Star-formation in NGC 4038/4039 from broad and narrow band photometry: cluster destruction?

¹ European Southern Observatory, Karl-Schwarzschild-Str. 2, 85748 Garching, Germany e-mail: smengel@eso.org
² Max-Planck-Institut für extraterrestrische Physik, Giessenbachstraße, 85748 Garching, Germany e-mail: mlehnert@mpe.mpg.de
³ University of Oxford, Dept. of Astrophysics, Denys Wilkinson Building, Keble Road, GB-Oxford OX1 3RH, UK e-mail: thatte@astro.ox.ac.uk
⁴ Also at Department of Physics, University of California at Berkeley, 366 Le Conte Hall, Berkeley, CA 94720-7300, USA e-mail: genzel@mpe.mpg.de

Received: 19 February 2005
Accepted: 18 May 2005

Abstract

Accurately determining the star formation history in NGC 4038/4039 – “The Antennae” – is hampered by variable and sometimes substantial extinction. We therefore used near infrared broad- and narrow-band images obtained with ISAAC at the VLT and with SOFI at the NTT to determine the recent star formation history in this prototypical merger. In combination with archival HST data, we determined ages, extinction, and other parameters for single star clusters, and properties of the cluster population as a whole.
About 70% of the K_s-band detected star clusters with masses ≥10⁵  are younger than 10 Myr (this is approximately an e-folding time for cluster ages), which we interpret as evidence of rapid dissolution but not free expansion.
The total mass of K-band selected clusters is about 5 to  and represents about 3-6% of the total molecular gas. However, this takes only the detected clusters into account, and in view of the rapid dissolution, means that this is only a lower limit to the total mass of stars produced in clusters during the burst. Studies of cluster formation in other galaxies recently suggested short cluster dissolution timescales, too, which means that star formation rates may have been severely underestimated in the past.
Extinction is strongly variable and very high in some regions, but around  mag on average. Even though most clusters are detected at least in I-band, only the information about individual cluster ages and extinction allows us to avoid uncertainties of orders of magnitude in star formation rate estimates determined from optical fluxes. From the distribution of individual cluster extinction vs. age, which is significantly higher for clusters below 8-9 Myr than for older clusters, we infer that this is the time by which a typical cluster blows free of its native dust cocoon.