Kelsey E. Johnson

The initial mass function and massive star evolution in the Ob associations of the northern Milky-Way

We investigate the massive star content of Milky Way clusters and OB associations in order to answer three questions: (1) How coeval is star formation? (2) How constant is the initial mass function (IMF)? (3) What is the progenitor mass of Wolf-Rayet stars? Our sample includes NGC 6823/Vul OB1, NGC 6871/Cyg OB3, Berkeley 86/Cyg OB1, NGC 6983/Cyg OB1, NGC 7235, NGC 7380/Cep OB1, Cep OB5, IC 1805/Gas OB6, NGC 1893/Aug OB2, and NGC 2244/Mon OB2. Large-field CCD imaging and multiobject, fiber spectroscopy has resulted in UBV photometry for >14000 stars and new spectral types for approximate to 200 stars. These data are used to redetermine distances and reddenings for these regions and to help exclude probable nonmembers in constructing the H-R diagrams. We reanalyze comparable data previously published on Cyg OB2, Tr 14/16, and NGC 6611 and use all of these to paint a picture of star formation and to measure the IMFs. We find the following: (1) Most of the massive stars are born during a period Delta tau 7 M. A comparison with similarly studied OB associations in the Magellanic Clouds reveals no difference in IMF slope, and hence we conclude that starformation of massive stars in clusters proceeds independently of metallicity, at least between z = 0.02 and z = 0.002. The masses of the highest mass stars are approximately equal in the Milky Way, LMC, and SMC associations, contrary to the expectation that this value should vary by a factor of 3 over this metallicity range. We conclude that radiation pressure on grains must not limit the mass of the highest mass star that can form, in accord with the suggestion of Wolfire & Cassinelli that the mere existence of massive stars suggests that shocks or other mechanisms have disrupted grains in star-forming events. (3) The four Wolf-Rayet stars in our sample have come from stars more massive than 40 M.; one WC star and one late-type WN star each appear to have come from very massive (approximate to 100 M.) progenitors.

An actively accreting massive black hole in the dwarf starburst galaxy Henize 2-10

Supermassive black holes are now thought to lie at the heart of every giant galaxy with a spheroidal component, including our own Milky Way. The birth and growth of the first ‘seed’ black holes in the earlier Universe, however, is observationally unconstrained and we are only beginning to piece together a scenario for their subsequent evolution. Here we report that the nearby dwarf starburst galaxy Henize 2-10 (refs 5 and 6) contains a compact radio source at the dynamical centre of the galaxy that is spatially coincident with a hard X-ray source. From these observations, we conclude that Henize 2-10 harbours an actively accreting central black hole with a mass of approximately one million solar masses. This nearby dwarf galaxy, simultaneously hosting a massive black hole and an extreme burst of star formation, is analogous in many ways to galaxies in the infant Universe during the early stages of black-hole growth and galaxy mass assembly. Our results confirm that nearby star-forming dwarf galaxies can indeed form massive black holes, and that by implication so can their primordial counterparts. Moreover, the lack of a substantial spheroidal component in Henize 2-10 indicates that supermassive black-hole growth may precede the build-up of galaxy spheroids.

THE IMPORTANCE OF NEBULAR CONTINUUM AND LINE EMISSION IN OBSERVATIONS OF YOUNG MASSIVE STAR CLUSTERS

In this spectroscopic study of infant massive star clusters, we find that continuum emission from ionized gas rivals the stellar luminosity at optical wavelengths. In addition, we find that nebular line emission is significant in many commonly used broadband Hubble Space Telescope (HST) filters including the F814W I-band, the F555W V-band, and the F435W B-band. Two young massive clusters (YMCs) in the nearby starburst galaxy NGC 4449 were targeted for follow-up spectroscopic observations after Reines et al. discovered an F814W I-band excess in their photometric study of radio-detected clusters in the galaxy. The spectra were obtained with the Dual Imaging Spectrograph (DIS) on the 3.5 m Apache Point Observatory (APO) telescope and have a spectral range of ~3800-9800 A. We supplement these data with HST and Sloan Digital Sky Survey photometry of the clusters. By comparing our data to the Starburst99 and GALEV evolutionary synthesis models, we find that nebular continuum emission competes with the stellar light in our observations and that the relative contribution from the nebular continuum is largest in the U- and I-bands, where the Balmer (3646 A) and Paschen jumps (8207 A) are located. The spectra also exhibit strong line emission including the [S III] λλ9069, 9532 lines in the HST F814W I-band. We find that the combination of nebular continuum and line emission can account for the F814W I-band excess previously found by Reines et al. In an effort to provide a benchmark for estimating the impact of ionized gas emission on photometric observations of young massive stellar populations, we compute the relative contributions of the stellar continuum, nebular continuum, and emission lines to the total observed flux of a 3 Myr old cluster through various HST filter/instrument combinations, including filters in the Wide Field Camera 3. We urge caution when comparing observations of YMCs to evolutionary synthesis models since nebular continuum and line emission can have a large impact on magnitudes and colors of young (5 Myr) clusters, significantly affecting inferred properties such as ages, masses and extinctions.

MID-INFRARED EVIDENCE FOR ACCELERATED EVOLUTION IN COMPACT GROUP GALAXIES

Compact galaxy groups are at the extremes of the group environment, with high number densities and low velocity dispersions that likely affect member galaxy evolution. To explore the impact of this environment in detail, we examine the distribution in the mid-infrared (MIR) 3.6-8.0 micron colorspace of 42 galaxies from 12 Hickson compact groups in comparison with several control samples, including the LVL+SINGS galaxies, interacting galaxies, and galaxies from the Coma Cluster. We find that the HCG galaxies are strongly bimodal, with statistically significant evidence for a gap in their distribution. In contrast, none of the other samples show such a marked gap, and only galaxies in the Coma infall region have a distribution that is statistically consistent with the HCGs in this parameter space. To further investigate the cause of the HCG gap, we compare the galaxy morphologies of the HCG and LVL+SINGS galaxies, and also probe the specific star formation rate (SSFR) of the HCG galaxies. While galaxy morphology in HCG galaxies is strongly linked to position with MIR colorspace, the more fundamental property appears to be the SSFR, or star formation rate normalized by stellar mass. We conclude that the unusual MIR color distribution of HCG galaxies is a direct product of their environment, which is most similar to that of the Coma infall region. In both cases, galaxy densities are high, but gas has not been fully processed or stripped. We speculate that the compact group environment fosters accelerated evolution of galaxies from star-forming and neutral gas-rich to quiescent and neutral gas-poor, leaving few members in the MIR gap at any time.

LEGACY EXTRAGALACTIC UV SURVEY (LEGUS) WITH THE HUBBLE SPACE TELESCOPE. I. SURVEY DESCRIPTION

The Legacy ExtraGalactic UV Survey (LEGUS) is a Cycle 21 Treasury program on the Hubble Space Telescope aimed at the investigation of star formation and its relation with galactic environment in nearby galaxies, from the scales of individual stars to those of ~kiloparsec-size clustered structures. Five-band imaging from the near-ultraviolet to the I band with the Wide-Field Camera 3 (WFC3), plus parallel optical imaging with the Advanced Camera for Surveys (ACS), is being collected for selected pointings of 50 galaxies within the local 12 Mpc. The filters used for the observations with the WFC3 are F275W(λ2704 A), F336W(λ3355 A), F438W(λ4325 A), F555W(λ5308 A), and F814W(λ8024 A); the parallel observations with the ACS use the filters F435W(λ4328 A), F606W(λ5921 A), and F814W(λ8057 A). The multiband images are yielding accurate recent (lesssim50 Myr) star formation histories from resolved massive stars and the extinction-corrected ages and masses of star clusters and associations. The extensive inventories of massive stars and clustered systems will be used to investigate the spatial and temporal evolution of star formation within galaxies. This will, in turn, inform theories of galaxy evolution and improve the understanding of the physical underpinning of the gas-star formation relation and the nature of star formation at high redshift. This paper describes the survey, its goals and observational strategy, and the initial scientific results. Because LEGUS will provide a reference survey and a foundation for future observations with the James Webb Space Telescope and with ALMA, a large number of data products are planned for delivery to the community.

Stellar Winds and Embedded Star Formation in the Galactic Center Quintuplet and Arches Clusters: Multifrequency Radio Observations

A multifrequency, multiconfiguration study has been made of the compact radio sources in the Galactic center Quintuplet and Arches stellar clusters using the Very Large Array. Ten radio sources have been detected in the Quintuplet cluster. The majority of these radio sources have rising spectral indices and are positionally coincident with young massive stars that are known to have powerful stellar winds. We conclude that the three most compact of these sources are produced by stellar wind emission; thus, mass-loss rates can be derived and have an average value of 3 × 10-5 M⊙ yr-1. The remainder of the sources are likely to be a combination of stellar wind emission and free-free emission from surrounding ionized gas. In three cases, the radio sources have no stellar counterpart, and the radio emission is thought to arise from compact or ultracompact H II regions. If so, these sources would be the first detections of embedded massive stars to be discovered in the Galactic center clusters. The radio nebula associated with the Pistol star resembles the nebula surrounding the luminous blue variable star η Car and may be related to the stellar wind of the Pistol star. Ten compact radio sources are also detected in the Arches cluster and are interpreted to be stellar wind sources, consistent with previous findings. Several of the sources show moderate variability (10%–30%) in their flux density, possibly related to a nonthermal component in the wind emission. A number of radio sources in both clusters have X-ray counterparts, which have been interpreted to be the shocked, colliding winds of massive binary systems.

A New View of the Super Star Clusters in the Low-Metallicity Galaxy SBS 0335-052

We present a study of the individual super star clusters (SSCs) in the low-metallicity galaxy SBS 0335-052 using new near-infrared (near-IR) and archival optical Hubble Space Telescope observations. The physical properties of the SSCs are derived from fitting model spectral energy distributions (SEDs) to the optical photometry, as well as from the H? and Pa? nebular emission lines. Among the clusters, we find a significant age spread that is correlated with position in the galaxy, suggesting that successive cluster formation occurred in SBS 0335-052 triggered by a large-scale disturbance traveling through the galaxy at a speed of ~35 km s?1. The SSCs exhibit I-band (~0.8??m) and near-IR (~1.6-2.1 ?m) excesses with respect to model SEDs that are fit to the optical data. We hypothesize that the I-band excess is dominated by a photoluminescent process known as Extended Red Emission; however, this mechanism cannot account for the excesses observed at longer near-IR wavelengths. From the cluster SEDs and colors, we find that the primary origin of the near-IR excess observed in the youngest SSCs (3 Myr) is hot dust emission, while evolved red supergiants dominate the near-IR light in the older (7 Myr) clusters. We also find evidence for a porous and clumpy interstellar medium surrounding the youngest, embedded SSCs: the ionized gas emission underpredicts the expected ionizing luminosities from the optical stellar continuum, suggesting that ionizing photons leak out of the immediate vicinity of the clusters before ionizing hydrogen. The corrected, intrinsic ionizing luminosities of the two SSCs younger than ~3 Myr are each ~5 ? 1052s?1, which is equivalent to each cluster hosting ~5000 O7.5 V stars. The inferred masses of these SSCs are ~106 M ?.

HIERARCHICAL STRUCTURE FORMATION AND MODES OF STAR FORMATION IN HICKSON COMPACT GROUP 31

The handful of low-mass, late-type galaxies that comprise Hickson Compact Group 31 (HCG 31) is in the midst of complex, ongoing gravitational interactions, evocative of the process of hierarchical structure formation at higher redshifts. With sensitive, multicolor Hubble Space Telescope imaging, we characterize the large population of < 10 Myr old star clusters (SCs) that suffuse the system. From the colors and luminosities of the young SCs, we find that the galaxies in HCG 31 follow the same universal scaling relations as actively star-forming galaxies in the local universe despite the unusual compact group environment. Furthermore, the specific frequency of the globular cluster system is consistent with the low end of galaxies of comparable masses locally. This, combined with the large mass of neutral hydrogen and tight constraints on the amount of intragroup light, indicate that the group is undergoing its first epoch of interaction-induced star formation. In both the main galaxies and the tidal-dwarf candidate, F, stellar complexes, which are sensitive to the magnitude of disk turbulence, have both sizes and masses more characteristic of z = 1-2 galaxies. After subtracting the light from compact sources, we find no evidence for an underlying old stellar population in F—it appears to be a truly new structure. The low-velocity dispersion of the system components, available reservoir of H I, and current star formation rate of ~10 M ☉ yr–1 indicate that HCG 31 is likely to both exhaust its cold gas supply and merge within ~1 Gyr. We conclude that the end product will be an isolated, X-ray-faint, low-mass elliptical.

The Very Young Starburst Merger System NGC 1741

We use Hubble Space Telescope (HST) Faint Object Camera (FOC) ultraviolet (UV) and WFPC2 optical images in conjunction with UV spectroscopic observations taken with the Goddard High Resolution Spectrograph to examine the star formation history and properties of the interacting galaxy system NGC 1741 in the Hickson Compact Group 31. The high spatial resolution afforded by HST has allowed us to identify a large number of starburst knots, or super?star clusters (SSCs), in the starburst regions of this system. Photometry of these SSCs in the UV and optical bands indicates that most of these objects have ages of a few Myr, with a few up to ?100 Myr, and masses between 104 and 106 M?. The estimated age is confirmed by a spectral synthesis analysis of one knot for which we have obtained a UV spectrum. The V-band luminosity function of the SSCs is well represented by a power law (L) ~ L-? with an index of -1.85, with no evidence of a turnover brighter than the completeness limit. These properties are in good agreement with those found for SSCs in other starburst galaxies. Our results support the suggestion that some of these SSCs may be extremely young globular clusters formed in a relatively recent starburst episode that has been triggered by a merger event.

INTRAGROUP AND GALAXY-LINKED DIFFUSE X-RAY EMISSION IN HICKSON COMPACT GROUPS

Isolated compact groups (CGs) of galaxies present a range of dynamical states, group velocity dispersions, and galaxy morphologies with which to study galaxy evolution, particularly the properties of gas both within the galaxies and in the intragroup medium. As part of a large, multiwavelength examination of CGs, we present an archival study of diffuse X-ray emission in a subset of nine Hickson compact groups (HCGs) observed with the Chandra X-Ray Observatory. We find that seven of the groups in our sample exhibit detectable diffuse emission. However, unlike large-scale emission in galaxy clusters, the diffuse features in the majority of the detected groups are linked to the individual galaxies, in the form of both plumes and halos likely as a result of vigourous star formation or activity in the galaxy nucleus, as well as in emission from tidal features. Unlike previous studies from earlier X-ray missions, HCGs 31, 42, 59, and 92 are found to be consistent with the LX -T relationship from clusters within the errors, while HCGs 16 and 31 are consistent with the cluster LX -? relation, though this is likely coincidental given that the hot gas in these two systems is largely due to star formation. We find that LX increases with decreasing group H I to dynamical-mass ratio with tentative evidence for a dependence in X-ray luminosity on H I morphology whereby systems with intragroup H I indicative of strong interactions are considerably more X-ray luminous than passively evolving groups. We also find a gap in the LX of groups as a function of the total group specific star formation rate. Our findings suggest that the hot gas in these groups is not in hydrostatic equilibrium and these systems are not low-mass analogs of rich groups or clusters, with the possible exception of HCG 62.