A Census of Massive Stars Across the Hertzsprung-Russell Diagram of Nearby Galaxies: What We Know and What We Dont

arXiv:0903.0155v2 [astro-ph.SR] 21 Sep 2009 Hot And Cool: Bridging Gaps in Massive Star Evolution ASP Conference Series, Vol. xxx, 2009 C. Leitherer, Ph. D. Bennett, P. W. Morris & J. Th. van Loon, eds. A Census of Massive Stars Across the Hertzsprung-Russell Diagram of Nearby Galaxies: What We Know and What We Don’t Philip Massey Lowell Observatory, 1400 W. Mars Hill Road, Flagstaff, AZ 86001 USA Abstract. When we look at a nearby galaxy, we see a mixture of foreground stars and bona fide extragalactic stars. I will describe what we need to do to get meaningful statistics on the massive star populations across the H-R diagram. Such a census provides the means of a very powerful test of massive star evolutionary theory. 1. Introduction In this review, I will briefly discuss what we do (and don’t!) know about the population of massive stars across the H-R diagram (HRD) of nearby, Local Group galaxies. If we are accurate census takers, then we can use the numbers of differing types of massive stars (O-type, Luminous Blue Variable stars, F and G supergiants, red supergiants, Wolf-Rayet stars) to provide a sanity check on the predictions of stellar evolution theory, and in particular to see if the relative numbers change as a function of metallicity in the way that they should. For this, we use those galaxies of the Local Group which are currently forming massive stars. These span a range of a factor of 20 or so in metallicity as measured by the oxygen content of their HII regions (Massey 2003), and thus allow a more powerful test than merely that offered by comparing the content of (say) the SMC to the Milky Way, where the metallicities differ only by a factor of 4 or so. We expect that metallicity z matters, primarily as the mass loss rates scale as something like z0.7 (Vink et al. 2001) during the hydrogen-burning main- sequence phase, and the amount of mass loss during this phase plays a dominant role in the evolution of massive stars. Our modern understanding of this traces back to Peter Conti, who proposed (Conti 1976) that Wolf-Rayet stars were the result of mass loss. As a massive star evolves, it loses mass, and eventually reveals He and N (the products of CNO burning) at the surface, at which point the star is spectroscopically identified as a WN star. Further mass-loss eventually reveals C and O, the products of He-burning, and the star is called a WC. Conti and collaborators eventually argued that the metallicity dependence of the mass loss rates were responsible for the very different relative number of WC- and WN-type Wolf-Rayets seen in external galaxies (Vanbeveren & Conti 1980; see also Massey & Johnson 1998 and Massey 2003). We will discuss this more extensively below, but this is the sort of test such a census allows. This is a good moment to conduct such tests. On the one hand, our Local Group Galaxies survey (Massey et al. 2006, 2007a, 2007b) has used the KPNO and CTIO 4-m telescopes to perform photometry of stars throughout the star- 1 2 Massey forming galaxies of the Local Group. On the other hand, evolutionary theory now includes the important effects of rotation at various metallicities (Meynet & Maeder 2003, 2005), and thanks to the generosity of our colleagues in Geneva the models are freely available. Still, nothing worth doing is easy, and we will find that while there are a lot of things that we do know now, there is still quite a bit that we don’t. There is still plenty of fun to be had. 2. Massive Star Evolution: An Observer’s View I’d like to start by sharing my (limited) understanding of what stellar evolu- tionary theory predicts, using as our baseline the Geneva evolutionary tracks for z = 0.020 (solar) of Meynet & Maeder (2003) and an initial rotation of 300 km s−1. This way when we talk about a census of stars across the HRD we will all have in mind the same mental picture of where these stars are coming from. First, let us consider a star of very high mass, 120M⊙. In Fig. 1(a) we plot its path in the HRD, with the main-sequnce (MS), Luminous Blue Variable (LBV), and Wolf-Rayet (WR) phases (WN, WC) indicated. At the beginning the star is spectroscopically an O3 V, but the star then enters the WN phase while still a main-sequence (core-H burning) object. In Fig. 1(b) we show what the models predict will happen with a 40M⊙star: it will go through its main-sequence phase, starting out as an O5 V star and evolving through an F and G supergiant stage. The star then becomes hotter again, and eventually evolves to a WN- and then a WC-type WR. Still, making it as far as a WC is a close thing; if the calculation is done without including rotation, the star never makes it to the WC stage, as shown in Fig. 1(c). Finally, let us consider the evolution of a 20M⊙star. It begins its main- sequence life as an O8 V, and then evolves through the yellow (F and G-type) supergiant stage, finally reaching the red supergiant (RSG) stage. 3. O-type Stars It would be extremely attractive to be able to use t

…(Full text truncated)…

This content is AI-processed based on ArXiv data.