1: \documentclass{emulateapj}
2: %\documentclass[12pt,preprint]{aastex}
3: %\usepackage{amsmath}
4: %\usepackage{graphicx}
5: \usepackage{psfig}
6:
7: %\usepackage{epsf}
8:
9:
10: \def\ltsima{$\; \buildrel < \over \sim \;$}
11: \def\simlt{\lower.5ex\hbox{\ltsima}}
12: \def\gtsima{$\; \buildrel > \over \sim \;$}
13: \def\simgt{\lower.5ex\hbox{\gtsima}}
14:
15: \def\gs{\mathrel{\raise0.35ex\hbox{$\scriptstyle >$}\kern-0.6em
16: \lower0.40ex\hbox{{$\scriptstyle \sim$}}}}
17: \def\ls{\mathrel{\raise0.35ex\hbox{$\scriptstyle <$}\kern-0.6em
18: \lower0.40ex\hbox{{$\scriptstyle \sim$}}}}
19: \def\kms{km$\,$s$^{-1}$}
20: \def\micron{$\mu$m}
21: \def\andxi{AndXI}
22: \def\andxii{AndXII}
23: \def\andxiii{AndXIII}
24: \def\K{{$\,$K}}
25: \def\sec{{$\,$s}}
26:
27: %\shortauthors{Chapman et al.}
28: %\shorttitle{halo}
29:
30: \makeatletter
31: \newenvironment{inlinetable}{%
32: \def\@captype{table}%
33: \noindent\begin{minipage}{0.999\linewidth}\begin{center}\footnotesize}
34: {\end{center}\end{minipage}\smallskip}
35:
36: \newenvironment{inlinefigure}{%
37: \def\@captype{figure}%
38: \noindent\begin{minipage}{0.999\linewidth}\begin{center}}
39: {\end{center}\end{minipage}\smallskip}
40: \makeatother
41:
42: \newcommand{\andix}{Andromeda {\sc IX}}
43: \newcommand{\ha}{{\rm H$\alpha$ }}
44: \newcommand{\hb}{{\rm H$\beta$ }}
45: \newcommand{\hbns}{{\rm H$\beta$}}
46: \newcommand{\hahb}{{\rm H$\alpha$/H$\beta$ }}
47: \newcommand{\hahbns}{{\rm H$\alpha$/H$\beta$}}
48: \newcommand{\hi}{{\rm H{\sc i} }}
49: \newcommand{\mol}{{\rm H$_2$ }}
50: \newcommand{\hans}{{\rm H$\alpha$}}
51: \newcommand{\hii}{{\rm H\,{\sc ii} }}
52: \newcommand{\hiins}{{\rm H\,{\sc ii}}}
53: \newcommand{\msun}{${\rm M_{\sun}}$}
54:
55: \slugcomment{{\sc Accepted in ApJ, July issue: } }
56:
57: \begin{document}
58:
59: \title{Strangers in the night: Discovery of a
60: dwarf spheroidal galaxy on its first Local Group infall}
61: \author{S.\ C.\ Chapman\altaffilmark{1,2,3},
62: J.\ Pe\~narrubia\altaffilmark{2},
63: R.\ Ibata\altaffilmark{4},
64: A.\ McConnachie\altaffilmark{2},
65: N.\ Martin\altaffilmark{5},
66: M.\ Irwin\altaffilmark{1},
67: A.\ Blain\altaffilmark{3},
68: G.~F.~Lewis\altaffilmark{6},
69: B.\ Letarte\altaffilmark{3},
70: K.\ Lo\altaffilmark{7},
71: A.\ Ludlow\altaffilmark{2},
72: K.\ O'neil\altaffilmark{7}
73: }
74: \altaffiltext{1}{Institute of Astronomy, Madingley Road, Cambridge, CB3 0HA, U.K.}
75: \altaffiltext{2}{University of Victoria, Victoria BC, V8P 1A1 Canada}
76: \altaffiltext{3}{California Institute of Technology, Pasadena, CA\,91125}
77: % \texttt{schapman@ast.cam.ac.uk}
78: \altaffiltext{4}{
79: Observatoire de Strasbourg, 11, rue de l'Universit\'e, F-67000, Strasbourg, France}
80: \altaffiltext{5}{Max-Planck-Institut fur Astronomie, Konigstuhl 17
81: 69117 Heidelberg, Germany}
82: \altaffiltext{6}{
83: Institute of Astronomy, School of Physics, A29, University of Sydney, NSW
84: 2006, Australia}
85: \altaffiltext{7}{NRAO, P.O.\ Box 2, Green Bank, WV 24944}
86:
87:
88: \begin{abstract}
89: We present spectroscopic observations of the \andxii\ dwarf spheroidal galaxy
90: using DEIMOS/Keck-II,
91: showing it to be moving rapidly through the Local Group
92: (-556~km/s heliocentric velocity, -281~km/s relative to Andromeda),
93: falling into the Local Group from $\sim$115 kpc
94: beyond Andromeda's nucleus.
95: \andxii\ therefore represents a dwarf galaxy plausibly falling
96: into the Local Group
97: for the first time, and never having experienced a dense galactic
98: environment. From Green
99: Bank Telescope observations, a limit on the H{\sc I} gas mass of $<3\times10^3$~M$_\odot$
100: suggests that \andxii's gas could have been removed prior to experiencing the
101: tides of the Local Group galaxies.
102: Orbit models suggest the dwarf is close to the escape velocity of
103: M31 for published mass models. \andxii\ is
104: our best direct evidence for the late infall of satellite galaxies,
105: a prediction of cosmological simulations.
106: \end{abstract}
107:
108:
109: \keywords{galaxies: spiral --- galaxies: individual (Andromeda) --- galaxies:
110: individual (Andromeda) --- galaxies: evolution --- Local Group}
111:
112: \section{Introduction}
113: \label{txt:intro}
114:
115: Dwarf galaxies and stellar streams represent the visible remnants of the
116: merging process by which the halos of galaxies are built up. They can be used
117: to unravel the hierarchical formation of their host galaxies \citep{whiterees}.
118: %many small galaxies were incorporated into large galaxies.
119: %
120: Dwarf galaxies are also the systems with the highest mass-to-light ratio
121: found in the Universe (e.g., Mateo et al.\ 1998),
122: which makes them one of the best
123: laboratories for investigating the nature of dark matter
124: (e.g., Pe\~narrubia et al.\ 2007).
125:
126: In the Cold Dark Matter paradigm a few Myrs after the Big Bang, the
127: distance between proto-dwarf galaxies and their hosts were much smaller.
128: Early structure formation left dwarf galaxies over a large range
129: of distances from the host galaxy with small radial velocity components
130: (they were located approximately at their orbital apocenter). The largest dark
131: matter over-density in the local volume attracted these systems so that they
132: eventually merged moving on high eccentric orbits. In this picture, those
133: substructures initially lying close to the parent galaxy
134: were the first ones to be accreted, whereas those with large initial
135: separations either are accreting now or have yet to accrete. An important
136: prediction implicit in this scenario is that all accreting substructures are
137: bound to the host galaxy; only strong interactions, like three-body
138: encounters, might provide enough energy for them to escape the host systems.
139: Also interesting is the fact that the late accretion events are expected to
140: move on highly eccentric orbits, with large radial velocity components with
141: respect to the host. Moreover, these systems correspond to the objects that
142: formed furthest from the host galaxy, thus sampling a
143: satellite galaxy population that has mostly
144: lived in a very different environment from the Local Group.
145: %for most of their time and that
146: It is only now, after they have been
147: accreted, when strong gravitational fields start to alter their original
148: properties (mass, dark matter and distributions, gas fraction, etc.).
149: High-speed dwarf galaxies, moving with velocities close to the escape
150: velocity, therefore represent an interesting theoretical and observational
151: target that might provide important insights into the effects of dynamical
152: evolution on the properties of dwarf galaxies.
153:
154: It is believed that the satellites of the Milky Way and M31
155: have been accreted at relatively early times and their properties have been
156: molded by their interactions with their parent galaxies.
157: There will undoubtedly be
158: mergers in the future of %bound
159: satellite systems within the Local Group (e.g., the Magellanic Clouds
160: -- Kallivayalil, van der
161: Marel \& Alcock 2006; Pedreros, Costa \& Mendez 2006).
162: There is also the case of the rather extreme orbit of the
163: Leo~I dwarf spheroidal \citep{mateo}, which is
164: now moving rapidly outwards from the Local Group on the limit of being bound.
165: Recent observations \citep{sohn} support a picture where Leo~I has been
166: tidally disrupted on several perigalactic passages of a massive galaxy.
167: %
168: Here we report on kinematic observations of \andxii,
169: a dwarf spheroidal which is a candidate for first infall into the Local
170: Group (LG).
171:
172:
173: \section{Observations}
174:
175: The dwarf spheroidal galaxy \andxii\ lies at 105~kpc (projected) from
176: the nucleus of M31, the faintest dwarf
177: galaxy in the M31 outer halo MegaCam/CFHT survey data \citep{martin06,ibata07},
178: with an overdensity of Red Giant Branch (RGB) stars with similar metallicity
179: (Fig.~1).
180: \andxii\ was followed
181: up with imaging from a SUPRIMECam/Subaru survey of M31 halo
182: substructures (Martin et al.\ in prep) to
183: improve the distance estimate to \andxii\ and better understand its
184: stellar populations.
185: Spectra for candidate \andxii\ stars were obtained with the DEIMOS spectrograph on the Keck-II telescope (Figs.~1 \& 2).
186: Multi-object Keck observations
187: with DEIMOS \citep{deep2} were made on 2006 Sept 21--24,
188: in photometric conditions and excellent seeing of 0.6\arcsec.
189: We used the 600~line/mm grating, achieving resolutions $\sim3.5$\AA\ and
190: probing the observed wavelength range from 0.56--0.98~$\mu$m.
191: Exposure time was 140 min, split into 20-min integrations. Data
192: reduction followed standard techniques using the DEIMOS-DEEP2
193: pipeline \citep{deep2},
194: debiassing, flat-fielding, extracting, wavelength-calibrating and
195: sky-subtracting the spectra.
196:
197: The radial velocities of the stars were
198: then measured with respect to a Gaussian
199: model of the Calcium-II triplet (CaT) absorption lines \citep{wilkinson}.
200: By fitting the three strong CaT lines separately
201: (Fig.~2), an estimate of the radial
202: velocity accuracy was obtained %for each radial velocity measurement
203: (Table~1), with typical
204: uncertainties of $5$\ km/s to $12$\ km/s, before accounting for systematic
205: errors from sky lines (which can add another $5-10$\ km/s).
206: Close to the center of \andxii\
207: there is a clear kinematic grouping of eight stars with cross-correlation
208: peaks greater than 0.1 at a velocity of
209: $\sim$-556~km/s heliocentric which also lie on the \andxii\ RGB (Fig.~1).
210: This corresponds to -281~km/s relative to M31 after removing Milky Way motions.
211: Although we targeted 49 stars in the vicinity of the \andxii\ overdensity,
212: no other stars are associated to this dwarf by their kinematics;
213: all other stars lie more than $5\sigma$ away from the kinematic
214: grouping of stars associated with the \andxii\ CMD. The \andxii\
215: stars have a
216: velocity dispersion of 5~km/s before accounting for instrumental errors,
217: however the velocity errors are sufficiently large
218: that a meaningful limit on the mass of \andxii\ is difficult to constrain.
219: An average metallicity of [Fe/H]=-1.9$\pm$0.2 is estimated for \andxii\
220: from the equivalent widths of the CaT lines of the combined spectrum in
221: Fig.~2, adopting the technique described in Ibata et al.\ (2005).
222: However, the 5 highest S/N spectra have a more metal-rich median,
223: [Fe/H]=-1.7, closer to the photometric [Fe/H]=-1.5 (Martin et al.\ 2006).
224:
225: We also searched for neutral hydrogen at rest frequency 1420.406~MHz,
226: shifted to the systemic velocity of \andxii.
227: Using the Green Bank Telescope (Nov.~29, 2006)
228: and the 12.5~MHz bandwidth of the spectrometer,
229: we integrated on \andxii\ using interspersed on/off observations totalling 1\,hr each.
230: The 9.8\arcmin\ beam size implies that the full stellar extent of \andxii\ is
231: well within the beam.
232: There is no detection down to 2.53\,mJy RMS, suggesting a limit on the H{\sc I}
233: gas mass of $<3\times10^{3}$~M$_\odot$ for an assumed linewidth of 4~km/s
234: (comparable to the raw dispersion of the RGB stars).
235:
236: %so it must have been stripped
237: %before it got where it is now falling in to the local group, or it is
238: %primordial and its Pop III stars emptied it of gas
239:
240: \section{Results}
241:
242: Our new data measure the radial motion and constrain the distance to \andxii.
243: %Martin et al.\ (2006) used an averaging approach of three
244: %dwarf galaxies to set a plausible range
245: %of distances: they superposed the CMDs of \andxi, \andxii, and \andxiii,
246: %and estimated the average distance from fitting to the combined RGB.
247: %
248: Since our spectroscopy tells us which amongst the candidates
249: are definite \andxii\ RGB stars, we can assess the distance directly.
250: We first fit isochrones from the library of Girardi et al.\ (2004) to the
251: RGB, assuming the oldest age in the library (log(age)= 10.25);
252: younger ages make the fits worse, and the implied
253: radial distance larger.
254: Template isochrones were first extinction corrected using the
255: Schlegel, Finkbeiner \& Davis (1998) extinction maps, E(B-V)=0.11.
256: We then attempted to fit isochrones of [Fe/H]=-2.3,-1.7,-1.5 and -1.3,
257: over a range in distance modulus (DM) from
258: 24.1 to 24.9 (M31 has a DM=24.47 and a
259: $I_{0,\rm Vega, TRGB}$=20.54 -- McConnachie et al.\ 2005).
260: %in the Sloan filter system.
261: The best fitting isochrones are overlaid in Fig.~1,
262: with [Fe/H]=-1.5, DM=24.5 (D$_{\rm hel}$=810~kpc) being the best overall fit
263: to the RGB and likely horizontal branch stars.
264: Fits to the RGB shape and horizontal branch region were worse with other
265: metallicities. Further analysis of the stellar populations in \andxii\ will
266: be presented in Martin et al.\ (in prep).
267: To assess a likely error range on the distance, we then employed
268: the ``Tip of the Red Giant Branch" (TRGB) technique
269: (e.g., \citealt{mcconnachie04}).
270: An absolute upper limit to the radial distance can be obtained by
271: assuming the brightest RGB star ($I_{0,\rm Vega}$=20.95) is at the TRGB.
272: This implies D$_{max}=950$ kpc (Fig.~3).
273: To constrain the minimum distance,
274: we must assess what the maximum possible offset could be from this
275: brightest RGB star.
276: In the absence of knowledge of the luminosity function of
277: \andxii, we analyse the well populated RGB of
278: the dwarf spheroidal, AndII \citep{mcconnachie05}.
279: Our Girardi et al.\ (2004) isochrone fit to the RGB suggests that our
280: 8 confirmed members of \andxii\ lie within the top magnitude of the RGB.
281: With 1000 random samples of 8 of the brightest stars from the top magnitude
282: of the AndII RGB, we constrain a probability distribution for the true TRGB
283: for \andxii\ (corresponding to offsets up to 0.5-mag)
284: and thereby the likely range of distances (Fig.~3).
285: The median distance for \andxii\ is 830$^{+50}_{-50}$~kpc
286: (interquartile range), in excellent agreement with the simple isochrone fit
287: performed earlier, and
288: updates the derived parameters of \andxii:
289: %with m-M=24.87,
290: M$_v$ becomes -6.9, and the half-light radius, r$_{hb}$,
291: becomes 137~pc (previously estimated as 125~pc).
292:
293: %\noindent{\bf The orbit of \andxii:}
294:
295: Fig.~3 summarizes our model of orbits for \andxii.
296: Integrating the orbit back in time necessitates that \andxii\ likely
297: came from beyond the Local Group's virial radius.
298: We adopt a recently published mass model for M31 \citep{geehan05},
299: dominated by a NFW \citep{nfw}
300: dark matter halo with total virial mass of
301: $\sim1\times10^{12}$ M$_\odot$.
302: An object traveling at -281\ km/s with respect to M31 at time zero
303: (today) is considered over all possible heliocentric distances in the
304: gravitational potential of M31.
305: \andxii\ likely reached distances close to
306: 725~kpc beyond M31 in the previous 10\,Gyr
307: (a substantial part of the journey to the
308: neighboring galaxy concentration around M81), and is an excellent
309: candidate for a dwarf falling into the Local Group for the first time.
310:
311: The high radial velocity of \andxii\ together with the large separation to
312: its apparent host, M31, suggest that \andxii's orbit might be highly
313: eccentric, close to the limit of being bound to M31 and the Local Group.
314: Figure~4 shows \andxii\ relative to the circular and escape velocities of
315: M31 in the Geehan et al.\ (2005) model. \andxii\ is the most extreme satellite of M31, even more so than recently discovered satellites
316: AndXIV (Majewski et al.\ 2007) and AndXI, AndXIII (Martin et al.\ 2006,
317: Chapman et al.\ 2007).
318: In a CDM framework, subhalos cannot be
319: accreted exceeding the escape velocity unless they suffered strong 3-body
320: interactions.
321: In this framework, \andxii\ %falling in for the first time
322: places a lower limit on the mass of M31 slightly larger than
323: estimates from the present models (Fig.~4; Geehan et al.\ 2005).
324: There is also
325: an unknown proper (transverse) motion to add to the radial velocity
326: -- it is unlikely for \andxii\ to be approaching
327: head-on, and therefore might be falling into the Local Group
328: faster than the limiting v$_{hel}=-556$~km/s.
329:
330:
331: \section{Discussion}
332:
333: Theoretical considerations have suggested that even at the present day,
334: galaxies should still be
335: falling into the Local Group from surrounding overdense structures.
336: Benson (2005) measured the orbital parameters of
337: infalling substructures from N-body simulations.
338: There was little evidence for any mass dependence in the distribution of
339: orbital parameters, and so the results are expected to be applicable to \andxii.
340: The orbital parameters in the Benson (2005) simulations were
341: determined at the point where the substructure first entered the virial
342: radius of the larger halo.
343: % which is likely where we find \andxii\ at the present.
344:
345: %Benson (2005) found 18\% of substructures were on unbound orbits
346: %and calculate that most of these substructures lost sufficient energy/angular
347: %momentum during their first passage through the halo to become bound.
348:
349: The large velocity and likely distance behind M31 suggest that
350: \andxii\ is falling into the Local Group for the first time.
351: %We refer to Fig.~3 which shows the distance that \andxii\ would have
352: %traversed in the last 10~Gyr, likely its location of formation.
353: %The error in radial distance implies a range from 0.8 to 1.0~Mpc.
354: %
355: The most likely place where systems like \andxii\ form is in filaments.
356: Ludlow et al.\ (in prep) have looked explicitly at high velocity structures
357: falling into halos of similar total mass to the Local Group at late times.
358: Substructures which have yet to fall into the surrogate Local Group
359: clearly lie within a filamentary morphology.
360:
361: At late times (today), one observes subhalos in N-body cosmological
362: simulations with high velocities because a large fraction of them move
363: on parabolic orbits.
364: The ones that move outwards are those that accreted
365: many Gyrs ago and are now close to escaping, perhaps the situation
366: with Leo~I \citep{mateo} and AndXIV (Majewski et al.\ 2007).
367: The ones that move inwards, like \andxii, are those that were accreted
368: later.
369:
370: %Since \andxii\ is now falling so quickly through the Local Group, likely
371: %from behind M31,
372: %the age of the Universe is not sufficient to allow it to have started in
373: %the dense Local Group environment, traveled outwards, and be on its
374: %second passage.
375: %reached apocenter and returned.
376:
377: %GAS
378: %about how to form a dSph from a dIrr using tidal stripping, stirring and
379: %RAM pressure stripping in the halo of a massive galaxy clearly don't apply to
380: %AndXII and are probably wrong (see their Nature paper if nothing else).
381:
382: If \andxii\ has not yet experienced the strong gravitational environment
383: of the Local Group, its H{\sc I} gas content could be an interesting
384: evolutionary constraint. Mayer et al.\ (2006, 2007) show that
385: assuming a low density of the hot Galactic corona consistent with
386: observational constraints, dwarfs with V$_{\rm peak}<$30km/s will be completely stripped of their gas content
387: on orbits with pericenters of 50~kpc or less.
388: In these objects most of the gas is removed or becomes ionized at the
389: first pericenter passage, explaining the early truncation of the star formation
390: observed in Draco and Ursa Minor.
391: Galaxies on orbits with larger pericenters and/or falling into
392: the Local Group at late times (like \andxii)
393: should retain significant amounts of the centrally concentrated gas.
394: These dwarfs would continue to form stars over a longer period.
395: %
396: While the dark matter mass in \andxii\ is uncertain, it is plausibly $>10^6$
397: M$_\odot$, with an
398: initial gas mass of $>10^4$ M$_\odot$.
399: If \andxii\ has not yet passed through the potential of the Local Group,
400: and no longer has a sizeable gas supply ($<3\times10^3$~M$_\odot$ from
401: our GBT measurement) it
402: must have been stripped of its neutral H{\sc I} through other means,
403: perhaps some of it through its Pop~III stars.
404:
405: While the precise direction of origin is uncertain because of the unknown
406: tangential velocity,
407: \andxii\ could have come roughly from the direction of the Sculptor group
408: and beyond towards the M81 group,
409: although it could not have traversed the distance from the M81 group
410: ($\sim3.5$~Mpc \citealt{karachentsev02}) in the age of the Universe.
411: The discovery of \andxii\ presents the best piece of evidence to date for the late accretion of satellites
412: and sets a new benchmark for testing the mass of Local Group galaxies and simulations of galaxy formation.
413:
414: \medskip
415: \section*{acknowledgements}
416: SCC acknowledges a fellowship from the Canadian Space Agency and an NSERC
417: discovery grant.
418: JP acknowledge Julio Navarro for finanacial support.
419: %GFL ackowledges support through ARC DP0343508
420: Data presented herein were obtained using the W.\ M.\ Keck
421: Observatory, which is operated as a scientific partnership among
422: Caltech, the University of California and NASA. The Observatory was
423: made possible by the generous financial support of the W.\ M.\ Keck
424: Foundation.
425:
426:
427: %-----------------------------------------------------------------------
428: %REFERENCES
429: %-----------------------------------------------------------------------
430:
431: \begin{thebibliography}{}
432:
433: %\bibitem[Abadi et al.(2003)]{abadi}
434: %Abadi, M., Navarro, J., Steinmetz, M., Eke, V.,
435: %%Simulations of galaxy formation in a $\Lambda$-CDM universe II: the fine structure of simulated galactic disks,
436: %2003, ApJ, 597, 21
437:
438: \bibitem[Benson (2005)]{benson}
439: Benson A.
440: %Orbital parameters of infalling dark matter substructures.
441: 2005, MNRAS, 358, 551
442:
443: \bibitem[Chapman et al.(2005)]{chapman05}
444: Chapman, S.\ C.; Ibata, R.; Lewis, G.\ F.; Ferguson, A.; Irwin, M.; McConnachie, A.; Tanvir, N.
445: %A Keck DEIMOS Kinematic Study of Andromeda IX: Dark Matter on the Smallest Galactic Scales.
446: 2005, \apj\ 632, L87
447:
448: \bibitem[Chapman et al.(2007)]{chapman07}
449: Chapman S.\ C.; Martin N., Ibata, R.; Blain A., Ferguson A., Lewis G., Irwin M., McConnachie A., Pe\~narrubia J.
450: 2007, MNRAS, submitted
451:
452: \bibitem[Cote et al.(2000)]{cote00}
453: Cote P., Mateo M., Sargent W., Olszewski E.,
454: 2000, ApJ, 537, 91
455:
456: \bibitem[Faber et al.(2003)]{deep2}
457: Faber S., A. C. Phillips, R. I. Kibrick, B. Alcott, S. L. Allen, {\it et al.,}
458: 2003, {\it Proceedings of the SPIE} 4841, 1657
459:
460: \bibitem[Geehan et al.(2005)]{geehan05}
461: Geehan J., Fardal M.,
462: Babul A., Guhathakurta P.
463: %Investigating the Andromeda stream - I. Simple analytic bulge-disc-halo model for M31.
464: 2006, MNRAS, 366, 996
465:
466: \bibitem[Girardi et al.(2004)]{girardi04}
467: Girardi,ÊL.; Grebel,ÊE.ÊK.; Odenkirchen,ÊM.; Chiosi,ÊC.
468: 2004, A\&A, 422, 205
469:
470: %\bibitem[Ibata et al.(2001)]{ibata}
471: %Ibata R., M. Irwin, G. F. Lewis, A. Ferguson, N. Tanvir,
472: %2001, Nature 412, 49
473:
474: \bibitem[Ibata et al.(2005)]{ibata05}
475: Ibata R., et al.\ 2005, ApJ, 634, 287
476:
477: \bibitem[Ibata et al.(2007)]{ibata07}
478: Ibata R., et al.\ 2007, submitted, astroph/07041318
479:
480: \bibitem[Kallivayalil et al.(2006)]{kallivayalil}
481: Kallivayalil N., van der Marel R., Alcock C.,
482: 2006, ApJ, 652, 1213
483:
484: \bibitem[Karachentsev et al.(2002)]{karachentsev02}
485: Karachentsev I., Dolphin A., Geisler D., {\it et al.,}
486: %The M81 group of galaxies: New distances, kinematics and structure.
487: 2002 \aap\ 383, 125
488:
489: %\bibitem[Klypin et al.(2002)]{klypin}
490: %Klypin A., Zhao H., Somerville R.,
491: %%$\Lambda$CDM-based Models for the Milky Way and M31. I. Dynamical Models.
492: %2002, \apj\ 573, 597
493:
494: \bibitem[Majewski et al.(1999)]{majewski}
495: Majewski S., Siegel M., Kunkel W., Reid I., Johnston K., et al.,
496: 1999, AJ, 118, 1819
497:
498: \bibitem[Majewski et al.(2007)]{majewski07}
499: Majewski S., et al., 2007, ApJ, submitted
500:
501: \bibitem[Martin et al.(2006)]{martin06}
502: Martin N., Ibata R., Irwin M., Chapman S., Lewis G., {\it et al.,}
503: % Ferguson, A.\ M.\ N., Tanvir, N., McConnachie, A. W.
504: %Discovery and analysis of three faint dwarf galaxies and a globular cluster in the outer halo of the Andromeda galaxy.
505: 2006, MNRAS, 371, 1983
506:
507: \bibitem[Mateo et al.(1998)]{mateo}
508: Mateo M., Olszewski E., Vogt S., Keane M.,
509: %The Internal Kinematics of the Leo I Dwarf Spheroidal Galaxy:
510: %Dark Matter at the Fringe of the Milky Way
511: 1998, \aj\ 116, 2315
512:
513: \bibitem[Mayer et al.(2007)]{mayer07}
514: Mayer L., et al.,
515: 2007, Nature, 445, 738
516:
517: \bibitem[Mayer et al.(2006)]{mayer}
518: Mayer L., et al.,
519: 2006, MNRAS, 369, 1021
520:
521: \bibitem[McConnachie et al.(2004)]{mcconnachie04}
522: McConnachie A., Lewis G., Irwin M., Ibata R., Chapman S. {\it et al.,}
523: 2004, MNRAS, 351, 94L
524:
525: \bibitem[McConnachie et al.(2005)]{mcconnachie05}
526: McConnachie A., Irwin M., et al.,
527: %Distances and metallicities for 17 Local Group galaxies
528: 2005, MNRAS, 356, 979
529:
530: \bibitem[Navarro et al.(1997)]{nfw}
531: Navarro J., Frenk C., White S.,
532: 1997, \apj\ 490, 493
533:
534: \bibitem[Pedreros et al.(2006)]{pedreros}
535: Pedreros M., Costa E., Mendez, R.\ A.
536: 2006, AJ, 131, 1461
537:
538: \bibitem[Pe\~narrubia et al.(2007)]{penarrubia}
539: Pe\~narrubia J., McConnachie A., Navarro J.,
540: 2007, ApJ, in press., astroph/0701780
541:
542: \bibitem[Sohn et al.(2007)]{sohn}
543: Sohn S., Majewski S., R. R. Munoz, W. E. Kunkel, K. V. Johnston,
544: {\it et al.,}
545: % J. C. Ostheimer, P. Guhathakurta, R. J. Patterson, M. H. Siegel, M. C. Cooper
546: %Exploring Halo Substructure with Giant Stars X. Extended Dark Matter or Tidal Disruption?: The Case for the Leo I Dwarf Spheroidal Galaxy
547: 2007, \apj\ in press
548:
549: \bibitem[Wilkinson et al.(2004)]{wilkinson}
550: Wilkinson M., Kleyna J., N. Evans, G. Gilmore, M. Irwin,
551: {\it et al.,} % Grebel, E.
552: %Kinematically Cold Populations at Large Radii in the Draco and Ursa Minor Dwarf Spheroidal Galaxies.
553: 2004, \apj\ 611, 21
554:
555: \bibitem[White \& Rees (1978)]{whiterees}
556: White S., Rees M.
557: 1978, MNRAS, 183, 341
558:
559: \end{thebibliography}
560:
561: \begin{table*}
562: \begin{center}
563: \caption{Positions and parameters of the
564: eight stars confirmed as associated to \andxii.}
565: \label{tableSat}
566: \begin{tabular}{llcccc}
567: \hline
568: {RA} & {dec} & vel (km/s) & vel-error & $g$-mag & $i$-mag \cr
569: \hline
570: 0 47 31.04 & 34 24 12.1 & -557.2 & 6.9 & 23.028 & 21.642 \cr
571: 0 47 24.69 & 34 22 23.9 & -565.8 & 10.1 & 23.129 & 21.658 \cr
572: 0 47 27.76 & 34 22 6.2 & -556.9 & 4.9 & 23.214 & 21.744 \cr
573: 0 47 28.63 & 34 22 43.1 & -561.0 & 5.1 & 23.382 & 21.974 \cr
574: 0 47 31.34 & 34 22 57.6 & -555.0 & 7.4 & 23.504 & 22.156 \cr
575: 0 47 26.65 & 34 23 22.8 & -548.1 & 10.5 & 23.880 & 22.542 \cr
576: 0 47 27.18 & 34 23 53.5 & -550.4 & 7.0 & 23.883 & 22.532 \cr
577: 0 47 30.60 & 34 24 20.3 & -559.5 & 11.2 & 23.903 & 22.657 \cr
578: \hline
579: \end{tabular}
580: \end{center}
581: \end{table*}
582:
583:
584: \eject
585: \centerline{\hbox{\psfig{file=f1a.ps,angle=0,width=3.5in}
586: \psfig{file=f1b.ps,angle=-90,width=3.5in}}}
587: \noindent
588: {\bf Figure 1.}
589: {\bf LEFT PANEL:}
590: On the top is shown the radial velocity histogram binned by 10~km/s.
591: The 8 stars in \andxii\ are bold. Stars at $>$150~km/s are likely
592: Galactic, while the remaining few stars are likely
593: members of the outer halo of M31.
594: On the left, the color-magnitude diagram of the field of \andxii\ with the
595: 8 stars in \andxii\ as large squares, the 49 stars with Keck radial velocities shown as squares, and all other stars within 3$'$ radius of
596: \andxii\ shown as open triangles. Stars selected for Keck spectroscopic
597: followup were taken from the CFHT-MegaCam image. All other stars
598: are shown from the Subaru SUPRIMECAM image.
599: Best fitting isochrones from the library of Girardi et al.\ (2004) are overlaid,
600: with [Fe/H]=-1.5, log(age)=10.25, and distance moduli of 24.3, 24.5, and 24.7.
601: On the right is shown the velocity errors for the
602: 49 stars with radial velocity measurements.
603: %The fiducial RGBs correspond to, from left to right, NGC~6397,
604: %NGC~1851, 47~Tuc, NGC~6553 which have metallicity of [Fe/H]$=
605: %$-1.91, -1.29, -0.71, and -0.2,
606: %respectively. These fiducials have been shifted to the
607: %HB distance of \andxii\ (810~kpc -- Martin et al.\ in prep).
608: {\bf RIGHT PANEL:} A representation of the dwarf galaxy
609: \andxii\ showing all stars with
610: photometric metallicity $-2.3<$[Fe/H]$<-1.3$ (Table~1 stars
611: highlighted, as well as the other 41 stars with DEIMOS velocities {\it not}
612: lying in \andxii) in the CFHT MegaCam M31 survey \citep{martin06,ibata07}.
613:
614: \medskip
615:
616: \centerline{\hbox{\psfig{file=f2.ps,angle=0,width=5.5in}}}
617: \noindent
618: {\bf Figure 2.} The eight individual spectra, and the combined spectrum
619: of stars from Table~1
620: (smoothed to the instrumental resolution)
621: spanning the $i$ = 21.6--22.6 range.
622: The CaT lines are shown in the restframe, with spectra shifted to zero
623: velocity.
624: \medskip
625:
626: \centerline{\hbox{\psfig{file=f3.ps,angle=0,width=6.0in}}}
627: \noindent{\bf Figure 3.}
628: {\bf Top panel:}
629: The distance \andxii\ would have reached from M31 in the last
630: 10 Gyr, 725~kpc for the most probable radial distance from the
631: TRGB resampling distribution.
632: {\bf Bottom panel:}
633: The heliocentric distance probability curve for
634: \andxii, derived using our TRGB analysis and calibrated to an isochrone
635: fit to the RGB and HB.
636: \medskip
637:
638: \centerline{\hbox{\psfig{file=f4.ps,angle=0,width=6.0in}}}
639: \noindent{\bf Figure 4.}
640: The escape velocity is shown for M31 assuming the Geehan et al.\ (2005) mass model (solid line), derived from the circular velocity (dashed line),
641: as a function of three dimensional distance of M31 from its satellites
642: (data from Cote et al.\ 2000, and McConnachie et al. 2005).
643: \andxii\ is shown with a circle.
644: Recently discovered satellites AndXIV (Majewski et al.\ 2007) and
645: AndXI, AndXIII (Martin et al.\ 2006, Chapman et al.\ 2007) are
646: highlighted.
647:
648:
649:
650:
651: \end{document}
652:
653:
654:
655: