Oleg Y. Gnedin

Formation and Dynamical Evolution of Globular Clusters

Tidal Shocks and the Destruction of globular clusters in our Galaxy

Gnedin & Ostriker (1997)   To investigate the dynamical evolution of the Galactic Globular Cluster System, we used a Fokker-Planck code which includes two-body stellar relaxation, tidal truncation of the clusters, compressive gravitational shocks during the passage through the Galactic disk, and the tidal shock perturbations near the Galactic bulge. The results suggest that more than half of the present sample of globular clusters will be destroyed within the next Hubble time. The destruction rate varies from 52% to 86% depending on the Galactic model.

Gnedin, Hernquist & Ostriker (1999)   We derived the formalism for the tidal field exerted by a spherically symmetric galaxy having an extended mass distribution, and calculated tidal perturbations and heating of stars in a globular cluster or a satellite galaxy orbiting in the external potential. Heating on highly eccentric orbits dominates as the adiabatic corrections strongly damp the energy change on low eccentricity orbits. The results are illustrated for the example of globular cluster NGC 6712. For the orbital eccentricities higher than 0.9 the future lifetime of NGC 6712 is less than 1010 yr.

Gnedin, Lee, & Ostriker (1999)

Vital diagram of globular clusters:

Table of the Galactic globular cluster parameters   (Table 1 of Gnedin & Ostriker 1997)

Table of the destruction times for globular clusters   (Table 3 of Gnedin & Ostriker 1997)

Dynamical effects on the Globular Cluster Luminosity Function

Gnedin (1997)   For the three best studied systems, in the Milky Way, M31, and M87, there is a statistically significant difference between the inner and outer populations of globular clusters. In all cases the inner clusters are on average brighter than the outer clusters (0.24 < Deltam0 < 0.79) and have a smaller dispersion in magnitudes (0.09 < Deltasigma < 0.61). The differences are of the type that would be expected if the inner population had been depleted by tidal shocks.

Ostriker & Gnedin (1997)   Assuming that the initial Globular Cluster Luminosity Function (GCLF) was identical in all galaxies, we can correct the observed GCLF for dynamical evolution and use it as a distance indicator. This new method gives dmM31 = 24.03 ± 0.23, dmM87 = 30.81 ± 0.17.

Formation of Globular Clusters in Hierarchical Cosmology

Kravtsov & Gnedin (2003)   We study the formation of globular clusters (GCs) in a Milky Way-size galaxy using a high-resolution cosmological simulation. The clusters in our model form in the strongly baryon-dominated cores of supergiant molecular clouds in the gaseous disks of high-redshift galaxies. The properties of clusters are estimated using a physically-motivated subgrid model of the isothermal cloud collapse. The first clusters in the simulation form at z~12, while the best conditions for GC formation appear to be at z~3-5. Most clusters form in the progenitor galaxies of the virial mass >10^9 Msun and the total mass of the cluster population is strongly correlated with the mass of its host galaxy with a fraction ~2x10^-4 of the galactic baryons being in the form of GCs. In addition, the mass of the GC population and the maximum cluster mass in a given region strongly correlate with the local average star formation rate. We find that the mass, size, and metallicity distributions of the cluster population identified in the simulation are remarkably similar to the corresponding distributions of the Milky Way globulars. We find no clear mass-metallicity or age-metallicity correlations for the old clusters. The zero-age cluster mass function can be approximated by a power-law, dN/dM~M^-alpha, with alpha~2, in agreement with the mass function of young stellar clusters in starbursting galaxies. However, the shape of the zero-age mass function may be better described by the high-mass tail of a lognormal distribution which peaks at \~10^3 Msun. We discuss in detail the origin, universality, and dynamical evolution of the globular cluster mass function. Our results indicate that globular clusters with properties similar to those of observed clusters can form naturally within young dense gaseous disks at z>~3 in the LCDM cosmology.

The Unique History of the Globular Cluster Omega Centauri

Gnedin et al. (2002)   Using current observational data and simple dynamical modeling, we demonstrate that Omega Cen is not special among the Galactic globular clusters in its ability to produce and retain the heavy elements dispersed in the AGB phase of stellar evolution. Multiple stellar populations, observed in Omega Cen, cannot be explained if it had formed as an isolated star cluster. The formation within a progenitor galaxy of the Milky Way is more likely, although the unique properties of Omega Cen still remain a mystery.

Table of the escape velocity and velocity dispersion from the photometric models

Table of the central velocity dispersion of the Galactic globular clusters   (Table 2 of Pryor & Meylan 1993)

Table of main-sequence lifetimes for the low-metallicity AGB stars

Data on Galactic Globular Clusters compiled by George Djorgovski
Catalogue of Milky Way Globular Cluster Parameters compiled by Bill Harris

Extragalactic Globular Cluster Systems review by Harris
N-body models of star clusters by Manybody.org
Gaseous models of star clusters
Starburst 99 model
Star Clusters Young & Old Newsletter
The Local Group by Lucio Mayer

Last modified: November 20, 2003
Oleg Gnedin