Research Byte
Published in the RSAA Lunations
Vol1 Issue32 1–30 September 2022
Whilst they are often more difficult to measure than strong emission lines, absorption lines offer a wealth of information on the complex star formation histories of individual galaxies. These lines trace the chemical composition of the stellar content, rather than the hot gas inside a galaxy. Individual features show strong correlations to population parameters, such as the Hydrogen Balmer lines with stellar age, and Fe4668 and Fe5270 with total metallicity. The other common population parameter is the elemental abundance, often measured with respect to the metallicity. A particularly valuable example is the [α/Fe] abundance ratio, as this provides a good insight into the length of star formation in a given population.
The study of these spectral features stretches back decades. In 2000, Trager et al. found an empirical scaling relation between [α/Fe], and stellar velocity dispersion σ. Since its discovery, this relation has proven to be a useful test for simulations and semi-analytic models of galaxy formation, although some difficulties persist in reproducing it across different mass scales. Over the past few years in my PhD, it's this relation specifically that I've been focussing on, using data from the SAMI Galaxy Survey.
Prior to the advent of large-scale spectroscopic surveys, the behaviour of the [α/Fe]-σ relation with respect to morphology and environment was poorly understood at the lower end of the mass scale. Simultaneously disentangling the influence of the environment from any morphology-dependent factors requires a considerable number of galaxies to achieve any kind of statistical significance. This is what I first set out to look at, and found that whilst galaxies in high density environments were α-enhanced over their counterparts in low density environments, the effect was concentrated in low-mass galaxies. Additionally, the highest offset was found in galaxies visually classified as elliptical (https://doi.org/10.1093/mnras/stab3477).
There are, of course, multiple ways to classify galaxies. Alongside the traditional Hubble Sequence, one alternative is to combine a measurement of the motion of the stellar content, such as the spin proxy λ_R, with a quantitative describing the galaxy shape, giving a kinematic classification. This is often used to divide galaxies into "fast rotators" with ordered rotation, and "slow rotators" with more complex velocity fields. These slow rotators are thought to have substantially different formation pathways than rotation-dominated systems, giving rise to this kinematics-based distinction. Perhaps surprisingly, when we considered the total stellar populations of slowly-spinning galaxies (both those classified as slow rotators, and those considered to be fast rotators), we found their residual α-enhancement to be indistinguishable (https://doi.org/10.1093/mnras/stac1221). This would indicate that there is little difference in the integrated star formation histories of these types of galaxies.
So far, we've assumed that galaxies have a uniform star formation history across their visible extent. What if that weren't the case? This is where galaxy surveys such as SAMI and HECTOR are invaluable. By observing the galaxy using a cleverly-constructed bundle of optical fibres, we can obtain an image containing a spectrum at every pixel. Of course, depending on the specific research area, we don't necessarily need all of this information. Since the galaxies I'm looking at are typically spherically symmetric, we can focus on the radial change in stellar populations, and stack pixels azimuthally to increase the signal to noise ratio.
What do we find? There's almost no change from the global properties. For our slowly-spinning sample (λ_Re<0.4), both the slow and fast rotators have strongly negative metallicity gradients, and relatively flat age and [α/Fe] profiles. An interesting result (for me at least!), but for now it's back to the literature to see how this compares to the simulations.
Peter Watson
Figure: The increase in [α/Fe] as a function of velocity dispersion σ, and environmental density Σ_5. We show both the raw values, and a locally smoothed plot underneath.
