Previous Topic: Age Estimation Lecture Index Next Topic: Other Globular Cluster Systems Next (Working) Topic: Population Synthesis

ASTR 5610, Majewski [SPRING 2016]. Lecture Notes

ASTR 5610 (Majewski) Lecture Notes


GLOBULAR CLUSTERS AS STELLAR POPULATION TRACERS: THE HALO


Galactic Spatial Distribution

The Globular cluster system of the Milky Way globular cluster system has been traditionally divided (Zinn 1985) into a disk and halo system, near [Fe/H] = -1.0.

From Carney 2001, in Star Clusters, SAAS-FEE Advanced Course No. 28, eds. B.W. Carney and W.E. Harris, Springer-Verlag. Originally in Zinn (1985).
The disk system is clear as a flattened distribution.

From Majewski (1999, in Globular Clusters , eds. C. Martinez-Roger, I. Perez Fournon & F. Sanchez) based on data from Harris (1996).
The halo system is more or less spherically distributed.

From Harris 2001, in Star Clusters, SAAS-FEE Advanced Course No. 28, eds. B.W. Carney and W.E. Harris, Springer-Verlag. Originally in Zinn (1985).
The outer halo (RGC > 8 kpc) distribution follows an RGC-3.5 power law.

From Harris 2001, in Star Clusters, SAAS-FEE Advanced Course No. 28, eds. B.W. Carney and W.E. Harris, Springer-Verlag. Originally in Zinn (1985).
The census of clusters is probably only incomplete for small, distant clusters (but just five new ones in the last 20 years) and extremely dust-obscured clusters.


Metallicity Trends

Kinman (1959a,b) systematically studied the chemodynamical properties of globular clusters. A few general trends became lore:

All of the above trends seem to fit nicely within the Eggen, Lynden-Bell & Sandage (1962) picture of a rapid collapse with progressive enrichment and spin-up.

But as more clusters were added to the database, the picture became more complex.


Searle & Zinn (1978)

The growing realization that things were more complex than the simple ELS collapse picture culminated with the paper by Searle & Zinn (1978).

These authors noted three observations about the Milky Way globular clusters.

  1. All globulars are internally chemically homogeneous ("proof" that they each formed quickly and are SSPs).
  2. There is no metallicity gradient in the outer (RGC > 8 kpc) cluster system.
  3. Searle & Zinn's (1978, ApJ, 225, 357) analysis of the metallicity distribution with Galactocentric radius.
    From this finding, Searle & Zinn infer:

  4. There is a significant spread in the second parameter effect in the outer halo system.
  5. Searle & Zinn's (1978, ApJ, 225, 357) analysis of the metallicity versus HB type for clusters as a function of Galactocentric radius.
    A more modern version of the above analysis, shows the distances of the extreme second parameter clusters, and includes clusters farther than Searle & Zinn could study:

    Note the distances of the second parameter clusters, which put them predominantly in the outer halo. From Carney 2001, in Star Clusters, SAAS-FEE Advanced Course No. 28, eds. B.W. Carney and W.E. Harris, Springer-Verlag. Originally in Zinn (1985).
    From this finding, Searle & Zinn infer:

  6. If you compare the cluster metallicity distribution function (MDF) to the results of a ``closed box model" for evolution of the halo (clusters and subdwarfs), you find that you need to have the proto-galaxy evolve completely -- i.e. all gas converted to stars -- while in halo forming phase to match what you see.
  7. Searle & Zinn's (1978, ApJ, 225, 357) analysis of the metallicity distribution function observed versus that modeled by various closed-box chemical evolution scenarios. The best match is obtained for when all gas gets used up in the formation of the clusters.

    From this finding, Searle & Zinn infer:

Under the assumption that age is the second parameter, the last point suggests a slower formation of the outer halo MWGC system.

The lack of a metallicity gradient also suggests a more chaotic formation, inconsistent with ELS rapid collapse.

Searle & Zinn concluded that the formation of the halo must have been a slower, more prolonged process that was underway even as the inner Galaxy continued to do its own thing (e.g., ELS collapse).

The "protogalactic fragments" are vaguely described by S&Z (although gas-rich irregular galaxies mentioned), but many today interpret them to have been dwarf galaxies (maybe like the SMC).

Sandage (1990, JRASC) has proposed a modified version of the ELS model.


Early Clues of Cluster Accretion


The Modern Halo Cluster Formation Context

Recall Zinn's (1993) division of clusters into disk, young halo and old halo.


These groups show differences in dynamics (note analysis based solely on radial velocities):

Specifically:

A halo formation model like Zinn's, in which both accretion and collapse have played a role and have left behind distinctive populations, are part of a new wave of dual halo models that are now popular (and perhaps prefaced by Sandage's [1990] proposal of "ELS + noise").


Models of Satellite Accretion

Click here (surface density coded) and here (color coded by stripping time) to see a model of how satellite accretion occurs in a Milky Way Galaxy (animation by Kathryn Johnston).

Recall the reason for formation of two tails:

Here is yet another model showing dark matter (animation by Stelios Kazantzidis).


Evidence Within the Halo Field Star Sample

If the cluster system owes its origin partly from the accretion of satellite galaxies (protogalactic fragments), then so too should the halo field star population.

So it is relevant at this point to look at some properties of the halo field star population to see what they tell us about an accretion origin.

First, what about the metallicity distribution?


Is there a metallicity gradient in the outer halo field stars?


The clusters show a metallicity-velocity relationship for the Old Halo sample. Is anything like that seen for the halo field stars?

Three different assessments of the run of rotational speed with metallicity from studies of nearby, high proper motion stars. From Carney (2001, in Star Clusters, Springer-Verlag).


In the Zinn Young Halo clusters, a distinct kinematical signature was found -- a retrograde rotation. What about field stars?


Age Spread in Halo Field Stars

The "young halo" clusters show evidence of actually having an age spread and being younger on average. What about the field stars?


Finally, we have seen evidence for possible dynamical families of globulars clusters in the halo. Can we see evidence for this in the halo field stars?


Direct Proof of Halo and Cluster Build-Up From Current Satellites

We now have, of course, direct evidence that the Milky Way -- at least in its outer parts -- has had some contribution of halo stars AND clusters from accreting dwarf satellites.

The best example of a dwarf galaxy of the Milky Way creating a stream is Sagittarius, the core of which was discovered in 1994 by Ibata et al. (1994, Nature).
2MASS density map of the Sgr dSph by Majewski et al. (2003, ApJ).
A cartoon representation of the Sgr dSph galaxy as mapped by 2MASS M giant stars from Majewski et al. (2003). Figure by David Law based on the model with a triaxial Milky Way potential by Law & Majewski (2010, ApJ, 714, 229). Click here for a movie flyaround of the Sgr configuration.
Diagram showing the Law & Majewski 2010 model of the Sgr system (colored points) in Galactocentric Cartesian coordinates (XGC, ZGC). The color of individual debris particles represents the time at which the particle became unbound from the Sgr dwarf. The "X" indicates the location of the Galactic center, while the circled "X" indicates the location of the Sun. The horizontal dashed line represents the Galactic disk plane, the length of the line is chosen to indicate a diameter 30 kpc. Overplotted against the model are ID numbers corresponding to the locations in the XGC-ZGC plane of Milky Way stellar clusters and satellite galaxies. (The locations of some ID numbers have been adjusted slightly to minimize confusion from overlapping labels.

This analysis, which accounts for position and dynamics, finds the following globular clusters to likely be associated with Sgr:

Arp 2 (object 45, in the Sgr core (black blob))

M54 (object 42, in the Sgr core (black blob))

NGC 5634 (object 27)

Terzan 8 (object 46, in the Sgr core (black blob))

Whiting 1 (object 3)

Berkeley 29 (object 11)

NGC 5053 (object 22)

Palomar 12 (object 53)

Terzan 7 (object 44, in the Sgr core (black blob))

The tidal tail of the Carina dSph being mapped by Ricardo Munoz, SRM, et al. The left panel shows survey area for Carina dSph stars with those spectroscopically confirmed to be part of the system shown as dots. Note the tail-like distribution. Right panel shows an N-body model matching all of the properties of the system as an isodensity plot overlaying the discovered members. Figures from Munoz et al. (2008, ApJ, 679, 346).
The model of ω Cen disruption and the stars identified by SRM group to be identified with its debris, as we saw earlier. From Majewski et al. (2012, ApJL).
SDSS "field of streams", by Belokurov et al. (2006).
There is even clear evidence if gas accretion into the halo, in the form of the Magellanic Stream. However, no stars have yet to be identified with the Magellanic Stream.

Magellanic Stream as mapped by Nidever et al. (2008).
Finally, here is a model of the formation of the halo of a Milky Way-like galaxy via accretion of multiple satellites, by Bullock & Johnston. The model is "cosmologically based", in the sense that cosmological N-body simulations were used to determine the mass spectrum of contributors to this particular halo, which was then enhanced with more detail (higher resolution) simulations of the accretion process. The rings represent the central disk. Note the changing spatial scale. Here is a fly thorugh of the halo created. Diamonds mark "intact" satellites in the present era.


Previous Topic: Age Estimation Lecture Index Next Topic: Other Globular Cluster Systems Next (Working) Topic: Population Synthesis

All material copyright © 2003,2006,2008,2010,2012,2016 Steven R. Majewski. All rights reserved. These notes are intended for the private, noncommercial use of students enrolled in Astronomy 551 and Astronomy 5610 at the University of Virginia.