1. Introduction The formation and destruction of star cluster populations in galaxy mergers Slide 1 of 22 Star clusters in galaxy mergers – 7 April 2011 Diederik Kruijssen Astronomical Institute Utrecht; Leiden Observatory Summer 2011: MPA Garching Collaborators and advisors – Henny Lamers (Utrecht), Inti Pelupessy (Leiden), Nate Bastian (Excellence Cluster Munich), Simon Portegies Zwart (Leiden), Vincent Icke (Leiden) Thomas Maschberger (Grenoble), Cathie Clarke (Cambridge), Nick Moeckel (Cambridge), Ian Bonnell (St. Andrews) Models Mergers Antennae Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory

2. Introduction Slide 2 of 22 Star clusters in galaxy mergers – 7 April 2011 Models Mergers Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Galaxy mergers  Crucial in hierarchical cosmology (White & Rees 1978)  Seen to occur at all redshifts  Also affect the Local Group Antennae McConnachie et al. (2009)

3. Introduction Slide 3 of 22 Star clusters in galaxy mergers – 7 April 2011 Models Mergers Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters  Are the cradle of at least some stars (Elmegreen, Lada/Lada, Bastian, Gieles, Kroupa, Bressert)  “Can be used to trace galaxy formation/evolution” (Schweizer, Ashman/Zepf, Larsen)  IF we understand the multi-scale physics Antennae Fall & Zhang 2001 log Mass log Number

4. Introduction Slide 4 of 22 Star clusters in galaxy mergers – 7 April 2011 Models Mergers Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory To trace galaxy evolution with clusters…  We need to understand the impact of the galactic environment on cluster populations  Observations reveal many clusters in mergers (Holtzman, Whitmore)  Born in the starbursts during merger process (Barnes, Hernquist, Mihos)  Are all young, some very massive (> 10 7 M  ) (Schweizer, Bastian)  Possibly young globular clusters? (Ashman & Zepf 1992) Antennae

5. Introduction Slide 5 of 22 Star clusters in galaxy mergers – 7 April 2011 Models Mergers Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in changing environments  Star clusters do not live forever  Disrupted by steady tidal field and tidal shocks (Spitzer, Gieles, Gnedin/Ostriker, Heggie, Baumgardt, Portegies Zwart)  High gas densities in mergers  tidal shocks  Mergers should also show enhanced disruption  Does cluster formation or disruption dominate? Strength of cluster disruption and formation processes vary in time and space, and depend on the galactic environment Antennae

6. Slide 6 of 22 Mergers Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Introduction Models Use numerical simulations to model co-evolution of clusters and galaxies Antennae

7. Models: galaxy simulation code Stars  We use the N-body/SPH treecode Stars (Pelupessy et al. 2004)  Contains:  Self-gravity  Star formation  Several forms of feedback  Multiphase interstellar medium (explicitly computed)  Cooling curves & cosmic ray heating  Photometry  Radiative transfer & extinction Slide 7 of 22 Mergers Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Introduction Models Pelupessy et al. 2004  Stars  Gas  Dark matter halo Antennae

8. Slide 8 of 22 Introduction Models: star cluster formation and evolution  Sub-grid, simple cluster formation within star particle  Assumes fraction of star formation in clusters Models Mergers Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Antennae

9. Slide 9 of 22 Introduction Models: star cluster evolution code SPACE  We use the semi-analytic star cluster model SPACE (Kruijssen & Lamers 2008; Kruijssen 2009)  Contains:  Stellar evolution (Padova/Marigo et al. 2008)  Stellar remnants  Disruption by two-body relaxation.  Disruption by tidal shocks.  Evolution of the stellar mass function.  Model described and validated in Kruijssen et al. 2011 (MNRAS in press) ArXiV:1102.1013 Models Mergers Related to the tidal field Consistent with N-body sims Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Antennae

10. Slide 10 of 22 Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Introduction Does cluster formation or destruction dominate in galaxy mergers? Models Mergers Antennae Kruijssen et al. in prep.

11. Slide 11 of 22 Introduction Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Does cluster formation or destruction dominate in galaxy mergers? Models Mergers t (Gyr) SFR Tidal heating N Antennae

12. Slide 12 of 22 Introduction Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Evolution of the mass function of surviving clusters  Low-mass clusters are more efficiently disrupted than massive clusters  Mass function develops a characteristic mass  Young massive clusters in nearby mergers could be young GCs  Can be generalised to ‘violent birth’ (Elmegreen 2010) Models Mergers Log(M) dN/dlogM Antennae

13. Slide 13 of 22 Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Introduction How is the cluster age distribution affected by a changing environment? Models Mergers Antennae Observed slope of -1 over the entire age range, independently of mass (Whitmore et al. 2007)

14. Log(Age) Log(dN/dt) Slide 14 of 22 Introduction Age distribution for a constant disruption rate Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Models Mergers Lifetime of lowest-mass cluster Boutloukos & Lamers (2003) Lamers et al. (2005) Slope is zero Slope is steeper than -1 Antennae

15. Age distribution for a changing disruption rate  Motion of clusters with respect to primordial region, i.e. “Cluster migration”  Disruption rate decreases with age (Elmegreen & Hunter 2010, Kruijssen et al. 2011) Log(Age) Log(dN/dt) Slide 15 of 22 Introduction Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Log(Age/yr) = 6Log(Age/yr) = 7Log(Age/yr) = 7.5 Cluster migration Models Mergers Antennae

16. Age distribution for a changing disruption rate  Cluster population with a range of disruption rates is subject to “Natural selection”  Disruption rate decreases with age (Elmegreen & Hunter 2010, Kruijssen et al. 2011) Slide 16 of 22 Introduction Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Log(Age) Log(dN/dt) Log(Age/yr) = 6Log(Age/yr) = 7Log(Age/yr) = 7.5 Natural selection Models Mergers Antennae

17. Slide 17 of 22 Introduction Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Changing environment with high density contrast in Antennae galaxies Models Mergers Antennae

18. Slide 18 of 22 Introduction Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Simulating the Antennae galaxies  …is interesting for two reasons  We can test the validity of the model  We can possibly explain observed cluster age distribution  Use the initial conditions from Karl et al. 2010 (Karl et al. 2010) Models Mergers Antennae

19. Slide 19 of 22 Introduction Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Morphological comparison Models Mergers Antennae Kruijssen & Bastian in prep.

20. Slide 20 of 22 Introduction Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Age distributions Models Mergers Log(Age) dN/dt Log(Ambient gas density) Log(Age) Infant mortality??? Antennae Kruijssen & Bastian in prep.

21. Slide 21 of 22 Introduction Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Virial ratios in star formation simulations (Bonnell et al. 2003; 2008) Models Mergers x (pc) y (pc) N Virial ratio Infant mortality  “Cruel cradle effect”? Antennae Kruijssen et al. 2011 to be submitted to MNRAS

22. Slide 22 of 22 Introduction Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Conclusions  Dense starburst environments may yield a net destruction of star clusters  Disruption rate decreases with age: cluster migration + natural selection  These processes explain the age distribution of clusters in the Antennae  Accounting for the variation of the disruption rate in time and space enables the tracing of galaxy evolution with star clusters  Instead of ‘infant mortality’, young clusters may be disrupted by the ‘cruel cradle effect’ Models Mergers Antennae

23. Slide 23 of 22 Introduction Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Models Mergers Antennae Models: star cluster evolution code SPACE  Disruption by shocks is derived from tidal field tensor (Prieto & Gnedin 2008, Kruijssen et al. in prep. 2010)  Shock disruption time depend on the density  Need mass-radius relation: N-body lifetimes (Gyr) Theory lifetimes (Gyr) no shocksincluding shocks

24. Slide 24 of 22 Introduction Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Models Mergers Antennae Models: star cluster evolution code SPACE  Evolution of stellar mass function from energy considerations (Kruijssen 2009)

25. Slide 25 of 22 Introduction Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Models Mergers Antennae Proof of concept: follow individual clusters in galaxy interaction

26. Slide 26 of 22 Introduction Diederik Kruijssen – Astronomical Institute Utrecht; Leiden Observatory Star clusters in galaxy mergers – 7 April 2011 Models Mergers Antennae Mergers  Other simulations while varying:  Galaxy mass ratio  Virial mass  Disc scale length  Gas fraction  Bulge fraction  Orbital geometry  Always net destruction, typical number of survivors 2–20% of pre-merger  Number of survivors decreases with increases starburst intensity