Long-term and large-scale viscous evolution of dense planetary rings

Abstract

Planetary rings are common in the outer Solar System but their origin and long-term evolution is still a matter of debate. It is well known that viscous spreading is a major evolutionary process for rings, as it globally redistributes the disk’s mass and angular momentum, and can lead to the disk’s loosing mass by infall onto the planet or through the Roche limit. However, describing this process is highly dependent on the model used for the viscosity. In this paper we investigate the global and long-term viscous evolution of a circumplanetary disk. We have developed a simple 1D numerical code, but we use a physically realistic viscosity model derived from N-body simulations ( Daisaka et al., 2001), and dependent on the disk’s local properties (surface mass density, particle size, distance to the planet). Particularly, we include the effects of gravitational instabilities (wakes) that importantly enhance the disk’s viscosity. This method allows to study the global evolution of the disk over the age of the Solar System. Common estimates of the disk’s spreading time-scales with constant viscosity significantly underestimate the rings’ lifetime. We show that, with a realistic viscosity model, an initially narrow ring undergoes two successive evolutionary stages: (1) a transient rapid spreading when the disk is self-gravitating, with the formation of a density peak inward and an outer region marginally gravitationally stable, and with an emptying time-scale proportional to 1 / M 0 2 (where M 0 is the disk’s initial mass), (2) an asymptotic regime where the spreading rate continuously slows down as larger parts of the disk become non-self-gravitating due to the decrease of the surface density, until the disk becomes completely non-self-gravitating. At this point its evolution dramatically slows down, with an emptying time-scale proportional to 1/ M 0, which significantly increases the disk’s lifetime compared to the case with constant viscosity. We show also that the disk’s width scales like t 1/4 with the realistic viscosity model, while it scales like t 1/2 in the case of constant viscosity, resulting in much larger evolutionary time-scales in our model. We find however that the present shape of Saturn’s rings looks like a 100 million-years old disk in our simulations. Concerning Jupiter’s, Uranus’ and Neptune’s rings that are faint today, it is not likely that they were much more massive in the past and lost most of their mass due to viscous spreading alone.