James E. Pringle

The rapidly evolving dust and gas extinction observed towards WD 1145+017 has opened a real-time window on to the mechanisms for destruction-accretion of planetary bodies on to white dwarf stars, and has served to underline the importance of considering the dynamics of dust particles around such objects. Here it is argued that the interaction between (charged) dust grains and the stellar magnetic field is an important ingredient in understanding the physical distribution of infrared emitting particles in the vicinity of such white dwarfs. These ideas are used to suggest a possible model for WD 1145+017 in which the unusual transit shapes are caused by opaque clouds of dust trapped in the stellar magnetosphere. The model can account for the observed transit periodicities if the stellar rotation is near 4.5 h, as the clouds of trapped dust are then located near or within the co-rotation radius. The model requires the surface magnetic field to be at least around some tens of kG. In contrast to the eccentric orbits expected for large planetesimals undergoing tidal disintegration, the orbits of magnetospherically-trapped dust clouds are essentially circular, consistent with the observations.

The rapidly evolving dust and gas extinction observed towards WD 1145+017 has opened a real-time window onto the mechanisms for destruction-accretion of planetary bodies onto white dwarf stars, and has served to underline the importance of considering the dynamics of dust particles around such objects. Here it is argued that the interaction between (charged) dust grains and the stellar magnetic field is an important ingredient in understanding the physical distribution of infrared emitting particles in the vicinity of such white dwarfs. These ideas are used to suggest a possible model for WD 1145+017 in which the unusual transit shapes are caused by opaque clouds of dust trapped in the stellar magnetosphere. The model can account for the observed transit periodicities if the stellar rotation is near 4.5 h, as the clouds of trapped dust are then located near or within the co-rotation radius. The model requires the surface magnetic field to be at least around some tens of kG. In contrast to the eccentric orbits expected for large planetesimals undergoing tidal disintegration, the orbits of magnetospherically-trapped dust clouds are essentially circular, consistent with the observations.

Standard, planar accretion discs operate through a dissipative mechanism, usually thought to be turbulent, and often modelled as a viscosity. This acts to take energy from the radial shear, enabling the flow of mass and angular momentum in the radial direction. In a previous paper we discussed observational evidence for the magnitude of this viscosity, and pointed out discrepancies between these values and those obtained in numerical simulations. In this paper we discuss the observational evidence for the magnitude of the dissipative effects which act in non--planar discs, both to transfer and to eliminate the non--planarity. Estimates based on the model by Ogilvie (1999), which assumes a small--scale, isotropic viscosity, give alignment timescales for fully ionized discs which are apparently too short by a factor of a few compared with observations, although we emphasize that more detailed computations as well as tighter observational constraints are required to verify this conclusion. For discs with low temperature and conductivity, we find that the timescales for disc alignment based on isotropic viscosity are too short by around two orders of magnitude. This large discrepancy suggests that our understanding of viscosity in quiescent discs is currently inadequate.