See discussions, stats, and author profiles for this publication at: [Link]

net/publication/1796110

The Andromeda Stream

Article  in  Publications of the Astronomical Society of Australia · January 2004

DOI: 10.1071/AS04003 · Source: arXiv

CITATIONS READS
13 180

7 authors, including:

Geraint F. Lewis Scott Chapman

The University of Sydney Dalhousie University
420 PUBLICATIONS   11,486 CITATIONS    563 PUBLICATIONS   18,401 CITATIONS   

SEE PROFILE SEE PROFILE

N. Tanvir
University of Leicester
749 PUBLICATIONS   17,851 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

SGRB (gravitational wave : GW) and LGRB View project

South Pole Telescope (SPT) Submillimeter Galaxy (SMG) View project

All content following this page was uploaded by Scott Chapman on 12 March 2015.

The user has requested enhancement of the downloaded file.

 The 5th Workshop on Galactic Chemodynamics - Swinburne University (9-11 July 2003)
To be published in the Publications of the Astronomical Society of Australia in 2004
B.K. Gibson and D. Kawata, eds.

THE ANDROMEDA STREAM

G. F. Lewis1 , R. A. Ibata2 , S. C. Chapman3 , A. M. N. Ferguson4 , A. W. McConnachie5 , M. J. Irwin5 & N. Tanvir6

1
School of Physics, University of Sydney, Sydney, NSW, 2006
2
Observatoire de Strasbourg, 11, rue de l’Université, F-67000, Strasbourg
3
California Institute of Technology, Pasadena, CA 91125, U.S.A
4
Max-Planck-Institut fur Astrophysik, Karl-Schwarzschild-Str. 1, Postfach 1317, D-85741, Garching, Germany
5
Institute of Astronomy, Madingley Road, Cambridge, CB3 0HA, U.K.
arXiv:astro-ph/0401092v1 7 Jan 2004

6
Physical Sciences, University of Hertfordshire, Hatfield, AL10 9AB, U.K.

Abstract
The existence of a stream of tidally stripped stars from the Sagittarius Dwarf galaxy demonstrates that
the Milky Way is still in the process of accreting mass. More recently, an extensive stream of stars has
been uncovered in the halo of the Andromeda galaxy (M31), revealing that it too is cannibalizing a small
companion. This paper reports the recent observations of this stream, determining it spatial and kinematic
properties, and tracing its three-dimensional structure, as well as describing future observations and what
we may learn about the Andromeda galaxy from this giant tidal stream.
Keywords: Galaxy: kinematics and dynamics — methods: N-body simulations
1 Introduction extremely low stellar density on the sky. Furthermore, it is
very important to know if the situation of the Milky Way is
ΛCDM has become the preferred model of cosmological
unique or if its current appetite is representative of galaxies
structure formation, providing a convincing description of
in general. We need, therefore, to turn our attention to the
the large scale distribution of matter in the Universe. On
search for tidal features in external galaxies. While this in-
galactic scales, however, this paradigm has proved some-
creases the apparent stellar density of tidal debris, distance
what unsatisfactory, predicting a myriad of satellite sys-
rapidly blurs individual stars into uniform surface bright-
tems accompanying the Milky Way which do not appear to
ness and dynamical measures become extremely time con-
be there (Klypin et al. 1999). One clear prediction of this
suming. Hence, the search for tidal debris should be aimed
model, however, is that large galaxies like the Milky Way
at our nearest companions if we are to use their structural
grew over time via the accretion of smaller systems.
and kinematic properties in a fashion similar to that of the
Numerical simulations of such accretion events reveal
Sagittarius stream.
that they leave long lived fossil signatures within the halo
of the cannibalizing galaxy in the form of extensive tidal
streams that can completely wrap the host (e.g. John-
ston 1998). The detection and characterisation of this fossil 2 The Andromeda Stream
record will unravel the recent accretion history of a galaxy. 2.1 Wide Field Photometry
While there is some evidence that the Magellanic Clouds
may represent a case of ongoing accretion, displaying an ex- M31, being our extragalactic neighbour, has been the tar-
tensive gaseous tail, the first clear detection of current can- get of numerous observational programs. With the advent
nibalization in the halo of the Milky Way came with the of CCDs earlier work focused on deep pencil-beam studies
discovery of the Sagittarius Dwarf galaxy in 1994 (Ibata, which indicated a complex metallicity mix in M31’s halo.
Gilmore & Irwin 1994). Since its initial detection, studies In contrast, this present study employed the Wide Field
have revealed more and more stellar debris located farther Camera on the Isaac Newton Telescope to obtain a deep
and farther from the main body of the dwarf (e.g. Majewski but panoramic survey of the stellar populations in the halo
et al. 1999) . Ibata et al. (2001) found a stream of carbon of M31. Covering 0.3 deg2 per pointing, the initial survey
stars lying over a great circle on the sky which intersects in the year 2000 tiled a region out to 4o (55kpc) of the
with the current location of the Sagittarius Dwarf. Further- Southern portion of the halo of M31, covering an area of
more, this stream is also aligned with Sagittarius’s proper ∼10deg2 to i = 23.5 and V = 24.5 (Ibata et al. 2001).
motion, clearly demonstrating it represents material asso- An examination of the halo stellar density by eye clearly
ciated with the dwarf galaxy. Intriguingly, the collimated reveals that it is not smooth, showing significant substruc-
nature of the stream strongly suggests that the dark matter ture, particularly an apparent stream of stars stretching
halo is spherical, at odds with theoretical expectations of a to the south of the main body of Andromeda. This fea-
strongly flattened, triaxial distribution. These results were ture is significantly enhanced if a selection is made of metal
recently confirmed by Majewski et al. (2003) using a larger rich red giant branch (RGB) stars (see Figure 1; note, this
sample of stars drawn from the 2-Micron All Sky Survey figure contains further observations with the INT WFC,
(2MASS). mapping out the northern sector of the halo also). As well
While the study of Sagittarius has proved quite suc- as the prominent tidal stream, these data also reveal com-
cessful, our location within the Milky Way limits our ob- plex structure in the halo of M31, including the northern
servational prospects, with any tidal debris presenting an spur and a significant over density of stars in the vicinity of
the giant globular cluster G1 (Ferguson et al. 2002). The
2 Lewis et al.

Figure 1: RGB stars as derived from the INT WFC Figure 2: The location of the CFHT 12K fields relative
survey. The Andromeda Stream of stars is clearly vis- to the central regions of M31. Three fields crossing
ible extending to the south. the plane of M31 have been removed as they were not
employed in the analysis of McConnachie et al. (2003)
due to severe crowding issues.

extensive substructure suggests that M31 may have under-

gone more recent accretion events, resulting in the rather
complex metallicity distribution within its halo. of ∼900kpc from us, 120kpc farther than M31 itself. In
the North, the fields are actually in front of M31, but are
curving away from us northwards.
2.2 The distance to the Stream
In unraveling the history of the stream, it is important to 2.3 Stellar Velocities
determine its three-dimensional structure. To this end, the
stream was targeted with the 12K camera on the Canada While the spatial structure of the Andromeda Stream re-
France Hawaii Telescope (CFH12K). Covering a total of veals important clues to its origin and evolution, fully un-
∼3deg2 , 14 fields were obtained, starting ∼5 degrees below raveling its dynamical history requires a measurement of
the plane of M31, and ending ∼2.5 degrees above the plane the stellar kinematics. As the wide field imaging has re-
(Figure 2). An examination of colour-magnitude diagrams solved the stream into individual stars the determination
of the CFH12K fields clearly reveal the presence of the tidal of stellar velocities is possible with 8-m class telescopes. To
stream of stars as a distinct feature to the very southern ex- this end, a series of observations were undertaken using the
tremity of the survey. The situation is, however, different in DEIMOS spectrograph on the 10-m Keck2 telescope. Em-
the north where the signature of the stream quickly peters ploying multislit masks, resulting with ∼ 100 spectra over
out, disappearing in the northernmost field; this reflects the a 16′ × 5′ region, four fields along the stream were targeted.
structure seen in the INT imaging survey described previ- The RGB sources possessed 20.5 < i < 22.0, resulting in a
ously. velocity accuracy of 5-10km/s.
The superb quality of the CFH12K data, however, pro- These observations revealed a strong velocity gradient
vided an opportunity to measure the distance to the stream along the stream, with the southernmost region of the stream
at various locations along its length. This was achieved by traveling toward us at the systemic velocity of M31, while
first determining the location of the tip of the red giant in the vicinity of M31 it is approaching at 300km/s with
branch (TRGB) in the main body of M31 at a range of respect to M31 (Ibata et al., submitted). Coupled with the
metallicities. By comparing the location of the TRGB in distance determinations outlined above, these kinematics
the fields along the stream with the main body of M31, the provide strong constraints on the orbital properties of the
position of each field relative to M31 could be measured stream; Figure 4 presents two preliminary orbit fits to the
via a cross-correlation of the stellar luminosity functions, extant kinematic and spatial data of stars in the stream.
as any offsets would primarily be due to a difference in dis-
tance (McConnachie et al. 2003). The results of this study
2.4 Interpretation
are presented in Figure 3, revealing that the Andromeda
stream curves away from us below the disk of M31. The The stream orbits allow us constrain the mass distribution
most southern extremity of the survey lies at a distance in the M31 halo. A full analysis will require N-body simula-
 The Andromeda Stream 3

Figure 3: The distance to the Andromeda Stream as

determined from calibrating the tip of the red giant Figure 4: Two preliminary fits to the orbit of the An-
branch and cross-correlating the luminosity function dromeda stream. The green model utilises only the
of stars along the Andromeda Stream. southern data; the overall fit to the kinematic and spa-
tial data is good but the resulting orbit is strongly ra-
dial. The red model, incorportating the entire data
set, is a poor representation of the data, but its orbit
tions to be carried out in different halo potentials, with the is similar to the rosette orbit of the Sagittarius dwarf.
resulting stellar stream compared to the observed position,
distance and velocity data. However, a good approximation
can be obtained by simply comparing the locus and velocity
profile of test particle orbits, a task which is computation- the form of the dark matter halo of M31.
ally much cheaper. For this approximation to work, the
dwarf galaxy progenitor of the stream must be of relatively
low mass < 109 M⊙ , for the self-gravity of the stream to Acknowledgments
be unimportant. By comparing the best-fit orbit to the
data as a function of halo potential, the most likely value GFL thanks Swinburne University’s Astronomy Group for
of the total mass of M31 is Mtot = 6.5 ± 1.6 × 1011 M⊙ out hosting a fun and informative meeting, although Sydney
to 145kpc. With this, Andromeda has a very similar halo would have been warmer.
mass to the Milky Way. However, this result is somewhat
uncertain; using only the southern data to contstrain the
form of the potential reduces the halo mass by a factor of References
∼ 4 [see Ibata et al. (Submitted) for more details].
Ferguson, A. M. N., Irwin, M. J., Ibata, R. A., Lewis, G. F.,
& Tanvir, N. R. 2002, AJ, 124, 1452
Ibata, R. A., Gilmore, G., & Irwin, M. J. 1994, Nature,
3 Future Observations 370, 194
In September 2003, further DEIMOS spectroscopy was un- Ibata, R., Lewis, G. F., Irwin, M., Totten, E., & Quinn, T.
dertaken at Keck2. To increase the observational multiplex- 2001, ApJ, 551, 294
ing to roughly 1000 stars per mask, a narrow-band filter was Ibata, R., Irwin, M., Lewis, G., Ferguson, A. M. N., &
employed to capture only the region of the CaII NIR triplet. Tanvir, N. 2001, Nature, 412, 49
In this manner we targeted a total of ∼ 2 × 104 stars. This Johnston, K. V. 1998, ApJ, 495, 297
data will be supplemented with a CFHT/MEGACAM sur- Klypin A., Kravtsov A. V., Valenzuela O., Prada F. 1999,
vey of the southern quadrant of M31, covering a total of ApJ, 522, 82
almost 80 square degrees, calibrating the entire extent of Majewski, S. R., Siegel, M. H., Kunkel, W. E., Reid, I. N.,
the tidal stream and allowing us to probe the halo between Johnston, K. V., Thompson, I. B., Landolt, A. U., &
M31 and M33. Palma, C. 1999, AJ, 118, 1709
This extensive dataset will allow a more complete de- Majewski, S. R., Skrutskie, M. F., Weinberg, M. D. & Os-
termination of the orbital properties of the Andromeda theimer, J. C., 2003, astro-ph/0304198
stream and characterise its dynamical evolution and ulti- McConnachie, A. W., Irwin, M. J., Ibata, R. A., Ferguson,
mate demise. As with the studies of the Sagittarius tidal A. M. N., Lewis, G. F., & Tanvir, N. 2003, MNRAS,
stream (e.g. Ibata et al. 2001), this approach will reveal 343, 1335

View publication stats