Volume 463, Number 3, March I 2007
|Page(s)||921 - 935|
|Published online||19 December 2006|
Slow evolution of elliptical galaxies induced by dynamical friction
III. Role of density concentration and pressure anisotropy
Università degli Studi di Milano, Dipartimento di Fisica, via Celoria 16, 20133 Milano, Italy e-mail: [Serena.Arena;Giuseppe.Bertin]@unimi.it
Accepted: 27 November 2006
Context.Dynamical friction is expected to play an important role in a variety of astrophysical contexts, yet we still lack a quantitative understanding of this basic mechanism and of its effects. In fact, numerical simulations have shown that in inhomogeneous systems the classical idealized description, given by Chandrasekhar (1943), may have severe limitations, possibly related to the global nature of the processes that are involved.
Aims.In this paper we address two issues. Firstly, we study how dynamical friction depends on the density concentration and on the pressure anisotropy of the host galaxy. For the purpose, we consider models characterized by a “realistic” distribution function and compare the behavior of dynamical friction in these systems to that found in other simpler models (often used in the past because of their mathematical convenience). Secondly, we study the response of the galaxy to the infall, by dynamical friction, of heavy objects (“satellites”) taken in a variety of initial configurations.
Methods.The investigation is carried out by using a numerical laboratory set up in previous papers of this series. The process of dynamical friction is studied in terms of its strength (i.e., its efficiency to drag a satellite toward the galaxy center) and in terms of its ability to circularize the orbit of the satellite under friction. The response of the galaxy is studied in terms of the induced modifications to the galaxy density distribution and shape and of the changes produced to its phase space properties.
Results.(1) We find that, within the range of our models, the pressure anisotropy present in the host galaxy has little effect on dynamical friction. Instead, the shape of the galaxy density profile is very important. The classical idealized description, although with an effectively smaller Coulomb logarithm, appears to be applicable to galaxy models characterized by a broad core (such as a polytrope) but not to concentrated models. Correspondingly, in contrast to the behavior found in models with a broad core, the orbits of satellites captured in concentrated models are not circularized by dynamical friction. To a large extent, these results confirm trends already known in the literature; in this respect, we also confirm the value of some simple modifications to the classical formulae proposed in the literature. (2) The induced evolution in the host galaxy reflects the initial conditions adopted for the captured satellite. Satellites spiraling in on quasi-circular orbits tend to modify the pressure tensor of the host galaxy in the tangential direction, while satellites captured along quasi-radial orbits tend to induce pressure anisotropy in the radial direction. While satellites captured along quasi-circular orbits make a galaxy change its shape from spherical to oblate, satellites captured along quasi-radial orbits tend to induce a shape in the host galaxy of the prolate type. This result suggests that the shape of early-type galaxies may just result from the characteristics of occasional mergers rather than being directly related to the effectiveness of the radial-orbit instability during the process of formation via collisionless collapse, as often argued in the past.
Key words: methods: N-body simulations / galaxies: elliptical and lenticular, cD / galaxies: evolution / galaxies: kinematics and dynamics
© ESO, 2007
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.