Distribution of star formation rates during the rapid assembly of NGC 1399 as deduced from its globular cluster system

¹ European Southern Observatory (ESO), Karl-Schwarzschild-Straße 2, 85748 Garching bei München, Germany
e-mail: cschulz@eso.org; mhilker@eso.org
² Helmholtz-Institut für Strahlen- und Kernphysik (HISKP), Universität Bonn, Nussallee 14–16, 53115 Bonn, Germany
e-mail: pavel@astro.uni-bonn.de;jpflamm@astro.uni-bonn.de
³ Charles University in Prague, Faculty of Mathematics and Physics, Astronomical Institute, V Holešovičkách 2, 180 00 Praha 8, Czech Republic

Received: 13 May 2016
Accepted: 23 June 2016

Abstract

Ultra-compact dwarf galaxies (UCDs) share many properties with globular clusters (GCs) and are found in similar environments. Here, a large sample of UCDs and GCs around NGC 1399, the central giant elliptical of the Fornax galaxy cluster, is used to infer their formation history and also to shed light on the formation of NGC 1399 itself. We assumed that all GCs and UCDs in our sample are the result of star cluster (SC) formation processes and used them as tracers of past star formation activities. After correcting our GC/UCD sample for mass loss, we interpreted their overall mass function to be a superposition of SC populations that formed coevally during different formation epochs. The SC masses of each population were distributed according to the embedded cluster mass function (ECMF), a pure power law with the slope − β. Each ECMF was characterized by a stellar upper mass limit, M_max, which depended on the star formation rate (SFR). We decomposed the observed GC/UCD mass function into individual SC populations and converted M_max of each SC population to an SFR. The overall distribution of SFRs reveals under which conditions the GC/UCD sample around NGC 1399 formed. Considering the constraints set by the age of the GCs/UCDs and the present stellar mass of NGC 1399, we found that the formation of the GCs/UCDs can be well explained within our framework with values for β below 2.3. This finding agrees very well with the observation of young SCs where β ≈ 2.0 is usually found. Even though we took into account that some of the most massive objects might not be genuine SCs and applied different corrections for the mass loss, we found that these considerations do not influence much the outcome. We derived the peak SFRs to be between approximately 300 and 3000 M_⊙ yr^-1, which matches the SFRs observed in massive high-redshift sub-millimeter galaxies and an SFR estimate inferred from NGC 1399 based on the so-called downsizing picture, meaning that more massive galaxies must have formed over shorter periods of time. Our findings give rise to the interpretation that NGC 1399 and its GC/UCD system formed in a relatively short, intense starburst early on.