Death in the Sky: M31 Shreds its Satellites

An international team of astronomers has identified two new tidal streams in M31, the Andromeda galaxy. They are more-or-less intact remnants of dwarf galaxies that M31 has otherwise ripped to shreds.

One team – using the Suprime-Cam camera on Subaru – discovered two new dwarf galaxy shards by mapping the sky density of red giants in M31’s outskirts; the other – using the DEIMOS spectrograph on Keck II – separated the M31 red giant wheat from the Milky Way chaff.

In a project led by collaborators Mikito Tanaka and Masashi Chiba of Tohoku University, Japan, the astronomers used the Subaru 8-meter telescope and Suprime-Cam camera to map the density of red giants in large portions of M31, including the hitherto uncharted north side. This led to the discovery of two tidal streams to the northwest (streams E and F) at projected distances of 60 and 100 kiloparsecs (200,000 and 300,000 light-years) from M31’s nucleus. The study also confirmed a few previously known streams, including the little-studied diffuse stream to the southwest (stream SW), which lies at a projected distance of 60 to 100 kiloparsecs (200,000 to 300,000 light years) from M31’s nucleus.

The Spectroscopic and Photometric Landscape of Andromeda’s Stellar Halo (SPLASH) collaboration, a large survey of red giants in M31 lead by Puragra Guhathakurta, professor of astronomy and astrophysics at the University of California, Santa Cruz, has followed up with a spectroscopic survey of several hundred red giants in Streams E, F, and SW, using the Keck II 10-meter telescope and DEIMOS spectrograph at the W. M. Keck Observatory in Hawaii. Analysis of the spectra from this survey yields estimates of the line-of-sight velocity of the stars, which in turn allows M31 red giants to be distinguished from foreground stars (in the Milky Way). The spectral data confirmed the presence of coherent groups of M31 red giants moving with a common velocity.

Distribution of line-of-sight velocities in the Stream SW field (Raja Guhathakurta)

Stars spread over the vast reaches of a halo in a big galaxy like the Milky Way or M31 are characterized by old age, few elements other than helium and hydrogen (i.e. low metallicities; astronomers call all elements other than hydrogen and helium “metals”), and high velocities. The exceptional nature of these halo stars, when compared to stars in a galaxy’s disk, reflects the early dynamics and element formation of the galaxy when its appearance differed significantly from what we see today. Consequently, the halo provides important insights into the processes involved in the formation and evolution of a massive galaxy. In the best Big Bang model we have today – ΛCDM (Lambda Cold Dark Matter) – the outer halos are built up through the merger and dissolution of smaller, dwarf, satellite galaxies. “This process of galactic cannibalism is an integral part of the growth of galaxies,” said Guhathakurta.

The smooth, well-mixed population of halo stars in these large galaxies represents the aggregate of the dwarf galaxy victims of this cannibalism process, while the dwarf galaxies that are still intact as they orbit their large parent galaxy are the survivors of this process.

“The merging and dissolution of a dwarf galaxy typically lasts for a couple billion years, so one occasionally catches a large galaxy in the act of cannibalizing one of its dwarf galaxy satellites,” Guhathakurta said. “The characteristic signature of such an event is a tidal stream: an enhancement in the density of stars, localized in space and moving as a coherent group through the parent galaxy.”

Tidal streams are important because they represent a link between the victims and survivors of galactic cannibalism – an intermediate stage between the population of intact dwarf galaxies and the well-mixed stars dissolved in the halo.

The Andromeda galaxy is a unique test bed for studying the formation and evolution of a large galaxy, said Guhathakurta, “Our external vantage point gives us a global perspective of the galaxy, and yet the galaxy is close enough for us to obtain detailed measurements of individual red giant stars within it.”

One of the next steps will be to measure the detailed elemental compositions (“chemical properties”, in astronomer-speak) of red giants in these newly discovered tidal streams in M31. Comparing the chemical properties of tidal streams, intact dwarf satellites, and the smooth halo will be of particular significance, Guhathakurta said. Mikito Tanaka put it this way: “Further observational surveys of an entire halo region in Andromeda will provide very useful information on galaxy formation, including how many and how massive individual dwarf galaxies as building blocks are and how star formation and chemical evolution proceeded in each dwarf galaxy.”

At the present time, detailed studies of the chemical properties of tidal streams, intact dwarf satellites, and smooth stellar halos are possible only in the Milky Way and M31 galaxies and their immediate surroundings. Existing telescopes and instruments are simply not powerful enough for astronomers to carry out such studies in more distant galaxies. This situation will improve greatly with the advent of the planned Thirty Meter Telescope later in this decade, Guhathakurta said.

Tanaka’s team published their survey results in a recent Astrophysics Journal (ApJ) paper (the preprint is arXiv:0908.0245), and Guhathakurta’s team presented their results on the newly discovered tidal streams earlier this month at the 215th meeting of the American Astronomical Society in Washington, D.C.; they hope to have an ApJ paper on these results published later this year. You can read an earlier SPLASH paper, “The SPLASH Survey: A Spectroscopic Portrait of Andromeda’s Giant Southern Stream”, published in ApJ (the preprint is arxiv:0909.4540).

Sources: University of California, Santa Cruz, National Astronomical Observatory of Japan.

4 Replies to “Death in the Sky: M31 Shreds its Satellites”

  1. Roger Penrose proposed something called the Weyl curvature hypothesis. This says that the Weyl curvature, that portion of the Riemann curvature involved with tidal forces and the distention of material into ellipsoidal configurations, increases with time. This is a measure of the entropy of the universe, and the occurrence of event horizons in black holes produces such. These mergers of galaxies illustrate large tidal disruptions. There is a bit of a problem or question. If 80% of the mass of galaxies is dark matter, which interacts with everything very weakly, then the blob or halos of dark matter would largely pass through each other and continue their motions according to conservative dynamics. Something similar to this is seen in the Bullet galaxy cluster. Yet in these coalescences for the Weyl curvature hypothesis to be correct, and for these mergers to form larger elliptical galaxies, the non-conservative “friction” which results from luminous matter (LM) interactions must be sufficient to result in some long term gravitational pull on the dark matter (DM). In other world the DM + LM has sufficient attenuation or dissipation of energy so that by gravitational means the two halos of DM end up in a single “blob.” Without the frictional processes of LM the DM halos would pass through each other so there would be no coalescence and Penrose’s hypothesis is then false. Hence the existence of LM might be considered as a way in which the universe is “fine tuned” so that they Weyl curvature, call it C in accord with most general relativity literature, obeys dC/dt >= 0 and is identified with the second law of thermodynamics.


  2. Great article, lousy illustration (that accompanied the PR, at top). I would urge the curious to take a look at Fig 1 in the Tanaka paper (pg 6). This labelled image shows (among other things) Streams A-D and the ‘Giant Stream’ thought to have resulted from M110s recent plunge through the disk of M 31. The bright greenish blob in the lower left corner is M 33. A stream of stars and neutral hydrogen (not shown) between M 31 and M 33 has also been discovered, pointing at a past encounter. A properly labelled image illustrates just how complex the outer regions of the Andromeda Galaxy are.

    LC, it should probably be noted that in this case, dwarf galaxies being accreted onto the halo of our galaxy and Andromeda will probably not result in the formation of an elliptical galaxy, as they (of course) collectively lack the mass. Interactions between the more massive galaxies in the Local Group (MWG, M 31, and M 33) will probably eventually result in the formation-transition to an elliptical.

    You bring up a good point regarding LM-DM interactions. Over the past decade or so, observations have inferred that these systems are dark matter dominated, with mass-to-light ratios from 100 to 1000:1. It’s thought that this enormous dark matter content is responsible for holding these galaxies together as best they can as they travel through the disk and halo of our galaxy. Stuff that keeps astronomers on their toes – a few Milky Way dwarf galaxies exhibit M/L ratios closer to the value for our galaxy.

  3. Penrose is a mystic, so it is hard for a layman to know when his results are physics (say, Penrose tilings as found in quasi-crystals) or bogus (say, the idea that the Church-Turing thesis doesn’t apply for biological brains, so that the algorithmic hierarchy of computer science collapses and all sorts of funky stuff including time travel is possible, et cetera.)

    I’m not trying to poison the well, just reporting a genuine difficulty.

    I’m sorry, but yhe Weyl curvature hypothesis reads to me like the second category, as we can’t identify a microscopic law on energy level availability with a law on macroscopic matter configurations. That the arrow of time from the 2nd law of thermodynamics (2LOT) is aligned with the arrow of cosmological time is simply because inflationary cosmology expands our universe which drives the 2LOT (by dissipation “out there”).

    Speaking of entropy, it’s laws and finetuning, I’m currently reading Boussou et al’s latest paper on the causal entropic principle. (A.K.A. the anthropic principle in an environmental form.) They predict all sorts of seeming finetuning, such as we live in a at universe at the onset of vacuum domination, without any need for specific mechanism such as inflation.

    Besides 6 direct falsification tests integrating at 10^-7 level (that is finetuning indeed!), they find that either we are among the least complex observers possible or we are among the most complex. Um, yes. :-/

    Either way, entropy and the cosmological constant, not curvature, is correlated by that. (pp 34-35.)

    As no other theory predicts anything here (the old “there can be only one” unique outcome of fundamental physics conspicuously still among them) but causal entropic principles now falsifiably is tested on 7 physics finetunings (including the old CC of garden variety of AP), I feel fairly confident with the results.

    [Btw, I also like the way this leads up to their conclusion that “this [“that captures the rarity of vacua with particle physics and cosmological parameters that admit complex phenomena such as observers”] could explain the origin of a kind of ur-hierarchy in physics, from which all other hierarchies may be derived either by symmetry arguments of [sic] anthropic correlations.”

    Then spontaneous symmetry and symmetry breaking to get laws, and entropy or in other words inflation to get space, time and finetuning, that’s all the initial condition and process we observers need. :-D]

  4. These dwarf galaxies will no of course result in M31 becoming a dwarf galaxy of course. Yet these probably have their own DM halos, which will not automatically coalesce with M31 halo. The luminous material may stick due to “friction” from the collision, but since DM is very weakly interacting these halos will continue on their “merry way,” so to speak. Now inside a halo Gauss’ law gives a gravitational force that is F ~ -x, it is not hard to show, but this will still act as a conservative force, or one shich conserves energy. So I would think, at least naively, orbits of halos would more or less continue to orbit along their Newtonian paths. The Bullet cluster illustrates by gravitational lensing how DM halos are leaving behind the LM material of two galaxies.

    This has for me raised some curious questions. It might mean that a galaxy which grows by cannibalizing dwarf galaxies or by coalescences ends up with a high LM to DM ratio if the DM halos end up in more or less their original orbits. This would I think change the over all dynamics or configuration of a galaxy as observed from its LM material. Either that, or the accumulation of LM into larger galaxies provide enough gravitational potential in a region to hold at least some of this DM in a growing “master halo.”

    Penrose has gotten somewhat mystical, in particular with his Platonic ideas of mind, mathematics and matter. He also sort of went down a curious path with ideas of quantum consciousness and state reductions by quantum gravity. Yet that aside, his Weyl curvature conjecture is worth considering. I suspect that some how DM halos in galaxy clusters do eventually coalesce into larger ones, and either decay, say by neutralino decays (a supersymmetric field which might compose much of DM) or eventually end up being consumed by the galactic black hole at the center. Erick Verlinde has written an interesting article on the “entropy force of gravity.” and this is creating a bit of a firestorm these recent weeks. I think this might address some of your issues with the Weyl curvature conjecture.

    Thanks for the Bousso paper, somehow this escaped my attention. All of these do suggest some element of tuning. There is a converse relationship between symmetry and complexity. The “master symmetry,” what ever that might be, preserves the structure of a vacuum configuration — preserving a void so to speak. Yet there are scalar fields called dilatons which define a conformal symmetry in this group. A quantum fluctuation may break this master group so that this dilaton freezes out this symmetry at different length scales. This results in a symmetry breaking decomposition of this master group, with the emergence of structure and effectively what might be called classical information. This then suggests some converse relationship between a master symmetry and the level of complexity which can exist at lower energy. I suspect there is some extremal principle here, where the tuning is some variational principle on a partition function which results in a greatest upper bound on the measure of complexity which can exist in a local region.


Comments are closed.