Jurassic Universe: Interpreting a Fossil Galaxy from the Deep Past

Title: “Segue 1: An Unevolved Fossil Galaxy from the Early Universe”

Authors: Anna Frebel, Joshua D. Simon, and Evan N. Kirby

First Author’s Institution: Kavli Institute for Astrophysics and Space Research and Department of Physics, Massachusetts Institute of Technology, Cambridge, MA 02139, USA

Status: Published in ApJ [open access]

What if we could jump into a time machine and travel 13 billion years into the past, right up to the point when the very first galaxies in our Universe were forming from some of the earliest stars in existence? Right then and there, with a few observations, we could answer so many questions about how our Universe formed, how galaxies form, and how our own Milky Way came to be! Unfortunately, such a time machine has yet to be invented, but we do know of some reasonable alternatives — namely, the ultra-faint dwarf (UFD) galaxies: the dimmest, smallest, and most metal-poor galaxies thus far detected, and alleged “fossils” from the distant cosmological past.

Ultra-faint dwarfs: the tools of galactic paleontology

For a galaxy, the epithet “ultra-faint dwarf” is certainly not the most impressive — indeed, many UFDs possess luminosities below that of a typical globular cluster (see Figure 1)! However, there is way more to these UFDs than meets the eye (both literally and figuratively): as some of the oldest and most dark matter dominated bodies in the local Universe, UFDs serve as invaluable probes of cosmology, providing a glimpse into the state of the Universe during the epoch of reionization, constraining models of galaxy formation, potentially solving our problem of “missing satellites”, and, as we’ll see in today’s paper, presenting us with a direct means of studying the oldest stars in our Galaxy and the oldest galaxies in the Universe.

Figure 1. A census of dwarf galaxies orbiting the Milky Way, as plotted in a fantastic review on ultra-faint dwarfs by Joshua Simon (2019); the circles represent dwarf galaxies and the diamonds represent globular clusters. The intrinsic brightness of these bodies is plotted on the vertical axis (in terms of absolute magnitude in the photometric V-band), while their effective sizes are shown on the horizontal axis (in terms of the half-light radius, the distance within which half of a body’s light is emitted). The UFDs populate the low-brightness, low-radius region of the dwarf galaxy distribution; comparing the vertical-axis positions of the diamonds and the lowest circles, we see that many UFDs are less luminous than the average globular cluster, with a few globular clusters even possessing larger radii than some dwarves! Adapted from Figure 2 in Simon (2019).

Before we dive into the meat of today’s paper, let’s take a closer look at the ultra-faint dwarf galaxy Segue 1 — the main character of the paper and one of the most extreme UFDs — to get a sense of the scales we’re dealing with. At around 23 kiloparsecs (~75,000 light-years) from the Sun, Segue 1 is close enough to allow for spectroscopic observations of some of its brightest stars; unfortunately, though, its effective radius of 30 pc (~4.4 arcseconds on the sky) and luminosity of only 300 suns demand high sensitivities from our instruments — for those keeping score at home, Segue’s radius is about 17 times smaller than that of the Milky Way’s bulge and its luminosity is about one billion times fainter than that of the entire Milky Way (ultra-faint indeed!). However, with a mass of 600,000 solar masses, Segue possesses a mass-to-light ratio on the order of 2000 M/L (detailed studies place this number at around 3400 M/L). This is hundreds of times larger than the Milky Way’s mass-to-light; if this excess mass is contributed by dark matter, then to state that Segue is strongly dark-matter-dominated would be an understatement. While Segue’s high dark matter content makes it an appealing laboratory for the study of dark matter on small scales, it’s the galaxy’s peculiar stellar population that holds the key to uncovering Segue’s past.

Digging up clues to Segue 1’s past

This nicely segues into our main discussion of the paper at hand. In order to better understand the formation history of Segue 1, the authors conducted a thorough spectroscopic investigation of six red giant stars within the galaxy, combining their data with observations of a seventh red giant made by Norris et al. (2010). This data set likely covers all of Segue’s red giants, the brightest stars in the galaxy and the only stars in Segue for which high-resolution spectra (and thus high-quality metallicity and chemical abundance information) could be obtained. To probe the nature of these stars, the authors focused primarily on four observable properties: stellar metallicity, α-enrichment, carbon-enhancement, and neutron-capture abundances.

For the seven stars measured, the authors find stunningly low metallicities, with iron-to-hydrogen ratios reaching values of over 6000 times less than that of the Sun. Indeed, Segue 1 contains the highest fraction of extremely metal-poor stars yet seen in a galaxy, implying that very little chemical evolution has taken place within Segue’s stellar population. 

Further, in typical galaxies, higher-metallicity stars tend to possess lower abundances of “alpha” elements (elements like magnesium and calcium that are formed in the alpha process and expelled in the core-collapse (“Type II”) supernovae of massive stars) and higher abundances of “iron peak” elements (elements like cobalt and nickel that are mainly produced in the “Type Ia” supernovae of exploding white dwarfs). In Segue’s red giants, we see no such trend between α-enrichment and metallicity; in fact, Segue 1 is the only known galaxy for which this α-metallicity correlation is absent — this is illustrated in Figure 2. Chemical enrichment by Type Ia supernovae would raise the abundance of iron relative to that of alpha elements, yielding a downward trend in stars of a particular galaxy plotted on an [α/Fe] vs. [Fe/H] plot. Remarkably, Segue’s stars practically lie on a horizontal line in Figure 2, implying that no Type Ia supernovae ever exploded in Segue, or at least none of their chemical byproducts were ever incorporated into Segue’s stars — all the chemical enrichment in Segue occurred before low-mass stars had the chance to evolve into white dwarfs!

Figure 2. A comparison of α-element abundances (shown on the vertical axis, represented here by the combined abundances of magnesium, calcium, and titanium relative to the abundance of iron) and iron abundances (shown on the horizontal axis) for stars in various dwarf galaxies. The six solid red markers indicate the six Segue 1 red giants measured by the authors, with the open red marker indicating the red giant measured by Norris et al.; barring the outlier from Norris, α-abundance remains exceptionally steady across each of the red markers. By contrast, the abundance patterns of the larger “classical dwarf” galaxies, some of whose stars are plotted as open black circles, exhibit downward-sloping trends. The other colored markers correspond to stars in other ultra-faint dwarves, which have abundance patterns intermediate between those of the Segue stars and the classical dwarf stars, possessing subtle, but non-negligible, downward trends. Evidently, the enrichment pattern of Segue’s stars is wholly unique — and extremely peculiar! Adapted from Figure 9 in the paper.

Additional chemical signatures support the idea that the stellar population in Segue evolved only from massive stars and their subsequent Type II supernovae. For instance, the high carbon content of Segue’s red giants suggests that their stellar ancestors underwent processes typical of massive stars, and the distinct lack of neutron capture elements (high-mass nuclei typically formed in intermediate-mass stars via mergers with neutrons) among Segue 1’s stellar population shows that the galaxy stopped forming stars early in its lifetime, before such heavy elements could be synthesized in later-generation stars.

Interpreting the galactic fossil record

Putting all these clues together, the authors conclude not only that Segue 1 is one of the first galaxies that formed in the Universe, but also that, following its assembly from primordial stars, Segue underwent only one generation of star formation before promptly being shut off — possibly because its tiny gravitational pull was too weak to hold onto star-forming gas ejected by supernovae, or perhaps because the reionization of the Universe evaporated the dense clouds that would’ve collapsed into stars  — thus implying that the Segue we see today still resembles the Segue whose star formation was quenched 13 billion years ago: in other words, Segue 1 bears a “fossil imprint” of the early Universe! If Segue 1’s surprisingly simple formation history is typical of the earliest galaxies, then the unique combination of high α-abundances and low neutron-capture abundances found in Segue’s stellar population may be a tell-tale signature of stars formed in the first galaxies.  

Clearly, an astounding amount of information is hidden within these ultra-faint dwarf galaxies! Unfortunately, observing UFDs still remains a challenge; however, as technology improves in the coming decades, these small, dim galaxies will certainly have their time to shine.

Astrobite edited by Macy Huston and Jenny Calahan

Featured image credit: Sloan Digital Sky Survey  (via AAS Nova)

About Ryan Golant

I'm a third-year Ph.D. student in astronomy at Columbia University. I'm broadly interested in plasma astrophysics and numerical simulation; for my thesis, I'm combining small-scale particle-in-cell (PIC) simulations with large-scale cosmological MHD simulations to probe the ultimate origins of the Universe's magnetic fields. I completed my undergraduate at Princeton University, but I'm originally from Northern Virginia. Outside of astronomy, I enjoy playing violin and video games, learning about art history, and watching cat videos.

Leave a Reply