[Previous] | [Session 55] | [Next]
R. M. Canup, Wm. R. Ward (Southwest Research Institute)
Large impacts are believed to be common in the late stages of accretion of terrestrial planets (e.g., Wetherill 1992, Agnor et al. 1999) and giant planetary cores (e.g., Lissauer et al. 1995). A case has been made that the likely cause of the 97-degree Uranian obliquity is a late impact by a roughly Earth-sized impactor (e.g, Lissauer and Safronov 1991, Slattery et al. 1992). It has also been suggested (e.g., Pollack et al. 1991) that the satellites of Uranus may have been a byproduct of this event; here we begin to investigate this hypothesis.
The majority of the simulations of potential Uranus-tilting impacts place material originating from the impactor into circumplanetary orbit (Slattery et al. 1992). The yield of orbiting material displays a variation with impact angular momentum similar to that exhibited by simulations of lunar-forming impacts (Canup et al. 2000), and reaches maximum values of 1 - 3 % of the total system mass (the mass of Uranus' 5 largest satellites is roughly 10-4MU). Using N-body accretion simulations, we find that satellite systems resembling that of Uranus result from a particulate disk with an initial mass similar to that of the current satellites and an outer disk edge of 20-25RU. This is likely a much more radially extended disk than would have been produced by an impact; however, such a disk would have viscously spread significantly before it cooled and condensed. If one assumes that accretion would be delayed until a balance between viscous energy dissipation and radiative losses yielded a temperature low enough for condensation, then a rough estimate can be made of the effective disk viscosity required to yield a sufficiently extended disk to account for the placement of the 5 outer satellites.