Research Interests

My research focuses on young, massive star clusters in starburst galaxies. These clusters are substantially more massive than the population of old globular clusters in the halo of our Galaxy, but in other respects are quite similar. The origin of Galactic globular clusters is a mystery, and has important implications for models of galaxy formation and stellar populations.

Young star clusters in the M82 starburst. Left: HST/ACS optical light (BVI) images. A prominent dust lane runs diagonally through the lower half of the image. Right: The same region in Keck/NIRC2 AO near-infrared (IHK) images. The near-IR light penetrates the dust to reveal swarms of stars around the very red clusters. Both images are approximately 10 x 10 arcseconds.

The origins of stars and the time evolution of galaxies are defining questions for modern astrophysics. While adequate formation theories have been developed for low mass stars, such as our own Sun, the formation processes for high mass stars remain mysterious. High mass stars exert significant influence on their surrounding through energy and momentum loss of their prodigious winds and subsequent explosion as supernovae. Observations indicate that high mass stars do not form in isolation, but rather in the dense central regions of star clusters. Rich young clusters, many still embedded in the molecular clouds from which they form, account for most of the star formation in the local Universe. Such clusters appear to form in great numbers in extreme environments; observations indicate that star formation in merging and interacting galaxies resolves into young massive clusters (YMCs) with masses as large as 10⁶ to 10⁷ solar masses. Over time, star clusters are subject to various internal and external mechanisms that tend to disrupt the clusters and disperse the member stars. Understanding cluster evolution, specifically their dissolution, is therefore key to understanding the origin of stellar populations and the time evolution of galaxies.

In my research, I seek answers to several outstanding questions about young massive star clusters:

1. What is the internal structure of young massive clusters?

Laser guide star adaptive optics systems, such as this one at Keck Observatory, provide the high angular resolution required for detailed structural study of extragalactic star clusters.
Photo: John McDonald, CFHT

This 1.6-micron AO image of the double cluster NGC1569 SSC-A resolves the cluster's substructure. Note that the image is only 1.2 x 1.2 arcseconds!

The observation that the largest YMCs are more than one hundred times as massive as the Orion Nebular Cluster prompts a basic question: are YMCs scaled-up Galactic clusters or are they clusters of clusters? Nearby YMCs offer the best opportunity to search for primordial structure before it is washed out by dynamical relaxation. The highest possible angular resolution is necessary for imaging studies: YMCs have typical radii of 2-6 pc, but at a distance of 10 Mpc, 5pc subtends an angle of only 0.1 arcseconds! Until the advent of adaptive optics (AO) systems, this work has been the exclusive domain of the Hubble Space Telescope. I use near-IR observations for imaging the youngest clusters, which are inevitably located in regions of high extinction and are thus invisible in optical light. The two images above show the same region in M82: clusters readily visible in the near-IR (IHK) image are highly reddened or totally obscured in the optical (BVI) image.

My images of the young massive clusters in the starburst galaxy NGC 1569 with adaptive optics at the Keck Observatory resolve the individual YMCs, allowing me to investigate cluster light profiles and characterize substructure. Of particular interst is cluster A, which is from spectroscopic studies to have two kinematic components. My NIRC2 AO images clearly resolve the substructure of the cluster, as seen in the image at right.

2. How are young massive clusters populating
galaxies with stars?

Cluster M82-a in a 1.6-micron AO image. This cluster, invisible at optical wavelengths, is surrounded by a swarm of stars. The compact light profile of the concentrated core of the cluster suggests the cluster is shedding stars into the field. The bright cluster at upper right is M82-F. Field of view is ~ 10x10 arcsec.

Most stars form in clusters or loose associations. Over time, massive star clusters are subject to various internal and external mechanisms which tend to disrupt and disperse the clusters. Low-mass clusters are particularly vulnerable to each of these processes, while stars in the low velocity tail of the distribution function in massive clusters can remain bound in spite of loss of as much as 70% of the cluster mass. The small galactocentric radii of many starburst clusters tends to make them very prone to disruption by dynamical friction and bulge shocking. Understanding cluster evolution, specifically their dissolution, is therefore key to understanding the origin of stellar populations within galaxies.

I am using AO imaging to study the light profile of extragalactic star clusters and color magnitude diagrams of stars in their vicinity. The high angular resolution afforded by AO imaging enables me to do photometry in crowded fields at the edges of cluster cores and in their extended halos.

3. How do the energy outputs of high-mass stars in YMCs influence the surrounding interstellar medium?

An ACS/NICMOS near-IR (IHK) mosaic of the nuclear region of the galaxy NGC 253. The bright nuclear region is characterized by strong hydrogen recombination emission, apparently excited by a population of young stars. The outlined region shows the border of the NICMOS images.

Young massive clusters contain significant populations of high-mass stars: a 10⁶ solar mass cluster will contain several hundred OB stars for a normal stellar mass distribution. Energy and momentum outputs from these stars and their subsequent supernova explosions have a profound impact on the surrounding interstellar medium. I am currently mentoring third-year UCLA graduate student Kathy Kornei on a project investigating the nature of a massive super star cluster in the nuclear region of the nearby starburst galaxy NGC 253. We are analyzing near-infrared echelle spectra obtained with the CTIO 4-meter telescope, in addition to archival ACS and NICMOS optical and infrared imaging from the Hubble Space Telescope. We are investigating the cluster's age, mass, luminosity, and extinction in order to place it in the context of super star clusters in other starbursts (Kornei & McCrady 2009).

4. Are YMCs the upper end of a distribution over mass or a distinct form of star formation?

In my most recent paper (McCrady & Graham 2007), I use high resolution echelle spectra to measure the velocity dispersion of two dozen YMCs in M82. I combine these data with sizes based on HST imaging to measure the dynamical masses of 15 clusters and determine the cluster mass function. This is the largest population of clusters in a single environment with dynamical mass measurements, free of the usual assumptions about the initial mass function (IMF) of member stars. My dynamical mass measurements facilitate comparison with population synthesis models to constrain the IMF of the clusters. I found a power-law distribution of clusters over mass, dN/dM ~ M^a where a = -1.9. This is very similar to observational results for the power law distributions of giant molecular clouds and star forming clumps, suggesting a possible link between the clumping of molecular gas in giant clouds and subsequent formation of massive star clusters.

ACS/NICMOS mosaic of the nuclear starburst region in M82. This near-IR (IHK) image reveals a population of roughly two dozen massive clusters, only a handful of which can be studied in optical light because of the galaxy's pervasive, obscuring dust.

.

5. What is the nature of mass segregation in YMCs?

The young massive cluster M82-F shows evidence for mass segregation: the red supergiant stars appear to be concentrated in the center based on a color gradient evident in images of the cluster (McCrady et al. 2005). This may be a general phenomenon in massive clusters, even where young ages seem to preclude dynamical segregation by mass through star-star scattering. To investigate mass segregation within YMCs, I am using integral field spectroscopy behind adaptive optics to characterize the radial distribution of the most massive stars in the clusters. Newly available integral field unit (IFU) spectrometers operating behind AO systems provide the opportunity to obtain spatially resolved spectroscopy at moderate spectral resolution (R ~ 3,500). In my current work at UCLA, I am using the OSIRIS IFU at on the Keck II telescope. OSIRIS observations of clusters in M82 and NGC 6946 are enabling me to probe the distribution of high-mass red supergiant stars via spectral signatures such as CO absorption bandheads.

IFU spectroscopy of a massive cluster in NGC 6946. The spectrum (bottom) is from the four central pixels of the cluster, imaged at top. The spectrum shows nine CO absorption bandheads, direct detection of the evolved red supergiant population in the cluster's core.

.