Flame Propagation on the Surfaces of Rapidly Rotating Neutron Stars during Type I X-ray Bursts

We present the first vertically resolved hydrodynamic simulations of a laterally propagating, deflagrating flame in the thin helium ocean of a rotating accreting neutron star. We use a new hydrodynamics solver tailored to deal with the large discrepancy in horizontal and vertical length scales typical of neutron star oceans, and which filters out sound waves that would otherwise limit our timesteps. We find that the flame moves horizontally with velocities of order $10^5$ cm s$^{-1}$, crossing the ocean in few seconds, broadly consistent with the rise times of Type I X-ray bursts. We address the open question of what drives flame propagation, and find that heat is transported from burning to unburnt fuel by a combination of top-to-bottom conduction and mixing driven by a baroclinic instability. The speed of the flame propagation is therefore a sensitive function of the ocean conductivity and spin: we explore this dependence for an astrophysically relevant range of parameters and find that in general flame propagation is faster for slower rotation and higher conductivity.

A numerical magnetohydrodynamic scheme using the hydrostatic approximation

In gravitationally stratified fluids, length scales are normally much greater in the horizontal direction than in the vertical one. When modelling these fluids it can be advantageous to use the hydrostatic approximation, which filters out vertically propagating sound waves and thus allows a greater timestep. We briefly review this approximation, which is commonplace in atmospheric physics, and compare it to other approximations used in astrophysics such as Boussinesq and anelastic, finding that it should be the best approximation to use in context such as radiative stellar zones, compact objects, stellar or planetary atmospheres and other contexts. We describe a finite-difference numerical scheme which uses this approximation, which includes magnetic fields.

Magnetic spots on hot massive stars

Hot luminous stars show a variety of phenomena in their photospheres and winds which still lack clear physical explanation. Among these phenomena are photospheric turbulence, line profile variability (LPV), non-thermal emission, non-radial pulsations, discrete absorption components (DACs) and wind clumping. Cantiello et al. (2009) argued that a convection zone close to the stellar surface could be responsible for some of these phenomena. This convective zone is caused by a peak in the opacity associated with iron-group elements and is referred to as the iron convection zone (FeCZ). Assuming dynamo action producing magnetic fields at equipartition in the FeCZ, we investigate the occurrence of subsurface magnetism in OB stars. Then we study the surface emergence of these magnetic fields and discuss possible observational signatures of magnetic spots. Simple estimates are made using the subsurface properties of massive stars, as calculated in 1D stellar evolution models. We find that magnetic fields of sufficient amplitude to affect the wind could emerge at the surface via magnetic buoyancy. While at this stage it is difficult to predict the geometry of these features, we show that magnetic spots of size comparable to the local pressure scale height can manifest themselves as hot, bright spots. Localized magnetic fields could be widespread in those early type stars that have subsurface convection. This type of surface magnetism could be responsible for photometric variability and play a role in X-ray emission and wind clumping.