Void galaxies: The true island universes

Why Pisces A and B could hold the key to understanding the early evolution of dwarf galaxies.

In the early years of the 20th century, one of the major points of contention in astronomy was the nature of the objects we now know as galaxies. Some, like Harlow Shapely, called them “spiral nebulae”, believing them to lie within the confines of the Milky Way, while others, like Heber Curtis, used the term “island universes”, describing a sea of galaxies drifting through the universe. The answer to this conundrum would have profound implications for the size of the universe and cosmology as a whole.

Today, we know that galaxies do lie outside the Milky Way — which is just the galaxy we happen to live in. However, they aren’t as much like islands as some might have believed. Many galaxies exist in galaxy groups or clusters, with tens or thousands of members. We lie inside the aptly-named Local Group, along with Andromeda and Triangulum, and we’re surrounded by a swarm of satellite galaxies.

A map of voids and superclusters around the Milky Way, centered on the Virgo Supercluster. The Local Void is too small to be seen. Image credit: Andrew Z. Colvin, under the Creative Commons Attribution-Share Alike 3.0 Unported license.

There are exceptions to this rule, of course; not all galaxies are parts of clusters. In fact, a few obscure cases lurk in voids, enormous regions tens of megaparsecs across that contain almost no galaxies and only small amounts of extragalactic gas. These void galaxies are difficult to find, especially if they’re dim, but understanding how they form and evolve can lead us to a treasure trove of information about how the universe behaves on large scales — starting, of course, with voids.

Today, we travel just beyond the Local Group to the edge of the Local Void. Here lie two dwarf galaxies, Pisces A and Pisces B. They’re small and dim, and we still don’t know much about them, but what we do know has profound implications for the study of the formation and evolution of dwarf galaxies.

Gas-poor dwarfs and neutral hydrogen

Figure 1, Tollerud et al. 2015. The original radio and optical images of Pisces A and B. The resolution is pretty poor, but still yielded decent distance measurements.

Pisces A and B were discovered only a few years ago, with data from the GALFA-HI survey conducted at the Arecibo Observatory (Tollerud et al. 2015). GALFA-HI measured emission in the 21 cm line, a key indicator of neutral hydrogen clouds. Since any significant HI emission can only come from gas-rich regions, it’s a poor tool for detecting gas-depleted dwarf galaxies near the Milky Way — and an excellent way to find galaxies with plenty of hydrogen gas and ongoing star formation.

After these two radio sources were identified, optical observations with the 3.5-meter WIYN telescope found visible counterparts. Hα spectroscopy found that each optical source had a near-identical radial velocity as the corresponding radio source, confirming that they were the same objects: dwarf galaxies, at distances ranging from 1.7–3.5 Mpc (Pisces A) and 3.5–8.9 Mpc (Pisces B), placing them at the edge of the Local Group.

The discrepancies in distances came about because the group used two different methods to determine them. If we assume that the galaxies are detached from the Local Group, we can use Hubble’s law and the measurements of their recessional velocities to figure out how far away they are. However, the team was also able to discern a few young, blue stars in the galaxies; from their apparent magnitudes, they came up with different — lower — values.

Part of Figure 3, Tollerud et al. 2015. Notice how Pisces A and B are substantially hydrogen-rich compared to most dwarf galaxies of the same mass, making them easier to spot on HI surveys.

Red giants: More reliable standard candles

A year later, the team used the Hubble Space Telescope to image the galaxies (Tollerud et al. 2016). Photometry from the observations led to a third, even more precise way of measuring the distances to the galaxies: using the tip of the red giant branch (TRGB, for short).

When a red giant reaches the onset of the triple-alpha process, its luminosity increases substantially via a sudden burst of fusion called a helium flash. While the flash itself is brief, the ensuing period of evolution lasts longer, and the star gets brighter and brighter, eventually reaching a point called the tip of the red giant branch. It turns out that these stars all have basically the same luminosity, regardless of mass, metallicity or composition. This makes them excellent standard candles, just like Cepheid variables.

Fig. 1 and 2, Tollerud et al. 2016. Hubble images of Pisces A and B. While the WIYN images were good enough to discern a few point sources, Hubble was able to provide images of significantly higher resolution.

The Hubble photometry was good enough that the TRGB method could be used, and provided distances consistent with the idea that the Pisces dwarfs lie outside the Local Group: 5.6 and 9.2 Mpc, respectively. With accurate distance measurements in hand, the group was able to determine more properties of the pair, including luminosities and total stellar masses (about 10 million solar masses for each).

What it all means

What’s exciting is that the team was now able to place these galaxies on a map. It turned out that they both lie on the edge of the Local Void. Unfortunately, we don’t have detailed bulk kinematic data in the right directions, so we can’t tell for sure that Pisces A and B are moving out of the void. However, we do have a few clues.

Fig. 12, Tollerud et al. 2016. A map of nearby galaxies to within about 10 Mpc of Earth. The Local Group sits, tiny and seemingly compact, in the center.

First, the pair experienced a burst of star formation not long ago. It’s quite plausible that this was triggered by a sudden collision with the filaments of gas that form the boundaries of the Local Void. Additionally, the HI distributions match those we would expect to see in void galaxies. Finally, both of these galaxies — especially Pisces A — are smaller than most star-forming galaxies in the Local Group, meaning they’re much more compact. This might indicate that only now are they entering the same evolutionary paths that their counterparts in the Local Group traveled long ago.

Fig. 13, Tollerud et al. 2016. Notice how Pisces A and B are much smaller than most star-forming galaxies, but much brighter than most passive (non-star forming) galaxies.

It’s this final notion that makes Pisces A and B particularly exciting. If they are indeed void galaxies, they must have spent most of their lives undisturbed in the relatively pristine environment of the Local Void. This means they might look and act a lot like normal dwarf galaxies used to, billions and billions of years ago. Less than 10 Mpc away, then, could lie clues to the early lives of many of our closest neighbors. That’s a pretty exciting thought.