Magnetohydrodynamic Models of Accretion Including Radiation Transport

PI Name: 
James Stone
Institution: 
Princeton University
Allocation Program: 
INCITE
Allocation Hours at ALCF: 
54 million
Year: 
2017
Research Domain: 
Physics

The effect of radiation on the structure and dynamics of accretion flows onto compact objects is important to systems ranging from X-ray binaries to active galactic nuclei. While it is known that magnetohydrodynamic (MHD) processes dominate the angular momentum transport in such flows, and while numerical MHD simulations have revealed many important insights, whether any of the expectations of theory in the regime of radiation dominated accretion are borne out remains to be explored. With this INCITE award, a research team will calculate the structure of such flows, and make predictions about how accreting sources evolve and affect their environment.

The team has developed new and accurate numerical algorithms for time-dependent radiation transport (RT) that can be integrated into existing MHD codes to study the physics of radiation dominated flows. These methods have been implemented in a new version of a compressible MHD code, athena++, whose features include mesh refinement and new physics such as algorithms for general relativistic (GR) MHD in stationary spacetimes. This code shows excellent single core performance, and excellent weak scaling on Mira.

This project will use athena++ for a 2-year campaign to run three-dimensional radiation MHD simulations of accretion flows onto compact objects. The first goal is to complete a survey of the structure and dynamics of radiation dominated accretion as the mass accretion rate is varied from highly super-Eddington (the radiation dominated, slim disk regime) to sub-Eddington (the standard, thin disk regime). The team will investigate whether thermal and/or viscous instabilities, predicted over 40 years ago, actually occur in global models of MHD turbulent disks.