Impact bombardment on the regular satellites of Jupiter and Uranus during an episode of giant planet migration

Abstract

The intensity and effects of early impact bombardment on the major satellites of the giant planets during an episode of giant planet migration is still poorly known. We use a combination of dynamical N-body and Monte Carlo simulations to determine impact probabilities, impact velocities, and expected masses that collide with these satellites to determine the chronology of impacts during the migration. Volatile loss through bombardment is typically 20% for Miranda, a few percents for the larger Uranian satellites and negligible for the Galilean satellites. Due to its small size and the high impact velocity there is a >99% chance that Miranda suffered a catastrophic impact that shattered the satellite. Subsequent re-accretion from a circum-Uranian ring could account for its peculiar surface morphology and low density. The probability to destroy Ariel and Umbriel is 15% and 1% for Titania and Oberon. Approximately 90% of the mass in planetesimals that passes through the Jovian and Uranian satellite systems (about $4 {\rm \ M_{\oplus}}$ and $2 {\rm \ M_{\oplus}}$ respectively) does so in about 15 Myr. This extremely rapid and intense bombardment causes repeated local crustal melting on all satellites. The combination of these effects results in an entirely different impact chronology than that of the inner solar system. We conclude that the simple extrapolation of the lunar chronology to the outer solar system satellites is not correct. The tail end (after 25 Myr) of the chronology function has an e-folding time of 100 Myr at Jupiter, but follows a cumulative Weibull distribution at Uranus, making direct comparisons between the gas and ice giant planets difficult. Based on our results the surfaces of the Uranian satellites, Callisto, and possibly Ganymede, are all about the same age, and are roughly 150 Myr younger than the timing of the dynamical instability.