Further Definition of the Mass-Metallicity Relation in Globular Cluster Systems Around Brightest Cluster Galaxies

TL;DR

This study analyzes the mass-metallicity relation in globular cluster systems of fifteen Brightest Cluster Galaxies, revealing a consistent scaling for metal-poor clusters and no trend for metal-rich ones, supporting self-enrichment formation models.

Contribution

It extends previous work by combining data from multiple galaxies to refine the mass-metallicity relation, especially for metal-poor clusters, and validates the relation's independence from host galaxy properties.

Findings

Metal-poor clusters follow Z ~ M^{0.30±0.05} scaling.

No significant trend for metal-rich clusters.

Results support self-enrichment models of globular cluster formation.

Abstract

We combine the globular cluster data for fifteen Brightest Cluster Galaxies and use this material to trace the mass-metallicity relations (MMR) in their globular cluster systems (GCSs). This work extends previous studies which correlate the properties of the MMR with those of the host galaxy. Our combined data sets show a mean trend for the metal-poor (MP) subpopulation which corresponds to a scaling of heavy-element abundance with cluster mass Z ~ M^(0.30+/-0.05). No trend is seen for the metal-rich (MR) subpopulation which has a scaling relation that is consistent with zero. We also find that the scaling exponent is independent of the GCS specific frequency and host galaxy luminosity, except perhaps for dwarf galaxies. We present new photometry in (g',i') obtained with Gemini/GMOS for the globular cluster populations around the southern giant ellipticals NGC 5193 and IC 4329. Both galaxies have rich cluster populations which show up as normal, bimodal sequences in the colour-magnitude diagram. We test the observed MMRs and argue that they are statistically real, and not an artifact caused by the method we used. We also argue against asymmetric contamination causing the observed MMR as our mean results are no different from other contamination-free studies. Finally, we compare our method to the standard bimodal fitting method (KMM or RMIX) and find our results are consistent. Interpretation of these results is consistent with recent models for globular cluster formation in which the MMR is determined by GC self-enrichment during their brief formation period.