Galactic environments, clustering and mergers

Graeme Ing, HET604, Swinburne Astronomy Online

Introduction

We observe galaxies distributed across the universe in a hierarchy of groupings, from a small group of under ten galaxies, to clusters of a thousand or more; up to huge structures known as superclusters, home to tens of thousands of galaxies. This article will examine the properties of each of these groupings in order to answer the question: Which of these environments is most conducive to two galaxies merging after a collision?

Isolated Galaxies

Though galaxies tend to be gregarious, a small percentage of galaxies exist in isolation, not as part of a group or cluster. One definition of an isolated (“field”) galaxy is one uninfluenced gravitationally by a neighbour in the last “few” billion years. Varela et al. (2003) observed that isolates were largely Sc-type spiral galaxies (Sc-types exhibit widely spread, loose spiral arms) and likely formed alone in low-density regions from clumps of dust and gas left behind after the formation of groups and clusters. They preferred this model to that of a galaxy gravitationally ejected from its home group, out into the intergalactic medium.

Groups

We refer to the smallest association of galaxies as a Group. Typically, it consists of 10 to 50 member galaxies, in a region 1 to 2 Megaparsecs (Mpc) across. The total mass of a group is usually 10¹² to 10¹³ times the mass of the Sun (solar masses). (WikiWeb) Group members are gravitationally bound, i.e. mutual gravitational attraction holds them together in the same region of space. We can sub-categorize groups into loose and compact depending upon their size and density.

In a loose group, galaxies are sprawled across large distances. Our galaxy, the Milky Way is a member of such a loose group: the Local Group. The Local Group currently comprises about 40 members, mostly minor or dwarf galaxies attracted by gravity to the three largest members: the Milky Way, M31 in Andromeda and M33 in Triangulum. (Freedman & Kaufmann 2001) It has a diameter of ~2 Mpc and the distance between the Milky Way and M31 is almost half this distance. According to Sparke and Gallagher (2005), the Local Group masses greater than 3x10¹² solar masses, enough to cause the group to slowly collapse inward under gravity.

A compact group is much smaller and denser, typically consisting of 4 to 7 members covering only a few hundred Kiloparsecs (Kpc); a fraction the volume of the Local Group. Member galaxies are often separated by little more than their own diameter! Hickson (1999) conducted the first major study of compact groups in 1982, and his first discovery, Stephan’s Quartet is one of the best examples. For this reason, many compact groups are known as Hickson Compacts. According to Hickson, compact groups are the densest known galactic systems, and may represent the next evolutionary state of a loose group that collapses in upon itself over time.

The velocity dispersion (the spread of the velocities of galaxies within the group) of groups is low, <200 kms^-1 on average. Hickson (1999) reasoned that due to the low velocities and high density within a compact group, galactic interactions and mergers are commonplace. He also observed that compact groups contained significantly more elliptical galaxies than in other groups. Since one theory of giant elliptical formation speculates that these huge and dense galaxies form from the repetitive merging of other galaxies, this could be further evidence of galactic merging in a compact group.

According to Barnes (1985), galaxies with massive extended haloes of dark matter assist in galactic mergers. The high mass of the dark matter gravitationally slows interacting galaxies sufficiently for them to enter into a close orbit around one another, and finally merge when their orbits decay further under the drag of the dark matter. Other evidence, however, suggests that massive dark matter halos actually prevent mergers.

We also observe an extended halo of gas surrounding many giant elliptical galaxies. This hot gas radiates in X-ray wavelengths at temperatures of 10⁷ or 10⁸ K. The likely cause of such high temperatures is kinetic energy released in a galactic collision, and it is possible that this halo represents the original size of a group or subgroup, whose member galaxies fell into and merged with the elliptical. A Fossil Group is such a group remnant, that takes the form of an isolated giant elliptical galaxy surrounded by a hot halo of gas. Perhaps it consumed its other group members long ago. (CosmosWeb; Freedman & Kaufmann 2001)

Clusters

The next largest galactic construct is a cluster. In 1958, George Abell constructed a catalogue of thousands of clusters, classifying them into four types: Regular and irregular, and rich and poor. Abell classified the richness of a cluster by the number of galaxies inside of the “Abell Radius”, a sphere of 1.5 Mpc radius centered on the cluster core. (Abell et al. 1989) Whilst his classification is the most common method of distinguishing clusters, other classification schemes include Zwicky, who subdivided clusters into compact, medium-compact and open; Rood and Sastry, whose classification is based upon the distribution of the ten brightest galaxies in a cluster; and Bautz-Morgan, a method based upon the brightness ratio between the two most dominant galaxies in the cluster. (Keel 2003)

In general, clusters number 50 to 1000 galaxies; have a diameter of 2 to 10 Mpc and a total mass of 10¹⁴ to 10¹⁵ solar masses. (WikiWeb) Many clusters are so large that the universe is of insufficient age for outlying galaxies to have fallen into the centre under gravity, preventing us from predicting the future morphology of large clusters. (Sparke & Gallagher 2005) Clusters are said to have a high “mass to light ratio” meaning that their huge mass cannot be accounted for by direct observation. We suspect that the distribution of mass in a typical cluster is 1-2% stars, 11% intracluster gas and 85-87% dark matter. (IAXWeb)

A poor cluster is little larger than a group and the distinction between them is often unclear. Typical membership is 50 to 100, (compare to the Local Group’s 40), and poor clusters are the most dominant clusters in the viewable universe. By comparison, a rich cluster can number as many as 1000 galaxies within the same volume of space as a poor cluster. A particularly rich cluster is Coma, 90 Mpc away, suspected of numbering 10,000 galaxies! (Freedman & Kaufmann 2001) The nearest rich cluster is Virgo, 16 Mpc away, comprised of over 2000 galaxies in a diameter of ~3 Mpc centered on the giant elliptical M87.

A regular cluster is generally spherical or symmetrical in shape, often with a well-defined nucleus of closely grouped galaxies. The nucleus galaxies are usually giant elliptical, classified cD, or S0 galaxies, whilst the outer lying members of the cluster are predominantly spirals. According to Waller and Hodge (2003), the nucleus of the Coma cluster contains 3000 galaxies per cubic Mpc, the galactic equivalent of a globular star cluster. The average distance between these galaxies would be only 150,000 light years, about the distance between the Milky Way and its very close satellite galaxy, the Large Magellanic Cloud.

An irregular cluster, as the name suggests, has no uniform shape. They are usually smaller in density and members (less than 1000) than regular clusters, with a lower total mass of ~10¹² to 10¹⁴ solar masses. Whilst they lack a well-defined nucleus of galaxies, they may possess multiple regions of higher galactic density. The ratio between ellipticals and spirals is higher than in a regular cluster, tending toward more spiral and irregular galaxies than ellipticals. (Waller & Hodge 2003)

The velocity dispersion of clusters ranges from 100 kms^-1 to in excess of 1000 kms^-1, with a median of ~800 kms^-1. (Sparke & Gallagher 2005) We expect regular interactions and collisions in rich regular clusters due to their density, particularly in the nucleus. Whilst the existence of giant ellipticals may provide evidence of past collisions and mergers, the high velocity of member galaxies suggests that mergers are unlikely. At high relative velocities, interacting galaxies are more likely to pass through each other, distorting their shapes and streaming their galactic disks into the intracluster medium. These galaxies probably lose insufficient kinetic energy in the collision to merge.

Sometimes several galaxies in a cluster gravitationally attach themselves to a nearby dominant galaxy. This small grouping of galaxies is termed a sub cluster. The Virgo cluster has three sub clusters centered on M87, M86 and M49. The Local Group has a similar structure, known as a subgroup, taking the form of several dwarf galaxies orbiting M31 in Andromeda. (Sparke & Gallagher 2005)

Superclusters

A supercluster is a huge cluster of clusters, typically containing 10 clusters. A supercluster may have a diameter of 10 to 100 Mpc. and a total mass of 10¹⁶ solar masses. These structures are so large that whilst they are likely gravitationally bound, the universe is not yet old enough for the member clusters and galaxies to have settled into equilibrium. A member cluster might require in excess of 300 billion years traversing or orbiting a supercluster, which might explain why superclusters are irregular in shape and possessing no definite nucleus. (Waller & Hodge 2003)

Our Local Group is part of the Virgo supercluster, thought to encompass all galaxies within 20 Mpc, comprising ~100 clusters. In turn, we consider Virgo part of the immense Hydra-Centaurus supercluster. (Waller & Hodge 2003) Figure 1 gives an indication of the size and location of all clusters and superclusters within 80 Mpc of the Milky Way.

Like a subcluster, a denser region of a supercluster is termed a galaxy cluster cloud. This is a rich grouping of several clusters.

Figure 1: Nearby superclusters (Hudson M., 1993, MNRAS 265 42)

Filaments, sheets and voids

On a universal scale, superclusters appear to form along dense edges or lines surrounding largely empty holes called voids. We refer to these edges as filaments, sheets or Walls and they are probably the largest structures in the universe, hundreds of Mpc in length and ~20 Mpc in thickness. The voids that they surround are generally spherical with radius 10 to 50 Mpc, and devoid of anything other than isolated field galaxies or very poor clusters. Voids represent 90% of volume in the universe. This bubble like structure to the universe is likely a direct result of the condensing of matter from tiny density fluctuations, and its subsequent expansion, in the early universe. (CosmosWeb)

Conclusion

It is clear that the distribution of galaxies in the universe is not uniform and that most galaxies formed in groupings according to the irregular clumping of matter in the early universe. Whilst many of these groupings are gravitationally bound, the universe is likely too young for gravity alone to have caused galaxies to collect into groups.

Interactions between galaxies may be commonplace in the densest groups and clusters, but the outcome of such interactions depends largely upon the galaxy’s velocity and it’s mass of dark matter. A low kinetic energy collision, typically less than 200kms^-1 is required for the interacting galaxies to enter into a mutual, decaying orbit that ends in a merger. The dark matter halos of the galaxies play a large part in inducing drag during the collision, and the hot gas evident at X-ray wavelengths indicates the large energies output from galactic collisions. Collisions with galactic velocities in excess of 200kms^-1 usually result in too much kinetic energy, allowing the galaxies to pass through each other without merging, though the disruption to each galaxy can be devastating.

Whilst there is evidence of mergers within the densest rich clusters, compact groups provide the most conducive environment for galactic mergers due to galactic density and very low velocity dispersions.

References

Abell G., Corwin H., Olowin R., 1989, “A catalog of rich clusters of galaxies”, 1989ApJS…70…1A

Barnes J., 1985, “The dynamical state of groups of galaxies”, 1985MNRAS 215 517B

CosmosWeb: Cosmos: SAO Encyplopedia, http://cosmos.swin.edu.au/

Freedman R., Kaufmann W., 2001, “Universe, 6^th edition”

Hickson P., 1999, “Compact groups of Galaxies”, 1999IAUS 186 367H

IAXWeb: “Clusters of galaxies: Introduction”

Cambridge University Institute of Astronomy X-Ray group 2005

http://www-xray.ast.cam.ac.uk/xray_introduction/Clusters_intro.html

Keel, 2003, “Groups and clusters of galaxies”, http://www.astr.ua.edu/keel/galaxies/clusters.html

Sparke L., Gallagher J., 2005, “Galaxies in the Universe”

Varela J., Moles M., Marquez I., Galletta G., Masegosa J., Bettoni D., 2003, “Properties of isolated disk galaxies”, 2004A&A…420..873V

Waller W., Hodge P., 2003, “Galaxies and the cosmic frontier”

WikiWeb: “Galaxy groups and clusters”, http://en.wikipedia.org/wiki/Galactic_groups