Michael A. Beasley

Globular clusters in Virgo ellipticals: Unexpected results for giants and dwarfs from Advanced Camera for Surveys imaging

We have analyzed archival Hubble Space Telescope Advanced Camera for Surveys images in g and z of the globular cluster (GC) systems of 53 ellipticals in the Virgo Cluster, spanning massive galaxies to dwarf ellipticals (dEs). Several new results emerged. (1) In the giant ellipticals (gEs) M87 and NGC 4649, there is a correlation between luminosity and color for individual blue (metal-poor) GCs, such that more massive GCs are more red (metal-rich). A plausible interpretation of this result is self-enrichment, and a speculative suggestion is that these GCs once possessed dark matter halos. (2) The dispersion in color is nearly twice as large for the metal-rich GCs as for the metal-poor GCs. However, there is evidence for a nonlinear relation between g - z and metallicity, and the dispersion in metallicity may be the same for both subpopulations. (3) Very luminous, intermediate-color GCs are common in gEs. These objects may be remnants of many stripped dwarfs, analogs of ω Cen in the Galaxy. (4) There is a continuity of GC system colors from gEs to some dEs; in particular, many dEs have metal-rich GC subpopulations. We also confirm the GC color-galaxy luminosity relations found previously for both metal-poor and metal-rich GC subpopulations. (5) There are large differences in GC specific frequency among dEs, independent of the presence of a nucleus and the fraction of metal-rich GCs. Over -15 < MB < -18 we find little correlation between specific frequency and MB. But we do find evidence for two separate SN classes of dEs: those with B-band SN ~ 2 and those with populous GC systems that have SN ranging from ~5 to 20 with median SN ~ 10. Together, these points suggest multiple formation channels for dEs in the Virgo Cluster. (6) The peak of the GC luminosity function (GCLF) is the same for both gEs and dEs. This is contrary to expectations of dynamical friction on massive GCs, unless the primordial GCLF varies between gEs and dEs. Among gEs the GCLF turnover varies by a surprisingly small 0.05 mag, an encouraging result for its use as an accurate standard candle.

A robust method for the analysis of integrated spectra from globular clusters using Lick indices

ABSTRACT We define a method for analysis of the integrated spectra of extra-galactic globularclusters that provides more reliable measures of age, metallicity and α-element abun-dance ratio than have so far been achieved. The method involves the simultaneousfitting of up to 25 Lick indices in a χ 2 -fitting technique that maximises the use of theavailable data. Here we compare three sets of single stellar population (SSP) modelsof Lick indices to the high signal-to-noise, integrated spectra of 20 Galactic globu-lar clusters. The ages, [Fe/H] and α-element abundance ratios derived from the SSPmodels are compared to the results of resolved stellar population studies from theliterature. We find good consistency with the published values, with an agreement ofbetter than 0.1 dex in all three derived parameters. The technique allows identificationof abundance ratio anomalies, such as the known nitrogen over-abundance in Galac-tic globular clusters, and the presence of anomalous horizontal-branch morphologies.It also minimises the impact on the derived parameters of imperfect calibration tothe Lick system, and reduction errors in general. The method defined in this work istherefore robust with respect to many of the difficulties that plague the application ofSSP models in general, and is, consequently, well suited to the study of extra-galacticglobular cluster systems.Key words: stars: abundances – globular clusters: general – methods: analytical

Extragalactic Globular Clusters: Old Spectroscopic Ages and New Views on Their Formation

We present the results of a meta-analysis of Keck spectra of extragalactic globular clusters (GCs) in a sample of eight galaxies, ranging from dwarf galaxies to massive elliptical galaxies. We infer ages for the metal-poor and metal-rich GCs in these galaxies through comparisons to Galactic GCs. Both subpopulations appear to be no younger than their Galactic counterparts, with ages 10 Gyr. This is the largest sample of galaxies for which ages have been constrained spectroscopically. Our results support the formation of most GCs in massive galaxies at high redshift. We propose a scenario for the formation of GC subpopulations that synthesizes aspects of both accretion and in situ approaches in the context of galaxy formation through hierarchical merging.

The Chemical Properties of Milky Way and M31 Globular Clusters. I. A Comparative Study

A comparative analysis is performed between high-quality integrated spectral indices of 30 M31 globular clusters, 20 Milky Way globular clusters, and a sample of field and cluster elliptical galaxies. We find that the Lick CN indices in the M31 and Galactic clusters are enhanced relative to the bulges of the Milky Way, M31, and elliptical spheroids, in agreement with Burstein and coworkers. Although not particularly evident in the Lick CN indices, the near-UV cyanogen feature (?3883) is strongly enhanced with respect to the Galactic globular clusters at metallicities -1.5 -0.8, we observe no systematic differences in the H?, H?, or H? indices between the M31 and Galactic globular clusters, in contrast to previous studies. The elliptical galaxy sample lies offset from the loci of the globular clusters in both the cyanogen-[MgFe] and Balmer-line?[MgFe] planes. Six of the M31 clusters appear young and are projected onto the M31 disk. Population synthesis models suggest that these are metal-rich clusters with ages 100?800 Myr, metallicities -0.20 ? [Fe/H] ? 0.35, and masses 0.7? ~ 7.0 ? 104 M?. Two other young clusters are Hubble V in NGC?205, observed as a template, and an older (~3 Gyr) cluster some 7 kpc away from the plane of the disk. The six clusters projected onto the disk show signs of rotation similar to the H I gas in M31, and three clusters exhibit thin disk kinematics, according to Morrison and coworkers. Dynamical mass estimates and detailed structural parameters are required for these objects to determine whether they are massive open clusters or globular clusters. If they are the latter, our findings suggest globular clusters may trace the buildup of galaxy disks. In either case, we conclude that these clusters are part of a young, metal-rich disk cluster system in M31, possibly as young as 1 Gyr old.

Kinematics of globular cluster systems and the formation of early-type galaxies

We numerically investigate the kinematic properties of globular cluster systems (GCSs) in E/S0 galaxies formed from dissipationless merging of spiral galaxies. The metal-poor globular clusters (MPCs) and metal-rich clusters (MRCs) in the merger progenitors are initially assumed to have spatial distributions consistent with the Milky Way GCS. Our principal results, which can be tested against observations, are as follows. Both MPCs and MRCs in elliptical galaxies formed from major mergers can exhibit significant rotation at large radii (∼20 kpc) due to the conversion of initial orbital angular momentum into intrinsic angular momentum of the remnant. MPCs show higher central velocity dispersions than MRCs for most major merger models. V m /σ 0 (where V m and σ 0 are the GCS maximum rotational velocity and central velocity dispersion, respectively) ranges from 0.2 to 1.0 and from 0.1 to 0.9 for the MPCs and MRCs, respectively, within 6R e for the remnant elliptical. For most merger remnant ellipticals, V m /σ 0 of GCSs within 6R e is greater than that of the field stars within 2R e . The radial profiles of rotational velocities and velocity dispersions of the GCSs depend upon the orbital configuration of the merger progenitors, their mass ratios and the viewing angle. For example, more flattened early-type galaxies, formed through mergers with small mass ratios (∼0.1), show little rotation in the outer MRCs. Two-dimensional (2D) velocity dispersion distributions of the GCSs of merger remnant ellipticals are generally flattened for both MPCs and MRCs, reflecting the fact that the GCSs have anisotropic velocity dispersions. The 2D distributions of the line-of-sight velocity of the GCSs in some remnant ellipticals show minor-axis rotation, particularly for MRCs. The kinematic properties of MPCs in merger remnant ellipticals strongly resemble those of the surrounding dark matter. This implies that the kinematics of MPCs in such galaxies can be used to probe the kinematic properties of their dark matter haloes. We discuss these results in the context of GC and galaxy formation. We note a possible difference in the GC kinematics between field and cluster Es and explain how GC kinematics may help us understand the origin of SO galaxies.

Gemini/GMOS spectra of globular clusters in the Leo group elliptical NGC 3379

The Leo group elliptical NGC 3379 is one of the few normal elliptical galaxies close enough to make possible observations of resolved stellar populations, deep globular cluster (GC) photometry and high signal-to-noise ratio GC spectra. We have obtained Gemini/GMOS spectra for 22 GCs associated with NGC 3379. We derive ages, metallicities and α-element abundance ratios from simple stellar population models using the recent multi-index χ 2 minimization method of Proctor & Sansom. All of these GCs are found to be consistent with old ages, i.e. 10 Gyr, with a wide range of metallicities. This is comparable to the ages and metallicities that Gregg et al. found a couple of years ago for resolved stellar populations in the outer regions of this elliptical. A trend of decreasing α-element abundance ratio with increasing metallicity is indicated. The projected velocity dispersion of the GC system is consistent with being constant with radius. Non-parametric, isotropic models require a significant increase in the mass-to-light ratio at large radii. This result is in contrast to that of Romanowsky et al., who recently found a decrease in the velocity dispersion profile as determined from planetary nebulae (PN). Our constant dispersion requires a normal-sized dark halo, although without anisotropic models we cannot rigorously determine the dark halo mass. A two-sided χ 2 test over all radii gives a 2σ difference between the mass profile derived from our GCs compared to the PN-derived mass model of Romanowsky et al. However, if we restrict our analysis to radii beyond one effective radius and test if the GC velocity dispersion is consistently higher, we determine a > 3σ difference between the mass models, and hence we favour the conclusion that NGC 3379 does indeed have dark matter at large radii in its halo.

Is the initial mass function of low surface brightness galaxies dominated by low-mass stars?

The rotation curves of low surface brightness (LSB) galaxies suggest that they possess significantly higher mass-to-light (M/L) ratios than their high surface brightness counterparts, indicating that LSB galaxies may be dark matter dominated. This interpretation is hampered by the difficulty of disentangling the disc and dark halo contributions from the disc dynamics of LSB galaxies. Recently, Fuchs has attempted such a disentanglement using spiral arm density wave and swing amplification theory, allowing an independent measurement of the disc mass; this work suggests that LSB discs are significantly more massive than previously believed. This would considerably reduce the amount of matter required in the dark haloes in fitting the rotation curves. Interestingly, the high mass-to-light ratios derived for the discs appear inconsistent with standard stellar population synthesis models. In this paper, we investigate whether the high M/L ratios for the Fuchs LSB discs might be understood by adopting a very ‘bottom heavy’ initial mass function (IMF). We find that an IMF with a power-law exponent of around α = 3.85 (compared with the standard Salpeter IMF, α = 2.35) is sufficient to explain the unusually high M/L ratios of the Fuchs sample. Within the context of the models, the blue colours [(B − R)0 < 1.0] of the sample galaxies result from being metal-poor ([Fe/H] =− 1.5 ∼− 1.0) and having undergone recent (∼1‐3 Gyr ago) star formation. Ke yw ords: stars: luminosity function, mass function ‐ galaxies: fundamental parameters ‐ galaxies: stellar content.

Old Globular Clusters Masquerading as Young in NGC 4365

High signal-to-noise ratio, low-resolution spectra have been obtained for 22 globular clusters (GCs) in NGC 4365. Some of these were selected as probable representatives of an intermediate-age (2–5 Gyr), extremely metal-rich GC subpopulation. The presence of such a subpopulation had been inferred from the unusual optical and near-infrared color distributions of GCs in this otherwise typical Virgo elliptical galaxy. However, ages derived from Lick indices are consistent with uniformly old mean ages for all GCs in our sample. The metallicities of the clusters show some evidence of a trimodal distribution. The most metal-poor and metal-rich peaks are consistent with the values expected for an elliptical galaxy of this luminosity, but there appears to be an additional, intermediate-metallicity peak lying between them. New Hubble Space Telescope photometry is consistent with this result. A plausible scenario is that in earlier data these three peaks merged into a single broad distribution. Our results suggest that it is difficult to identify intermediate-age GC subpopulations solely with photometry, even when both optical and near-infrared colors are used.

The Chemical Properties of Milky Way and M31 Globular Clusters. II. Stellar Population Model Predictions

We derive ages, metallicities, and abundance ratios ([α/Fe]) from the integrated spectra of 23 globular clusters in M31 by employing multivariate fits to two different stellar population models. We also perform a parallel analysis on 21 Galactic globular clusters as a consistency check and in order to facilitate a differential analysis. Our analysis shows that the M31 globular clusters separate into three distinct components in age and metallicity; we identify an old, metal-poor group (seven clusters), an old, metal-rich group (10 clusters), and an intermediate-age (3–6 Gyr), intermediate-metallicity ([Z/H] ~ -1) group (six clusters). This third group is not identified in the Galactic globular cluster sample. We also see evidence that the old, metal-rich Galactic globular clusters are 1–2 Gyr older than their counterparts in M31. The majority of globular clusters in both samples appear to be enhanced in α-elements, but the degree of enhancement is rather model-dependent. The intermediate-age globular clusters appear to be the most enhanced, with [α/Fe] ~ 0.4. These clusters are clearly depressed in CN with respect to the models and the bulk of the M31 and Milky Way sample. Compared with the bulge of M31, M32, and NGC 205, these clusters most resemble the stellar populations in NGC 205 in terms of age, metallicity, and CN abundance. We infer horizontal branch morphologies for the M31 clusters using the Rose Ca II index and demonstrate that blue horizontal branches are not leading to erroneous age estimates in our analysis. We discuss and reject as unlikely the hypothesis that these objects are in fact foreground stars contaminating the optical catalogs. The intermediate-age clusters have generally higher velocities than the bulk of the M31 cluster population. Spatially, three of these clusters are projected onto the bulge region, and the remaining three are distributed at large radii. We discuss these objects within the context of the build-up of the M31 halo and suggest that these clusters possibly originated in a gas-rich dwarf galaxy, which may or may not be presently observable in M31.

Stellar Populations of Globular Clusters in the Elliptical Galaxy NGC 1407

We present high-quality Keck spectroscopic data for a sample of 20 globular clusters (GCs) in the massive E0 galaxy NGC 1407. A subset of 20 line-strength indices of the Lick/IDS system has been measured for both the GC system and the central integrated starlight of the galaxy. Ages, metallicities, and [α/Fe] ratios have been derived using several different approaches. The majority of GCs in NGC 1407 studied are old, follow a tight metallicity sequence reaching values slightly above solar, and exhibit mean [α/Fe] ratios of ~0.3 dex. In addition, three GCs are formally derived to be young (~4 Gyr), but we argue that they are actually old GCs hosting blue horizontal branches. We report, for the first time, evidence for the existence of two chemically distinct subpopulations of metal-rich (MR) GCs. We find that some MR GCs exhibit significantly larger [Mg/Fe] and [C/Fe] ratios. Different star formation timescales are proposed to explain the correlation between Mg and C abundances. We also find striking CN overabundances over the entire GC metallicity range. It is interesting to note that the behavior of C and N in metal-poor GCs clearly deviates from that in MR GCs. In particular, for MR GCs, N increases dramatically while C essentially saturates. This may be interpreted as a consequence of the increasing importance of the CNO cycle with increasing metallicity.