions. Our numerical simulations of radially spreading particles with different sizes support this result. Based on the results, we proposed a procedure for evaluating the subsurface structures of celestial bodies from the images of ejecta curtains obtained from space-impact experiments.

In most of the previous studies related to collisional disruption of planetesimals in the gravity regime, Smoothed Particle Hydrodynamics (SPH) simulations have been used. On the other hand, impact simulations using grid-based hydrodynamic code have not been sufficiently performed. In the present study, we execute impact simulations in the gravity regime using the shock-physics code iSALE, which is a grid-based Eulerian hydrocode. We examine the dependence of the critical specific impact energy Q_RD* on impact conditions for a wide range of specific impact energy (Q_R) from disruptive collisions to erosive collisions, and compare our results with previous studies. We find collision outcomes of the iSALE simulation agree well with those of the SPH simulation. Detailed analysis mainly gives three results. (1) The value of Q_RD* depends on numerical resolution, and is close to convergence with increasing numerical resolution. The difference in converged value of Q_RD* between the iSALE code and the SPH code is within 30%. (2) Ejected mass normalized by total mass (M_ej/M_tot) generally depends on various impact conditions. However, when Q_R is normalized by Q_RD* that is calculated for each impact simulation, M_ej/M_tot can be scaled by Q_R/Q_RD*, and is independent of numerical resolution, impact velocity and target size. (3) This similarity law for Q_R/Q_RD* is confirmed for a wide range of specific impact energy. We also derive a semi-analytic formula for Q_RD* based on the similarity law and the crater scaling law. We find that the semi-analytic formula for the case with a non-porous object is consistent with numerical results.

Regular satellites of giant planets are formed by accretion of solid bodies in circumplanetary disks. Planetesimals that are moving on heliocentric orbits and are sufficiently large to be decoupled from the flow of the protoplanetary gas disk can be captured by gas drag from the circumplanetary disk. In the present work, we examine the distribution of captured planetesimals in circumplanetary disks using orbital integrations. We find that the number of captured planetesimals reaches an equilibrium state as a balance between continuous capture and orbital decay into the planet. The number of planetesimals captured into retrograde orbits is much smaller than those on prograde orbits, because the former ones experience strong headwind and spiral into the planet rapidly. We find that the surface number density of planetesimals at the current radial location of regular satellites can be significantly enhanced by gas drag capture, depending on the velocity dispersions of planetesimals and the width of the gap in the protoplanetary disk. Using a simple model, we also examine the ratio of the surface densities of dust and captured planetesimals in the circumplanetary disk, and find that solid material at the current location of regular satellites can be dominated by captured planetesimals when the velocity dispersion of planetesimals is rather small and a wide gap is not formed in the protoplanetary disk. In this case, captured planetesimals in such a region can grow by mutual collision before spiraling into the planet, and would contribute to the growth of regular satellites.

Disruptive collisions have been regarded as an important process for planet formation, while non-disruptive, small-scale collisions (hereafter called erosive collisions) have been underestimated or neglected by many studies. However, recent studies have suggested that erosive collisions are also important to the growth of planets, because they are much more frequent than disruptive collisions. Although the thresholds of the specific impact energy for disruptive collisions (Q_RD^*) have been investigated well, there is no reliable model for erosive collisions. In this study, we systematically carried out impact simulations of gravity-dominated planetesimals for a wide range of specific impact energy (Q_R) from disruptive collisions (Q_R ~ Q_RD^*) to erosive ones (Q_R << Q_RD^*) using the smoothed particle hydrodynamics method. We found that the ejected mass normalized by the total mass (M_ej/M_tot) depends on the numerical resolution, the target radius (R_tar) and the impact velocity (v_imp), as well as on Q_R, but that it can be nicely scaled by Q_RD^* for the parameter ranges investigated (R_tar = 30-300 km, v_imp = 2-5 km/s). This means that M_ej/M_tot depends only on Q_R/Q_RD^* in these parameter ranges. We confirmed that the collision outcomes for much less erosive collisions (Q_R < 0.01 Q_RD^*) converge to the results of an impact onto a planar target for various impact angles and that M_ej/M_tot = C * QR/QRD* holds. For disruptive collisions (Q_R ~ Q_RD^*), the curvature of the target has a significant effect on Mej/Mtot. We also examined the angle-averaged value of M_ej/M_tot and found that the numerically obtained relation between angle-averaged M_ej/M_tot and Q_R/Q_RD^* is very similar to the cases for 45-degree impacts. We proposed a new erosion model based on our numerical simulations for future research on planet formation with collisional erosion.

Sufficiently massive growing giant planets have circumplanetary disks, and the capture of solid bodies by the disks would likely influence the growth of the planets and formation of satellite systems around them. In addition to dust particles that are supplied to the disk with inflowing gas, recent studies suggest the importance of capture of planetesimals whose motion is decoupled from the gas, but orbital evolution of captured bodies in circumplanetary gas disks has not been studied in detail. In the present work, using three-body orbital integration and analytic calculations, we examine orbital characteristics and subsequent dynamical evolution of planetesimals captured by gas drag from circumplanetary gas disks. We find that the semi-major axes of the planet-centered orbits of planetesimals at the time of permanent capture are smaller than about one third of the planet's Hill radius in most cases. Typically, captured bodies rapidly spiral into the planet, and the rate of the orbital decay is faster for the retrograde orbits due to the strong headwind from the circumplanetary gas. When a planetesimal captured into a retrograde orbit suffers from sufficiently strong gas drag before spiraling into the planet, its orbit turns to the prograde direction at a radial location that can be explained using the Stokes number. We also find that those captured into certain types of orbits can survive for a long period of time even under gas drag both in the prograde and retrograde cases, which may be important for the origin of irregular satellites of giant planets.

When gas giant protoplanets grow sufficiently massive, circumplanetary disks would form. While solid bodies captured by the circumplanetary disks likely contribute to the growth of the planets and regular satellites around them, some of captured bodies would remain in planet-centered orbits after the dispersal of the disk. We examine capture and subsequent orbital evolution of planetesimals in waning circumplanetary gas disks using three-body orbital integration. We find that some of captured planetesimals can survive in the circumplanetary disk for a long period of time under such weak gas drag. Captured planetesimals have semi-major axes smaller than about one third of the planet's Hill radius. Distributions of their eccentricities and inclinations after disk dispersal depend on the strength of gas drag and the timescale of disk dispersal, and initially strong gas drag and quick disk dispersal facilitates capture and survival of planetesimals. However, in such a case, final orbital eccentricities and inclinations of captured bodies remain rather large. Although our results suggest that some of the present irregular satellites of gas giant planets with small semi-major axes would have been captured by gas drag, other mechanisms are required to fully explain their current orbital characteristics.