Radial-orbit instability of a family of anisotropic Hernquist models with and without a supermassive black hole

Abstract

We present a method to investigate the non-radial stability of a spherical anisotropic system that hosts a central supermassive black hole (SBH). Such systems have never been tested before for stability, although high anisotropies have been considered in the dynamical models that were used to estimate the masses of the central putative SBHs. A family of analytical anisotropic spherical Hernquist models with and without a black hole were investigated by means of N-body simulations. A clear trend emerges that the SBH has a significant effect on the overall stability of the system, that is, an SBH with a mass of a few per cent of the total mass of the galaxy can prevent or reduce the bar instabilities in anisotropic systems. Its mass determines not only the strength of the instability reduction, but also the time in which this occurs. These effects are most significant for models with strong radial anisotropies. Furthermore, our analysis shows that unstable systems with similar SBHs but with different anisotropy radii evolve differently: highly radial systems become oblate, while more isotropic models tend to form into prolate structures. In addition to this study, we also present a Monte Carlo algorithm to generate particles in spherical anisotropic systems.

An iterative method for the construction of N-body galaxy models in collisionless equilibrium

“…The stability appears to be excellent in most cases, something that is also confirmed by other measures, such as the time evolution of kinetic and potential energies. Only the B2 model performs slightly worse, an outcome that we blame on the dominance of radial orbits in the dark matter component of this model even in the very centre (similar as in H2), and some of these orbits can be affected by the radial orbit instability (Buyle et al 2007). …”

Completely analytical families of anisotropic -models

“…Our distribution functions are in exact equilibrium. They can be coupled with a three‐dimensional Monte Carlo simulator to provide valid initial conditions for N ‐body and Monte Carlo simulations (Buyle et al 2006), and so avoid the subsequent development being influenced by artefacts of the start.…”

“…As the importance of numerical simulations increases, the issue of controlling numerical errors in them becomes equally important. One can avoid invalid initial conditions for N ‐body and Monte Carlo simulations simply by using a three‐dimensional Monte Carlo simulator on an exact distribution function to derive the initial conditions (Buyle et al 2006).…”

The Dynamical Structure of Dark Matter Halos With Universal Properties

“…The DFs can also serve to generate initial conditions with Monte Carlo simulators (e.g. Kazantzidis et al 2004;Buyle et al 2007b) to investigate the physical processes within these equilibrium models by means of controlled numerical N-body simulations. Finally, our al-gorithm and base functions can be used in other dynamical studies, using different data moments than the density.…”