Dynamical Modeling of Dense Star Clusters with a Parallel Monte Carlo Code

Project: Research project

Project Details


We will investigate the dynamical evolution of dense star clusters, such as globular clusters in the Milky Way and other galaxies, using numerical simulations on parallel supercomputers. Our new parallel Monte Carlo code allows us to model clusters with realistic numbers of stars (N ~ 10^5 - 10^6), both single and in binaries, and with accurate treatments of all dynamical and stellar evolution processes. Even a small initial binary fraction (e.g., 10% of stars in binaries) can play a key role in supporting a cluster against gravothermal collapse for many relaxation times. Inelastic encounters between binaries and single stars or other binaries provide a very significant energy source for the cluster. These dynamical interactions also lead to the production of large numbers of important sources containing, e.g., neutron stars and black holes in bright X-ray binaries, millisecond radio pulsars, and double white dwarfs in tight orbits.

Our new parallel code is based on Henon's Monte Carlo algorithm for computing the dynamical evolution of dense stellar systems in the Fokker-Planck approximation. This new code allows us to calculate accurately the entire evolution of even the largest globular clusters (e.g., 47 Tuc) in typically just a few hours of computing time, allowing us to explore and understand the complex dynamical and stellar processes at play over the full range of relevant physical parameters. The results of this investigation will provide a solid theoretical framework for the interpretation of data from many current and future NASA missions such as HST, Chandra, Fermi, Spitzer, and JWST. Such a framework is currently lacking because of the severe computational limits of other methods like direct N-body integrations.
Effective start/end date9/1/148/31/18


  • NASA Goddard Space Flight Center (NNX14AP92G/000005)

Fingerprint Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.