Horizon Scale Simulations of Low Luminosity AGN

The emission coincident with the two targets for horizon scale measurements of supermassive black holes, Sagittarius A* in our own galaxy and the supermassive black hole at the center of M87, is believed to be provided by either an accretion disk or the associated relativistic jet. To accurately model the observational signature of the regions closest to the event horizon, general relativistic magnetohydrodyanmic simulations (GRMHD) are required. However, since both sources are emitting well below the Eddington limit, they fall into a class of accretion systems known as radiatively inefficient accretion flows (RIAFs). The low density and high temperature in such systems imply that the Coulomb collision time is long and thus that collisionless plasma effects need to be considered. In particular, to accurately model the plasma it is needed to account for the two-temperature nature of the system. Previous work dealt with this problem by using phenomenological prescriptions to "paint" electron temperatures onto single-fluid GRMHD simulations. We, for the first time, sought to motivate some of these prescriptions by dynamically evolving the electron temperature in GRMHD simulations using a sub-grid model motivated by kinetic calculations of electron-ion heating in hot, magnetized plasmas.

image
betah Tecontour
Our model for electron heating is motivated by kinetic results for the damping of Alfvenic waves in turbulent, collisionless plasmas, where it is found that electrons (protons) are primarily heated in strongly (weakly) magnetized regions with plasma β<∼1 (β>∼1). In a typical RIAF disk, this implies that hot electrons will be primarily concentrated in the coronal and jet regions. Our results show exactly that, as we plot polar slices of β, the fraction of heat given by electrons, f_e, and σ = b2/(ρ2 c2)for a 3D GRMHD simulation with our electron heating model (left) and the resulting electron temperatures relative to the electron rest mass energy (right).
spec
Spectral energy decomposition resulting from a 3D GRMHD simulation of Sagittarius A* that self-consistently evolves the electron temperature using our model for beta-dependent electron heating. The spectrum includes contributions from both synchrotron emission, absorption, and Compton scattering. The shaded region represents the 1-σ variation in time over the course of ∼ 1 day. Our model well reproduces the spectral slope near 230 GHz and the level of infrared variability but underproduces the low frequency radio slope and lacks the strong flares seen by Chandra. The latter may point to the need for nonthermal particles. Anisotropic electron conduction along field lines has a negligible effect on the emission due to the field being primarily aligned with the φ-direction; mostly perpendicular to the temperature gradients which are primarily in the r-θ plane.