CaltechTHESIS
  A Caltech Library Service

The Role of Stellar Feedback in Star Cluster Formation

Citation

Grudić, Michael Yvan (2019) The Role of Stellar Feedback in Star Cluster Formation. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/7FBG-TE87. https://resolver.caltech.edu/CaltechTHESIS:06072019-111646558

Abstract

A methodology for numerical magnetohydrodynamics simulations of star cluster formation, accounting for all mechanisms of stellar feedback from massive stars, is developed and used to address a range of problems regarding the formation of stars and star clusters in giant molecular clouds (GMCs). These studies culminate in a new theoretical framework that connects properties of GMCs to those of the star clusters that form in them.

The simulation methodology is established and tested, and the problem of the star formation efficiency (SFE) of molecular clouds is addressed. It is found that SFE is set by the balance of feedback and gravity, with very weak dependence upon other factors. A simple dimensional scaling law with cloud surface density emerges from the complex interplay of different feedback physics. Parameter space is found where feedback must fail, and the SFE is high, and the implications of this prediction are explored.

The star clusters formed in the simulations are found to resemble observed young, massive star clusters in the form of their surface brightness profiles, leading to the hypothesis that this structure is a result of the star formation process. It is shown that the shallow, power-law density profiles characteristic of young clusters is predicted by the hierarchical star formation scenario. It is shown that the SFE law, when coupled to an analytic cloud collapse model, predicts that gas should be exhausted by highly-efficient star formation at a stellar surface density of ∼ 105 − 106 Msun pc-2, consistent with the maximum observed.

A new suite of simulations is developed to specifically model GMCs in the Milky Way. It is found that the picture of feedback-disrupted star formation is able to account for both the normalization and the scatter in the measured SFE of GMCs in the Milky Way, the first theory to do so.

The uncertainty in the simulated SFE due to the choice of feedback prescription is quantified, by running a controlled methods study of several different prescriptions in the literature. In the cloud model simulated, the choice of prescription affects the simulated SFE at the factor of ∼ 3 level, explaining discrepancies in the literature and identifying the small-scale details of massive star formation as the main uncertainty in cluster formation simulations.

Finally, the simulation suite is extended to model massive GMCs in local spiral galaxies, and to simulate 10 random realizations at each point in parameter space, mapping out the stochastic nature of star cluster formation in GMCs. A model is calibrated to the simulation results, taking the cloud bulk properties as input parameters, and predicting the detailed properties of the star clusters formed in it. A star cluster catalogue is synthesized from observed GMCs in M83, and good agreement is found with observed star cluster properties, including the fraction of stars in bound clusters, the maximum cluster mass, and the distribution of cluster sizes.

Item Type:Thesis (Dissertation (Ph.D.))
Subject Keywords:Star formation; stellar feedback; magnetohydrodynamics; numerical methods; Star clusters; Galaxy formation; astrophysics
Degree Grantor:California Institute of Technology
Division:Physics, Mathematics and Astronomy
Major Option:Physics
Awards:Robert F. Christy Prize for an Outstanding Doctoral Thesis in Theoretical Physics, 2019. James A. Cullen Memorial Fellowship Fund, 2018.
Thesis Availability:Public (worldwide access)
Research Advisor(s):
  • Hopkins, Philip F.
Group:Astronomy Department
Thesis Committee:
  • Kirby, Evan N. (chair)
  • Hillenbrand, Lynne A.
  • Murray, Norman
  • Hopkins, Philip F.
Defense Date:22 May 2019
Record Number:CaltechTHESIS:06072019-111646558
Persistent URL:https://resolver.caltech.edu/CaltechTHESIS:06072019-111646558
DOI:10.7907/7FBG-TE87
Related URLs:
URLURL TypeDescription
https://doi.org/10.1093/mnras/sty035DOIArticle in Chapter 2
https://doi.org/10.1093/mnras/sty2303DOIArticle in Chapter 3
https://doi.org/10.1093/mnras/sty3386DOIArticle in Chapter 4
ORCID:
AuthorORCID
Grudić, Michael Yvan0000-0002-1655-5604
Default Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:11708
Collection:CaltechTHESIS
Deposited By: Michael Grudic
Deposited On:10 Jun 2019 22:26
Last Modified:10 Mar 2020 20:20

Thesis Files

[img]
Preview
PDF - Final Version
See Usage Policy.

10MB

Repository Staff Only: item control page