2006/08/12 by D. Heinzeller, S. Mineshige, K. Ohsuga · 2 citations
Physics and Astronomy · #Accretion (finance) #Anisotropy #Astrophysical Phenomena and Observations #Astrophysics and Cosmic Phenomena #Doppler effect #Galaxies: Formation, Evolution, Phenomena #Photon #Radiation #Radiative transfer #Spectral energy distribution #Thermal #Thermal radiation #astro-ph
paper · pdf · doi:10.1111/j.1365-2966.2006.10926.x
published as Mon.Not.Roy.Astron.Soc.372:1208-1216,2006 · 10 pages, 8 figures, accepted for publication in MNRAS
arxiv created 2006/08/12 · openalex publication_date 2006/09/19 · arxiv updated 2009/12/01 · openalex created_date 2016/06/24 · openalex updated_date 2026/08/05
Spectral properties of super-Eddington accretion flows are investigated by means of a parallel line-of-sight calculation. The subjacent model, taken from the two-dimensional radiation hydrodynamic simulations by Ohsuga et al. (2005), consists of a disc accretion region and an extended atmosphere with high-velocity outflows. The non-grey radiative transfer equation is solved, including relativistic effects, by applying the flux-limited diffusion approximation. The calculated spectrum is composed of a thermal, blackbody-like emission from the disc which depends sensitively on the inclination angle, and of high-energy X-ray and gamma-ray emission from the atmosphere. We find mild beaming effects in the thermal radiation for small inclination angles. If we compare the face-on case with the edge-on case, the average photon energy is larger by a factor of ∼1.7 due mainly to Doppler boosting, while the photon number density is larger by a factor of ∼3.7 due mainly to anisotropic matter distribution around the central black hole. This gives an explanation for the observed X-ray temperatures of ULXs which are too high to be explained in the framework of intermediate-mass black holes. While the main features of the thermal spectral component are consistent with more detailed calculations of slim accretion discs, the atmosphere induces major changes in the high-energy part, which cannot be reproduced by existing models. We also conclude that, in order to interpret the observational data properly, simple approaches like the Eddington–Barbier approximation cannot be applied.