Searching for Fossil Star Streams in Nearby Galaxies

Fig. 1: An illustration of what happens when a dwarf satellite falls within the gravitational influence of a massive galaxy. The dwarf galaxy, visible in its original configuration (a ball of stars) at the top of the picture, falls towards the massive galaxy along the dashed line and is torn apart by immense tidal forces. The stars ripped from the dwarf spread out in long streamers along its path. Theory predicts that Milky Way-like galaxies will have experienced a hundred such cannibalization events over their lifetime (courtesy Kathryn Johnston, Wesleyan University).

Fig. 2: The surface density of red giant branch stars around M31 mapped using the INT Wide Field Camera. A true colour image of the galaxy is superposed at the center of the map to illustrate sizes. Our map covers a region of roughly 100 x100 kpc centered on M31. The giant stellar stream and putative loop of stars from NGC205 are indicated. Many other inhomogeneities are seen in the stellar distribution.

Fig. 3: A similar map made for the other Local Group spiral, M33. In this case, the map covers a region of roughly 40 x 40 kpc. Contrary to the M31, M33 exhibits no irregularities in its stellar distribution at large radius. Instead the density of stars falls off very smoothly with radius.

Galaxies are the basic structural units of the Universe and are composed of a minority component of normal matter, i.e. stars and gas, and a majority component of cold `dark matter -- the presence of which is inferred only from its gravitational effects. Over the past decades, astronomers have strived to construct a model of galaxy formation which can simultaneously account for large scale distribution of galaxies in the Universe, as well as the detailed properties of individual galaxies. Supercomputer simulations have enabled theorists to follow the evolution of matter in the Universe from a short time after the Big Bang until the present epoch (see linkPfeil.gifThe largest N-body simulation of the universe). These simulations indicate that galaxy formation in the presence of cold dark matter occurs from the bottom up; that is, the first galaxies to form are small dwarf systems and these subsequently merge together to form progressively larger systems. In this scenario, galaxy formation can be viewed as a continual process. Large galaxies, like our own Milky Way, are predicted to have cannibalized roughly a hundred small dwarf galaxies which have fallen within their gravitational influence (a process which is expected to continue to the present day). During destruction, stars from the satellite galaxy are pulled out into long tidal streams which can remain coherent for several billion years and leave a long-lasting observable signature of the mass accretion (see Figure 1).

This picture of galaxy formation makes a number of well-defined predictions which can be tested against observations. There is a particular need to test the model on small 'galactic-sized' scales where complicated astrophysics, such as star formation and feedback, play an important role alongside gravity. The most detailed information we can gain about how a galaxy has formed and evolved comes from resolving it into individual stars and studying the distributions of these stars in space, as well as in age, heavy metal content and velocity. Old and intermediate-age stars are especially useful since they probe conditions more than 5 billion years ago when they were born and when galaxies were in an early stage of evolution. Studies of this `fossil record' have traditionally focused on our own Milky Way, where the quantity and quality of available observations is unsurpassed. With new telescopes and instruments, it is becoming possible to extend this type of work to other nearby galaxies. This is of great importance in order to understand both how representative our Milky Way is and how the nature of the assembly process varies with galaxy mass, morphological type and local environment.

Scientists at the Max-Planck-Institute for Astrophysics are participating in an international collaboration to explore the fossil record of galaxy formation and evolution in our two nearest giant neighbours, M31 and M33. Located at roughly 800 kiloparsecs (2.5 million lightyears) from us, these galaxies are close enough to resolve into individual stars with ground-based telescopes such as the Isaac Newton Telescope on La Palma. The researchers are conducting wide-field imaging surveys of red giant branch stars in the faint outer regions of these galaxies and finding some surprising results. Figure 2 shows that the stellar distribution in the far outer regions of M31 is distinctly inhomogeneous. There is a large stream of stars extending to the south-east; distance estimates indicate this stream begins 100 kpc behind M31. There is also a loop of stars which appears to emanate from the dwarf satellite system NGC205, as well as a variety of other stellar overdensities. The Advanced Camera for Surveys on board the Hubble Space Telescope has been used to obtain detailed follow-up observations of many of these features and reveals distinct variations in the age and metallicity mix of the constituent stars. Taken together, these results indicate that the outer regions of M31 contain tidal debris from at least one, and possibly more, cannibalized dwarf galaxies. On the other hand, an identical study of the lower mass galaxy M33 reveals a very smooth stellar distribution at large radius and no obvious signs of any substructure (see Figure 3). Contrary to simple expectations, M33 appears to be a system which may not have accreted any significant (luminous) mass for much of its lifetime.

Understanding the origin of the differences between M31 and M33, as well as which behaviour is more typical, will require both additional observations and improved theoretical predictions. The MPA researchers have already been awarded observing time with ESO's VLT and Subaru's SuprimeCam to conduct similar observations for a larger sample of nearby galaxies.


Annette Ferguson

Further Information:

linkPfeilExtern.gifIsaac Newton Group of Telescopes

linkPfeilExtern.gifThe Hubble Space Telescope

linkPfeilExtern.gifThe Subaru Telescope

A. M. N. Ferguson, M. J. Irwin, R.A. Ibata, G. F. Lewis & N. R. Tanvir: Evidence for Stellar Substructure in the Halo and Outer Disk of M31, 2002, AJ, 124, 1452

A. M. N. Ferguson, R. A. Johnson, D. C. Faria, M. J. Irwin, R. A. Ibata, K. V. Johnston, G. F. Lewis & N. R. Tanvir: The Stellar Populations of M31 Halo Substructure, 2005, ApJL, submitted

A. M. N. Ferguson, M. J. Irwin, R.A. Ibata, G. F. Lewis, A. McConnachie & N. R. Tanvir: A Global Map of the Stellar Populations In and Around M33, 2005, MNRAS, submitted