The Anomalous Drift of Comet ISON (C/2012 S1) due to Sublimating Volatiles near Perihelion

Prior to perihelion passage on 28 November 2013, the observed right ascension (RA) and declination (Dec) coordinates of comet C/2012 S1 (ISON) significantly lagged the predicted JPL (# 53) ephemeris. We show that this "braking effect" is due to a dynamic pressure exerted by sublimating gases on the sunward side of the nucleus [1]. Comet ISON was observed November 23 through November 28 using the SCUBA-2 sub-millimeter camera on the James Clerk Maxwell Telescope (JCMT). Imaging is achieved simultaneously at wavelengths of 850 μm and 450 μm, with RA and Dec determined from the central peak in the coma brightness [2]. When comet ISON was first detected at 850 μm, the 1-mm-sized dust particles were tightly bound to the comet nucleus until at least November 23. Three days later, the dust was less tightly bound, elongated and diffuse, spread out over as much as 120 arc seconds (80,000 km) in the anti-solar direction, suggesting a fragmentation event. We compute the average braking velocity of the nucleus of comet ISON by first measuring the distance between the central RA position and the predicted JPL ephemeris. We then calculate the change in this distance between subsequent observations, and divide this value by the elapsed time between the two observations to yield an average drift velocity of the nucleus over this time interval. We assume that comet ISON, like a number of Jupiter Family Comets visited by spacecraft [3], has low thermal inertia. Thus, the sublimating gases are emitted predominantly on the sunward side of the nucleus. Additionally, we assume that water ice dominates the sublimating gases [4]. We then calculate the pressure on the surface of the nucleus due to the emitted gases using the procedure described in [1]. We match the average drift velocity of the nucleus due to this sublimation pressure with the observed average drift velocity from the JCMT observations, which is sensitive to the size of the body, allowing us to estimate the size of the nucleus (or its fragments) shortly before perihelion. References: [1] Steckloff et al. (2014) Submitted Nature [2] Keane et al. (2014) Submitted Ap. J. [3] Groussin et al. (2013) Icarus 222, 580-594 [4] Combi et al. (2014) Ap. J. 788:L7 (5pp)