Co-moving frame radiative transfer in spherical media with arbitrary velocity fields

¹ Dept. of Physics and Astronomy, University of Oklahoma, 440 W. Brooks, Rm 131, Norman, OK 73019, USA e-mail: baron@nhn.ou.edu
² Computational Research Division, Lawrence Berkeley National Laboratory, MS 50F-1650, 1 Cyclotron Rd, Berkeley, CA 94720-8139, USA
³ Laboratoire de Physique Nucléaire et de Hautes Énergies, CNRS-IN2P3, University of Paris VII, Paris, France
⁴ Hamburger Sternwarte, Gojenbergsweg 112, 21029 Hamburg, Germany e-mail: yeti@hs.uni-hamburg.de

Received: 8 January 2004
Accepted: 30 July 2004

Abstract

Recently, with the advances in computational speed and availability there has been a growth in the number and resolution of fully 3D hydrodynamical simulations. However, all of these simulations are purely hydrodynamical and there has been little attempt to include the effects of radiative transfer except in a purely phenomenological manner because the computational cost is too large even for modern supercomputers. While there has been an effort to develop 3D Monte Carlo radiative transfer codes, most of these have been for static atmospheres or have employed the Sobolev approximation, which limits their applicability to studying purely geometric effects such as macroscopic mixing. Also the computational requirements of Monte Carlo methods are such that it is difficult to couple with 3D hydrodynamics. Here, we present an algorithm for calculating 1D spherical radiative transfer in the presence of non-monotonic velocity fields in the co-moving frame. Non-monotonic velocity flows will occur in convective, and Raleigh-Taylor unstable flows, in flows with multiple shocks, and in pulsationally unstable stars such as Mira and Cepheids. This is a first step to developing fully 3D radiative transfer than can be coupled with hydrodynamics. We present the computational method and the results of some test calculations.