Exploiting the Power of Metallicity Studies in the UV: Dissecting the Nearby Spiral Galaxy M83

S. Hernandez (sveash[at]stsci.edu)

Metallicities of galaxies at all redshifts are critical for deciphering a plethora of physical and evolutionary processes that take place among and inside galaxies, including star formation, stellar feedback, and interstellar/intergalactic chemical enrichment. In the last decades it has become evident that a large fraction of the stellar mass in the universe was assembled in intense episodes of star formation (SF) at high (z > 1) redshift. Studies of local starbursts and star-forming galaxies can be performed at exquisite signal-to-noise, spatial and spectral resolution, which are not achievable at higher redshift. Such studies provide an invaluable tool for creating a baseline in understanding how gas and stellar properties evolve through cosmic time. 

In this Newsletter I present a brief summary of our latest findings in the nearby spiral galaxy M83 (Hernandez et al. 2019). Using young star clusters, we performed a detailed metallicity gradient analysis. Our measurements provide us with hints to the chemical history, as well as physical properties of this face-on, star-forming galaxy. We identify two possible breaks in the metallicity gradient of M83 at galactocentric distances of R ∼ 0.5 and 1.0 R25.The metallicity gradient of this galaxy follows a steep-shallow-steep trend, a scenario predicted by three-dimensional (3D) numerical simulations of disk galaxies. We propose a scenario where the first steep gradient originated by recent star-formation episodes and a relatively young bar (<1 Gyr). The shallow gradient, on the other hand, is created by the effects of dilution of outflowing enriched gas mixing with low-metallicity material present at larger radial distances. And finally, the second break and last steep gradient mark the farthest galactocentric distances where the outward flow has penetrated. 

Metallicity of Galaxies

With the purpose of understanding how galaxies are formed and evolve with time, astronomers continuously search for new tools that can provide clues to their past. One of these tools that can lead to strong constraints on the history of cosmic chemical enrichment of galaxies is the measurement of their chemical composition. It is through such studies that we can learn about the evolution of the Universe from a primordial metal-free system to the present-day chemically diversified environment.

The metallicity of a galaxy, [Z], is mainly controlled by two components: 1) chemically processed material in stars, and 2) exchanges between the galaxy and the intergalactic medium (IGM). Given the dependence of metallicity on these two factors, studies focused on metallicity relations and metallicity gradients in galaxies hold a wealth of information on their formation and evolution.

Back in 1979, Lequeux discovered that the masses and metallicities of several star-forming, irregular, and blue compact dwarf galaxies, correlated with each other in the sense that more massive galaxies appeared to have higher metallicities. This same correlation was later confirmed by several other independent studies making this mass-metallicity relation (MZR) widely used for studying star-formation episodes, galactic winds, and chemical enrichment in general. Similarly important, the metallicity gradients in individual galaxies track some of the complex dynamics taking place within them, such as outflows and infall of material. Given the importance of accurate metallicities, it is crucial to obtain reliable metallicity indicators in order to correctly interpret the processes influencing these relations.

In the last decade, new techniques have been developed to study the chemical content of nearby galaxies using blue supergiant stars (BSGs) and red supergiant stars (RSGs), as well as the integrated light of star clusters. These are in addition to the more classical techniques where nebular emission lines at optical and infrared wavelengths are used to infer the metallicity of H Ⅱ regions or absorption-line spectra arising from heavy elements against bright background UV sources are analyzed to make a census of the metals in the interstellar medium (ISM) of galaxies. With current telescopes, star clusters well outside of the Local Group and at distances of several megaparsecs (Mpc), appear not to be resolved into individual stars. This facilitates measurement of their chemical composition through the analysis of their integrated light (IL) spectra. This kind of spectroscopy has shown ample potential; however, until recently most efforts have focused on the optical and infrared regimes. Young stellar populations, for instance, when observed in the near-infrared are strongly dominated by RSGs, which facilitates their analysis through modeling of the whole population as a single RSG star. When studying globular clusters (GCs), on the other hand, one needs to make the assumption that all stars in the cluster are coeval and implement full population synthesis techniques to perform a similarly detailed abundance analysis. By way of contrast, the UV spectral coverage for IL metallicity analyses of clusters is rather unexplored.

Piecing together the History of M83

composite image
Figure 1: A colour-composite image of M83 observed with the 8.2-meter Subaru Telescope (NAOJ), the 2.2-meter Max Planck-ESO telescope and the Hubble Space Telescope. Red circles locate the clusters studied in the optical regime with VLT/X-Shooter. Cyan circles show the clusters observed with HST/COS. The subpanel above shows the VLT/X-Shooter observations (in black), plotted with the best fit model (red) for star cluster 1182. Image processing and copyright: Robert Gendler.

In our latest publication, Hernandez et al. (2019), we performed a metallicity study of the nearby (4.9 Mpc) face-on star-forming galaxy M83. This was accomplished by merging ground-based optical observations from the X-Shooter spectrograph on the Very Large Telescope (VLT) in Chile with space-based far-ultraviolet observations with the Cosmic Origin Spectrograph (COS) onboard the Hubble Space Telescope (HST). Our sample of young star clusters was comprised of 23 targets covering the inner disk of the galaxy (Figure 1). Our metallicity measurements confirm a relatively steep gradient in the inner disk of the galaxy. 

To obtain a more complete picture of the metallicity trends in M83, we combined our measurements with those from Bresolin et al. 2007 and 2009 covering the outer disk of M83 as shown in Figure 2. Based on the trends in this figure, we proposed that M83 exhibits a metallicity gradient with two possible breaks at R ∼ 0.5 R25 and R ∼ 1.0 R25. If the metallicity breaks are genuine, the metallicity gradient of this galaxy follows a steep-shallow-steep trend, a scenario predicted by three-dimensional (3D) numerical simulations of disk galaxies. The first break is located near the corotation radius. This first steep gradient may have originated by recent starformation episodes and a relatively young bar (<1 Gyr) possibly triggered by an interaction or merger. In the numerical simulations the shallow gradient is created by the effects of dilution of chemically enriched material outflowing and mixing with lower metallicity gas present at larger galactic distances. The second break and last steep gradient mark instead the farthest galactocentric distances where the outward enriched gas flow has penetrated the disk. 

Our work has demonstrated that the integrated-light method can be used as an alternative metallicity tool to nebular and interstellar gas techniques, applicable not only in the optical and infrared, but also in the UV. By successfully expanding the metallicity analysis of stellar clusters to UV wavelengths we have made it possible to simultaneously study the multi-phase gas and stellar properties. Such analysis can bridge our understanding of galactic outflows, stellar evolution and stellar feedback not only locally, but also at higher redshifts (z ~ 2) where the rest-frame UV light is shifted into the optical/IR wavelengths.

Measurements from COS
Figure 2: Left panel: Oxygen abundance as a function of galactocentric distance normalized to isophotal radius, R25. In red we show the cluster measurements for the VLT/X-Shooter sample; in cyan we display the results from the targets observed with HST/COS. In yellow circles and green triangles, we show the measurements from H Ⅱ regions by Bresolin et al. 2007 and 2009, respectively. Right panel: Zoomed-in version of the left panel in the center of the galaxy. For both panels, thick dashed black lines show linear regressions at three different galactocentric distance ranges: <0.5 R25, 0.5–1.0 R25, and >1.0 R25. For comparison we show a single power-law fit as a thin black line and the linear regressions estimated by Bresolin et al. (2009) as solid blue lines.


Bresolin, F., 2007, ApJ, 656, 186
Bresolin, F., Ryan-Weber, E., Kennicutt, R. C., & Goddard, Q. 2009, ApJ, 695, 580 
Hernandez, Svea, Larsen, Søren, Aloisi, Alessandra, Berg, Danielle A., Blair, William P., Fox, Andrew J., Heckman, Timothy M., James, Bethan L., Long, Knox S., Skillman, Evan D., & Whitmore, Bradley C., 2019 ApJ, 872, 116H