Particle sizes in Saturn’s rings from UVIS stellar occultations 2. Outlier Populations in the C ring and Cassini Division

Abstract

The Cassini Ultraviolet Imaging Spectrograph (UVIS) included a High-Speed Photometer (HSP), which observed hundreds of stellar occultations by Saturn’s ring system across a range of viewing geometries (Colwell et al. 2010). The unocculted time series data from the HSP follow Poisson counting statistics, such that the second and third central moments of the unocculted data should be equal to the mean photon rate of the star. When the star is occulted by the rings, the presence of ring particles introduces a correlation in the previously uncorrelated photons. This causes each of the central moments to deviate from their expected value. In particular, the second central moment, μ2, commonly known as the variance, becomes greater than the mean (Colwell et al. 2018). The third central moment, μ3, also deviates from this expectation upon occultation, but the nature of the deviation varies with optical depth. In particular, μ3 is less than its Poisson expectation for τ<0.33 and greater than this expectation for τ>0.33. The introduction of outlier features, namely small gaps (‘ghosts’, Baillié et al. (2013)) and clumps produce variable effects on the behavior of both μ2 and μ3, which is also dependent on optical depth.We compare the higher-order moments from Monte Carlo Simulations of a simplified ring system to those from the UVIS data to gain insight into the nature of such outliers in the C ring and Cassini Division. The behavior of the data in the C ring plateaus indicates that a small population of ∼ 10-m ghosts exists in this region, with a frequency of about 1 ghost per km of radial extent. In the background C ring, we find that particle sizes are positively correlated with optical depth and the data cannot be explained with a simple power-law size distribution. Instead, we are able to describe the behavior of the higher-order moments using a bent-power law size distribution of particles. We find similar behavior in the C ring ramp and Cassini Division ramp. However, we cannot rule out nor confirm the presence of either ghosts or clumps in the background Cassini Division. In the Triple Band, the local ∼ 10-m ghost frequency oscillates between about 1 and 4 ghosts per km.