0709.1510/ms.tex
1: %\documentclass[aps,preprint]{revtex4}
2: \documentclass[aps,prd,twocolumn,superscriptaddress,showpacs,balancelastpage]{revtex4}
3: \usepackage{graphicx,subfigure}
4: 
5: \def\btt#1{{\tt$\backslash$#1}}
6: \input epsf
7: \def\plotone#1{\centering \leavevmode
8: \epsfxsize= 1.0\columnwidth \epsfbox{#1}}
9: \def\plottwo#1{\centering \leavevmode
10: \epsfxsize= 1.0\columnwidth \epsfbox{#1}}
11: \def\plotfiddle#1#2#3#4#5#6#7{\centering \leavevmode
12: \vbox to#2{\rule{0pt}{#2}}
13: \special{psfile=#1 voffset=#7 hoffset=#6 vscale=#5 hscale=#4 angle=#3}}
14: %\def\plotrotate#1{\centering \leavevmode
15: %\epsfxsize= 0.8\columnwidth  \epsfbox{#1  {angle=-90}}}
16: \def\plotrotate#1{\centering 
17: %\leavevmode
18: %\epsfxsize= 1.7\columnwidth 
19: \epsfbox{#1 angle=-90}}
20: \def\bff{}
21: %\newcommand\apj[3]{ {\it Astrophys. J.} {\bf #1}, #2 (19#3) }
22: \def\apjl{Astrophys. J. Lett. }
23: %\newcommand\prd[3]{ {\it Physical Review D} {\bf #1}, #2 (19#3) }
24: %\newcommand\prl[3]{ {\it Physical Review Letters} {\bf #1}, #2 (19#3) }
25: %\newcommand\np[3]{ {\it Nucl.~Phys.} {\bf #1}, #2 (19#3) }
26: \def\mnras{Mon. Not. R. astron. Soc. }
27: \def\araa{Annu. Rev. Astron. Astrophys. }
28: \def\aj{Astron. J. }
29: \def\asap{Astron. Astrophys. }
30: 
31: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32: %%%%%%%%%%%%%%%%%%%   begin local macros %%%%%%%%%%%%%%%%%%%%%%
33: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34: 
35: \font\FermiPPTfont=cmssbx10 scaled 1440
36: \font\FermiSmallfont=cmssq8 scaled 1200
37: \def\LANLppthead#1#2{
38: \null 
39: \begin{center}\vskip -1.0truein{\hbox to 7.5truein {
40: \hfill
41: \vbox to 1in {\vfill \FermiSmallfont
42:               \hbox{#1}
43:               \hbox{#2}
44:               \vfill}
45: }}\vskip-0.0truein\end{center}}
46: \def\be{\begin{equation}}
47: \def\ee{\end{equation}}
48: \def\bea{\begin{eqnarray}}
49: \def\eea{\end{eqnarray}}
50: \def\sm{{\rm M}_\odot}
51: \def\uline#1{$\underline{\smash{\hbox{#1}}}$}
52: 
53: \def\muk{\mu{\rm K}}
54: 
55: \def\Psusy{P_{\rm susy}} % Used in this paper
56: 
57: %\def\ang{\,{\rm\AA}}
58: %\def\flux{\,{\rm erg\,cm^{-2}\,arcsec^{-2}\,\AA^{-1}\,s^{-1}}}
59: \def\GeV{\,{\rm GeV}}
60: \def\kms{\,{\rm km\,s^{-1}}}
61: \def\VEV#1{\left\langle #1\right\rangle}
62: \def\fun#1#2{\lower3.6pt\vbox{\baselineskip0pt\lineskip.9pt
63:   \ialign{$\mathsurround=0pt#1\hfil##\hfil$\crcr#2\crcr\sim\crcr}}}
64: \def\C{{\cal C}}
65: \def\hyi{H\thinspace{$\scriptstyle{\rm I}$}~}
66: \def\hii{H\thinspace{$\scriptstyle{\rm II}$}~}
67: 
68: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
69: %%%%%%%%%%%%        MST local macros      %%%%%%%%%%%%%%%%%%%%%%%%%%%
70: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
71: \def\figcapsize{\footnotesize \baselineskip=0.1cm}
72: \def\reff#1{$^{{#1})}$}
73: \def\cc{{\rm~cm}^{-3}}
74: \def\hmpc{{\, {\rm h}^{-1}~\rm Mpc}}
75: \def\hkpc{{\, {\rm h}^{-1}~\rm kpc}}
76: \def\kev{{\rm~keV}}
77: \def\km{{\rm~km}}
78: \def\kpc{{\rm~kpc}}
79: \def\mpc{{\rm~Mpc}}
80: \def\msun{{\,M_\odot}}
81: \def\deg{^\circ}
82: \def\avrg#1{{\langle #1 \rangle}}
83: \def\bra#1{{\langle #1 \vert}}
84: \def\ket#1{{\vert #1 \rangle}}
85: \def\abs#1{{\vert #1 \vert}}
86: \def\eps{\varepsilon}
87: \def\pomega{\varpi}
88: \def\lbar{{\mathchar'26\mskip-9mu\lambda}}
89: \def\vs{\vskip 16pt}
90: \def\oneskip{\vskip 16pt}
91: \def\half{{\textstyle{1\over2}}}
92: \def\la{\mathrel{\mathpalette\fun <}}
93: \def\ga{\mathrel{\mathpalette\fun >}}
94: \def\fun#1#2{\lower3.6pt\vbox{\baselineskip0pt\lineskip.9pt
95:   \ialign{$\mathsurround=0pt#1\hfil##\hfil$\crcr#2\crcr\sim\crcr}}}
96: %%%%%%%%%
97: \def\boldsymbol{\bf}
98: 
99: \def\Vmax{V_{\rm max} }
100: \def\rmax{r_{\rm max}}
101: \def\rhos{\rho_s} 
102: \def\rs{r_s} 
103: \def\Aeff{A_{\rm eff}}
104: \def\texp{t_{\rm exp}}
105: \def\BV{B_{\rm V}} 
106: \def\BG{B_{\rm G}}
107: \def\Gev{{\rm GeV}}
108: \def\s{{\rm s}}
109: \def\cm{{\rm cm}}
110:  \def\Eth{E_{\rm th}}
111:  \def\Mchi{M_{\chi}}
112: 
113: \def\boldcdot{\mathbin{{\boldsymbol\cdot}}}
114: \def\boldnabla{{\boldsymbol\nabla}}
115: \font\BF=cmmib10
116: \def\Vmax{V_{\rm max}} 
117: \def\rmax{r_{\rm max}}
118: 
119: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
120: %%%%%%%%%%%%        end local macros      %%%%%%%%%%%%%%%%%%%%%%%%%%%
121: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
122: 
123: \begin{document}
124: 
125: % LANL LA-UR requirement
126: \LANLppthead{LA-UR 07-4969}{astro-ph/yymmnnn}
127: 
128: \title{The Most Dark Matter Dominated Galaxies: Predicted Gamma-ray Signals 
129: \\ from the Faintest Milky Way Dwarfs}
130: 
131: % Author list and affiliations
132: \author{Louis E. Strigari}
133: \email{lstrigar@uci.edu}
134: \affiliation{Center for Cosmology, Department of Physics and Astronomy, 
135: University of California, Irvine, CA 92697,
136:  USA}
137: %\affiliation{McCue Fellow} 
138: 
139: \author{Savvas M. Koushiappas}
140: %\email{smkoush@lanl.gov}
141: \affiliation{Theoretical Division,
142:   \& ISR Division, MS B227, Los Alamos National Laboratory, Los
143:   Alamos, NM 87545, USA}
144: 
145: \author{James S. Bullock}
146: %\email{bullock@uci.edu}
147: \affiliation{Center for Cosmology, Department of Physics and Astronomy, 
148: University of California, Irvine, CA 92697,
149:  USA}
150: 
151: \author{Manoj~Kaplinghat} 
152: %\email{mkapling@uci.edu}
153:  \affiliation{Center for Cosmology, Department of Physics and Astronomy,  University of
154: California, Irvine, CA 92697, USA}
155: 
156: \author{Joshua D. Simon}
157: %\email{jsimon@astro.caltech.edu}
158: \affiliation{Department of Astronomy, California Institute of Technology, 
159: 1200 E. California Blvd., MS105-24, Pasadena, CA 91125 USA}
160: 
161: \author{Marla Geha}
162: %\email{marla.geha@nrc-cnrc.gc.ca}
163: \affiliation{National Research Council of Canada, Herzberg Institute of Astrophysics,
164: 5071 West Saanich Road, Victoria, BC V9E, 2E7, Canada}
165: 
166: \author{Beth Willman} 
167: %\email{bwillman@cfa.harvard.edu}
168: \affiliation{Harvard-Smithsonian Center for Astrophysics, 60 Garden St. 
169: Cambridge, MA, 02138}
170: 
171: \date{\today}
172: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
173: \begin{abstract} 
174: We use kinematic data from three new, nearby, extremely low-luminosity 
175: Milky Way dwarf galaxies (Ursa Major II, Willman 1, and Coma
176: Berenices) to  constrain the properties of their dark matter halos,
177: and from these make predictions for the  $\gamma$-ray flux from
178: annihilation of dark matter particles in these halos. We show that
179: these $\sim 10^3$ L$_\odot$ dwarfs are the most dark matter dominated
180: galaxies in the Universe, with total masses within 100 pc in excess of
181: $10^6$ M$_\odot$. Coupled with their 
182: relative proximity, their large  masses imply that they should have
183: mean $\gamma$-ray fluxes comparable to or greater than any  other
184: known satellite galaxy of the Milky Way. Our results are robust to
185: both variations of  the inner  slope of the density profile and the
186: effect of tidal interactions. The fluxes could be boosted by up to two
187: orders of magnitude if we include the density enhancements caused by
188: surviving dark matter substructure. 
189: 
190: \end{abstract} 
191: 
192: \pacs{95.35.+d,14.80.Ly,98.35.Gi,98.62.Gq}
193: 
194: \maketitle
195: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
196: \section{Introduction} 
197: The census of the Local Group has changed dramatically in the last few
198: years. Prior to the turn of the century, there were only eleven known
199: satellite galaxies of the Milky Way (MW), with a discovery rate of
200: roughly one new Local Group satellite per decade \cite{Mateo:1998wg}.   
201: However, the Sloan Digital Sky Survey (SDSS) has been able to uncover 
202: a population of extremely  low-luminosity  satellite galaxies, which has roughly 
203: doubled the number of known satellites \citep{discovery,Zucker:2006bf,Belokurov:2006ph}. 
204: Determining how these new satellites fit in a given model for dark matter and 
205: cosmology presents a very exciting theoretical challenge.
206: 
207: The Cold Dark Matter (CDM) model predicts the existence of hundreds of
208: MW satellites that are expected to host galaxies at the faint
209: end of the luminosity  function \cite{missingsatellites}. The ability
210: of gas to cool and form stars in these low mass dark matter halos 
211: depends on a number of complex physical processes, such as  supernova
212: feedback, the photoionizing background, as well as mass loss
213: due to tidal interactions \cite{dsp}. Despite the broad range of
214: observed luminosities, the dark matter masses for all of the pre-SDSS
215: satellites are  constrained to within relatively narrow range,
216: approximately $\sim [1-6] \times 10^7$ M$_\odot$ within their inner
217: 600 pc  \cite{Walker:2007ju,Strigari:2007ma}.  
218: Understanding this strong luminosity bias at the low mass end is
219: crucial to deciphering the formation of these dwarf spheroidal (dSph) 
220: galaxies as well as to constraining the nature of dark matter.
221: 
222: In this paper we show that three new and nearby members of the Local Group
223: discovered by the SDSS (Willman 1, Coma Berenices, and Ursa Major II)
224: are likely to have masses comparable to their 
225: more luminous counterparts. Initial estimates have already shown that
226: these galaxies have mass-to-light  ratios similar to or larger than
227: the pre-SDSS dwarfs \cite{Martin:2007ic,Simon:2007dq}.  
228: With luminosities more than two orders of magnitude less than the
229: pre-SDSS dwarfs,  these new satellites are not only interesting in
230: the context of galaxy formation at the  lowest-mass scales, but also
231: for indirect dark matter detection. The new dwarfs are very faint, but 
232: they contain large amounts of dark matter and are located quite nearby, 
233: making them ideal sites to search for signals of dark matter annihilation.
234: 
235: Current and future observatories, including  
236: space-based experiments, such as GLAST \cite{GLAST}, as well as a suite of 
237: ground-based  Cerenkov detectors, such as STACEE \cite{Hanna:2002bf}, 
238: HESS \cite{Hofmann:2003kx}, MAGIC \cite{Cortina:2005pt}, 
239: VERITAS \cite{Weekes:2001pd}, CANGAROO \cite{Yoshikoshi:1999rg}, \& 
240: HAWK \cite{Sinnis:2005un}, will search for the signal 
241: of $\gamma$-rays from dark matter annihilations. 
242: The prospects for $\gamma$-ray detection from 
243: dark matter in well-known MW satellites with these observatories has been the
244: subject of many previous studies 
245: \cite{previousdwarfs,Bergstrom:2005qk,Strigari:2006rd}.
246: All of these systems are interesting targets not only because of their
247: large  mass-to-light ratios, but also because they are expected to
248: have very  low intrinsic  $\gamma$-ray emission. This is in contrast
249: to the situation at the Galactic center,   where astrophysical
250: backgrounds hinder the prospects of extracting the signal from  
251: dark matter annihilation \cite{galacticcenter}. Moreover, the known
252: location of MW  satellites makes a search of dark matter annihilation
253: well-defined, unlike the  search of completely dark substructure,
254: which would rely on  
255: serendipitous discovery \cite{microhalos,Koushiappas:2006qq,Diemand:2006ik}. 
256: 
257: From the mass modeling of the dark matter halos, we provide the 
258: first determination of the $\gamma$-ray
259: signal from dark matter from Ursa Major II,  Willman 1, and Coma
260: Berenices (Coma  hereafter). These galaxies provide promising targets for $\gamma$-ray
261: detection for three  reasons: 1) they are the among the closest dark matter dominated systems,  
262: 2) they are expected to be  free from intrinsic $\gamma$-ray emission, 
263: and 3) present data on their stellar kinematics suggest that their
264: dark matter halos are as massive as the more well-known population of
265: MW satellites.  
266: 
267: This paper is organized as follows. In section~\ref{sec:theoreticalmodeling}, 
268: we review  the theoretical modeling of dwarf dark matter halos and the 
269: calculation of the $\gamma$-ray flux. In section~\ref{sec:likelihoodsection}, 
270: we present the likelihood function for determining the flux, and outline the 
271: theoretical priors in the modeling. In sections~\ref{sec:results}
272: and~\ref{sec:conclusions} we present the results and  discussions. 
273: 
274: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
275: %%%%%%
276: \section{Theoretical Modeling} 
277: \label{sec:theoreticalmodeling}
278: 
279: When modeling the stellar distribution of a dwarf galaxy, it is important to 
280: determine the effect of external tidal forces on the dynamics of the system. 
281: For the MW satellites we study, we can obtain an estimate of 
282: the external tidal force by comparing its magnitude to the internal 
283: gravitational force. The internal gravitational force is $\sim \sigma^2/R$ and the 
284: external tidal force from the MW potential is $\sim (220 \, {\rm km}/{\rm s})^2R/D^2$, 
285: where $D$ is the distance to the dwarf from the center of the MW, and
286: $220 \, {\rm km}/{\rm s}$ is the MW rotation speed at the distance of
287: the dwarf. 
288: The MW satellites are characterized by scale radii of $R \sim
289: 10-100$ pc and velocity dispersions $\sigma \sim 5-10$  km s$^{-1}$.
290: For a typical distance of $D \sim 40$ kpc, the internal gravitational forces are thus 
291: larger by $\sim 100$. 
292: 
293: An additional estimate of tidal effects can be obtained by comparing 
294: the internal crossing times of the stars in the galaxy to the orbital
295: time scale of the system in the external potential of the host. 
296: For a galaxy with scale radii and velocity dispersions given
297: above, 
298: an estimate of the crossing time is given by  $R/\sigma \sim 1-20$
299: Myr. Assuming a rotation speed of $\sim 220$ km s$^{-1}$ at the 
300: distance of these dSphs ($\sim 40$ kpc), their orbital
301: time scale  in the MW potential is $\sim$ Gyr. 
302: 
303: From the above estimates we conclude that {\em it is highly 
304: unlikely that these galaxies are presently undergoing significant tidal stripping.}
305: These galaxies may have been tidally stripped before (for example, if their orbit 
306: took them closer to the center of the MW), however the stellar core that
307: has survived is faithfully tracing the local potential \cite{tides,Penarrubia:2007zx}. 
308: Of the galaxies we consider, only Ursa Major II shows strong evidence of past 
309: tidal interaction, as it is located on the same great circle as the Orphan Stream 
310: discovered by SDSS~\cite{Zucker:2006bf,Belokurov:2006ph}.
311: Thus we can proceed ahead with confidence in modeling the surviving stellar cores
312: as systems in dynamical equilibrium.  
313: 
314: Line-of-sight velocities are widely used to determine the
315: properties of the  dark matter halos of dSphs 
316: \cite{Lokas:2004sw,Mashchenko:2005bj,Gilmore:2007fy} assuming
317: spherical symmetry. 
318: For a system in dynamical equilibrium, the spherically symmetric Jeans
319: equation gives the  stellar line-of-sight velocity dispersion at a
320: projected radius, $R$,  from the center of the galaxy,  
321: \be
322: \sigma_{t}^{2}(R) = \frac{2}{I(R)} \int_{R}^{\infty} 
323: \left ( 1 - \beta \frac{R^{2}}{r^2} \right )
324: \frac{\rho_{\star} \sigma_{r}^{2} r}{\sqrt{r^2-R^2}} dr, 
325: \label{eq:sigma}
326: \ee 
327: where the three-dimensional velocity dispersion, $\sigma_{r}$ is obtained 
328: from 
329: \be 
330: r \frac{d(\rho_{\star} \sigma_r^2)}{dr} =  - \rho_{\star}(r) V_c^2(r)
331:         - 2 \beta(r) \rho_{\star} \sigma_r^2.
332: \label{eq:jeans}         
333: \ee 
334: Here $\beta$ is the stellar velocity anisotropy, and $\rho_\star(r)$
335: is the density profile for the  stellar distribution, which is
336: obtained from the projected stellar distribution, $I(R)$. 
337: The stellar distributions of the dSphs are typically fit with either  
338: Plummer or a King profiles \cite{King:1962wi}. The primary difference
339: between  these fits is that Plummer profiles are described by a  
340: single parameter, $r_P$, and fall-off as a power law in the outer 
341: regions of the galaxy, while King profiles are described by a core
342: radius,  $r_K$, and a cut-off radius, $r_{cut}$, and fall-off exponentially   
343: in the outer regions. For the stellar distributions of Ursa Major II
344: and Coma,  we use the Plummer fits compiled in \cite{Simon:2007dq},
345: and for Willman 1 we use the King profile fit from
346: \cite{Martin:2007ic}. These quantities, along with the 
347: distance to each galaxy and their respective luminosities, are given in 
348: Table~\ref{tab:plummerking}. 
349: 
350: %%%%%%%%%%%%%   BEGIN TABLE I  %%%%%%%%%%%%%%%%%%%%
351: \begin{table*}
352: \begin{ruledtabular}
353: \begin{tabular}{lccccc}
354:   dSph   &  Distance (kpc) & Luminosity ($10^3$ L$_\odot$) & Core Radius (kpc)  
355:   & Cut-off Radius (kpc) & Number of stars \\
356: \hline
357: Ursa Major II   & 32 &2.8& 0.127 (P)&  ---&20 \\
358: Coma Berenices  & 44 &2.6& 0.064 (P) &  ---&59\\
359: Willman 1   & 38 &0.9& 0.02 (K) & 0.08 (K)& 47 \\
360: Ursa Minor & 66 & 290& 0.30 (K) & 1.50 (K) &187\\   
361: \end{tabular}
362: \end{ruledtabular}
363: \caption{ \label{tab:plummerking} The distance, luminosity, core 
364: and cut-off radii (Plummer (P), King (K)) for each of the dSphs we study. 
365: The last column gives the total number of stars used in the analysis.}
366: \end{table*}
367: % End table%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
368: 
369: We model the distribution of dark matter in the dSphs with radial
370: density profiles of the form
371: \be
372: \rho(r) = \frac{\rho_s}{ \tilde{r}^\gamma (1 + \tilde{r} ) ^{3-\gamma}}, 
373: \end{equation}
374: where $\rho_s$ is the characteristic density, $\tilde{r}=r / r_s$, and $r_s$ is the scale radius. 
375: Numerical simulations bound $\gamma$ in the range $[0.7-1.2]$,  and
376: the outer slope  is $\sim -3$ \cite{slopes}. 
377: In order to compare the mass distribution in dSphs to dark matter
378: halos in N-body simulations,  it is useful to work in terms of the two
379: parameters $\Vmax$ and $\rmax$, the maximum circular velocity  
380: and the radius at which it is obtained. For example, for  $\gamma =
381: (0.8,1,1.2)$,  $\rmax/r_s = (2.61,2.16,1.72)$ . 
382: Thus, for a particular value of $\gamma$, the density profiles of dark
383: matter halos can be described by either $\rho_s$-$r_s$ or similarly by
384: $V_{\rm max}$-$r_{\rm max}$.  
385: 
386: With the parameters of the halo density profile specified, the
387: $\gamma$-ray  flux from dark matter annihilations is given by 
388: \begin{equation}
389: \Phi = \frac{1}{2} P \int_0^{\xi_{max}} \sin \xi d \xi \int_{{\eta_{-}}}^{{\eta_{+}}} 
390: \left[ \frac{\rho_s}{\tilde{r}^\gamma ( 1 + \tilde{r})^{3-\gamma}} \right]^2 d \eta, 
391: \label{eq:gammarayflux}
392: \end{equation}
393: where $\eta_\pm = D \cos \xi \pm \sqrt{r_t^2 - D^2 \sin^2 \xi}$,  $D$
394: is the distance to the galaxy, $\xi$ is the angular distance from the
395: center of the galaxy,  and $r_t$ is the tidal radius for the dark
396: matter halo. 
397: Note that in the limit of $D \gg r_s$, the flux
398: scales as $\int \rho(r)^2 dr/D^2$, and in the particular case where $\gamma = 1$, 
399: $\Phi \sim \rho_s^2 r_s^3/D^2$, with $\sim 90\%$ of the flux originating within $r_s$.
400: 
401: In Eq.~(\ref{eq:gammarayflux}), the properties of the
402: dark matter particle are determined by 
403:  \be P= \frac{\langle  \sigma v \rangle}{M_x^2}  \int_{\Eth}^{\Mchi}
404: \frac{dN}{dE} dE.
405: \label{eq:Psusy}
406: \ee 
407: Here, $\Eth$ is a threshold energy, $\Mchi$ is the mass of the dark matter particle, 
408: $\langle \sigma v \rangle$ is the annihilation cross section, and the spectrum of 
409: the emitted $\gamma$-rays is given by $dN/dE$. Unless otherwise noted,  we assume 
410: $P \approx 10^{-28} {\rm cm^3} {\rm s}^{-2} {\rm GeV}^{-2}$, which corresponds to the most optimistic 
411: supersymmetric dark matter models. However, we stress that the derived results can be rescaled to any 
412: dark matter model, by a simple rescaling of $P$. 
413: 
414: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
415: \section{Likelihood Function and Priors}
416: \label{sec:likelihoodsection}
417: Observed stellar line-of-sight velocities place strong constraints on  several 
418: important parameters describing the dark matter halos of dSphs. Two  examples 
419: of these parameters are the halo mass and density at a characteristic halo radius, 
420: corresponding to about twice the King core radius (or about 600 pc for a typical 
421: dSph \cite{Walker:2007ju,Strigari:2007ma}). More relevant to $\gamma$-rays is the 
422: quantity $\rho_s^2 r_s^3$ (the $\gamma$-ray luminosity is ${\cal L} \sim \rho_s^2 r_s^3$), 
423: which is typically determined to within a factor $3-6$ with the line-of-sight velocities of 
424: several hundred stars \cite{Bergstrom:2005qk,Strigari:2006rd}. The constraints on these 
425: parameters can be strengthened by including relations between similar parameters 
426: observed in numerical simulations. When combined with the observational constraints, 
427: these empirical relations in numerical simulations constitute a theoretical prior, delineating 
428: a preferred region of the parameter space of the dark matter distribution in dSphs 
429: \cite{Strigari:2006rd,Penarrubia:2007zz}. In this section, we discuss the implementation of 
430: this prior, and derive the general form of the likelihood function we use to constrain the 
431: $\gamma$-flux from dark matter annihilations.
432: 
433: We assume the line-of-sight velocities are drawn from a Gaussian
434: distribution, centered on the true value of the mean velocity,
435: $u$. This has been shown to be a good description of  the well-studied
436: dwarfs with line-of-sight velocities of several hundred stars \cite{Walker:2005:nt}.
437: Given the set of theoretical parameters, the probability to obtain the set of observed 
438: line-of-sight velocities, $\vec{\bf x}$, is 
439: \be
440: P({\bf \vec{x}}|{\vec{\bf \theta}} ) 
441: = \prod_{i=1}^n 
442: \frac{1}{\sqrt{2 \pi ( \sigma_{t,\imath}^2 + \sigma_{m,\imath}^2})}
443: \exp \left [ - \frac{1}{2}\frac{(v_\imath - u)^2}{\sigma_{t,\imath}^2 
444: + \sigma_{m,\imath}^2} \right ], 
445: \label{eq:likelihood} 
446: \ee 
447: Here $\vec{\theta}$ is the set of parameters that describe the dSph, and
448: the sum is over the observed total number of stars.
449: The dispersion in the velocities thus has two sources:  1) the intrinsic  
450: dispersion, $\sigma_{t,\imath}(\vec{\theta})$, which is a function of the position
451: of the $\imath^{th}$ star,  and 2) the uncertainty stemming from the
452: measurement, $\sigma_{m,\imath}$.
453: 
454: We can simplify Eq.~(\ref{eq:likelihood}) by assuming that the
455: measurement uncertainties are small  relative to the intrinsic 
456: dispersion.  This is a good approximation for well-studied dwarfs,
457: which have intrinsic dispersions  $\sim 10 \, {\rm km} \, {\rm
458: s}^{-1}$, and measurement errors $\sim 1 \, {\rm km} \, {\rm s}^{-1}$
459: \cite{dwarfdata}. Under this approximation Eq.~(\ref{eq:likelihood})
460: becomes
461: \begin{equation}
462: P({\bf \vec{x}}|{\vec{\bf \theta}}) =
463: %{\cal L}(\vec{\theta})=
464: \prod_{\imath=1}^{N_B}
465: \frac{1}{\sqrt{2\pi\sigma_{t,\imath}^2}}
466: \exp\left[-\frac{1}{2}
467: \frac{N_\imath\hat{\sigma}_{t,\imath}^2}{\sigma_{t,\imath}^2}\right]\,,
468: \label{eq:bin-approx} 
469: \end{equation}
470: where the sum is now over the number of bins, $N_B$, for which the
471: velocity dispersion  is determined. 
472: The velocity dispersion in the $\imath^{th}$ bin is $\hat{\sigma}_{t,\imath}^2$, and
473: the number of stars in the $\imath^{th}$ bin is 
474: $N_\imath$. We can use Eq.~({\ref{eq:bin-approx}) if the 
475: observations are given by line-of-sight velocity dispersions, 
476: and if the measurement errors are small in comparison to 
477: the intrinsic dispersion. This is the case for Ursa Minor, as discussed
478: below. 
479: 
480: We describe the dark matter halos in terms of the parameters 
481: $\vec{\theta} = (\Vmax, \rmax,\beta$). We assume that $\beta$ is constant
482: as a function of radius, and let it vary over the range $[-5:1]$. 
483: We integrate Eqs.~(\ref{eq:likelihood}) and (\ref{eq:bin-approx}) over 
484: these parameters and 
485: define the likelihood function for a fixed $\gamma$-ray flux, $f$,  as   
486: \bea  
487: {\cal L}(f)
488: &\propto& \int P({\bf \vec{x}} | V_{\rm max}, r_{\rm max},\beta)
489: P(V_{\rm max} , r_{\rm max})  \nonumber \\ &\times& \delta (\Phi(\Vmax,\rmax)-f)
490: dV_{\rm max} d r_{\rm max} d \beta.
491: \label{eq:Lflux}
492: \eea 
493: Here we have assumed a uniform prior on $\beta$, and  
494: the prior probability distribution for $V_{\rm max} , r_{\rm max}$ is 
495: given by $P(V_{\rm max} , r_{\rm max})$. This prior distribution is 
496: determined by the $\Vmax$-$\rmax$ relation from CDM simulations. 
497: 
498: In order to determine $P(V_{\rm max} , r_{\rm max})$,  we need both its mean
499: relation and its halo-to-halo scatter.
500: For dark subhalos that have been strongly affected by tidal interactions, 
501: the $V_{\rm max} - r_{\rm max}$ relation is strongly
502: dependent on the nature  of the potential of the host system, as
503: these systems undergo varying amounts of mass loss  as they evolve
504: within the host halo. For example, Bullock and Johnston~\cite{BJ07} 
505: have embedded a disk potential in
506: a MW size host halo, and found that the $\Vmax-\rmax$ relation of subhalos 
507: takes the form $\log (r_{\rm max}) = 1.35 [
508: \log (V_{\rm max}) -1] - 0.196$.  We obtain a similar slope by
509: examining the subhalos in the dark matter-only Via Lactea  simulation
510: of MW substructure \cite{Diemand:2006ik}, however differences
511: in the assumed  cosmological parameters, and the absence of a disk
512: potential in Via Lactea, translate into  differences in the
513: normalization of the $V_{\rm max}-r_{\rm max}$ relation. 
514: For Via Lactea subhalos, we find  the normalization is reduced by $\sim
515: 30\%$, which implies reduced halo concentrations (larger $r_{\rm max}$
516: for fixed $V_{\rm max}$).
517: 
518: We model the scatter in  $V_{\rm max}$ as a log-normal distribution,
519: with $\sigma_{\log V_{\rm max}} \simeq 0.20$. This
520: provides a conservative  estimate for the scatter in $V_{\rm max}$ as
521: a function of $r_{\rm max}$ for nearly the entire  range of the
522: subhalo mass function. At the extremely high end of the subhalo mass function 
523: ($V_{\rm max} \gtrsim 20 \, {\rm km} \, {\rm s}^{-1}$),  the scatter
524: increases because in this range it is dominated by a small
525: number of very massive systems that have been accreted into the host
526: halo very recently. 
527: This increase  in the scatter is simply because
528: Via Lactea is only one realization of a substructure population in a MW halo.
529: We find that by excluding the extreme
530: outliers in the Via Lactea mass function, the scatter is similar to
531: the low mass regime. This is a similar result to those obtained in semi-analytic 
532: models of many realizations of the subhalo population 
533: \cite{Zentner:2003yd,van den Bosch:2004zs}.
534: 
535: % Mass Table%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
536: \begin{table}
537: \begin{ruledtabular}
538: \begin{tabular}{lcc}
539:   dSph   &  Mass $<$ 100 pc [$10^6$ M$_\odot$] & V$_{\rm max}$ [km s$^{-1}$]\\
540: \hline
541: Ursa Major II    & $3.1_{-1.8}^{+5.6}$& $23_{-10}^{+69}$\\
542: Coma Berenices & $1.9_{-1.0}^{+2.1}$&$19_{-9}^{+53}$ \\
543: Willman 1   & $1.3_{-0.8}^{+1.5}$ &$27_{-15}$ \\
544: Ursa Minor & $2.3_{-1.2}^{+1.9}$& $30_{-16}^{+12}$\\
545: \end{tabular}
546: \end{ruledtabular}
547: \caption{\label{tab:masstable}
548: The masses within 100 pc and the maximum circular velocities
549: of the Milky Way satellites we study. Error bars indicate the $90\%$ c.l. regions. 
550: No upper limit could be obtained for the maximum circular velocity of 
551: Willman 1. 
552:  }
553: \end{table}
554: % vmax's. w1 > 12:27., com 10-72,19: um2 13-92,23
555: % End table%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
556: 
557: %%%%%%%%%%%%%%%%%%%%%%%%%%%
558: 
559: %%%%%%%%%%%%% FIGURE 1 %%%%%%%
560: \begin{figure}[hbtp]
561: \begin{center}
562: \includegraphics[height=8.cm]{fig1.ps}  
563: \caption{\small The $90\%$ confidence level 
564: region in the $\rho_s-r_s$ (top) $\rho_s^2 r_s^3-\rhos$ (bottom) parameter space 
565: for Coma , Ursa Major II, Willman 1, and Ursa Minor. 
566: We marginalize over the velocity anisotropy and have assumed an 
567: inner slope of $\gamma =1$. The best-fit values are indicated with points. 
568: \label{fig:rhosrs}}
569: \end{center}
570: \end{figure}
571: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%
572: 
573: \section{Results}
574: \label{sec:results}
575: \subsection{Flux estimates for smooth dark matter distributions}
576: 
577: We now quantify the prospects for detecting $\gamma$-rays from dark matter 
578: annihilation in the three new dSphs. We first assume that the dark matter is distributed 
579: smoothly, and we discuss the implications of a boost factor due to substructure in the 
580: next subsection. We use observations of these galaxies from the following references: 
581: Ursa Major II and Coma  \cite{Simon:2007dq}, Willman 1 \cite{Willmaninprep}. For these 
582: galaxies we have individual stellar velocities, so we use the likelihood function in 
583: Eq.~(\ref{eq:likelihood}). To make a connection to previous studies of dSphs, we compare 
584: the fluxes for these new dSphs to the flux from Ursa Minor, which is at a distance $D=66$ 
585: kpc, and has a luminosity of $L=2.9 \times 10^5$ L$_\odot$. Ursa Minor has the largest flux of 
586: any of the well-known dwarfs \cite{Strigari:2006rd}. We describe the stellar distribution of 
587: Ursa Minor with a King profile, with $r_K = 0.30$ kpc and $r_{cut} =1.50$ kpc 
588: \cite{Munoz:2005be}.  For Ursa Minor we use the measured velocity dispersion from a 
589: sample of 187 stars distributed evenly in 11 bins \cite{Munoz:2005be}, and we use the 
590: likelihood function in Eq.~(\ref{eq:bin-approx}). 
591: 
592: %%%%%%%%%%%%%%%% FIGURE 2 %%%%%%%%%%%%
593: \begin{figure}[hbtp]
594: \begin{center}
595: \includegraphics[height=8cm]{fig2.ps}   
596: \caption{\small The probability distributions of the angular size subtended by $r_s$ 
597: for each galaxy. We 
598: marginalize over the velocity anisotropy and $\rho_s$. The inner 
599: slope is fixed to $\gamma=1$. 
600: \label{fig:rs}}
601: \end{center}
602: \end{figure}
603: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
604: 
605: In the top panel of Fig.~(\ref{fig:rhosrs}), we show the $90\%$ c.l. 
606: region in the $\rhos -\rs$ plane for each galaxy, where the best-fit 
607: values are denoted by points. Here we use the $\Vmax-\rmax$ prior from 
608: section~\ref{sec:likelihoodsection}, and we take the inner slope of the dark matter
609: halo profile to be $\gamma = 1$. In the bottom panel 
610: of Fig.~(\ref{fig:rhosrs}) we show the $90\%$ confidence level region in the 
611: $\rhos^2 \rs^3-\rhos$ plane. As seen in Fig.~(\ref{fig:rhosrs}), the range of 
612: values that $\rhos^2 \rs^3$ can take in each dSph is reduced with the inclusion 
613: of more stars (e.g. Ursa Minor vs any one of the other three dSphs). 
614: 
615: The constrained regions in Fig.~(\ref{fig:rhosrs}) can be used to determine the
616: masses, maximum circular velocities, and $\gamma$-ray flux 
617: probability distributions for each galaxy. In
618: Table~\ref{tab:masstable}, we show  
619: the masses within 100 pc and the maximum circular velocities for each galaxy. The 
620: error bars indicate the $90\%$ c.l. regions. 
621: 
622: To determine the flux distributions, we must 
623: first specify a solid angle for integration. For optimal detection
624: scenarios, the solid angle should  encompass the region with the
625: largest signal-to-noise. For the present work, we will integrate over
626: a region where 90\% of the flux originates. As discussed above, for
627: the particular case where 
628: $\gamma=1$, 90\% of the flux originates  within $\rs$. Therefore, in
629: order to estimate the solid angle of  integration, we have to
630: first determine the maximum likelihood values of $\rs$. This is  done
631: by marginalizing over $\rhos$ and $\beta$ with the $\Vmax - \rmax$
632: prior.  The distributions of angular sizes are then obtained from   
633: $\mu = \tan^{-1}[ \rs / D]$, where $D$ is the distance to the dSph.  
634: As shown in Fig.~(\ref{fig:rs}), we find that given their similar 
635: size and roughly similar distances, all three dSphs will emit 90\% of 
636: their $\gamma-$ray flux within a region of $\sim 0.2$ degrees,
637: centered on each dSph (for $\gamma=1$).  
638: Ursa Minor is the most physically extended galaxy, subtending the
639: largest projected area on the sky.  
640: 
641: %%%%%%%%%%%%%%%%%% FIGURE 3 %%%%%%%%%%%%
642: \begin{figure}[hbtp]
643: \begin{center}
644: \includegraphics[height=8cm]{fig3.ps}  
645: \caption{\small The probability distributions for the $\gamma$-ray fluxes 
646: from Coma, Ursa Major II, Willman 1, and Ursa Minor, marginalizing over 
647: the velocity anisotropy, $\rho_s$, and $r_s$. We assume 
648: $P= 10^{-28} \, {\rm cm}^{3} \, {\rm s}^{-1} \, {\rm GeV}^{-2}$ and an inner 
649: slope of $\gamma = 1.0$. We have assumed no boost from 
650: halo substructure, which increases these fluxes by a factor $\sim 10-100$.  
651: }
652: \label{fig:flux}
653: \end{center}
654: \end{figure}
655: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
656: 
657: It is important to determine whether each of the galaxies will be detected as point sources, 
658: or whether they will be resolved as extended objects. To determine this we compare their 
659: angular size to the angular resolution of $\gamma$-ray telescopes. GLAST will have a single 
660: photon angular resolution of $\sim 10$ arcminutes for energies greater than 1 GeV, similar to 
661: the angular resolution of ground-based detectors (such as VERITAS) for energies greater than 
662: few tens of GeV. In the case where the detected number of photons is $N_\gamma > 1$, the 
663: angular resolution of a detector is improved by a factor of $1 / \sqrt{N_\gamma}$. Therefore, 
664: these galaxies can be resolved as extended objects, which in principle would allow a measured 
665: flux to determine the distribution of dark matter in the halo itself. 
666: 
667: Fig.~(\ref{fig:flux}) depicts the resulting flux probability distribution for the three new 
668: dSphs and Ursa Minor. These 
669: are obtained by marginalizing over $\beta$, $\rho_s$, and $r_s$ and including 
670: the $V_{\rm max}$-$r_{\rm max}$ prior. 
671: We set the inner slope to $\gamma = 1$, and integrate the flux over the solid angle 
672: that corresponds to 0.2 degrees from the center of the galaxy. We assume a value of 
673: $P=P_0=10^{-28} \cm^3 \s^{-1} \GeV^{-2}$, but the result can be scaled to any 
674: dark matter candidate with a different value of $P$ 
675: by simply multiplying the flux distribution by a factor of $P/P_0$. 
676: 
677: %%%%%%%% FIGURE 4 %%%%%%%%%%%%%%%%
678: \begin{figure}[hbtp]
679: \begin{center}
680: \includegraphics[height=5.cm]{fig4.ps}
681: \caption{\small The $\gamma$-ray flux probability distributions for Coma  and Ursa Minor 
682: for inner slopes of 0.8 (long-dashed), 1.0 (solid), and 1.2 (short-dashed). We marginalize over 
683: the same quantities as in Fig.~(\ref{fig:flux}). The value of $P$ is the same as in Fig.~(\ref{fig:flux}).
684: \label{fig:slope}
685: }
686: \end{center}
687: \end{figure}
688: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
689: 
690: The relative proximity of the three new dSphs, and their comparable sizes, results in 
691: $\gamma$-ray fluxes that are roughly similar. For $P\approx P_0$, the likelihood peaks at 
692: approximately $\Phi_0 \approx 10^{-10} \cm^{-2} \s^{-1}$, with a spread of nearly an order of 
693: magnitude. Thus Ursa Major II, Coma, and Ursa Minor all have comparable fluxes, 
694: and Willman 1 has a most likely flux that is about three times larger than Ursa Major II 
695: or Coma. 
696: 
697: %{\it The three new dSphs considered here are a factor of at
698:  % least three brighter than  Ursa Minor, which was previously
699: %  considered as the brightest dSph in the MW \cite{Strigari:2006rd}. } 
700: %This is mainly because the new dSphs have similar masses 
701: %to Ursa Minor, but are at closer distances. 
702: 
703: \subsection{The effects of the inner slope and substructure boost factors}
704: 
705: Understanding the distribution of dark matter in the inner regions of the dSphs also 
706: has important implications for detection of a $\gamma$-ray flux. 
707: However, when varying the inner slope, we must also be certain to vary all of the other
708: halo parameters so as to remain consistent with the line-of-sight observations. 
709: In order to quantify the effects of varying the inner slope, we marginalize over $\Vmax$, $\rmax$ and $\beta$ for 
710: profiles with different values of $\gamma$. In Fig.~(\ref{fig:slope}) we show the effects of 
711: varying $\gamma$ for Coma  and Ursa Minor. The shifts in the flux distribution function are 
712: not only a result of varying the inner slope but also come from the constraints imposed 
713: by the data on the density profile parameters $\rhos$ and $r_s$.  The relative amount of
714: the shifts can be understood by considering the best-fitting values of $r_s$ (where the
715: majority of the $\gamma$-ray flux comes from) as compared to the core radii of the 
716: systems. When the core radius is similar to $r_s$, the shifts are larger for varying 
717: $\gamma$, as in the case of Ursa Minor. However, when the core radius is much smaller 
718: than the fitted values of $r_s$, variations in the inner slope are less significant than the 
719: change induced by $\rhos$, as is the case for Coma. 
720: 
721: The presence of substructure in dark matter halos is firmly established on theoretical and 
722: numerical grounds. Dark matter halos are approximately self-similar, and substructure is 
723: expected to be present in all dark matter halos with mass greater than the cut-off scale 
724: in the primordial power spectrum, set by the kinetic decoupling temperature of the 
725: dark matter particle (for detecting the smallest dark matter halos in the Milky Way, see 
726: \cite{Koushiappas:2006qq}). It is therefore natural to expect that these galaxies contain 
727: substructure if they consist of CDM. 
728: 
729: The density enhancement over the smooth distribution of dark matter leads to an 
730: enhancement in the total annihilation rate, typically quantified in terms of a ``boost" factor. 
731: As was shown in \cite{Strigari:2006rd}, the boost factor cannot attain arbitrarily large values, 
732: but instead is bounded to be less than $\sim 100$, with the exact value depending on the 
733: cut-off scale in the CDM mass function. 
734: The boost is a multiplicative quantity, so the effect of dark substructure is simply 
735: accounted for by scaling the fluxes in Fig.~(\ref{fig:flux}) by the appropriate boost factor. 
736: 
737: \subsection{Detection prospects}
738: 
739: As is shown in Fig.~(\ref{fig:flux}), the flux probability
740: distribution functions peak around  
741: $\Phi_0 \approx 10^{-10} \cm^{-2} \s^{-1}$ without including any
742: enhancement to the  signal from substructure. Here, we will assume a
743: conservative value of 10 for the  boost factor, and discuss the
744: prospects for detecting the three new dSphs with $\gamma$-ray
745: instruments.  We can make simple estimates for the likelihood for
746: detection by adopting the specifications  
747: of particular $\gamma$-ray detectors. We will use two examples: a
748: space-based experiment,  
749: GLAST, and a ground-based Cerenkov telescope, VERITAS. 
750: For GLAST, if we assume an orbit-averaged effective area of $\Aeff
751: \approx 2 \times 10^3 \cm^2$  
752: and an exposure time of $\texp = 10$ years, their product is 
753: $B_{\rm G } = \Aeff \texp \approx 3 \times 10^{11} \cm^2 \s$.  A 50
754: hour exposure with  VERITAS ($\Aeff \approx 10^8 \cm^2$) has $B_{\rm
755:   V} \approx  2 \times 10^{13} \cm^2 \s$.  
756: Naively, for a fixed value of $P$, a ground-based detector seems more
757: sensitive because $\BV > \BG$. However, the backgrounds for a
758: ground-based detector are also larger and include a component from
759: the hadronization of cosmic rays in the atmosphere.
760: 
761: As an example, for a fixed value of $ P = P_0$, and a solid angle
762: that  corresponds to an angular size of 0.2 degrees, the number of
763: photons detected  by GLAST is $N_{\gamma, G} \approx 300$. The
764: dominant source of background  for GLAST is the Galactic diffuse
765: emission ($dN_B /dE = 1.2 \times 10^{-6} [\Eth/ \GeV]^{-1.1} \cm^{-2}  
766: \s^{-1} {\rm sr}^{-1} \GeV^{-1}$ \cite{Hunter:1997we}). If we assume an
767: energy threshold  
768: of $\Eth \approx 1 \GeV$, then the number of background photons is 
769: $N_B = B_{\rm G} \, dN_B/ dE \approx 250$, which means that the new satellites will 
770: be detected at approximately a $N_{\gamma,G} / \sqrt{N_B+N_{\gamma,G}}  
771: \approx 12 \sigma$ level. A similar estimate can be obtained for
772: VERITAS. The number  
773: of photons detected above $50\GeV$ in an instrument with an effective
774: area times exposure $\sim B_{\rm V}$ is  
775: $N_\gamma \approx 2 \times 10^4$. The dominant contribution to the
776: background are  photons that originate from neutral pion decays from
777: the nuclear interactions of cosmic  rays in the upper layers of the
778: atmosphere  
779: ($dN_B/dE = 3.8 \times 10^{-3} [\Eth / \GeV]^{-2.75} \cm^{-2} \s^{-1}
780: {\rm sr}^{-1} \GeV^{-1}$  
781: \cite{Nishimura:1980pz}). The number of background photons from pion decays is 
782: approximately $N_B \approx 2 \times 10^7$, therefore the 3 dSphs could be detected at 
783: a $\nu \approx 5 \sigma$ level. Understanding and discriminating against the background 
784: contamination in Cerenkov telescopes is very important in improving the prospects for 
785: detecting dSphs of the Milky Way. 
786: 
787: As shown in \cite{Strigari:2006rd}, the large number of stellar
788: velocities obtained in the older dSphs allow useful $\gamma$-ray
789: flux ratios between different dSphs to be determined. For the 
790: three new dSphs considered in this work, the kinematic  
791: data is not good enough to play the same game. Clearly, more stellar
792: velocities will shrink the allowed region of $\rhos^2 \rs^3$ parameter
793: permitting robust estimates of flux ratios between the galaxies
794: studied here and the rest  of the Milky Way satellites.  
795: 
796: \section{Discussion \& Conclusions}
797: \label{sec:conclusions}
798: In this paper, we have modeled the dark matter distribution in three recently-discovered 
799: Milky Way (MW) satellites (Ursa Major II, Willman 1, and Coma Berenices), and have presented 
800: the prospects for detecting $\gamma$-rays from dark  matter annihilations in their halos.  
801: We show that the expected flux from these galaxies 
802: is larger than the flux from any of the higher luminosity, more well-known (pre-SDSS)
803: dwarfs. There are two reasons for this surprising result: 1) the masses of
804: these new dwarfs within their stellar distributions are similar to the
805: masses of the well-known, larger luminosity  dwarfs, and 2) all 
806: three new galaxies are closer than the other well-known
807: dwarfs. The implied mass-to-light ratios, $\sim 1000$, of these new
808: dwarfs  makes them the most dark matter dominated galaxies in the
809: Universe.  
810: 
811: Our estimates show that it is unlikely that the observed stellar
812: distributions are presently undergoing tidal disruption. However,
813: this does not mean that they have been free from tidal interactions in
814: the past, but  rather that the surviving stellar core can be
815: faithfully modeled as a system in 
816: dynamical  equilibrium. By including the $\Vmax-\rmax$ CDM prior, we
817: have naturally accounted for tidal effects in the mass modeling, since
818: this $\Vmax-\rmax$ relation in fact comes  from dark matter halos that
819: have experienced tidal stripping.  
820: 
821: One of the galaxies we consider, Ursa Major II, may be a candidate for
822: past tidal disruption,  given that it is positioned on the same great
823: circle as the Orphan Stream of stars, which was also recently detected by
824: the SDSS~\cite{Zucker:2006bf,Belokurov:2006ph}.  
825: This is consistent with the findings of Simon and Geha
826: \cite{Simon:2007dq}, who have recently investigated the possibility of
827: tidal disruption in Ursa Major II, as well as all of the  other new
828: dwarfs, using proxies such as gradients in the observed velocity
829: distribution and metallicity of the stellar populations. Given the
830: total mass-to-light  ratio we have determined for Ursa Major II, tidal
831: stripping will have only been significant if its pericenter is $\sim$
832: 3 times closer than its  present distance. Future observations of the
833: stellar distributions in Ursa Major II, and all of the other new   
834: faint dwarfs, will be important in determining bound and unbound
835: stellar populations.  With a larger sample of stars from a galaxy such
836: as Ursa Major II, it will be possible to  remove unbound and
837: interloping stars with  techniques similar to those presented in
838: \cite{tides}. Upon removal of  stars unbound to the galaxy, these
839: authors show that in most cases the true bound  mass of the system can
840: be recovered to typically better than $25\%$. 
841: 
842: As a very conservative check for the effects of membership
843: uncertainties, we have redone the analysis for each of the galaxies by
844: just keeping the stars within the inner half of each galaxy, where the
845: surface densities are the most well-determined. We find that the peak
846: of the  flux likelihood is shifted by a small amount relative to the
847: 1-$\sigma$ widths in Fig.~(\ref{fig:flux}). Note however, that even
848: when including the entire population of stars in the observed
849: samples, in all cases equilibrium models provide adequate descriptions
850: of the dynamics of each system. 
851: 
852: Unresolved binary star systems also introduce a systematic that may
853: effect the fluxes we have presented.  Olszewski, Pryor, and Armandroff
854: \cite{Olszewski:1995gs} have determined the effect of binaries on the
855: velocity dispersion of two of the most luminous dwarfs, Draco and Ursa
856: Minor, by inferring the binary population of these systems  using
857: multiple epoch observations. They find a velocity dispersion of  $\sim
858: 1.5 \, {\rm km} \, {\rm s}^{-1}$ due to binaries, and a probability of   
859: $5 \%$ that binaries elevate the velocity dispersion to $4 \, {\rm km}
860: \, {\rm s}^{-1}$,  
861: which is still less than the velocity dispersion of the three new dwarfs. 
862: Thus if Ursa Major II, Willman 1, and Coma have binary fractions
863: similar to Draco and Ursa Minor, their observed velocity dispersions
864: are not significantly affected by  binaries. We note that this is
865: consistent with recent estimates of the binary fraction in low density
866: Galactic globular clusters \cite{Sollima:2007sc}.  
867: 
868: A strong test for the presence of binaries is to examine the
869: distribution of measured velocities. The velocity
870: distribution due to internal motion in binary systems should be flat
871: due to the observed broadness of the period distribution of binaries  
872: \cite{Duquennoy:1991zu}. The observed period distributions depend
873: on spectral type and age of the system (among other variables) but are
874: all broad with dispersions of about two orders of magnitude, which
875: seems to be consistent with theoretical expectations
876: \cite{Fisher:2003tg}. Further, there is no observational evidence or 
877: theoretical argument that suggests that the period distribution
878: should be sharply suppressed for all periods below $\sim 1000$ years 
879: (roughly velocities larger than $\sim 3$ km/s). Thus if there is a
880: large contribution from internal motion in binary stellar systems to
881: the intrinsic velocity dispersion of these dwarfs, we expect to see a
882: significant tail of high velocities. This is not observed and hence we
883: can be confident that measured velocity  dispersion is tracing the
884: total mass in the dwarf galaxy.  
885: 
886: The new, ultra-low luminosity galaxies represent an interesting
887: confluence of astronomical and $\gamma$-ray studies.  
888: Future kinematic studies of all of these new dwarfs will 
889: further reduce  the errors on the mass distributions, and sharpen the
890: predictions for $\gamma$-ray observatories searching for signatures of
891: dark matter annihilations. 
892: 
893: %\bigskip 
894: %\bigskip 
895: 
896: \section{Acknowledgments} 
897: We are grateful to Simon White for many useful discussions on likelihood 
898: functions for dwarf galaxies; Jay Strader and Connie Rockosi for
899: Willman 1 data;    Juerg Diemand, Michael Kuhlen, and Piero Madau for
900: making the data from  the  Via Lactea simulation publicly available;
901: and John Beacom, Nicolas Martin, and  Joe Wolf for useful
902: discussions. JSB, LES, and MK are supported in part by  NSF  grant
903: AST-0607746. Work at LANL was carried out under the auspices of the
904: NNSA  of the U.S. Department of Energy at Los Alamos National
905: Laboratory under  Contract No.  DE-AV52-06NA25396. JDS gratefully
906: acknowledges the support of a  Millikan Fellowship provided by
907: Caltech.  
908: 
909: \begin{thebibliography}{99}
910: 
911: %\cite{Mateo:1998wg}
912: \bibitem{Mateo:1998wg}
913:   M.~Mateo,
914:   %``Dwarf Galaxies of the Local Group,''
915:   Ann.\ Rev.\ Astron.\ Astrophys.\  {\bf 36}, 435 (1998)
916:   [arXiv:astro-ph/9810070].
917:   %%CITATION = ARAAA,36,435;%%  
918: 
919: \bibitem{discovery}
920: %\cite{Willman:2005cd}
921: %\bibitem{Willman:2005cd}
922:   B.~Willman {\it et al.},
923:   %``A New MW Dwarf Galaxy in Ursa Major,''
924:   Astrophys.\ J.\  {\bf 626}, L85 (2005)
925:   [arXiv:astro-ph/0503552]; 
926:   %%CITATION = ASJOA,626,L85;%%
927: %\cite{Irwin:2007jz}
928: %\bibitem{Irwin:2007jz}
929:   M.~J.~Irwin {\it et al.},
930:   %``Discovery of an Unusual Dwarf Galaxy in the Outskirts of the MW,''
931:   Astrophys.\ J.\  {\bf 656}, L13 (2007)
932:   [arXiv:astro-ph/0701154];
933:   %%CITATION = ASJOA,656,L13;%%
934: %\cite{Walsh:2007tm}
935: %\bibitem{Walsh:2007tm}
936:   S.~M.~Walsh, H.~Jerjen and B.~Willman,
937:   %``A Pair of Bootes: A New MW Satellite,''
938:   arXiv:0705.1378 [astro-ph]. 
939:   %%CITATION = ARXIV:0705.1378;%%  
940: 
941: %\cite{Zucker:2006bf}
942: \bibitem{Zucker:2006bf}
943:   D.~B.~Zucker {\it et al.},
944:   %``A curious MW satellite in Ursa Major,''
945:   Astrophys.\ J.\  {\bf 650}, L41 (2006)
946:   [arXiv:astro-ph/0606633]. 
947:   %%CITATION = ASJOA,650,L41;%%    
948:     
949: %\cite{Belokurov:2006ph}
950: \bibitem{Belokurov:2006ph}
951:   V.~Belokurov {\it et al.}  [SDSS Collaboration],
952:   %``Cats and Dogs, Hair and A Hero: A Quintet of New MW Companions,''
953:   Astrophys.\ J.\  {\bf 654}, 897 (2007)
954:   [arXiv:astro-ph/0608448]. 
955:   %%CITATION = ASJOA,654,897;%%  
956:   
957: \bibitem{missingsatellites}
958: %\cite{Kauffmann:1993gv}
959: %\bibitem{Kauffmann:1993gv}
960:   G.~Kauffmann, S.~D.~M.~White and B.~Guiderdoni,
961:   %``The Formation and Evolution of Galaxies Within Merging Dark Matter
962:   %Haloes,''
963:   Mon.\ Not.\ Roy.\ Astron.\ Soc.\  {\bf 264} (1993) 201.
964:   %%CITATION = MNRAA,264,201;%%
965: %\cite{Klypin:1999uc}
966: %\bibitem{Klypin:1999uc}
967:   A.~A.~Klypin, A.~V.~Kravtsov, O.~Valenzuela and F.~Prada,
968:   %``Where are the missing galactic satellites?,''
969:   Astrophys.\ J.\  {\bf 522}, 82 (1999)
970:   [arXiv:astro-ph/9901240].
971:   %%CITATION = ASJOA,522,82;%%
972: %\cite{Moore:1999nt}
973: %\bibitem{Moore:1999nt}
974:   B.~Moore, S.~Ghigna, F.~Governato, G.~Lake, T.~Quinn, J.~Stadel and P.~Tozzi,
975:   %``Dark Matter Substructure Within Galactic Halos,''
976:   Astrophys.\ J.\  {\bf 524}, L19 (1999).
977:   %%CITATION = ASJOA,524,L19;%%      
978:   
979: \bibitem{dsp}  
980: %\cite{Bullock:2000wn}
981: %\bibitem{Bullock:2000wn}
982:   J.~S.~Bullock, A.~V.~Kravtsov and D.~H.~Weinberg,
983:   %``Reionization and the abundance of galactic satellites,''
984:   Astrophys.\ J.\  {\bf 539}, 517 (2000)
985:   [arXiv:astro-ph/0002214];
986:   %%CITATION = ASJOA,539,517;%%
987: W.   A.   Chiu,   N.   Y.   Gnedin,  and   J.   P.   Ostriker,
988: Astrophys.   J.   {\bf   563},    21   (2001) [astro-ph/0103359];   
989: %\cite{Benson:2001au}
990: %\bibitem{Benson:2001au}
991:   A.~J.~Benson, C.~G.~Lacey, C.~M.~Baugh, S.~Cole and C.~S.~Frenk,
992:   %``The Effects of Photoionization on Galaxy Formation - I: Model and Results
993:   %at z=0,''
994:   Mon.\ Not.\ Roy.\ Astron.\ Soc.\  {\bf 333}, 156 (2002)
995:   [arXiv:astro-ph/0108217];
996:   %%CITATION = MNRAA,333,156;%%
997: R.  Barkana and A. Loeb, Astrophys.  J. {\bf 523},
998: 54 (1999) [arXiv:astro-ph/9901114]; 
999: A.  Dekel and J. Silk, Astrophys.  J. {\bf 303},
1000: 38  (1986);  
1001: %\cite{Cole:1994ab}
1002: %\bibitem{Cole:1994ab}
1003:   S.~Cole, A.~Aragon-Salamanca, C.~S.~Frenk, J.~F.~Navarro and S.~E.~Zepf,
1004:   %``A Recipe For Galaxy Formation,''
1005:   Mon.\ Not.\ Roy.\ Astron.\ Soc.\  {\bf 271}, 781 (1994)
1006:   [arXiv:astro-ph/9402001];
1007:   %%CITATION = MNRAA,271,781;%%
1008: %\cite{Somerville:1998bb}
1009: %\bibitem{Somerville:1998bb}
1010:   R.~S.~Somerville and J.~R.~Primack,
1011:   %``Semi-Analytic Modelling of Galaxy Formation: The Local Universe,''
1012:   Mon.\ Not.\ Roy.\ Astron.\ Soc.\  {\bf 310}, 1087 (1999)
1013:   [arXiv:astro-ph/9802268].
1014:   %%CITATION = MNRAA,310,1087;%%
1015: 
1016: %\cite{Walker:2007ju}
1017: \bibitem{Walker:2007ju}
1018:   M.~G.~Walker, M.~Mateo, E.~W.~Olszewski, O.~Y.~Gnedin, X.~Wang, B.~Sen and M.~Woodroofe,
1019:   %``Velocity Dispersion Profiles of Seven Dwarf Spheroidal Galaxies,''
1020:   arXiv:0708.0010 [astro-ph].
1021:   %%CITATION = ARXIV:0708.0010;%%  
1022:   
1023: %\cite{Strigari:2007ma}
1024: \bibitem{Strigari:2007ma}
1025:   L.~E.~Strigari, J.~S.~Bullock, M.~Kaplinghat, J.~Diemand, M.~Kuhlen and P.~Madau,
1026:   %``Redefining the Missing Satellites Problem,''
1027:   arXiv:0704.1817 [astro-ph].
1028:   %%CITATION = ARXIV:0704.1817;%%  
1029: 
1030: %\cite{Martin:2007ic}
1031: \bibitem{Martin:2007ic}
1032:   N.~F.~Martin, R.~A.~Ibata, S.~C.~Chapman, M.~Irwin and G.~F.~Lewis,
1033:   %``A Keck/DEIMOS spectroscopic survey of faint Galactic satellites:
1034:   %searching for the least massive dwarf galaxies,''
1035:   arXiv:0705.4622 [astro-ph].
1036:   %%CITATION = ARXIV:0705.4622;%%    
1037:   
1038: %\cite{Simon:2007dq}
1039: \bibitem{Simon:2007dq}
1040:   J.~D.~Simon and M.~Geha, Astrophys.\ J.\ in press, 
1041:   %``The Kinematics of the Ultra-Faint MW Satellites: Solving the
1042:   %Missing Satellite Problem,''
1043:   arXiv:0706.0516 [astro-ph].
1044:   %%CITATION = ARXIV:0706.0516;%%  
1045:   
1046: \bibitem{GLAST} S.~ Ritz., J.~Grindlay, 
1047: C.~Meegan, P.~F.~Michelson,  \& GLAST Mission Team 2005, Bulletin of the 
1048: American Astronomical Society, 37, 1198. 
1049: 
1050: %\cite{Hanna:2002bf}
1051: \bibitem{Hanna:2002bf}
1052:   D.~S.~Hanna {\it et al.},
1053:   %``The STACEE-32 ground based gamma-ray detector,''
1054:   Nucl.\ Instrum.\ Meth.\  A {\bf 491}, 126 (2002)
1055:   [arXiv:astro-ph/0205510].
1056:   %%CITATION = NUIMA,A491,126;%%
1057:    
1058: %\cite{Hofmann:2003kx}
1059: \bibitem{Hofmann:2003kx}
1060:   W.~Hofmann  [HESS Collaboration],
1061:   %``Status of the HESS project,''
1062: %\href{http://www.slac.stanford.edu/spires/find/hep/www?irn=5653185}{SPIRES entry}
1063: {\it Prepared for 28th International Cosmic Ray Conferences (ICRC 2003), Tsukuba, Japan, 31 Jul - 7 Aug 2003}.
1064: 
1065: %\cite{Cortina:2005pt}
1066: \bibitem{Cortina:2005pt}
1067:   J.~Cortina {\it et al.}  [MAGIC Collaboration],
1068:   %``Technical performance of the MAGIC telescope,''
1069: %\href{http://www.slac.stanford.edu/spires/find/hep/www?irn=6402666}{SPIRES entry}
1070: {\it Prepared for 29th International Cosmic Ray Conference (ICRC 2005), Pune, India, 3-11 Aug 2005}
1071: 
1072: %\cite{Weekes:2001pd}
1073: \bibitem{Weekes:2001pd}
1074:   T.~C.~Weekes {\it et al.},
1075:   %``VERITAS: The very energetic radiation imaging telescope array system,''
1076:   Astropart.\ Phys.\  {\bf 17}, 221 (2002)
1077:   [arXiv:astro-ph/0108478].
1078:   %%CITATION = APHYE,17,221;%%
1079:   
1080: %\cite{Yoshikoshi:1999rg}
1081: \bibitem{Yoshikoshi:1999rg}
1082:   T.~Yoshikoshi {\it et al.},
1083:   %``Present status of the 7-m to 10-m telescope of CANGAROO II,''
1084:   Astropart.\ Phys.\  {\bf 11}, 267 (1999).
1085:   %%CITATION = APHYE,11,267;%%  
1086: 
1087: %\cite{Sinnis:2005un}
1088: \bibitem{Sinnis:2005un}
1089:   G.~Sinnis  [HAWC Collaboration],
1090:   %``HAWC: A next generation VHE all-sky telescope,''
1091:   AIP Conf.\ Proc.\  {\bf 745} (2005) 234.
1092:   %%CITATION = APCPC,745,234;%%    
1093:   
1094: \bibitem{previousdwarfs}  
1095: %\cite{Baltz:1999ra}
1096: %\bibitem{Baltz:1999ra}
1097:   E.~A.~Baltz, C.~Briot, P.~Salati, R.~Taillet and J.~Silk,
1098:   %``Detection of neutralino annihilation photons from external galaxies,''
1099:   Phys.\ Rev.\  D {\bf 61}, 023514 (2000)
1100:   [arXiv:astro-ph/9909112]; 
1101:   %%CITATION = PHRVA,D61,023514;%%  
1102: %\cite{Evans:2003sc}
1103: %\bibitem{Evans:2003sc}
1104:   N.~W.~Evans, F.~Ferrer and S.~Sarkar,
1105:   %``A 'Baedecker' for the dark matter annihilation signal,''
1106:   Phys.\ Rev.\  D {\bf 69}, 123501 (2004)
1107:   [arXiv:astro-ph/0311145];
1108:   %%CITATION = PHRVA,D69,123501;%%  
1109: %\cite{Profumo:2006hs}
1110: %\bibitem{Profumo:2006hs}
1111:   S.~Profumo and M.~Kamionkowski,
1112:   %``Dark matter and the CACTUS gamma-ray excess from Draco,''
1113:   JCAP {\bf 0603}, 003 (2006)
1114:   [arXiv:astro-ph/0601249];
1115:   %%CITATION = JCAPA,0603,003;%% 
1116: %\cite{SanchezConde:2007te}
1117: %\bibitem{SanchezConde:2007te}
1118:   M.~A.~Sanchez-Conde, F.~Prada, E.~L.~Lokas, M.~E.~Gomez, R.~Wojtak and M.~Moles,
1119:   %``Dark matter annihilation in Draco: New considerations of the expected gamma
1120:   %flux,''
1121:   arXiv:astro-ph/0701426.
1122:   %%CITATION = ASTRO-PH/0701426;%%  
1123: 
1124: %\cite{Bergstrom:2005qk}
1125: \bibitem{Bergstrom:2005qk}
1126:   L.~Bergstrom and D.~Hooper,
1127:   %``Dark matter and gamma-rays from Draco: MAGIC, GLAST and CACTUS,''
1128:   Phys.\ Rev.\  D {\bf 73}, 063510 (2006)
1129:   [arXiv:hep-ph/0512317]. 
1130:   %%CITATION = PHRVA,D73,063510;%%
1131:   
1132: %\cite{Strigari:2006rd}
1133: \bibitem{Strigari:2006rd}
1134:   L.~E.~Strigari, S.~M.~Koushiappas, J.~S.~Bullock and M.~Kaplinghat,
1135:   %``Precise constraints on the dark matter content of MW dwarf
1136:   %galaxies for gamma-ray experiments,''
1137:   Phys.\ Rev.\  D {\bf 75}, 083526 (2007)
1138:   [arXiv:astro-ph/0611925].
1139:   %%CITATION = PHRVA,D75,083526;%%    
1140: 
1141: \bibitem{galacticcenter}
1142: %\cite{Hooper:2002ru}
1143: %\bibitem{Hooper:2002ru}
1144:   D.~Hooper and B.~L.~Dingus,
1145:   %``Limits on supersymmetric dark matter from EGRET observations of the
1146:   %galactic center region,''
1147:   Phys.\ Rev.\  D {\bf 70}, 113007 (2004)
1148:   [arXiv:astro-ph/0210617].
1149:   %%CITATION = PHRVA,D70,113007;%%
1150: 
1151: \bibitem{microhalos} 
1152: %\cite{CalcaneoRoldan:2000yt}
1153: %\bibitem{CalcaneoRoldan:2000yt}
1154:   C.~Calcaneo-Roldan and B.~Moore,
1155:   %``The surface brightness of dark matter: Unique signatures of neutralino
1156:   %annihilation in the galactic halo,''
1157:   Phys.\ Rev.\  D {\bf 62}, 123005 (2000)
1158:   [arXiv:astro-ph/0010056];
1159:   %%CITATION = PHRVA,D62,123005;%%
1160: %\cite{Tasitsiomi:2002vh}
1161: %\bibitem{Tasitsiomi:2002vh}
1162:   A.~Tasitsiomi and A.~V.~Olinto,
1163:   %``The detectability of neutralino clumps via atmospheric Cherenkov
1164:   %telescopes,''
1165:   Phys.\ Rev.\  D {\bf 66}, 083006 (2002)
1166:   [arXiv:astro-ph/0206040];
1167:   %%CITATION = PHRVA,D66,083006;%%
1168: %\cite{Stoehr:2003hf}
1169: %\bibitem{Stoehr:2003hf}
1170:   F.~Stoehr, S.~D.~M.~White, V.~Springel, G.~Tormen and N.~Yoshida,
1171:   %``Dark matter annihilation in the Milky Way's halo,''
1172:   Mon.\ Not.\ Roy.\ Astron.\ Soc.\  {\bf 345}, 1313 (2003)
1173:   [arXiv:astro-ph/0307026];
1174:   %%CITATION = MNRAA,345,1313;%%  
1175: %\cite{Koushiappas:2003bn}
1176: %\bibitem{Koushiappas:2003bn}
1177:   S.~M.~Koushiappas, A.~R.~Zentner and T.~P.~Walker,
1178:   %``The observability of gamma-rays from neutralino annihilations in Milky  Way
1179:   %substructure,''
1180:   Phys.\ Rev.\  D {\bf 69}, 043501 (2004)
1181:   [arXiv:astro-ph/0309464];
1182:   %%CITATION = PHRVA,D69,043501;%%
1183: %\cite{Baltz:2006sv}
1184: %\bibitem{Baltz:2006sv}
1185:   E.~A.~Baltz, J.~E.~Taylor and L.~L.~Wai,
1186:   %``Can Astrophysical Gamma Ray Sources Mimic Dark Matter Annihilation in
1187:   %Galactic Satellites?,''
1188:   arXiv:astro-ph/0610731;
1189:   %%CITATION = ASTRO-PH/0610731;%%
1190:   %%CITATION = ASJOA,657,262;%%
1191: %\cite{Pieri:2005pg}
1192: %\bibitem{Pieri:2005pg}
1193:   L.~Pieri, E.~Branchini and S.~Hofmann,
1194:   %``Difficulty of detecting minihalos via gamm rays from dark matter
1195:   %annihilation,''
1196:   Phys.\ Rev.\ Lett.\  {\bf 95}, 211301 (2005)
1197:   [arXiv:astro-ph/0505356];
1198:   %%CITATION = PRLTA,95,211301;%%
1199: %\cite{Koushiappas:2006qq}
1200: \bibitem{Koushiappas:2006qq}
1201:   S.~M.~Koushiappas,
1202:   %``Proper motion of gamma-rays from microhalo sources,''
1203:   Phys.\ Rev.\ Lett.\  {\bf 97}, 191301 (2006)
1204:   [arXiv:astro-ph/0606208].
1205:   %%CITATION = PRLTA,97,191301;%% 
1206: 
1207: %\cite{Diemand:2006ik}
1208: \bibitem{Diemand:2006ik}
1209:   J.~Diemand, M.~Kuhlen and P.~Madau,
1210:   %``Dark matter substructure and gamma-ray annihilation in the Milky Way
1211:   %halo,''
1212:   Astrophys.\ J.\  {\bf 657}, 262 (2007)
1213:   [arXiv:astro-ph/0611370].
1214:   
1215: \bibitem{tides} 
1216: S.~Piatek and C.~Pryor, \aj {\bf 109}, 1071 (1995);   
1217: %\cite{Klimentowski:2006qe}
1218: %\bibitem{Klimentowski:2006qe}
1219:   J.~Klimentowski, E.~L.~Lokas, S.~Kazantzidis, F.~Prada, L.~Mayer and G.~A.~Mamon,
1220:   %``Mass modelling of dwarf spheroidal galaxies: the effect of unbound stars
1221:   %from tidal tails and the MW,''
1222:   Mon.\ Not.\ Roy.\ Astron.\ Soc.\  {\bf 378}, 353 (2007)
1223:   [arXiv:astro-ph/0611296].
1224:   %%CITATION = MNRAA,378,353;%%  
1225:   
1226: %\cite{Penarrubia:2007zx}
1227: \bibitem{Penarrubia:2007zx}
1228:   J.~Penarrubia, J.~F.~Navarro and A.~W.~McConnachie,
1229:   %``The Tidal Evolution of Local Group Dwarf Spheroidals,''
1230:   arXiv:0708.3087 [astro-ph].
1231:   %%CITATION = ARXIV:0708.3087;%%  
1232: 
1233: %\cite{Lokas:2004sw}
1234: \bibitem{Lokas:2004sw}
1235:   E.~L.~Lokas, G.~A.~Mamon and F.~Prada,
1236:   %``Dark matter distribution in the Draco dwarf from velocity moments,''
1237:   Mon.\ Not.\ Roy.\ Astron.\ Soc.\  {\bf 363}, 918 (2005)
1238:   [arXiv:astro-ph/0411694].
1239:   %%CITATION = MNRAA,363,918;%%  
1240: 
1241: %\cite{Mashchenko:2005bj}
1242: \bibitem{Mashchenko:2005bj}
1243:   S.~Mashchenko, A.~Sills and H.~M.~P.~Couchman,
1244:   %``Constraining global properties of the Draco dwarf spheroidal galaxy,''
1245:   Astrophys.\ J.\  {\bf 640}, 252 (2006)
1246:   [arXiv:astro-ph/0511567].
1247:   %%CITATION = ASJOA,640,252;%%  
1248: 
1249: %\cite{Gilmore:2007fy}
1250: \bibitem{Gilmore:2007fy}
1251:   G.~Gilmore, M.~I.~Wilkinson, R.~F.~G.~Wyse, J.~T.~Kleyna, A.~Koch, N.~W.~Evans and E.~K.~Grebel,
1252:   %``The Observed properties of Dark Matter on small spatial scales,''
1253:   arXiv:astro-ph/0703308.
1254:   %%CITATION = ASTRO-PH/0703308;%%  
1255:   
1256: %\cite{King:1962wi}
1257: \bibitem{King:1962wi}
1258:   I.~King,
1259:   %``The structure of star clusters. I. An Empirical density law,''
1260:   Astron.\ J.\  {\bf 67}, 471 (1962).
1261:   %%CITATION = ANJOA,67,471;%%  
1262:   
1263: \bibitem{slopes}
1264: %\cite{Navarro:2003ew}
1265: %\bibitem{Navarro:2003ew}
1266:   J.~F.~Navarro {\it et al.},
1267:   %``The Inner Structure of LambdaCDM Halos III: Universality and Asymptotic
1268:   %Slopes,''
1269:   Mon.\ Not.\ Roy.\ Astron.\ Soc.\  {\bf 349}, 1039 (2004)
1270:   [arXiv:astro-ph/0311231];
1271:   %%CITATION = MNRAA,349,1039;%%  
1272: %\cite{Diemand:2005wv}
1273: %\bibitem{Diemand:2005wv}
1274:   J.~Diemand, M.~Zemp, B.~Moore, J.~Stadel and M.~Carollo,
1275:   %``Cusps in CDM halos: The density profile of a billion particle halo,''
1276:   Mon.\ Not.\ Roy.\ Astron.\ Soc.\  {\bf 364}, 665 (2005)
1277:   [arXiv:astro-ph/0504215].
1278:   %%CITATION = MNRAA,364,665;%%  
1279:       
1280: %\cite{Penarrubia:2007zz}
1281: \bibitem{Penarrubia:2007zz}
1282:   J.~Penarrubia, A.~McConnachie and J.~F.~Navarro,
1283:   %``The cold dark matter halos of local group dwarf spheroidals,''
1284:   arXiv:astro-ph/0701780.
1285:   %%CITATION = ASTRO-PH/0701780;%%  
1286:   
1287: %\cite{Walker:2005:nt} 
1288: \bibitem{Walker:2005:nt} 
1289: M.~G.~Walker, M.~Mateo, E.~W.~Olszewski, , R.~Bernstein, X.~Wang, 
1290: \& M.~Woodroofe, \aj, {\bf 131} 2114 (2006)     
1291: [arXiv:astro-ph/0511465]. 
1292: 
1293: \bibitem{dwarfdata}
1294: K.~B.~Westfall  {\it et al}, \aj, {\bf 131} 375 (2006)
1295: [arXiv:astro-ph/0508091];
1296: %\cite{Koch:2006in}
1297: %\bibitem{Koch:2006in}
1298:   A.~Koch {\it et al.},
1299:   %``Stellar kinematics and metallicities in the Leo I dwarf spheroidal galaxy
1300:   %-- wide field implications for galactic evolution,''
1301:   Astrophys.\ J.\  {\bf 657}, 241 (2007)
1302:   [arXiv:astro-ph/0611372].
1303:   %%CITATION = ASJOA,657,241;%%
1304:     
1305: \bibitem{BJ07} 
1306: J.~S.~Bullock and K.~V.~Johnston, Island Universes, 
1307: Astrophysics and Space Science  Proceedings~
1308: Springer, ~227 (2007).   
1309:     
1310: %\cite{Zentner:2003yd}
1311: \bibitem{Zentner:2003yd}
1312:   A.~R.~Zentner and J.~S.~Bullock,
1313:   %``Halo Substructure And The Power Spectrum,''
1314:   Astrophys.\ J.\  {\bf 598}, 49 (2003)
1315:   [arXiv:astro-ph/0304292].
1316:   %%CITATION = ASJOA,598,49;%%  
1317:   
1318: %\cite{van den Bosch:2004zs}
1319: \bibitem{van den Bosch:2004zs}
1320:   F.~C.~van den Bosch, G.~Tormen and C.~Giocoli,
1321:   %``The Mass Function and Average Mass Loss Rate of Dark Matter Subhaloes,''
1322:   Mon.\ Not.\ Roy.\ Astron.\ Soc.\  {\bf 359}, 1029 (2005)
1323:   [arXiv:astro-ph/0409201].
1324:   %%CITATION = MNRAA,359,1029;%%  
1325:   
1326: \bibitem{Willmaninprep}
1327: B.~Willman, M.~Geha, J.~Strader, and C.~Rockosi in preparation.
1328:   
1329: %\cite{Munoz:2005be}
1330: \bibitem{Munoz:2005be}
1331:   R.~R.~Munoz {\it et al.},
1332:   %``Exploring Halo Substructure with Giant Stars VIII: The Velocity Dispersion
1333:   %Profiles of the Ursa Minor and Draco Dwarf Spheroidals At Large Angular
1334:   %Separations,''
1335:   Astrophys.\ J.\  {\bf 631}, L137 (2005)
1336:   [arXiv:astro-ph/0504035]. 
1337:   %%CITATION = ASJOA,631,L137;%%    
1338:   
1339:  %\cite{Hunger:1997we}
1340: \bibitem{Hunter:1997we}
1341:   S.~D.~Hunter {\it et al.},
1342:   %``EGRET observations of the diffuse gamma-ray emission from the galactic
1343:   %plane,''
1344:   Astrophys.\ J.\  {\bf 481}, 205 (1997).
1345:   %%CITATION = ASJOA,481,205;%% 
1346:     
1347: %\cite{Nishimura:1980pz}
1348: \bibitem{Nishimura:1980pz}
1349:   J.~Nishimura {\it et al.},
1350:   %``Emulsion Chamber Observations Of Primary Cosmic Ray Electrons In The Energy
1351:   %Range 30-Gev - 1000-Gev,''
1352:   Astrophys.\ J.\  {\bf 238}, 394 (1980).
1353:   %%CITATION = ASJOA,238,394;%%   
1354: 
1355: %\cite{Olszewski:1995gs}
1356: \bibitem{Olszewski:1995gs}
1357:   E.~Olszewski, C.~Pryor and T.~Armandroff,
1358:   %``The Mass-to-Light Ratios of the Draco and Ursa Minor Dwarf Spheroidal
1359:   %Galaxies. II. The Binary Population and Its Effect in the Measured Velocity
1360:   %Dispersions of Dwarf Spheroidal Galaxies,''
1361:   Astron. J. {\bf 111}, 750 (1996)
1362:   [arXiv:astro-ph/9510155].
1363:   %%CITATION = ASTRO-PH/9510155;%%  
1364: 
1365: %\cite{Sollima:2007sc}
1366: \bibitem{Sollima:2007sc}
1367:   A.~Sollima, G.~Beccari, F.~R.~Ferraro, F.~Fusi Pecci and A.~Sarajedini,
1368:   %``The fraction of binary systems in the core of thirteen low-density
1369:   %Galactic globular clusters,''
1370:   arXiv:0706.2288 [astro-ph].
1371:   %%CITATION = ARXIV:0706.2288;%%
1372:   
1373: %\cite{Duquennoy:1991zu}
1374: \bibitem{Duquennoy:1991zu}
1375:   A.~Duquennoy and M.~Mayor,
1376:   %``Multiplicity among solar - type stars in the solar neighbourhood. 2.
1377:   %Distribution of the orbital elements in an unbiased sample,''
1378:   Astron.\ Astrophys.\  {\bf 248}, 485 (1991).
1379:   %%CITATION = AAEJA,248,485;%%
1380: 
1381: %\cite{Fisher:2003tg}
1382: \bibitem{Fisher:2003tg}
1383:   R.~T.~Fisher,
1384:   %``A Turbulent Interstellar Medium Origin of the Binary Period Distribution,''
1385:   Astrophys.\ J.\  {\bf 600}, 769 (2004)
1386:   [arXiv:astro-ph/0303280].
1387:   %%CITATION = ASJOA,600,769;%%
1388:   
1389:   %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1390:        
1391: \end{thebibliography}
1392: 
1393: \end{document}
1394: