Astronomy Colloquium

  The study of super-Eddington accretion is essential to ourunderstanding of the growth of super-massive black holes in the early universe, the accretion of tidally disrupted stars, and the nature of ultraluminous X-ray sources.  Unfortunately, this mode of accretion is particularly difficult to model because of the multidimensionality of the flow, the importance magnetohydrodynamic turbulence, and the dominant dynamical role played by radiation forces.  However, recent increases in computing power and advances in algorithms are facilitating major improvements in our ability to model the transportof radiation in numerical simulations of astrophysical plasmas.  I will briefly describe our new radiation transfer modules and discuss our efforts to model super-Eddington accretion flows.  I will focus on applications to ultraluminous X-ray sources, which must be radiating well above their Eddington luminosity unless they harbor intermediate mass black holes. I will argue that most of these sources can be (and likely are) "normal" ~10 solar mass black holes accreting and radiating with luminosities well above their Eddington "limit".