Astronomical observations of chemical elements heavier than lithium (known simply as "metals" in astrophysics) can tell us a lot about how galaxies evolve. For example, the total amount of metal in the interstellar gas of a galaxy correlates with the total number of stars that were formed. Also, the ratio of oxygen to iron abundance in stars (known as the oxygen enhancement, or simply [O/Fe]) is thought to act as a galactic clock, telling us how quickly a galaxy grew. Galaxies with high oxygen enhancement should have formed their stars rapidly, before the iron from type Ia supernovae (SNe-Ia) could pollute the star-forming gas. Galaxies with low oxygen enhancement, on the other hand, should have formed their stars over an extended period of time, with the youngest stars containing large amounts of SNe-Ia-produced iron.

However, despite this standard and straight-forward theoretical framework, sophisticated galaxy evolution models have been unable to reproduce, at the same time, the complex chemical patterns seen in different types of galaxy. Specifically, the metal abundances seen in the photospheres of stars in the Milky Way and those seen in integrated populations of old stars in elliptical galaxies could only be reproduced simultaneously by invoking certain physical processes that are not part of our canonical understanding of galaxy evolution.

Starting in 2010, a team of scientists from the MPA and University of Sussex embarked on a project to reconcile the chemical properties seen in these very different regions of the cosmos. Using their latest semi-analytic model and a state-of-the-art implementation of the metal enrichment of galaxies by stars, the team could reproduce the chemical properties of the gas in nearby star-forming galaxies, of Sun-like stars in the Milky Way, and of the old stars of elliptical galaxies. Crucially, this is all done simultaneously and without any radical departure from the standard framework of galaxy formation that has seen so much success in other areas of astrophysics.

Our galaxy, the Milky Way contains around 300 billion stars with various chemical properties, ranging from old, metal-poor stars to young, metal-rich stars (see Figure 1). The team found that the relation between iron abundance and oxygen enhancement for Sun-like stars in a sample of model Milky-Way-type galaxies shows good agreement with those observed in real Milky Way stars (see Figure 2). This tells us that the model is accurately representing the chemical evolution of the Milky Way for the last thirteen billion years.

The same model, with all the same assumptions about the physical processes occurring in galaxies, also reproduces the chemical trends seen in elliptical galaxies of different masses. In the real Universe, the most massive ellipticals (see, for example, the galaxy in the centre of Figure 3) are known to have a higher oxygen enhancement than lower-mass ellipticals (see, for example, the galaxy in the top right panel of Figure 3). In our model, we find the same correlation between mass and oxygen enhancement.

And it's for the reason we would expect: high-mass ellipticals have formed stars rapidly (before a lot of iron is produced), whereas low-mass ellipticals formed their stars over a more extended period of time (and so contain more iron). This result is a significant achievement in itself, as it shows the relationship between the mass, age and chemistry of ellipticals predicted by the model is similar to that really observed, without requiring any major changes to the standard galaxy formation paradigm.

So what is different about this new model that allows these results to be achieved? The team believes that the key is the assumptions made about the various metals ejected by different stars and the lifetimes of SN-Ia progenitors. In their model, the simulated metal yields depend on a star's mass and metallicity, and also take account of mass loss via stellar winds prior to the final supernova explosion. In addition, no more than half of the SNe-Ia progenitor systems should explode within four hundred million years of their birth, and only about one in a thousand of all the stars formed should produce a SN-Ia. None of these conditions is particularly controversial, and when combined with the detailed semi-analytic model the group obtained the results described above.

But this is not the end of the story! The team are now working on simultaneously reproducing the chemical properties of objects that are at even more extreme ends of the galaxy spectrum. These tests will show whether the same model can reproduce both the chemical evolution of very-low-mass dwarf galaxies and the iron content of the hot gas surrounding the most massive galaxy clusters. Such tests are also crucial to validating any galaxy formation model, and should teach us even more about the true nature of galaxies in the nearby Universe.

Rob Yates and Guinevere Kauffmann

References:

Yates R. M., Henriques B., Thomas P. A., Kauffmann G., Johansson J., White S. D. M., 2013, MNRAS, accepted http://arxiv.org/pdf/1305.7231v1.pdf