On the Origin of Galaxy Bars

Classifying galaxies based on their morphology (their shape or form) is a long-standing tradition in astronomy. The most famous classification scheme is that created by the famous astronomer Edwin Hubble: known as the Hubble sequence, it divides galaxies into either spirals or ellipticals, with further distinctions based on how spherical or elongated a galaxy is, or whether the spirals show the presence of a central bar. While classifying galaxies according to their forms can be a fun task in and of itself (check out how many people participate in Galaxy Zoo), characterizing galaxies based on their morphology can also give us clues about how they form and evolve. For example, spiral and elliptical galaxies tend to be found in different environments in the universe – ellipticals are preferentially found in dense regions, whereas almost all galaxies in very underdense regions are disks (a more generic name for spiral galaxies). That sort of environmental distinction brings up the question of galactic nature vs. nurture; do ellipticals prefer to form in dense regions, or does a high density environment turn galaxies into ellipticals?

Hubble images of six of the galaxies in the sample, showing examples of the different classes each galaxy was placed into. (Note, figure modified from original.)

In this study, the authors examine a sample of 257 disk galaxies, searching for the factors that influence the presence or absence of a central bar of stars, one of the defining features of the Hubble sequence. Using Hubble Space Telescope images, the authors classified every galaxy in the sample according to one of the following types, shown in the figure at right: unbarred spiral (126), long bar (28), short bar (20), clumpy (22), chain (12), or compact (49). The authors then looked for trends among the different classes that would explain why some had bars and some did not.

Among the features that they investigate are each galaxy’s kinematics, specifically its rotational velocity and its random velocity, also called velocity dispersion. To get the rotational and random velocities, the authors looked at several bright emission lines in each galaxy’s spectrum. This is a way to measure whether a galaxy is supported by rotational motion, like planets in Keplerian orbits around a star, or by pressure, like gas molecules bouncing around in a box. Galaxies that are supported by ordered rotational motion are referred to as kinematically “colder”, while those that are supported by random motions are kinematically “hotter”.

The main result of this study is that central bars are never found in disks that are supported by random velocity dispersion. The figure shown below displays this by placing the galaxies on a plot showing the Tully-Fisher relation. The Tully-Fisher relation is a correlation between a galaxy’s luminosity (which is proportional to stellar mass), and its rotational velocity; roughly, the luminosity is proportional to rotational velocity to the fourth power. What the plot shows is that bigger, kinematically cooler galaxies (the spirals and barred spirals in the sample) fall nicely into the region of the graph described by the Tully-Fisher relation. Those galaxies that lie outside the Tully-Fisher relation tend to be pressure supported and less massive.

A plot showing the measured rotational velocity of each galaxy in the sample versus its stellar mass. The dashed lines show the scatter around the Tully Fisher relation. Galaxies in the different sample classes tend to reside in different locations on the plot.

This result leads the authors to describe an evolutionary scenario for disk galaxies. The compact, clumpy galaxies in the sample are an early phase of spiral galaxies, undergoing large bursts of star formation in a dynamically unstable gas disk. As these young galaxies accrete more cold gas from their surroundings, they will become larger, rotationally supported disks, and will thus be found farther to the upper right on the plot of the Tully-Fisher relation. The authors note, however, that while their results indicate that stellar mass and rotational support appear to be necessary conditions for bar formation, they are not sufficient. There are still a large number of massive, rotationally supported galaxies in the sample that do NOT appear to have a central bar – and the reason remains unknown.


About Evan Schneider

I am a graduate student at the University of Arizona working on high resolution simulations of galactic winds, run with my new hydrodynamics code, Cholla. When I’m not doing astronomy, I enjoy essentially any outdoor activity, including hiking, rock climbing, and walking my dogs. On indoor days I can often be found reading, and of course, drinking coffee.

Leave a Reply