PADIS

Pubblicazioni Aperte DIgitali Sapienza > Fisica > ASTRONOMIA >

Please use this identifier to cite or link to this item: http://hdl.handle.net/10805/1304

Title: Globular Clusters and Galactic Nuclei
Authors: MASTROBUONO BATTISTI, ALESSANDRA
Tutor: Capuzzo Dolcetta, Roberto
Keywords: Stellar dynamics
Numeric astrophysics
Galaxy
Globular clusters
Nuclear star clusters
Dynamical evolution
Issue Date: 15-Dec-2011
Abstract: Dynamical evolution plays a key role in shaping the current properties of star clusters and star cluster systems. We present the study of stellar dynamics both from a theoretical and numerical point of view. In particular we investigate this topic on different astrophysical scales, from the study of the orbital evolution and the mutual interaction of GCs in the Galactic central region to the evolution of GCs in the larger scale galactic potential. Globular Clusters (GCs), very old and massive star clusters, are ideal objects to explore many aspects of stellar dynamics and to investigate the dynamical and evolutionary mechanisms of their host galaxy. Almost every surveyed galaxy of sufficiently large mass has an associated group of GCs, i.e. a Globular Cluster System (GCS). The first part of this Thesis is devoted to the study of the evolution of GCSs in elliptical galaxies. Basing on the hypothesis that the GCS and stellar halo in a galaxy were born at the same time and, so, with the same density distribution, a logical consequence is that the presently observed difference may be due to evolution of the GCS. Actually, in this scenario, GCSs evolve due to various mechanisms, among which dynamical friction and tidal interaction with the galactic field are the most important. On the other side, the collisionless stellar halo component stands unchanged, thus the difference between the two profiles may correspond to mass lost by the GCS to the galactic center. There the GCs merge and they contribute to the formation/accretion of a luminous and compact central Nuclear Star Cluster (NSC). This is known as the “merger model” for the formation of NSCs, observed at the center of many galaxies and also in the Milky Way (MW) center. In the second part of this work a new high performance code, NBSymple, is presented. NBSymple is an efficient N-body integrator implemented on a hybrid CPU+GPU platform, exploiting a double-parallelization on CPUs and on the hosted Graphic Processing Units (GPUs). The precision is guaranteed by direct summa- tion for force evaluation, and on the use of high order, symplectic time integration methods. The code allows the choice between two different symplectic integrators: a second-order algorithm (commonly known as leapfrog) and a much more accurate (but also time consuming) sixth-order method. The effect of the external galactic field is represented as an analytical approximation of its gravitational potential. The code has been widely tested and benchmarked. Moreover, it has been used for various applications (globular clusters quasi-radial orbit through a galactic massive central object, primordial evolution of young stellar clusters, etc.). NBSymple and another, publicly available, direct summation code, φGRAPE, have been used to explore the the previously described merger mode for the Galactic NSC formation. In particular, we used self consistent N-body simulations where the Galaxy was modeled using observational data about the Milky Way, and including the presence of the Galactic central supermassive black hole. We let decay 12 GCs initially located on different circular orbits at the same galactocentric distance. The merging of clusters in the central zone of the Galaxy and its following evolution due to two-body relaxation generates a NSC that actually resembles the one observed at the center of the MW. By mean of numerical simulations carried on with NBSymple, we investigated more in detail the dynamical evolution of GCs in the MW potential with particular attention to the formation of clumpy structures in the tidal tails that arise around the orbiting cluster. Although various hypothesis have been proposed, the formation process of these clumps is not yet clearly understood. Through a statistical analysis of the orbital properties of the stars that “escape” from the cluster we aimed to a better understanding of the on going process. Studying and comparing such simulations with observational data we could gain to a deeper knowledge of the shape Galactic potential and, more generally, of the Galactic dynamics.
Description: PhD Thesis
URI: http://hdl.handle.net/10805/1304
Research interests: Computational Astrophysics, Theoretical Astrophysics, Galactic Dynamics, Stellar Dynamics
Skills short description: My past and current research focuses on the field of stellar dynamics. In particular during my PhD I gained knowledge and abilities in the study of dynamical evolution of globular clusters in Milky Way-like potentials, paying particular attention to the possible consequences on the Galactic structure. For this purpose I used N-body direct summation parallel codes which I realized during my PhD. The codes exploit GPUs as computing accelerators, by mean of the CUDA C programming language and can run on multiple GPUs and CPUs systems.
Appears in PhD:ASTRONOMIA

Files in This Item:

File Description SizeFormat
Thesis_AMB.pdfTesi di Dottorato, XIV ciclo, Astronomia5.64 MBAdobe PDF

File del Curriculum Vitae:

CurriculumVitae.pdf 69.41 kBAdobe PDF


This item is protected by original copyright

Recommend this item

Items in PADIS are protected by copyright, with all rights reserved, unless otherwise indicated.

 

Valid XHTML 1.0! DSpace Software Copyright © 2002-2010  Duraspace - Feedback Sviluppo e manutenzione a cura del CINECA