MUSE reveals the Milky Way is not alone in harbouring chemically distinct thick and thin disks

By Nicholas Scott & Jesse Van de Sande
[email protected] & [email protected]

The Milky Way is by far the best-studied galaxy in the Universe, with observations of it and speculation on its nature dating back thousands of years, including amongst Indigenous Australians. The disk nature of the Galaxy has been evident for at least a hundred years, with early maps from William Herschel indicating a flattened structure. Discovery of the thick disk – distinct from the thin disk visible with the naked eye – only occurred a few decades ago (Gilmore & Reid, 1983). More recent observations (e.g. Fuhrmann 1998; Hayden et al. 2015) have shown the thick and thin disks are chemically distinct, with the thick disk consisting of stars with a high ratio of alpha elements to iron, [α/Fe].

Much effort has been expended in trying to reproduce the formation of this chemically distinct, double-disk structure in galaxy formation models. The leading model, often known as the ‘two in-fall’ model (Chiappini et al. 1997) posits an early period of very rapid star formation which formed the thick disk, followed by a break or pause in star formation, likely triggered by a merger. Gas accretion then reignited a more gradual epoch of star formation producing the thin disk. An alternative scenario is that the disk bimodality arises naturally from the different timescales of core collapse and Type Ia supernovae, combined with radial migration of stars (Schönrich & Binney, 2009). As both models were designed to explain the thin and thick disk, both exhibit very similar predictions for Galactic chemo-dynamical structure.

However, the Milky Way isn’t the only spiral galaxy in the Universe, and the formation theories developed for our Galaxy apply just as well to other similar spiral galaxies – so-called Milky Way Analogues (MWAs). The two theories predict very different rates of prevalence for chemically distinct thick/thin disks. In the two in-fall model chemically distinct thick disks should be rare, requiring a specific formation history to pause star formation, occurring in perhaps 5% of MWAs (Evans et al. 2020). In contrast the alternative model predicts that the double-disk structure should occur in essentially all MWAs. Therefore, by determining the prevalence of chemically distinct thick disks in MWAs we can discriminate between the two theories of thick and thin disk formation.

Figure 1: Pseudo-colour image of UGC 10738 derived from the mosaiced MUSE observations. UGC 10738 is an excellent analogue of the Milky Way, with a small classical bulge and prominent X-shaped structure indicative of a bar seen edge-on.

In Scott et al. (2021) we selected a sample of 9 nearby (D < 100 Mpc), highly-inclined disk galaxies that are close analogues of the Milky Way and observed them with MUSE on the VLT. Our selection was based on the stellar mass, apparent morphology and colour of the objects. Of the 9 objects, UGC 10738 (Fig. 1 above) was the outstanding candidate. It’s distinctive X-shaped central isophotes indicate a prominent bar like that found in our own Galaxy. Moreover, with an inclination of ~7 degrees there is minimal spatial contamination of the thin and thick disk stellar populations with the high spatial resolution of MUSE. At the distance of UGC 10738 (D ~99 Mpc) the spatial resolution of MUSE is 91 pc, sufficient to resolve the typical thin disk scale height of ~300 pc.

Our results confirm previous findings that the typical electron density in star-forming regions has decreased by a factor of 6-10 over the last 10 billion years (Fig. 2 below). Our sample excludes galaxies with outflows, indicating that contamination from compressed gas in outflows is unlikely to be the primary driver of the observed electron density evolution.

Figure 2: Maps of the mass-weighted average stellar metallicity (top), [α/Fe] (middle), and age (bottom) for UGC 10738 (left column) and the Milky Way (right column). For UGC 10738 these maps are measured from the MUSE observations. For the Milky Way they are derived from the models of Sharma et al. (2020). We find excellent qualitative agreement between the two, though note a difference in scale in the sense that UGC 10738 is approximately 50% larger than the Milky Way in terms of scale height and effective radius.

Utilising the widely-used pPXF software (Cappellari & Emsellem, 2004) we fit a set of single stellar population templates with variable age, metallicity ([Z/H] and [α/Fe]) to the spatially-resolved MUSE spectra. From the template weights we derive maps of the mass-weighted average age, [Z/H] and [α /Fe] as a function of projected radius and scale height in parsecs. We then compared these maps to equivalent maps for the Milky Way derived from the analytic model of Sharma et al. (2020). We found excellent agreement between the two galaxies, with a clear [α/Fe]-enhanced component at large scale heights indicative of a chemically distinct thick disk. Using the template weights we can go even further, comparing the distribution of stars (or stellar populations in the case of UGC 10738) in the [α/Fe]-[Z/H] plane as a function of spatial position within the two galaxies. Again we find good qualitative agreement between the observations UGC 10738 and the distributions for the Milky Way presented in Hayden et al. (2015).

We conclude that chemically distinct thick and thin disks are common, perhaps even ubiquitous in MWAs. This conclusion is supported by studies of edge-on S0s in the nearby Fornax cluster (Pinna et al. 2019, Poci et al. 2021) that also find a high frequency of [α/Fe]-enhanced thick disks. These observations highly favour the scenario of Schönrich & Binney (2009), suggesting our Galactic thin and thick disks emerged naturally as a result of supernovae enrichment and radial migration.


Michael Murphy is the Australian representative on the ESO Science Technical Committee. Contact: [email protected]

Sarah Sweet is the Australian representative on the ESO Users Committee. Contact: [email protected]

Stuart Ryder is a Program Manager with AAL. Contact: [email protected]

Guest posts are also welcome – please submit these to [email protected]