Available formats: HTML | PDF | (gzipped) PostScript
Young star clusters in the Antennae: a clue to their nature from evolutionary synthesis
Uta Fritze-v. Alvensleben
Received 12 November 1996 / Accepted 13 May 1998
We analyse the population of bright star clusters in the interacting galaxy pair NGC 4038/39 detected with HST WFPC1 by Whitmore & Schweizer (1995). Making use of our spectrophotometric evolutionary synthesis models for various initial metallicities we derive the ages of these star clusters and calculate their future luminosity evolution. This allows us to compare their luminosity function (LF ), evolved over a Hubble time, to globular cluster (GC ) LFs. Since effective radii are difficult to determine due to crowding of the clusters, the shape of the LF after a Hubble time may help decide whether the young clusters are young GCs or rather open clusters/OB associations. We subdivide the cluster population into subsamples with small and large effective radii. While the LF for the extended clusters looks exponential, that for clusters with small effective radii clearly shows a turn-over brighter than the completeness limit. For other possible subdivisions as to luminosity or colour no comparable differences are found. Evolving, in a first step, the LF from a common mean age of the young clusters of 0.2 Gyr to an assumed age of 12 Gyr, the LF for the subsample of clusters with small effective radii seems compatible with a Gaussian GCLF with typical parameters mag and mag, except for some overpopulation of the faint bins. These faintest bins, however, might be suspected to be subject to the strongest depopulation through effects of dynamical evolution not included in our models. We also follow the colour evolution of the young star clusters over a Hubble time and compare to observations on the Milky Way and other galaxies' GC systems.
For an ongoing starburst like the one in the NGC 4038/39 system age spread effects among the young star cluster population may not be negligible. In a second step, we therefore account for age spread effects by individually age-dating every cluster on the basis of its observed colours. This drastically changes the time evolution of the LF, confirming Meurer's (1995) conjecture. We find that - if age spread effects are properly accounted for - the LF of the entire young star cluster population, and in particular that of the brighter subsample, is in good agreement with an average Gaussian GCLF after a Hubble time. It shows a turn-over at mag with mag.
The age distribution reveals that the brightest GCs from the interacting galaxies' original population are also observed. They make up the bulk of the red subpopulation with (V-I)0 0.95. Their effective radii do not significantly differ from those of the young star cluster population, neither on average nor in their distribution.
We discuss the influence of metallicity, a possibly inhomogeneous internal dust distribution, as well as of internal - through stellar mass loss - and external dynamical effects on the secular evolution of the LF.
Referring YSC luminosities to a uniform age and combining with model M/L, we recover the intrinsic mass distribution of the YSC system. Its shape is compatible with a Gaussian which - according to Vesperini's (1997) dynamical modelling for the Milky Way GC system - represents a quasi-equilibrium distribution and will not be altered over a Hubble time of dynamical evolution, allthough a substantial number of clusters will be destroyed.
We briefly compare the young star cluster population of the Antennae to the older one in the merger remnant NGC 7252 and point out that the intercomparison of young cluster populations in an age sequence of interacting and merged galaxies may become an interesting approach to study in detail the role of external dynamical effects.
Key words: galaxies: star clusters galaxies: individual: NGC 4038/39, galaxies: interactions galaxies: starburst
© European Southern Observatory (ESO) 1998
Online publication: July 7, 1998