0706.0262/ms.tex
1: %% ApJ
2: %% \documentclass{aastex}
3: %% \documentclass[12pt,preprint]{aastex}
4: %% \documentclass[preprint2]{aastex}
5: \documentclass{emulateapj}
6: \bibliographystyle{apj}
7: %%\usepackage{color}
8: %%\slugcomment{Submitted to Astrophysical Journal}
9: \lefthead{Faltenbacher et al.} 
10: \righthead{Alignments between central and satellite halos}
11: \newcommand{\hMsol}{{\,h^{-1}\rm M}_\odot}
12: \newcommand{\hMpc}{{\,h^{-1}\rm Mpc}}
13: \newcommand{\hkpc}{{\,h^{-1}\rm kpc}}
14: \newcommand{\kpc}{{\,\rm kpc}}
15: \newcommand{\kms}{\>{\rm km}\,{\rm s}^{-1}}
16: \newcommand{\kmsmpc}{\>{\rm km}\,{\rm s}^{-1}\,{\rm Mpc}^{-1}}
17: \newcommand{\Rvir}{{\,r_{\rm vir}}}
18: \newcommand{\Om}{\Omega_{\rm m}}
19: \newcommand{\Ol}{\Omega_\Lambda}
20: \renewcommand{\vec}[1]{{\mathbf #1}}
21: \newcommand{\aII}{\langle|\vec{a}_{\rm GCS}\!\cdot\!\vec{r}|\rangle\,}
22: \newcommand{\axI}{\langle|\vec{a}_{\rm SAT}\!\cdot\!\vec{r}|\rangle\,}
23: \newcommand{\axB}{\langle|\vec{a}_3\!\cdot\!\vec{r}|\rangle\,}
24: \newcommand{\acI}{\langle|\vec{a}_{\rm GCS}\!\cdot\!\vec{a}_{\rm SAT}|\rangle\,}
25: \newcommand{\acB}{\langle|\vec{a}_{\rm GCS}\!\cdot\!\vec{a}_3|\rangle\,}
26: \newcommand{\vel}{\langle|\vec{v}  \!\cdot\!\vec{r}|\rangle\,}
27: \newcommand{\aIv}{\langle|\vec{a}_{\rm SAT}\!\cdot\!\vec{v}|\rangle\,}
28: \newcommand{\cIv}{\langle|\vec{a}_{\rm GCS}\!\cdot\!\vec{v}|\rangle\,}
29: %%\newcommand{\fig}{fig/}
30: \newcommand{\fig}{1E3-5E4_2sigma/}
31: %%
32: \begin{document}
33: %%
34: %%---------------------------------------------------------------------------
35: \title{Spatial and kinematic alignments between central and satellite halos} 
36: %%---------------------------------------------------------------------------
37: %%
38: \author {A. Faltenbacher\altaffilmark{1},  Y.P. Jing\altaffilmark{1}, 
39: Cheng Li\altaffilmark{1}, Shude Mao\altaffilmark{2},
40: H.J. Mo\altaffilmark{3}, Anna Pasquali\altaffilmark{4} and Frank C. van
41: den Bosch\altaffilmark{4}}
42: %%
43: \altaffiltext{1} {Shanghai Astronomical Observatory, Joint Center for
44: Cosmology and Astrophysics of the Max-Planck-Institut fuer Astrophysik
45: and the  Shanghai  Astronomical Observatory,  Nandan Road  80,  Shanghai
46: 200030, China}
47: \altaffiltext{2} {University of Manchester, Jodrell Bank Observatory,
48: Macclesfield, Cheshire SK11 9DL, UK}    
49: \altaffiltext{3}{Department of Astronomy, University of Massachusetts,
50: Amherst MA 01003-9305}
51: \altaffiltext{4} {Max-Planck-Institute for Astronomy, K\"onigstuhl 17,
52: D-69117 Heidelberg, Germany }
53: 
54: %%
55: \begin{abstract}
56: Based   on a cosmological  N-body  simulation  we analyze  spatial and
57: kinematic alignments of satellite halos  within  six times the  virial
58: radius of  group  size host halos ($\Rvir$).   We measure three
59: different types of  spatial   alignment: halo alignment    between the
60: orientation   of   the group    central substructure  (GCS)    and the
61: distribution   of  its   satellites, radial   alignment   between  the
62: orientation of   a satellite and  the  direction towards  its GCS, and
63: direct alignment  between the orientation of  the GCS and that  of its
64: satellites.  In analogy we use the  directions of satellite velocities
65: and  probe three further   types  of  alignment: the radial   velocity
66: alignment  between the satellite velocity  and connecting line between
67: satellite and GCS, the halo velocity alignment between the orientation
68: of the  GCS and satellite  velocities and  the auto velocity alignment
69: between the satellites orientations  and  their velocities.  We   find
70: that satellites are preferentially located along the major axis of the
71: GCS within at least  $6\Rvir$  (the range probed here).   Furthermore,
72: satellites preferentially point  towards the GCS.  The most pronounced
73: signal is detected on small scales but a detectable signal extends out
74: to $\sim  6\Rvir$.  The direct alignment signal  is weaker,  however a
75: systematic  trend is   visible  at  distances $\lesssim2\Rvir$.    All
76: velocity alignments are highly significant on  small scales.  The halo
77: velocity  alignment is constant   within $2\Rvir$ and declines rapidly
78: beyond.  The  halo  and the  auto velocity  alignments are maximal  at
79: small scales and disappear beyond  1 and $1.5\Rvir$ respectively. Our
80: results suggest that the halo alignment reflects the filamentary large
81: scale structure which  extends  far  beyond the  virial radii   of the
82: groups.  In contrast, the main  contribution  to the radial  alignment
83: arises from the adjustment of the satellite  orientations in the group
84: tidal  field.   The projected data  reveal good  agreement with recent
85: results derived from large galaxy surveys.
86: \end{abstract}
87: %%
88: %%
89: \keywords{dark matter --- galaxies: clusters: general --- galaxies:
90: kinematics and dynamics --- methods: numerical}
91: %%
92: %%--------------------
93: \section{Introduction}
94: %%--------------------
95: %%
96: Over  the  last  decades  observational  and  numerical  evidence  has
97: substantiated the  picture of  a filamentary large-scale  structure in
98: the universe.  In principle the large-scale tidal field is expected to
99: induce large-scale correlations between  the orientations of halos and
100: galaxies     that    are     embedded    within     these    filaments
101: \citep[e.g.,][]{2000ApJ...543L.107P,2000ApJ...545..561C,
102: 2000MNRAS.319..649H,         2001MNRAS.320L...7C,2001ApJ...559..552C,
103: 2002MNRAS.332..339P, 2002MNRAS.335L..89J}.   On the other  hand, the
104: subsequent accretion onto larger  systems, such as groups and clusters
105: of galaxies,  may alter the orientations of  these (sub-)structures in
106: response   to  the   local  tidal   field  \citep{1994MNRAS.270..390C,
107: 2005ApJ...629L...5L}.   Cosmological  N-body  simulations provide  a
108: valuable  tool  to  differentiate  the various  contributions  to  the
109: halo/galaxy alignments within overdense regions.
110: 
111: Observationally, various types of alignment between galaxies and their
112: environment have  been detected  on  a   wide  range in  scales,  from
113: super-cluster systems   down  to  the distribution of   the  satellite
114: galaxies in our  Milky Way (MW).   On cluster scales various different
115: types of alignment are discussed in the literature : alignment between
116: neighboring clusters \citep{1982A&A...107..338B,  1989ApJ...338..711U,
117: 1989ApJ...344..535W,     1994ApJS...95..401P,    2002ApJ...565..849C},
118: between brightest cluster galaxies (BCGs) and their parent clusters
119: \citep{1980MNRAS.191..325C,  1982A&A...107..338B, 1990AJ.....99..743S,
120: 1991A&A...243...38R,2002MNRAS.329L..47P}, between the orientation of
121: satellite    galaxies   and   the    orientation   of    the   cluster
122: \citep{1985ApJ...298..461D,  2003ApJ...594..144P},   and  between  the
123: orientation  of satellite  galaxies  and the  orientation  of the  BCG
124: \citep{1990AJ.....99..743S}.  According  to these studies  the typical
125: scales  over which  clusters reveal  signs for  alignment range  up to
126: $10-50\hMpc$, which can be most naturally explained by the presence of
127: filaments.
128: 
129: With  large galaxy  redshift  surveys, such  as  the two-degree  Field
130: Galaxy Redshift  Survey \citep[2dFGRS,][]{2001MNRAS.328.1039C} and the
131: Sloan Digital Sky  Survey \citep[SDSS,][]{2000AJ....120.1579Y}, it has
132: recently  also  become possible  to  investigate  alignments on  group
133: scales  using large  and homogeneous  samples.  This  has  resulted in
134: robust  detections of various  alignments: \cite{2005ApJ...628L.101B},
135: \cite{2006MNRAS.369.1293Y}  and  \cite{2007MNRAS.376L..43A} all  found
136: that satellite galaxies are preferentially distributed along the major
137: axes  of  their host  galaxies,  while \cite{2005ApJ...627L..21P}  and
138: \cite{2006ApJ...644L..25A} noticed that  satellite galaxies tend to be
139: oriented towards the galaxy at the center of the halo.
140: 
141: In   contradiction to   the  studies above, \cite{1969ArA.....5..305H}
142: found    that    satellites   around   isolated   late  type  galaxies
143: preferentially  lie  along the  minor  axis  of the  disc.  Subsequent
144: studies, however, were   unable  to confirm this so-called   `Holmberg
145: effect'\citep{1975AJ.....80..477H,                1979MNRAS.187..287S,
146: 1982MNRAS.198..605M,          1997ApJ...478L..53Z}.           Recently
147: \cite{2007arXiv0704.3441A} reported    a  Holmberg effect    at  large
148: projected distances around blue host galaxies, while on smaller scales
149: the satellites were found to  be aligned with the  major axis of their
150: host  galaxy  and   \cite{2005ApJ...627..647B} claim that   a  careful
151: selection of isolated late-type galaxies  reveals the the tendency for
152: the  satellites to    align  with the   minor  axis  of  the  galactic
153: disc.  Investigating  the companions of M31 \cite{2006AJ....131.1405K}
154: find little evidence  for a Holmberg effect.  Yet,  the Milky Way (MW)
155: seems to exhibit a  Holmberg effect even on small  scales, in that the
156: 11  innermost MW satellites  show   a pronounced planar   distribution
157: oriented       close   to    perpendicular     to     the    MW   disc
158: \citep{1982Obs...102..202L,  1994ApJ...431L..17M, 2005A&A...431..517K,
159: 2005A&A...437..383K, 2005MNRAS.363..146L}.
160: 
161: Numerical  simulations  have been employed  to    test alignment on  a
162: similar range in scales, from  super-clusters down to galaxy-satellite
163: systems.   All  studies  focusing on   cluster size  halos   report  a
164: correlation in  the orientations for distances  of at least $10\hMpc$;
165: some studies observe a positive alignment signal up to $100\hMpc$
166: \citep[e.g.,][]{2000MNRAS.319..614O,2002A&A...395....1F,
167: 2005MNRAS.362.1099F,2005ApJ...618....1H,2005ApJ...629..781K,
168: 2006MNRAS.365..539B}.  These findings are interpreted as the signature
169: of the filamentary   network which  interconnects  the  clusters.  The
170: preferential accretion along  these  filaments causes the  clusters to
171: point  towards each other.  Also,  for galaxy and  group-sized halos a
172: tendency to point toward neighboring halos is detected.  According to
173: \cite{2006MNRAS.370.1422A} the  alignments for such  intermediate mass
174: objects are  caused by  tidal fields  rather than  accretion along the
175: filaments.  Consequently, the mechanisms responsible for the alignment
176: of the orientations depend  on halo mass.  Further evidence for a mass
177: dependence   of alignment effects comes   from  the examination of the
178: halos' angular momenta.
179: \cite{2005ApJ...627..647B} and \cite{2007ApJ...655L...5A} find that the
180: spins of  galaxy  size  halos  tend to be   parallel  to the filaments
181: whereas the spins of group-sized halos tend  to be perpendicular. This
182: behavior may originate in the relative sizes  of halos with respect to
183: the surrounding filaments.  
184: 
185: On subhalo scales basically  three different types of  alignments have
186: been discussed: the alignment of the overall subhalo distribution with
187: the orientation of  the host halo \citep[e.g.,][]{2004ApJ...603....7K,
188: 2005ApJ...629..219Z,      2006ApJ...650..550A,    2007MNRAS.378.1531K,
189: 2007MNRAS.374...16L},  the alignment  of the  orientations of subhalos
190: among  each    other  \citep[e.g.,][]{2005ApJ...629L...5L}  and,  very
191: recently, the orientation of the satellites with respect to the center
192: of the host  \citep{2007arXiv0705.2037K,2007arXiv0707.1702P}.   Again,
193: accretion along the filaments and the impact of tidal fields have been
194: invoked as explanations  for the former  and the latter, respectively.
195: Thus, on  all scales tidal   fields and accretion along filaments  are
196: considered to be  the  main  contributers  to the observed   alignment
197: signals.  Here we attempt to isolate  the different contributions.  In
198: particular we focus on the continuous  transition from subhalo to halo
199: scales meaning we examine the  alignment of (sub)structure on distance
200: scales between 0.3  and 6  times the virial   of  groups sized  halos.
201: 
202: Faltenbacher et al. (2007, hereafter Paper~I)\nocite{2007ApJ...662L..71F} 
203: applied the halo-based  group finder  of \cite{2005MNRAS.356.1293Y} to
204: the  SDSS  Data Release  Four \citep[DR4;][]{2006ApJS..162...38A}  and
205: carried out a study of  the mutual alignments between central galaxies
206: (BCG) and their satellites in group-sized halos. Using the same
207: data set consisting of over  $60000$ galaxies three different types of
208: alignment  have been investigated : (1)   the `halo' alignment between
209: the orientations  of the  BCG  and associated  satellite distribution;
210: (2) the  `radial' alignment between  the direction  given  by the
211: BCG-satellite connection   line and  the satellite  orientation;  (3)
212: the `direct' alignment between the orientations of the BCG and the
213: satellites.  The study presented in  this paper focuses on  the same
214: types of alignment and  is aimed  to compare  the observational
215: results with theoretical expectations derived from N-body simulations.
216: 
217: There are a variety of dynamical processes which can contribute to the
218: alignments  of satellites associated  with groups,  the most important
219: are:  (1) a possible   pre-adjustment of satellites  in the filaments,
220: which for distances of  a few times  the virial radius commonly  point
221: radially towards the group; (2) the preferential accretion along those
222: filaments; (3)  the  change of the  satellite  orbits in the  triaxial
223: group potential well;  (4) the  continuous re-adjustment of  satellite
224: orientations as they orbit within the  group. Basically, the first two
225: points can be attributed to the large  scale environment of the groups
226: whereas the latter two are more closely associated  with the impact of
227: the  group potential  on  small scales.   The purpose   of the present
228: analysis is to separate the   different contributions to the  observed
229: alignment signals, therefore   we analyse the mutual orientations   of
230: satellites  within 6 times the virial  radius of the groups. 
231: Since the tidal forces are  closely related to   the dynamics of  the
232: satellites additional insight into  the generation of alignment can be
233: gained by considering  the  satellite velocities.  Therefore, we  also
234: investigate the direction of the satellite  velocities with respect to
235: their   orientations, which constitutes an   indirect way to infer the
236: impact of the dynamics onto the orientation of the satellites.  A more
237: direct  way to  work out the  interplay between  the dynamics  and the
238: orientations would  be to trace  the orbits  of individual satellites,
239: however such an approach  goes beyond the  scope  of the present
240: study.
241: 
242: The  paper is organized as  follows.  In \S~\ref{sec:sim} we introduce
243: the    simulation       and       describe     the     halo    finding
244: procedure. \S~\ref{sec:size} deals with some technical aspects, namely
245: the   determination  of the  size     and  	orientation  of    the
246: substructures.  In \S~\ref{sec:3D} we present the signals of the three
247: dimensional spatial and velocity alignments  and in \S~\ref{sec:2D} we
248: repeat the analysis based on projected data. Finally, we conclude with
249: a summary in
250: \S~\ref{sec:diss}. 
251: 
252: %%
253: %%-------------------------------------------
254: \section{Simulation and halo identification}
255: \label{sec:sim}
256: %%-------------------------------------------
257: %%
258: For the present analysis  we employ an  N-body simulation of structure
259: formation in   a flat $\Lambda$CDM   universe with   a matter  density
260: $\Omega_m=0.3$, a  Hubble parameter  $h=H_0/(100 \kmsmpc)=0.7$,  and a
261: Harrison-Zeldovich    initial power   spectrum  with     normalization
262: $\sigma_8=0.9$.  The  density  field is sampled by   $512^3$ particles
263: within a $100\hMpc$ cube resulting in  a mass resolution of $6.2\times
264: 10^8\hMsol$.   The softening length was  set  to $\epsilon = 10\hkpc$,
265: beyond which the gravitational force between  two particles is exactly
266: Newtonian. The density  filed is evolved with 5000  time steps from an
267: initial   redshift  of $z_i=72$  using  a  PPPM  method.  An extensive
268: description of the simulation can be found in
269: \cite{2002ApJ...574..538J} where it is quoted as LCDMa realization.  
270: 
271: As detailed in the following two paragraphs the host halos and its
272: satellites are found in two subsequent steps with two different
273: techniques, first the main halos are located thereafter the associated
274: satellite halos are detected. In  order to   identify  the host halos
275: we  first  run   a FoF   algorithm \citep{1985ApJ...292..371D} on  
276: the simulation output at $z=0$.  We set the  FoF linking length to 0.1
277: times  the mean particle  separation, which  selects  regions with  an
278: average overdensity  of $\sim3000$. Note that, this linking length is a
279: factor of two smaller than the commonly used value of 0.2,
280: consequently only the central part of the host halo (and occasionally
281: large substructures) are selected.   
282: Subsequently, the virial radius,
283: $\Rvir$, is  defined as the radius of  the sphere centered on the most
284: bound FoF  particle  which includes a  mean  density of  101 times the
285: critical density, and we simply define the virial mass of each halo as
286: the mass within $\Rvir$. If the virial  regions of two halos overlap,
287: the  lower mass halo is discarded.   In what follows  we only focus on
288: the 515 halos with  a virial mass in the  range from $10^{13} \hMsol$
289: to $5\times10^{14}\hMsol$  (corresponding  to  halos  with  more than
290: 16,000 particles).  Since  this is  the  typical mass scale of  galaxy
291: groups, we will refer to these halos as `groups'.
292: 
293: In a second step we search for self-bound (sub)structures using the SKID
294: halo finder \citep{2001PhDT........21S} applied to the particle 
295: distribution within group centric distances of $6\Rvir$.   As  discussed  in 
296: \cite{2006MNRAS.366.1529M}  SKID  adequately  identifies the  smallest
297: resolvable  substructures  when using a  linking length   $l$ equal to
298: twice the  softening  length, i.e.   four  times the spline  softening
299: length.   We    therefore adopt  $l=20\hkpc$.    Throughout  we   will
300: distinguish between  ``group central substructures'' (GCSs), which are
301: located at the center of our groups, and satellites  which are all the
302: other (sub)structures, no   matter whether they  lie within  or beyond
303: $\Rvir$.  According to this definition every group hosts one, and only
304: one, GCS at  its center while it  may have numerous satellites outside
305: the  volume occupied by  the   GCS.  Satellites  are allocated to  all
306: groups from  which they  are separated less   than $6\Rvir$. Hence,  a
307: satellite may be assigned to more than one GCS.
308: %%
309: %%--------------------------------------------
310: \section{Size and orientation of substructures}
311: \label{sec:size}
312: %%--------------------------------------------
313: %%
314: Before describing the computation  of the orientation we determine the
315: typical   sizes of the GCSs  and   the satellites. Knowledge about the
316: physical sizes of the (sub)structures provides a crucial link for the
317: comparison to observational data.
318: %%
319: \subsection{Sizes of  group central substructures} 
320: %%
321: %%
322: %%\clearpage
323: \begin{figure}
324: \plotone{f1.eps}
325: %%\plotone{\fig covi.Rsph6.mim200.eps}
326: \caption{\label{fig:covi}
327: Distribution of the spatial dispersion $r_\sigma$ of the group central
328: substructure (GCS) in units of  the virial radius. Satellites can only
329: be resolved at radii  larger  than the size of the GCS.}
330: \end{figure}
331: %%\clearpage
332: %%
333: The physical   interpretation    of the  size  of   the   GCS  is  not
334: straightforward. For one thing, it  depends on the SKID linking length
335: used.  However, for our  purposes it is sufficient  to notice that the
336: GCS represents the dense inner region of  the group which, largely due
337: to numerical reasons,    is free of substructure.   Consequently,  any
338: radial dependence  of satellite properties can only  be probed down to
339: the size of the GCS. In order to express the  sizes of the GCS and the
340: satellites we   use the  rms   of  the  distances between   the  bound
341: particles, $r_{\sigma}$.  The  advantage of this  size measure is that
342: it provides a direct  estimate of the  (momentary) size without having
343: to make any assumption regarding the  actual density distribution.  In
344: the case of an isolated NFW  halo $r_\sigma\approx0.5\Rvir$, with only
345: a  very    weak    dependence    on  the     concentration  parameter.
346: Figure~\ref{fig:covi} displays the $r_\sigma$  distribution of the  GCSs
347: in units   of the group's  virial  radius, $\Rvir$.   The distribution
348: peaks at $0.11\Rvir$ and has a mean of $0.13\Rvir$.
349: %%
350: %%----------------------------------
351: \subsection{Sizes of satellite halos} 
352: %%----------------------------------
353: %%
354: %%\clearpage
355: \begin{figure}
356: \plotone{f2.eps}
357: %%\plotone{\fig rkpc.Rsph6.mim200.eps}
358: \caption{\label{fig:rkpc}
359: Distribution of the radii of satellites found within the virial radius
360: of the  group.   In  this context radius    refers to listed  fraction
361: (0.5, 1.0)  of the  satellites  spatial dispersion $r_\sigma$.  For
362: example, the typical inner radii  probed by the $b=0.5r_\sigma$ sample
363: is $\sim30\hkpc$.}
364: \end{figure}
365: %%\clearpage
366: %%
367: The aim of the  present analysis is  twofold: (1) to assess the impact
368: of the  group tidal field  on the  satellite  orientations, and (2) to
369: compare the alignment signals in our N-body simulation to observations
370: of galaxy alignments.  The impact of the group tidal field is stronger
371: at larger satellite-centric radii.  On the  other hand, since galaxies
372: reside at the centers of their dark matter halos, the central parts of
373: the  satellites   are more of   interest  when comparing the alignment
374: signals with those observed for galaxies. To meet both requirements we
375: therefore measure the  orientation of the  satellite mass distribution
376: within  two radii.   In analogy to   the measurement of GCS  sizes, we
377: determine these   radii  with reference   to  the spatial   dispersion
378: $r_\sigma$.   More precisely, we choose  the particles  within 1.0 and
379: $0.5r_\sigma$  as the basic sets for   the subsequent determination of
380: the satellite     orientation   (see  Section~\ref{sec:ori}    below).
381: Figure~\ref{fig:rkpc} displays  the distributions of the corresponding
382: physical   sizes.   The  $0.5r_\sigma$    sample  probes the    matter
383: distribution of   the    satellites within  $\sim25\hkpc$,   which  is
384: comparable  to the sizes of elliptical   galaxies.  The mean, physical
385: radii  of  the $1.0r_\sigma$  sample is  $\sim50\hkpc$.  If not quoted
386: otherwise  we will display the results  for  the $0.5r_\sigma$ sample,
387: since this may most    closely resemble the properties of   observable
388: galaxy distributions (outside  of the  very  central part of the  host
389: halo).
390: %%
391: %%----------------------
392: \subsection{Orientation} 
393: \label{sec:ori}
394: %%----------------------
395: %%
396: %%\clearpage
397: \begin{figure}
398: \plotone{f3.eps}
399: %%\plotone{\fig shape.Rsph6.mim200.eps}
400: \caption{\label{fig:shape}
401: Distribution of satellite shapes, represented by the ratio of shortest
402: to the  longest semi-major axis of   the mass-ellipsoid ($s=c/a$).  The
403: colors  correspond to the samples with  different  truncation radii as
404: listed. With  increasing  size    the   halos become  rounder.     The
405: distribution of    the  $b=0.5r_\sigma$ sample  is   rather symmetric,
406: whereas for larger truncation radii there appears a slight asymmetry.}
407: \end{figure}
408: %%\clearpage
409: %%
410: There   are   a  few   different   ways   found   in  the   literature
411: \citep[e.g.,][]{2002sgdh.conf..109B,               2002ApJ...574..538J,
412: 2005ApJ...627..647B,  2005ApJ...629..781K,  2006MNRAS.367.1781A}  to
413: model  halos as  ellipsoids.  They  all differ  in details,  but most
414: methods  model halos  using the  eigenvectors from  some form  of the
415: inertia tensor.   The eigenvectors correspond to the  direction of the
416: major axes, and the eigenvalues  to the lengths of the semi-major axes
417: $a\geq b  \geq c$.  Following  \cite{2006MNRAS.367.1781A} we determine
418: the main  axes by iteratively  computing the eigenvectors of  the {\it
419: distance weighted} inertia tensor.
420: %%
421: \begin{equation}
422: I_{ij} = \sum_{k=1,N} {r_{ki}r_{kj}\over r^2_{k}}\ , 
423: \end{equation}
424: %% 
425: where $r_{ki}$ denotes  the $i$th component  of the position vector of
426: the $k$th particle with respect to the center of mass and
427: %%
428: \begin{equation}
429: \label{eq:weight}
430: r_k = \sqrt{{x^2\over a^2} + {y^2\over b^2} + {z^2\over c^2}}
431: \end{equation}
432: %%
433: is the elliptical distance  in the eigenvector coordinate  system from
434: the center to the $k$th particle.  The square roots of the eigenvalues
435: of the inertia tensor determine the axial ratios of the halo ($a:b:c =
436: \sqrt{\lambda_a}:\sqrt{\lambda_b}:\sqrt{\lambda_c}$). The iteration is 
437: initialized by computing the eigenvalues of the inertia tensor for the
438: spherically truncated halo.  In the following iterations the length of
439: the intermediate axis is kept unchanged and all bound particles within
440: the ellipsoidal window determined by  the eigenvalues of the foregoing
441: iteration are used for the computation of the new inertia tensor.  The
442: iteration is completed   when the eigenvectors    have converged.  The
443: direction   of    the resulting  major axis    is   identified  as the
444: orientation.  The advantage of keeping  the intermediate axis fixed is
445: that  the number of  particles within the  varying ellipsoidal windows
446: remains almost  constant.  Instead, if  the longest (shortest) axis is
447: kept constant the  number of particles  within the ellipsoidal windows
448: can decrease (increase) substantially during the iteration.
449: 
450: Note  that we apply this   truncation   to all (sub)structures,   both
451: satellites and  GCSs, and that  the orientation of each sub(structure)
452: is measured   within   this truncation  radius.   Throughout  we  only
453: consider those sub(structures)   that   comprise at least 200    bound
454: particles within the volume of the final ellipsoid (corresponding to a
455: lower limit  in mass of  $\approx 10^{11}\hMsol$).  For the satellites
456: this implies that a   smaller truncation radius  results in  a smaller
457: sample.  For example,   there are 772 $0.5r_\sigma$ satellites  within
458: the virial radii   of  our groups   whereas  the $1.0r_\sigma$  sample
459: comprises 1431 satellites.  Since all 515 GCSs  contain more than 200
460: particles within $0.5r_\sigma$ their sample size is independent of the
461: truncation radius used.
462: 
463: Figure~\ref{fig:shape}  displays the    distribution  of   the   shape
464: parameter $s=c/a$.   The  shading corresponds to  different truncation
465: radii as  listed.  There is a  weak  indication that satellites become
466: more spherical with  increasing truncation radii.  A  similar behavior
467: was found   for  isolated halos   \citep[e.g.,][]{2002ApJ...574..538J,
468: 2006MNRAS.367.1781A}.  As  discussed by \cite{2006MNRAS.367.1781A} the
469: exact determination   of individual shapes   may need as  many as 7000
470: particles, so  that  the resolution  of the present  simulation is not
471: suited for  the analysis of  (sub)structure shapes.   However, for the
472: determination of the orientations, which is the focus of this paper, a
473: particle limit of 200 can be considered conservative
474: \citep[cf.,][]{2002MNRAS.335L..89J, 2007arXiv0707.1702P}.   A study
475: examining the shapes of substructure in  a single high-resolution
476: Milky  Way-sized halo can be found in \cite{2007arXiv0705.2037K}.
477: %%
478: %%------------------------------------
479: \section{Three dimensional Alignments}
480: \label{sec:3D}
481: %%------------------------------------
482: %%
483: For both classes of objects, GCSs and satellites, the orientations are
484: determined according to    the  approach described  above.    A  third
485: orientation-like quantity is   given by  the   direction of the   line
486: connecting a    GCS-satellite  pair.   Throughout  we  refer   to  the
487: orientation  of  the GCS,  the satellite and   the  connecting line as
488: $\vec{a}_{\rm  GCS}$, $\vec{a}_{\rm SAT}$ and $\vec{r}$, respectively.
489: These quantities are unit vectors, such that the scalar product of two
490: vectors yields the cosine of the angle between them.  We will focus on
491: three   different types of alignment,   (1)  the {\it halo  alignment}
492: between the orientations of the GCSs and the connecting lines, (2) the
493: {\it radial alignment} between the  orientations of the satellites and
494: the  connecting lines and (3) the  {\it  direct alignment} between the
495: orientation of the  GCS and that  of its satellites.  In  addition, we
496: also consider   various   alignments based   on  the  proper velocity,
497: $\vec{v}$, of  the satellite with respect to  its GCS.  In particular,
498: we  discuss (4) the  {\it radial velocity alignment} between $\vec{v}$
499: and  $\vec{r}$,   (5)  the  {\it   halo   velocity alignment}  between
500: $\vec{a}_{\rm GCS}$  and $\vec{v}$,  and finally  (vi) the  {\it  auto
501: velocity alignment} between the orientations, $\vec{a}_{\rm SAT}$, and
502: velocities, $\vec{v}$, of the  satellites.  Here $\vec{v}$ is the unit
503: vector indicating  the direction of the  {\it proper}  velocity of the
504: satellite (including the Hubble flow) relative to the host.  Since all
505: the  other quantities also represent  unit vectors the scalar products
506: yield the cosines of the enclosed angles.
507: %%
508: %%-------------------------
509: \subsection{Halo alignment}
510: \label{sec:halo3D}
511: %%-------------------------
512: %%
513: %%\clearpage
514: \begin{figure}
515: \plotone{f4.eps}
516: %%\plotone{\fig a11.Rsph6Rmax6.bin12.mim200.eps}
517: \caption{\label{fig:a11}
518: Mean values of  the cosines of  the angles between the orientations of
519: the GCS  and  the connecting lines   to the satellites,  $\aII$,  as a
520: function  of  $r/\Rvir$   for the   $0.5r_\sigma$ sample.  The  dotted
521: horizontal   line   indicates   the mean  values     for an  isotropic
522: distribution. The error bars  indicate the $95\%$ bootstrap confidence
523: intervals within each distance bin.}
524: \end{figure}
525: %%\clearpage
526: %%
527: %%
528: %%The orientation of the GCS is correlated with the shape of the overall
529: %%halo    \citep[e.g.,][]{2005ApJ...627..647B}.     Consequently,    the
530: %%distribution of the  satellites is expected to correlate  with the GCS
531: %%since  the accretion  of satellites  and diffuse  dark  matter follows
532: %%similar    patterns   causing    similar   spatial    distributions.   
533: %%Alternatively, \cite{2006ApJ...637..561L} propose that the tidal field
534: %%of the host  is responsible for the correlation  between the satellite
535: %%distribution and the host halo, ultimately causing a similar alignment
536: %%between the orientation of the GCS and the satellite distribution.
537: 
538: In  order to measure the  alignment between the  GCS and the satellite
539: distribution we use $\vec{a}_{\rm GCS}$ and $\vec{r}$ (the orientation
540: of the GCS and the position of the satellite with respect to its GCS).
541: Figure~\ref{fig:a11} displays the  radial dependence  of $\aII$ within
542: $6\Rvir$, where $\langle\cdot\rangle$  denotes the mean value within a
543: bin of  $r/\Rvir$. The   error  bars  indicate the   $95\%$  bootstrap
544: confidence intervals based on 1000 bootstrap samples for each distance
545: bin.
546: %% The shaded areas  mark the parameter space between
547: %% the $16^{\rm  th}$ and $84^{\rm th}$  percentiles of the distributions
548: %% obtained from the 100 random  samples.   A signal outside this  shaded
549: %% region  therefore indicates that it  is inconsistent with no alignment
550: %% (i.e.,  with isotropy)  at   more than  68   percent  confidence.  The
551: %% different line  styles  correspond to different  truncation radii. The
552: %% solid line probes the alignment based on the  orientation of the inner
553: %% regions of the GCS ($0.5r_\sigma$) whereas the broken lines correspond
554: %% to GCS orientations at measured out to larger radii. Within the errors
555: %% the signals agree for all three cases, indicating that the orientation
556: %% of the GCS does not change significantly with radius.
557: %%
558: Over the  entire  range of distances   probed, the mean  values of the
559: cosines deviate  significantly from   a isotropic  distribution.   The
560: strength of    the alignment, i.e.    the  deviation  from $\aII=0.5$,
561: increases  with group   centric distance  and   reaches  a maximum  at
562: $\sim1.7\Rvir$.  The subsequent  decline,  however, is very  weak  and
563: even at   $6\Rvir$     the alignment   is     still very    pronounced
564: ($\aII\approx0.55$), with no  clear indication  of a  downward  trend.
565: The  fact  that there  is  strong  alignment over  such   a long range
566: suggests that the halo intrinsic alignment is closely connected to the
567: filamentary structure in which the groups are embedded in.  Since here
568: we focus  on the transition  between  group and environment  dominated
569: areas  we do  not  aim  to map out   the  entire range of  the  radial
570: alignment.  
571: %% However, we can use simple  geometrical arguments to argue
572: %% that  filaments have a  finite extent.   Consider  for simplicity that
573: %% filaments are infinitely long  and perfectly straight cylinders with a
574: %% uniform  distribution of satellites.   Then those satellites at larger
575: %% group-centric radii will extend a  smaller solid angle as viewed  from
576: %% the GCS. Consequently, $\aII \rightarrow 1$ as $r \rightarrow \infty$.
577: %% The  fact that, instead, $\aII$   decreases  (weakly) with  increasing
578: %% radius indicates that the filaments are not infinitely long.
579: 
580: The weakening of  the signal at small scales may  be attributed to the
581: fact that the information about  the filamentary origin is washed away
582: once  the satellites  start to  orbit  within the  groups (i.e.,  once
583: non-linear effects kick in). Yet,  the orientation of the group itself
584: is  closely correlated  with the  surrounding filamentary  network, so
585: that a residual alignment is maintained by the overall distribution of
586: satellites   orbiting   in   the   potential   well   of   the   group
587: \citep[cf.][]{1987ApJ...321..113S,                 2005ApJ...629..219Z,
588: 2007MNRAS.378.1531K}. Additionally,  if one assumes that filaments are
589: approximately cylindrical  in shape and  the  GCS is  aligned with the
590: orientation  of  the  cylinder,   then the   mean  angles between  the
591: orientation of  the GCS and  the satellites position become  larger at
592: smaller group-centric radii. In  fact,  at distances smaller than  the
593: radius     of  the   cylinder    the  distribution   will  converge to
594: isotropic. Finally, some contribution to the decrease of the alignment
595: strength on small  scales may come   from the fact  that satellites on
596: nearly radial orbits are filtered  out during their epicenter passage.
597: They get severely  stripped and consequently  the  number of particles
598: that  remains bound can    easily fall below the detection   criterion
599: (minimum of 200 particles), thus weakening the alignment signal.
600: %%
601: %%---------------------------
602: \subsection{Radial alignment}
603: \label{sec:radial3D}
604: %%---------------------------
605: %%
606: %%\clearpage
607: \begin{figure}
608: \plotone{f5.eps}
609: %%\plotone{\fig ax1.Rsph6Rmax6.bin12.mim200.eps}
610: \caption{\label{fig:ax1}
611: Same  as  Figure~\ref{fig:a11},  but  for  the  distributions of  cosines
612: between the satellite orientation and the  connecting line to the GCS,
613: $|\vec{a}_{\rm SAT}\!\cdot\!\vec{r}|$ for the $0.5$ and
614: $1.0r_\sigma$ samples.}
615: \end{figure}
616: %%\clearpage
617: %%
618: %%
619: The radial alignment, $\axI$, probes   the orientations of  individual
620: satellites,  $\vec{a}_{\rm SAT}$,  relative  to the direction pointing
621: towards their  GCS,  $\vec{r}$.  Figure~\ref{fig:ax1}  displays $\axI$
622: for  distances  up to  $6\Rvir$.  The  line styles represent different
623: truncation   radii of the  satellites.    Over   the entire range   of
624: group-centric   distances probed,   the  data  reveal  a   significant
625: anisotropic distribution.  The  signal   is most pronounced on   small
626: scales, where it also  shows  a strong  dependence on the   truncation
627: radii.   The $1.0r_\sigma$ sample,  which includes the behavior of the
628: outer   mass shells of  the  satellites,  clearly exhibits a  stronger
629: deviation from isotropy.  Within $\sim 1.5\Rvir$ there is a pronounced
630: decline of  the radial alignment signal,  while it  remains remarkably
631: constant   at larger radii.     For   distance in the  range   between
632: $2-6\Rvir$ we detect a weak but significant signal, $\acI\approx0.52$,
633: inconsistent with isotropy at 95\%  confidence level in good agreement
634: with
635: \cite{2007arXiv0704.2595H}.           In      a     recent      study,
636: \cite{2007arXiv0705.2037K}  detected no radial alignment for distances
637: $\gtrsim3\Rvir$.  However, their analysis  is based on a  resimulation
638: of  a  single galaxy-sized  host  halo.   Since  this halo   is rather
639: isolated, in that  it  has  not experienced  any major   merger  after
640: redshift $z=1.7$, it is likely  that its ambient   filaments have already  been
641: drained.
642: 
643: At large distances  satellites preferentially reside  in filaments (as
644: discussed in the context of Figure~\ref{fig:a11}) which point radially
645: towards    the    groups.    Consequently, the   signal     on  scales
646: $\gtrsim2\Rvir$ indicates   an   alignment   between  the    satellite
647: orientations  and the filaments in which  they are  embedded.  Such an
648: alignment may  be caused by accretion  of matter along those filaments
649: or by the local tidal fields generated by the mass distribution within
650: the filaments.  The group tidal field  is not likely to be responsible
651: for the observed large scale alignment signal due to its rapid decline
652: with   distance.   On small scales,    however,  the tidal  field  can
653: substantially  alter the orientations of  the satellites.  As shown by
654: \cite{1994MNRAS.270..390C} the time scale on which a prolate satellite
655: can adjust its  orientation to  the  tidal field  of  a group is  much
656: shorter than the Hubble time, but  longer than its intrinsic dynamical
657: time.   Therefore,   the  adjustment of   the  satellite  orientations
658: parallel to the  gradients of the group  potential offers a convincing
659: explanation for   a radial alignment  signal   on small scales.   This
660: perception  is  further supported by the   dependence of the alignment
661: strength on the  truncation radii of  the satellites.  For the largest
662: radii, which are  strongest  affected by  tidal forces,  the alignment
663: signal is strongest.
664: %%
665: %%---------------------------
666: \subsection{Direct alignment}
667: %%---------------------------
668: %%
669: %%\clearpage
670: \begin{figure}
671: \plotone{f6.eps}
672: %%\plotone{\fig ac1.Rsph6Rmax6.bin12.mim200.eps}
673: \caption{\label{fig:ac1}
674: Same as Figure~\ref{fig:a11},  but for the distributions of
675: cosines between the orientation of the satellites and the
676: GCS, $|\vec{a}_{\rm GCS}\!\cdot\!\vec{a}_{\rm SAT}|$.}
677: \end{figure}
678: %%\clearpage
679: %%
680: The  strong signals for  halo and  radial alignment   may lead  to the
681: expectation of a comparably pronounced signal for the direct alignment
682: between the  orientation of  the   GCS, $\vec{a}_{\rm GCS}$, and   the
683: orientations of its satellites,  $\vec{a}_{\rm SAT}$.  However, as can
684: be seen in Figure~\ref{fig:ac1}, the signal  is weak.  There is only a
685: weak trend for positive  alignment  up to $2\Rvir$.  The  significance
686: found at  distances between $1$ and $\sim2\Rvir$  seems to be somewhat
687: higher  ($\sim90\%$   confidence). Based    on an   analytical   model
688: \cite{2005ApJ...629L...5L}  predict   a   certain degree   of parallel
689: alignment between host and satellite orientations due to the evolution
690: of the satellites within the tidal shear field of host. The signal for
691: the direct  alignment may be  a  relic of  this effect.
692:  
693: To  summarize, we find positive alignment  signals for all three types
694: of alignment tested here.  However, they differ in strength and radial
695: extent. The halo alignment is the strongest and reaches far beyond the
696: virial radii of the groups ($\gtrsim6\Rvir$).  The radial alignment is
697: most pronounced at small scales, where it  reveals a strong dependence
698: on the radial  extent of the  satellite over which its orientation has
699: been measured. Although the radial alignment is weak beyond $\sim 1.5
700: \Rvir$,  the signal  stays  remarkably  constant out  to  $\sim6\Rvir$.   
701: Finally, the  least prominent signal  comes from the direct alignment.
702: This ranking of the alignment strengths is  in good agreement with the
703: observational results reported  in Paper~I.  
704: %%The radial  dependence of
705: %%the  halo alignment is readily  explained by the filamentary structure
706: %%of the group environment superimposed on the small scale signal caused
707: %%by  the satellites   orbiting  within the   triaxial group   potential
708: %%well. On  small  scales, the  radial alignment arises  from  the group
709: %%tidal  field, while on   large scales  it  most  likely  reflects that
710: %%satellites are aligned with the filaments in which they reside.
711: %%
712: %%
713: \subsection{Alignments based on subhalo velocities}
714: \label{sec:vel}
715: %%
716: If tidal forces give rise to the radial alignment  on small scales, as
717: displayed in Fig.~\ref{fig:ax1}, the  satellite orientations should be
718: related to their actual velocities and the local gradients of the host
719: potential.  For instance a satellite  moving radially towards the  GCS
720: will show  an  enhanced radial alignment  since  the  gradient of  the
721: potential and the  actual velocity are pointing  in the same direction
722: inducing  an orientation in radial direction.   On the other side the
723: orientations of satellites  moving perpendicular to  the gravitational
724: field (i.e.  tangentially with respect to the GCS) will lie in between
725: their velocities and  the gradients of the   potential well.  To  gain
726: some  more insight  into the  dynamical  origin of  the alignments, we
727: include  the directions  of  satellite velocities  into  the alignment
728: study.  We will  consider  three different  kinds  of alignments:  the
729: radial velocity alignment,  $|\vec{v}\cdot\vec{r}|$, the halo velocity
730: alignment  $|\vec{a}_{\rm  GCS}\cdot\vec{v}|$  and the  auto  velocity
731: alignment $|\vec{a}_{\rm SAT}\cdot\vec{v}|$.
732: %%
733: %%\clearpage
734: \begin{figure}
735: \plotone{f7.eps}
736: %%\plotone{\fig ioRsph6Rmax6.bin12.mim200.eps}
737: \caption{\label{fig:io} Ratio of inward moving satellites,
738: $f_{\rm in} = n_{\rm in}/ (n_{\rm out}  + n_{\rm in})$ for the
739: $0.5r_\sigma$ sample. Error bars are Poisson.}
740: \end{figure}
741: %%\clearpage
742: %%
743: To facilitate the interpretation  of the velocity alignments, we split
744: the  satellites  according  to  whether  their net  motion  is  inward
745: ($\vec{v} \! \cdot  \! \vec{r} < 0$) or outward ($\vec{v}  \! \cdot \! 
746: \vec{r} >  0$) with respect  to their group.   Figure~\ref{fig:io} shows
747: the fraction of inward moving  satellites, $f_{\rm in}$, as a function
748: of their  group centric distances.   Note that $f_{\rm in}$  reaches a
749: maximum around  $\sim 2\Rvir$, beyond which the  Hubble flow gradually
750: starts to  become more and  more important.  In fact,  at sufficiently
751: large radii, where the Hubble flow dominates, one expects that $f_{\rm
752: in}=0$, and all satellites reveal an outward motion.  For satellites
753: that are in virial equilibrium within the group potential (i.e., at $r
754: \la \Rvir$), one  expects roughly equal numbers of  inward and outward
755: moving  systems (i.e.,  $f_{\rm  in}=0.5$).  However,  on these  small
756: scales  one has  an  additional contribution  from  the infall  region
757: around  the  group,  causing  $f_{\rm  in} >  0.5$.   In  addition,  a
758: substantial fraction  of satellites  get stripped below  the detection
759: limit (200 particles)  at their peri-centric passage, so  that they no
760: longer   contribute   to   the   signal  on   their   outward   motion
761: \citep[cf.,][]{2007MNRAS.375..313F}.    At   $\Rvir$,   the   outgoing
762: satellite  fraction  is  about   40\%,  which  is  (within  the  errors)
763: consistent    with    the   value    $\sim    30\%$   determined    by
764: \cite{2005MNRAS.364..424W}.  
765: % The  majority  of these  outward  moving
766: %satelites must  have been closer to  the group center  before. 
767: If one assumes an average ratio of 6:1 between apo- and peri-center distances
768: for     typical    satellite     orbits    \citep{1998MNRAS.300..146G,
769: 1999ApJ...515...50V} the majority of these satellites must have passed the
770: central parts of the group before \citep[cf.,][]{2007astro.ph..3337D}.
771: 
772: %%
773: %%\clearpage
774: \begin{figure}
775: \plotone{f8.eps}
776: %%\plotone{\fig iovel.Rsph6Rmax6.bin12.mim200.eps}
777: \caption{\label{fig:vel}
778: Same as Figure~\ref{fig:a11},  but  for $\vel$.  The  upper, middle  and
779: lower panel displays the signal  for all, the  inward and the  outward
780: moving satellites, respectively.}
781: \end{figure}
782: %%\clearpage
783: %%
784: The upper panel of  Figure~\ref{fig:vel} displays the radial  velocity
785: alignment, $\vel$, as a function of $r/\Rvir$.  $\vel > 0.5$ indicates
786: that the distribution of angles between $\vec{r}$ and $\vec{v}$ is not
787: isotropic, instead, on   average they preferentially   point in radial
788: directions. This behavior is in agreement  with earlier studies of the
789: velocity anisotropy of subhalos  which   is usually expressed  by  the
790: anisotropy parameter $\beta=1-0.5(\sigma_t/\sigma_r)^2$
791: \citep[e.g.,][]{1987gady.book.....B}, where  $\sigma_t$ and $\sigma_r$
792: denote the  velocity dispersions of the  satellites  in the tangential
793: and radial  direction, respectively.  Note,  $\vel$ is closely related
794: to  $\beta$.  If one assumes a   relaxed (steady-state) halo the above
795: mentioned tendency towards  radial  motions translates  into a  higher
796: radial velocity dispersion compared to the tangential one $\sigma_r >
797: \sigma_{t1} = \sigma_t/\sqrt{2}$ (where $\sigma_{t1}$ and $\sigma_t$ 
798: are the  one  and  two  dimensional tangential  velocity  dispersions,
799: respectively  and tangential isotropy is assumed).   Thus $\vel > 0.5$
800: on    small  scales   ($r  \la  2\Rvir$)   suggest    that $\sigma_r  >
801: \sigma_t/\sqrt{2}$,  in  good  qualitative  agreement  with  numerical
802: simulations which have shown that $\beta >  0$ for subhalos within the
803: virial radius of their hosts.
804: \citep{1998MNRAS.300..146G, 2000ApJ...539..561C, 2004MNRAS.352..535D}.
805: 
806: In accordance with the spherical collapse model the signal extends out
807: to  $\sim   2\Rvir$,    which  roughly   reflects   the   distance  of
808: turnaround.  At $2.5\Rvir$  the  distribution  is close  to  isotropic
809: suggesting that at  these distances the impact  of the group potential
810: is   negligible and the  satellite  motions  are   dominated by  local
811: potential variations arising from the  filaments and dark matter halos
812: within these filaments.  Note  that the  presence of this  filamentary
813: structure  in   the   vicinity  of  groups is   clearly  evident  from
814: Figure~\ref{fig:a11}.   Finally, the  increase  of the radial velocity
815: alignment on large scales,   $\gtrsim4\Rvir$,  is simply due   to  the
816: Hubble flow (i.e., $\vel
817: \rightarrow  1$  at $r  \rightarrow  \infty$).   The  middle panel  of
818: Figure~\ref{fig:vel} shows $\vel$ for the inward moving satellites only.
819: The radial trend within $2\Rvir$  is somewhat enhanced compared to the
820: upper  panel. At  larger radii,  the inward  moving satellites  have a
821: velocity structure that is  consistent with isotropy.  The lower panel
822: of Figure~\ref{fig:vel}  reveals a marked difference in  the behavior of
823: $\vel$  for the outward  moving satellites.   It indicates  a slightly
824: radial trend for  satellites within $1\Rvir$ which is  much lower than
825: seen in the  upper two panels.  Within $1-2\Rvir$  it drops below 0.5,
826: indicating a preference for  tangential velocities.  Together with the
827: information  derived  from  Figure~\ref{fig:io}  this  suggests  that  a
828: substantial   fraction  of  outward   moving  satellites   located  at
829: $1-2\Rvir$  currently  are close  to  their  apo-center passage  after
830: having  crossed the  more central  regions of  the group.  Finally, on
831: large scales  the outward moving satellites clearly  reveal the Hubble
832: flow.
833: 
834: %%
835: %%\clearpage
836: \begin{figure}
837: \plotone{f9.eps}
838: %%\plotone{\fig c1v.Rsph6Rmax6.bin12.mim200.eps}
839: \caption{\label{fig:c1v} Same as Figure~\ref{fig:a11},  but for $\cIv$.}
840: \end{figure}
841: %%\clearpage
842: %%
843: Figure~\ref{fig:c1v} displays the  radial dependence  of $\cIv$  which
844: measures  the cosines of the angels   between the satellite velocities
845: and the  orientation of the GCS.  On  large  scales the radial outward
846: motion  caused by the Hubble flow  exceeds the  internal velocities of
847: the satellites  within  the filaments.   Since   the GCS is   strongly
848: aligned  with  these  filaments over  the  entire  radial range  shown
849: (cf. Figure~\ref{fig:a11}), one has that  $\cIv > 0.5$ on scales where
850: the Hubble flow    becomes important ($\gtrsim4\Rvir$).    The  strong
851: alignment signal on small scales indicates that the satellites tend to
852: move parallel to the orientation of the GCS.  According to
853: \cite{1997MNRAS.290..411T} and \cite{2006MNRAS.367.1781A} the
854: principal    axes of  the   velocity  anisotropy  tensor are  strongly
855: correlated  with     the    principal    axes of      the    satellite
856: distribution.   Therefore,   the     alignment   found    for   $\aII$
857: (Figure~\ref{fig:a11})  actually   implies  an  analogous   signal for
858: $\cIv$.   However, in  contrast to  the  halo  alignment, $\aII$,  the
859: velocity    halo   alignment,      $\cIv$,  only  extends     out   to
860: $\sim1\Rvir$.   Beyond this  radius   a  substantial  fraction of  the
861: satellites shows relatively large angles  between their velocities and
862: the orientation of  the GCS which is  consistent  with the picture  of
863: tangential motions associated    with the apo-center  passage   of the
864: satellites, as discussed in the context of Figure~\ref{fig:vel}.
865: 
866: %%
867: %%\clearpage
868: \begin{figure}
869: \plotone{f10.eps}
870: %%\plotone{\fig a1v.Rsph6Rmax6.bin12.mim200.eps}
871: \caption{\label{fig:a1v} Same as Figure~\ref{fig:a11},  but for the
872: distributions    of cosines   between   the  satellites  velocities and
873: positions, $|\vec{v}\!\cdot\!\vec{r}|$.}
874: \end{figure}
875: %%\clearpage
876: %%
877: Finally  we  consider  the  auto  velocity  alignment,  $\aIv$,  which
878: reflects  the  distribution  of  the  cosines  between  the  satellite
879: velocities   and   their   orientations,  $|\vec{v}\cdot\vec{a}_{\rm SAT}|$.   
880: Fig~\ref{fig:a1v} displays the variation of $\aIv$ with $r/\Rvir$.  
881: 
882: The   signal for  $\aIv$ shows  a   maximum at  $0.7\Rvir$.  At larger
883: distances  it decreases quickly.   Beyond $1.5\Rvir$  it is roughly in
884: agreement with an isotropic  distribution.  A possible reason  for the
885: slight central dip  is, that satellites  on their peri-center passages
886: move  perpendicular  to   the  gradients   of  the   group  potential.
887: Figure~\ref{fig:ax1},   however, revealed    a preferential     radial
888: orientation of   these   satellites.  Thus,   during   the peri-center
889: passages the angles between satellite  orientations and velocities can
890: become large. The degree of the radial  alignment depends on the ratio
891: between the internal  dynamical time of  the satellite,  with which it
892: can adjust  its   orientation, and the   duration  of  the peri-center
893: passage.   If the peri-center passage occurs  too quickly the time may
894: be too short for a `perfect' radial alignment
895: \citep[cf.,][]{2007arXiv0705.2037K}.  On  large  scales ($1-2\Rvir$) a
896: similar mechanism may  take place.  Above we  have  argued that within
897: this distance range a substantial fraction of  satellites are close to
898: their apo-center passage.  During this phase  the velocities are again
899: perpendicular to the  gradient of the  potential but, as indicated  by
900: Figure~\ref{fig:ax1}, the  satellites   are  oriented radially.    The
901: comparison between the signal for  $\axI$ and $\aIv$ suggests that, in
902: a statistical sense,  the (spatial)   radial alignment is   maintained
903: during the entire orbit of the satellite  within the potential well of
904: the groups, which in turn causes a suppression of  $\aIv$, at its apo-
905: and peri-center.
906: 
907: 
908: %%
909: %%
910: %%----------------------------
911: \section{Projected Alignments}
912: \label{sec:2D}
913: %%----------------------------
914: %%
915: To facilitate  a comparison with  observations, in particular with the
916: results presented  in Paper~I, we  repeat the foregoing analysis using
917: projected data, i.e. we project the particle distribution along one of
918: the   coordinate axes and  compute the  second moment of  mass for the
919: projected particle distribution. Accordingly,
920: for the  distances between  GCS and  satellites we  use  the projected
921: values (all satellites within a  sphere of $6\Rvir$  about the GCS are
922: projected), which we label as $R$ (the physical distances are labelled as
923: $r$).
924: 
925: Since   the  projections   along  the  three   Cartesian
926: coordinate axes are independent we   include all three projections  of
927: each host-satellite  in our 2D sample.   To reduce the contamination by
928: satellites   associated with massive  ambient  groups we exclude those
929: host-satellite systems where another  SKID group more massive than the
930: GCS (which  is  most likely the  center  of an ambient  host-satellite
931: system) is found  within a  sphere of  $6\Rvir$.   After rejection  of
932: `contaminated' groups   we  obtain 1034  and 543  satellites   for the
933: $1.0r_\sigma$ and $0.5r_\sigma$ samples with  3D distances to the  GCS
934: $\leq\Rvir$ (for all groups irrespective of their environment we found
935: 1431  and 772,   see  \S~\ref{sec:ori}.)  Furthermore, since   (due to
936: technical reasons) we project satellites located within a {\it sphere}
937: of $6\Rvir$ the projected volume at  large projected distances shrinks
938: substantially.  Therefore,  we analyze the  2D data  only for projected
939: distances $\lesssim3\Rvir$ which roughly  resembles the  projection of
940: all satellites within a cylinder with a radius  $3\Rvir$ and length of
941: $10\Rvir$ along the `line of sight'.  Thus,  in an approximate manner,
942: uncertainties in the  determination   of group membership   based on
943: redshift measurements are accounted for.
944: 
945: The resolution of  the simulation does not  permit to  probe alignment
946: below $0.3\Rvir$.      Other   authors  \citep[using   semi-analytical
947: techniques, e.g.,][]{2007MNRAS.378.1531K}  have  bypassed this problem
948: by introducing so-called {\it   orphan galaxies}, i.e.  galaxies which
949: are associated  with the once most bound  particle of a satellite halo
950: which subsequently has  become undetectable due   to the stripping  by
951: tidal forces.  Here we do  not adopt this technique  since it does not
952: provide us  with  information about  the  orientation of a  satellite.
953: Both approaches,  considering  only satellite  halos  with  a  minimum
954: number of  particles and  the introduction  of  orphan  galaxies, have
955: certain disadvantages.  The former does not account for galaxies which
956: are hosted by strongly  stripped subhalos  whereas the latter  ignores
957: the dynamical differences of galaxies and (once most bound) particles.
958: 
959: The  application of a fixed  lower  particle limit excludes satellites
960: from  the analysis which   still    constitute distinct objects.    In
961: particular satellites which  are strongly  tidally stripped  may  fall
962: below the selection criterion even if  the galaxy, which is assumed to
963: sit at the center, may still be observable.  Thus, we caution that our
964: satellite sample may be somewhat  biased toward more recently accreted
965: satellites  compared to a hypothetical  galaxy population. This effect
966: appears whenever a  fixed lower   particle  limit is imposed.
967: % It   is independent of the specific value chosen.
968: % which is chosen as the limit. 
969: 
970: %%\clearpage
971: \begin{figure}
972: \plotone{f11.eps}
973: %%\plotone{\fig sketch.eps}
974: \caption{\label{fig:sketch}
975: Illustration of the three angles $\theta$, $\phi$ and $\xi$, which are
976: used  for halo  alignment,  radial  alignment  and  direct  alignment,
977: respectively \citep[cf.,][]{2007ApJ...662L..71F}.}
978: \end{figure}
979: %%\clearpage
980: %%
981: In analogy to Paper~I we define  the angles $\theta$, $\phi$ and $\xi$
982: to  address the projected  halo,  radial and  direct  alignments (same
983: definitions    as   in \S\ref{sec:3D}  but   for   the   2D data,  see
984: Fig.~\ref{fig:sketch}) and  the projected orientations are referred to
985: as position angles  (PAs). It is not  straightforward  to derive galaxy
986: properties, such    as luminosity and   color, from  the   dark matter
987: distribution.  In particular, if the satellite  halo hosts a late type
988: galaxy, it is not obvious  how to accurately determine the orientation
989: of the disk (but see e.g., \citealt{2007MNRAS.378.1531K} and
990: \citealt{2007arXiv0704.3441A} for attempts). 
991: On  the    other side, if  one   focuses  on early type   galaxies the
992: orientation  of  the central dark matter   distribution is very likely
993: correlated  with the orientation  of   the stellar component (see  the
994: evidence from gravitational lensing, e.g.,
995: \citealt{2002sgdh.conf...62K}).  The   lower  particle  limit  for the
996: satellites  results  in   a   lower mass   of $10^{11}\hMsol$   within
997: $25\hkpc$.    Assuming  a  dynamical mass-to-light    ratio of  a  few
998: \citep[][]{2006MNRAS.366.1126C}  within  this radius  yields a stellar
999: component  which roughly resembles  $L_\ast$  galaxies. Therefore, our
1000: findings in the current paper may be  best compared with results based
1001: on bright  early-type satellite galaxies.  However, as we have pointed
1002: out   in   Paper~I, our observational   results   were only marginally
1003: dependent  on the luminosity/mass  of satellite galaxies. Therefore, a
1004: comparison with observations  based on somewhat fainter  satellites is
1005: viable as well.
1006: %%
1007: %%---------------------------
1008: \subsection{Halo alignment}
1009: %%---------------------------
1010: %%
1011: %%\clearpage
1012: \begin{figure}
1013: \plotone{f12.eps}
1014: %%\plotone{\fig SD2a11.Rsph6Rmax3.bin06.mim200.eps}
1015: \caption{\label{fig:D2a11}
1016: Mean   angle,  $\theta$,  between the   PA of  the    GCS and the line
1017: connecting the GCS   and  a satellite,   as a  function of   projected
1018: distance $R/\Rvir$  with equidistant  bins  of  $0.5\times\Rvir$.  The
1019: error bars give the $95\%$ bootstrap confidence intervals for the mean
1020: angles within   each  bin. The  short  horizontal   line on the   left
1021: indicates the signal for the innermost bin if only the satellites with
1022: in 3D  distances  $\leq1\Rvir$ are   projected. The corresponding   3D
1023: results are shown in Fig.~\ref{fig:a11}.}
1024: \end{figure}
1025: %%\clearpage
1026: %%
1027: 
1028: Figure~\ref{fig:D2a11}  shows   the  results  obtained  for  the angle
1029: $\theta$ between  the orientation of the  GCS  and the line connecting
1030: the GCS  with the satellite.  The   short horizontal line on  the left
1031: indicates  the  result for the innermost  bin  if only  the satellites
1032: within $1\Rvir$ are projected.  The sample shows $\langle\theta
1033: \rangle < 45^{\circ}$ for the entire distance range. The
1034: error bars give the $95\%$ bootstrap confidence intervals for the mean
1035: angles  within  each  bin. The alignment  strength  within  $\Rvir$ is
1036: $\sim42^\circ$, in good agreement with the findings of
1037: \cite{2005ApJ...628L.101B},  \cite{2006MNRAS.369.1293Y}. In Paper~I we
1038: found a mean value $\theta\approx41^\circ$  within $0.5\Rvir$ which is
1039: very close  to   the  values we   obtain  for the  innermost  bin,  in
1040: particular if only the   satellites within $1\Rvir$ (short  horizontal
1041: lines on the left) are projected. As also shown by
1042: \cite{2006ApJ...650..550A} the alignment signal extends beyond the
1043: virial radius.  The strongest amplitude  is  found outside the  virial
1044: radius   at  $\sim   1.7\Rvir$.   Currently  there  are  no  available
1045: observations  covering  the  same  distance  range.   The  analysis in
1046: Paper~I,  for instance, is based  on galaxies within the virial radius
1047: whereas      we   use  all   galaxies       with  projected  distances
1048: $\lesssim3\Rvir$. According to our findings a  search for alignment of
1049: satellite distribution in group environments for distances larger than
1050: $\Rvir$ may be a promising proposition. 
1051: %%
1052: %%---------------------------
1053: \subsection{Radial alignment}
1054: %%---------------------------
1055: %%
1056: %%\clearpage
1057: \begin{figure}
1058: \plotone{f13.eps}
1059: %%\plotone{\fig SD2ax1.Rsph6Rmax3.bin06.mim200.eps}
1060: \caption{\label{fig:D2ax1}
1061: Same as Figure~\ref{fig:D2a11}, but for the angle $\phi$. In addition
1062: the radial dependence  of  the $b=1 r_\sigma$  sample is  displayed as
1063: well. The corresponding 3D results are shown in Fig.~\ref{fig:ax1}.}
1064: \end{figure}
1065: %%\clearpage
1066: %%
1067: Figure~\ref{fig:D2ax1}  displays the mean  angle $\phi$ between the PA
1068: of  the  satellite  and the  line   connecting the satellite  with its
1069: GCS. For all group centric distances there  is a clear and significant
1070: signal for the  major axes of the satellites  to point towards the GCS
1071: (i.e., $\langle\phi\rangle < 45^\circ$).  The projection of only those
1072: satellites  within $1\Rvir$  increases   the central signal  by  about
1073: $1^\circ$ (differences between the innermost data points and the short
1074: horizontal lines). The mean angle for  the $0.5r_\sigma$ sample within
1075: the innermost bin is $\sim43^\circ$  and according to Paper~I the mean
1076: value for the red  SDSS satellites within  $0.5\Rvir$ is very close to
1077: this value. However, the  observations suggest a significant alignment
1078: for  red  galaxies  only  out to  $0.7\Rvir$  whereas  the N-body data
1079: indicate that radial     alignment  extends  beyond $3\Rvir$.      The
1080: discrepancy may be caused by the observational confinement to galaxies
1081: within the virial radius.
1082: %%
1083: %%---------------------------
1084: \subsection{Direct alignment}
1085: %%---------------------------
1086: %%
1087: %%
1088: %%\clearpage
1089: \begin{figure}
1090: \plotone{f14.eps}
1091: %%\plotone{\fig SD2ac1.Rsph6Rmax3.bin06.mim200.eps}
1092: \caption{\label{fig:D2ac1}
1093: Same  as Figure~\ref{fig:D2a11},  but  for the  angle $\xi$.}
1094: \end{figure}
1095: %%\clearpage
1096: %%
1097: Figure~\ref{fig:D2ac1} displays  the results for the direct alignment,
1098: based  on  the  angle  $\xi$ between  the   orientations  of GCSs  and
1099: satellites.  The  alignment signal  is significant at  a $\gtrsim95\%$
1100: confidence  level  for  distances  $\lesssim0.5\Rvir$.  In Paper~I  we
1101: obtained $\xi\approx44^\circ$ for  red  satellite with in   $0.5\Rvir$
1102: which indicates a somewhat weaker alignment  than we find here.  Since
1103: the  3D analysis   shows  no   increase  of  $\aII$  at small   scales
1104: (Figure~\ref{fig:a11}) the  central enhancement displayed  here has to
1105: be interpreted as a result of projection effects.
1106:  
1107: In  summary for all  three types of  alignments we find good agreement
1108: between numerical data   presented here and  the observational results
1109: from Paper~I.  In particular the relative strength among the different
1110: alignments is  well reproduced  in   the numerical  analysis.  Due  to
1111: limited resolution the range  below $1\Rvir$ is only  sparsely sampled
1112: thus   no  detailed information about the    radial  dependence of the
1113: alignment signal  on small scales can  be derived. However, the signal
1114: for $\theta$ increases with distance  which is only marginally implied
1115: by the  SDSS   results presented  in  Paper~I.   Also  for $\phi$, the
1116: 	 dependence  on   the    distance  disagrees between   simulations
1117: and observations. It  is   currently  unclear   whether this  is   due
1118: 		 to shortcomings from the numerical or observational side.
1119: %%
1120: %%------------------
1121: \section{Summary}
1122: \label{sec:diss}
1123: %%------------------
1124: %%
1125: Based  on   a  sample   of  515  groups   with  masses   ranging  from
1126: $10^{13}\hMsol$  to $5\times10^{14}\hMsol$  we  have investigated  the
1127: halo alignment,  $\aII$, the radial  alignment, $\axI$ and  the direct
1128: alignment $\acI$, between  the central region of each  group (the GCS)
1129: and  its  satellite  halos  out  to a  distance  of  $6\Rvir$.   Here
1130: $\vec{a}_{\rm GCS}$,  $\vec{a}_{\rm SAT}$  and  $\vec{r}$ denote  the  unit  vectors
1131: associated with  the orientation  of the GCS,  the satellites  and the
1132: line  connecting both  of them.   Additionally, we  have  employed the
1133: directions  of  the  satellite   velocities  $\vec{v}$  to  probe  the
1134: alignments $\vel$, $\cIv$ and $\aIv$,  referred to as radial, halo and
1135: auto velocity alignments, respectively. Our main results are:
1136: %%
1137: \begin{itemize}
1138: \item[(1)] Halo, radial and  direct alignment differ in strength.  The
1139:   halo alignment is strongest followed by the radial alignment. By far
1140:   the   weakest  and least significant  signal  comes  from the direct
1141:   alignment.  This sequence is found in the 3D analysis as well as for
1142:   the projected  data and  agrees well   with our  recent analysis  of
1143:   galaxy alignments in the SDSS (cf., Paper~I).
1144: %%
1145: \item[(2)] The signal  for the halo alignment, $\aII$, reaches  far 
1146:   beyond the virial radii of the groups ($>6\Rvir$) which we interpret
1147:   as evidence for large scale filamentary structure.
1148: %% 
1149: \item[(3)] The signal  for the radial alignment, $\axI$, is  largest
1150:   on small scales.  After a rapid  decline with distance it  flattens,
1151:   such that a   relatively small  $\axI\approx0.52$,  but  significant
1152:   deviation from isotropy  is  detected out to $\sim  6\Rvir$. Whereas
1153:   the small scale signal more likely  owes to the group's tidal field,
1154:   the weak   but significant  signal  on large   scales suggests  that
1155:   satellites tend to  be oriented  along the  filaments in which  they
1156:   reside.
1157: %%
1158: \item[(4)] The 3D signal for the direct alignment, $\acI$, shows a
1159:   weak trend for   parallel orientations on   scales $\lesssim2\Rvir$.
1160:   The   projected data indicate an   increasing  signal for  distances
1161:   $\lesssim0.5\Rvir$  which is likely   caused  by  projection effects. 
1162: %%
1163: \item[(5)]  All  kinetic alignment  signals  are  highly
1164:   significant  at small scales.   The  signal for $\vel$  is basically
1165:   constant within  $2.0\Rvir$, beyond which it  rapidly drops.  In the
1166:   subset     of  outward moving satellites    we  find  a tendency for
1167:   tangential motions which can be  attributed to the satellites  which
1168:   have been accreted earlier and are currently  passing their peri- or
1169:   apo-centers.  The signal for $\cIv$  is maximal at the center, drops
1170:   rapidly with distance  and disappears at $1\Rvir$.  Finally,  $\aIv$
1171:   shows a slight  dip at the  center, reaches a maximum at $0.7\Rvir$,
1172:   and  becomes consistent  with    isotropy at $1.5\Rvir$.  All  these
1173:   features  support the    interpretations advocated  for the  spatial
1174:   alignments.
1175: %%
1176: \end{itemize}
1177: %%
1178: The simulation analyzed here  clearly  demonstrates that tidal  forces
1179: cause a      variety of  alignments  among    neighboring, non-linear
1180: structures.   On  large scales, the   tidal forces are responsible for
1181: creating a  filamentary network, which  gives rise to a halo alignment
1182: out    to at least $6\Rvir$.   The   same tidal  forces  also cause an
1183: alignment between filaments and (sub)structures within the filaments
1184: \citep[cf.,][]{2006MNRAS.370.1422A,2007arXiv0704.2595H}
1185: which in turn   results in a  large  scale radial alignment  with  the
1186: virialized structures at  the nodes of  the cosmic  web.  Within these
1187: virialized structures,    tidal forces are  responsible for   a radial
1188: alignment of its substructures, similar to the tidal locking mechanism
1189: that affects the Earth-Moon system.  This is  further supported by the
1190: fact that the auto  velocity alignment $\aIv$ reveals  a dip  on small
1191: scale, indicating that at  peri-centric passage satellites tend  to be
1192: oriented    perpendicular   to   the    direction  of    their  motion
1193: \citep[cf.,][]{2007arXiv0705.2037K}.   This behavior also   explains,
1194: why the direct   spatial alignment, $\acI$,  is so  weak.   A possible
1195: direct alignment    originating  from the co-evolution    of group and
1196: satellites, as  proposed   by \cite{2005ApJ...629L...5L}, is   quickly
1197: erased  as the satellites  orbit in the  potential well  of the group.
1198: For  future   work it  will be  instructive   to  trace  the orbits of
1199: individual  satellites and consider more  closely how their shapes and
1200: orientations evolve with time.
1201: 
1202: The infall regions around virialized dark matter halos cause a radial
1203: velocity alignment out to $\sim 2\Rvir$, and  an enhancement of inward
1204: moving (sub)structures.  At around the same scale, the (sub)structures
1205: with  a  net outward  movement  have a tendency to  move tangentially.
1206: This most  likely reflects the  apo-centric passage  of substructures
1207: that  have  previously fallen through  the  virialized halo.  Within a
1208: virialized region, the orientation of orbits is naturally aligned with
1209: that of its GCS. Since (sub)structures reveal  at most a weak velocity
1210: bias with respect to dark matter particles
1211: \citep[e.g.,][]{2006MNRAS.369.1698F},  this   causes  a  strong    halo
1212: velocity alignment on scales $\la \Rvir$.  The halo velocity alignment
1213: is   also strong on large  scales  ($\ga 3\Rvir$),  which reflects the
1214: Hubble flow combined  with the filamentary, non-isotropic distribution
1215: of (sub)structures on these scales.  
1216: 
1217: A one-to-one comparison between the N-body  results discussed here and
1218: the    observations    presented    in   Paper~I is not
1219: straightforward. Although we have employed the same  mass range for the groups
1220: in both studies  the resolution of the  current simulation only allows
1221: to  resolve  satellites which are   expected to  host $\gtrsim L_\ast$
1222: galaxies.  These are  bright compared to  our SDSS sample for  which a
1223: lower magnitude  limit of  $^{0.1}M_r -  5\log h  \leq -19$  has  been
1224: adopted.  Nevertheless, the qualitative agreement between the relative
1225: strengths of   the   different   types   of   spatial alignment     is
1226: promising. Supplementary  to  the observational results of  Paper~I we
1227: find a strong halo  alignment and a  somewhat weaker  radial alignment
1228: out to at least $6\Rvir$ which we will investigate further.
1229: 
1230: Finally, the weak but significant detection of radial alignment out to
1231: $6\Rvir$ may  contaminate  the  cosmic  shear measurements   on  these
1232: scales.   This  correlation has to   be  considered, either by  simply
1233: removing  or  down-weighting pairs  of  galaxies  within this distance
1234: range \citep{2002A&A...396..411K, 2003MNRAS.339..711H}. This may be 
1235: particularly important for applications of weak gravitational lensing
1236: for the purposes of precision cosmology.
1237: %%
1238: %%-------------------------
1239: \acknowledgements
1240: %%-------------------------
1241: %%
1242: This work has is  supported by NSFC (10533030, 0742961001, 0742951001,
1243: 973 Program No. 2007CB815402) and  the Knowledge Innovation  Program of
1244: the  Chinese  Academy of  Sciences,   Grant No. KJCX2-YW-T05.  The CAS
1245: Research  Fellowship  for  International Young  Researchers  (AF), the
1246: local support of the Chinese Academy of  Sciences (HJM and SM) and the
1247: Alexander von  Humboldt Foundation  (SM)  is gratefully  acknowledged.
1248: HJM would like  to acknowledge the  support  of NSF ATP-0607535,  NASA
1249: AISR-126270, and NSF IIS-0611948 .
1250: %%
1251: %%\bibliography{lit}
1252: %%
1253: \begin{thebibliography}{75}
1254: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
1255: 
1256: \bibitem[{{Adelman-McCarthy} {et~al.}(2006)}]{2006ApJS..162...38A}
1257: {Adelman-McCarthy et~al.} 2006, \apjs, 162, 38
1258: 
1259: \bibitem[{{Agustsson} \& {Brainerd}(2006a)}]{2006ApJ...644L..25A}
1260: {Agustsson}, I. \& {Brainerd}, T.~G. 2006a, \apjl, 644, L25
1261: 
1262: \bibitem[Agustsson \& Brainerd(2006b)]{2006ApJ...650..550A}
1263: ---. 2006b, \apj, 650, 550 
1264: 
1265: \bibitem[{{Agustsson} \& {Brainerd}(2007)}]{2007arXiv0704.3441A}
1266: ---. 2007, arXiv:0704.3441v1 [astro-ph]
1267: 
1268: \bibitem[{{Allgood} {et~al.}(2006){Allgood}, {Flores}, {Primack}, {Kravtsov},
1269:   {Wechsler}, {Faltenbacher}, \& {Bullock}}]{2006MNRAS.367.1781A}
1270: {Allgood}, B., {Flores}, R.~A., {Primack}, J.~R., {Kravtsov}, A.~V.,
1271:   {Wechsler}, R.~H., {Faltenbacher}, A., \& {Bullock}, J.~S. 2006, \mnras, 367,
1272:   1781
1273: 
1274: \bibitem[{{Altay} {et~al.}(2006){Altay}, {Colberg}, \&
1275:   {Croft}}]{2006MNRAS.370.1422A}
1276: {Altay}, G., {Colberg}, J.~M., \& {Croft}, R.~A.~C. 2006, \mnras, 370, 1422
1277: 
1278: \bibitem[{{Arag{\'o}n-Calvo} {et~al.}(2007){Arag{\'o}n-Calvo}, {van de
1279:   Weygaert}, {Jones}, \& {van der Hulst}}]{2007ApJ...655L...5A}
1280: {Arag{\'o}n-Calvo}, M.~A., {van de Weygaert}, R., {Jones}, B.~J.~T., \& {van
1281:   der Hulst}, J.~M. 2007, \apjl, 655, L5
1282: 
1283: \bibitem[{{Azzaro} {et~al.}(2007){Azzaro}, {Patiri}, {Prada}, \&
1284:   {Zentner}}]{2007MNRAS.376L..43A}
1285: {Azzaro}, M., {Patiri}, S.~G., {Prada}, F., \& {Zentner}, A.~R. 2007, \mnras,
1286:   376, L43
1287: 
1288: \bibitem[{{Bailin} \& {Steinmetz}(2005)}]{2005ApJ...627..647B}
1289: {Bailin}, J. \& {Steinmetz}, M. 2005, \apj, 627, 647
1290: 
1291: \bibitem[Bailin et al.(2007)]{2007arXiv0706.1350B} Bailin, J., Power, C., 
1292: Norberg, P., Zaritsky, D., \& Gibson, B.~K.\ 2007, arXiv:0706.1350 
1293: 
1294: \bibitem[{{Basilakos} {et~al.}(2006){Basilakos}, {Plionis}, {Yepes},
1295:   {Gottl{\"o}ber}, \& {Turchaninov}}]{2006MNRAS.365..539B}
1296: {Basilakos}, S., {Plionis}, M., {Yepes}, G., {Gottl{\"o}ber}, S., \&
1297:   {Turchaninov}, V. 2006, \mnras, 365, 539
1298: 
1299: \bibitem[{{Binggeli}(1982)}]{1982A&A...107..338B}
1300: {Binggeli}, B. 1982, \aap, 107, 338
1301: 
1302: \bibitem[{{Binney} \& {Tremaine}(1987)}]{1987gady.book.....B}
1303: {Binney}, J. \& {Tremaine}, S. 1987, {Galactic dynamics} (Princeton:
1304:   Princeton University Press, p747.)
1305: 
1306: \bibitem[{{Brainerd}(2005)}]{2005ApJ...628L.101B}
1307: {Brainerd}, T.~G. 2005, \apjl, 628, L101
1308: 
1309: \bibitem[{{Bullock}(2002)}]{2002sgdh.conf..109B}
1310: {Bullock}, J.~S. 2002, in ''The Shapes of Galaxies and Their
1311:   Dark Matter Halos'', eds. P. Natarajan (Singapore: World Scientific)
1312:   p.109
1313: 
1314: \bibitem[{{Cappellari} {et~al.}(2006){Cappellari}, {Bacon}, {Bureau}, {Damen},
1315:   {Davies}, {de Zeeuw}, {Emsellem}, {Falc{\'o}n-Barroso}, {Krajnovi{\'c}},
1316:   {Kuntschner}, {McDermid}, {Peletier}, {Sarzi}, {van den Bosch}, \& {van de
1317:   Ven}}]{2006MNRAS.366.1126C}
1318: {Cappellari}, M., {Bacon}, R., {Bureau}, M., {Damen}, M.~C., {Davies}, R.~L.,
1319:   {de Zeeuw}, P.~T., {Emsellem}, E., {Falc{\'o}n-Barroso}, J., {Krajnovi{\'c}},
1320:   D., {Kuntschner}, H., {McDermid}, R.~M., {Peletier}, R.~F., {Sarzi}, M., {van
1321:   den Bosch}, R.~C.~E., \& {van de Ven}, G. 2006, \mnras, 366, 1126
1322: 
1323: \bibitem[{{Carter} \& {Metcalfe}(1980)}]{1980MNRAS.191..325C}
1324: {Carter}, D. \& {Metcalfe}, N. 1980, \mnras, 191, 325
1325: 
1326: \bibitem[{{Catelan} {et~al.}(2001){Catelan}, {Kamionkowski}, \&
1327:   {Blandford}}]{2001MNRAS.320L...7C}
1328: {Catelan}, P., {Kamionkowski}, M., \& {Blandford}, R.~D. 2001, \mnras, 320, L7
1329: 
1330: \bibitem[{{Chambers} {et~al.}(2002){Chambers}, {Melott}, \&
1331:   {Miller}}]{2002ApJ...565..849C}
1332: {Chambers}, S.~W., {Melott}, A.~L., \& {Miller}, C.~J. 2002, \apj, 565, 849
1333: 
1334: \bibitem[{{Ciotti} \& {Dutta}(1994)}]{1994MNRAS.270..390C}
1335: {Ciotti}, L. \& {Dutta}, S.~N. 1994, \mnras, 270, 390
1336: 
1337: \bibitem[{{Col{\'{\i}}n} {et~al.}(2000){Col{\'{\i}}n}, {Klypin}, \&
1338:   {Kravtsov}}]{2000ApJ...539..561C}
1339: {Col{\'{\i}}n}, P., {Klypin}, A.~A., \& {Kravtsov}, A.~V. 2000, \apj, 539, 561
1340: 
1341: \bibitem[{{Colless} {et~al.}(2001){Colless}, {Dalton}, {Maddox}, {Sutherland},
1342:   {Norberg}, {Cole}, {Bland-Hawthorn}, {Bridges}, {Cannon}, {Collins}, {Couch},
1343:   {Cross}, {Deeley}, {De Propris}, {Driver}, {Efstathiou}, {Ellis}, {Frenk},
1344:   {Glazebrook}, {Jackson}, {Lahav}, {Lewis}, {Lumsden}, {Madgwick}, {Peacock},
1345:   {Peterson}, {Price}, {Seaborne}, \& {Taylor}}]{2001MNRAS.328.1039C}
1346: {Colless}, M., {Dalton}, G., {Maddox}, S., {Sutherland}, W., {Norberg}, P.,
1347:   {Cole}, S., {Bland-Hawthorn}, J., {Bridges}, T., {Cannon}, R., {Collins}, C.,
1348:   {Couch}, W., {Cross}, N., {Deeley}, K., {De Propris}, R., {Driver}, S.~P.,
1349:   {Efstathiou}, G., {Ellis}, R.~S., {Frenk}, C.~S., {Glazebrook}, K.,
1350:   {Jackson}, C., {Lahav}, O., {Lewis}, I., {Lumsden}, S., {Madgwick}, D.,
1351:   {Peacock}, J.~A., {Peterson}, B.~A., {Price}, I., {Seaborne}, M., \&
1352:   {Taylor}, K. 2001, \mnras, 328, 1039
1353: 
1354: \bibitem[{{Crittenden} {et~al.}(2001){Crittenden}, {Natarajan}, {Pen}, \&
1355:   {Theuns}}]{2001ApJ...559..552C}
1356: {Crittenden}, R.~G., {Natarajan}, P., {Pen}, U.-L., \& {Theuns}, T. 2001, \apj,
1357:   559, 552
1358: 
1359: \bibitem[{{Croft} \& {Metzler}(2000)}]{2000ApJ...545..561C}
1360: {Croft}, R.~A.~C. \& {Metzler}, C.~A. 2000, \apj, 545, 561
1361: 
1362: \bibitem[{{Davis} {et~al.}(1985){Davis}, {Efstathiou}, {Frenk}, \&
1363:   {White}}]{1985ApJ...292..371D}
1364: {Davis}, M., {Efstathiou}, G., {Frenk}, C.~S., \& {White}, S.~D.~M. 1985, \apj,
1365:   292, 371
1366: 
1367: \bibitem[{{Dekel}(1985)}]{1985ApJ...298..461D}
1368: {Dekel}, A. 1985, \apj, 298, 461
1369: 
1370: \bibitem[{{Diemand} {et~al.}(2004){Diemand}, {Moore}, \&
1371:   {Stadel}}]{2004MNRAS.352..535D}
1372: {Diemand}, J., {Moore}, B., \& {Stadel}, J. 2004, \mnras, 352, 535
1373: 
1374: \bibitem[{{Diemand} {et~al.}(2007){Diemand}, {Kuhlen}, \&
1375:   {Madau}}]{2007astro.ph..3337D}
1376: {Diemand}, J., {Kuhlen}, M., \& {Madau}, P. 2007, arXiv:0705.2037v2 [astro-ph]
1377: 
1378: \bibitem[{{Faltenbacher} {et~al.}(2002){Faltenbacher}, {Gottl{\"o}ber},
1379:   {Kerscher}, \& {M{\"u}ller}}]{2002A&A...395....1F}
1380: {Faltenbacher}, A., {Gottl{\"o}ber}, S., {Kerscher}, M., \& {M{\"u}ller}, V.
1381:   2002, \aap, 395, 1
1382: 
1383: \bibitem[{{Faltenbacher} {et~al.}(2005){Faltenbacher}, {Allgood},
1384:   {Gottl{\"o}ber}, {Yepes}, \& {Hoffman}}]{2005MNRAS.362.1099F}
1385: {Faltenbacher}, A., {Allgood}, B., {Gottl{\"o}ber}, S., {Yepes}, G., \&
1386:   {Hoffman}, Y. 2005, \mnras, 362, 1099
1387: 
1388: \bibitem[{{Faltenbacher} \& {Diemand}(2006)}]{2006MNRAS.369.1698F}
1389: {Faltenbacher}, A. \& {Diemand}, J. 2006, \mnras, 369, 1698
1390: 
1391: \bibitem[{{Faltenbacher} \& {Mathews}(2007)}]{2007MNRAS.375..313F}
1392: {Faltenbacher}, A. \& {Mathews}, W.~G. 2007, \mnras, 375, 313
1393: 
1394: \bibitem[Faltenbacher et al.(2007)]{2007ApJ...662L..71F} Faltenbacher, A., 
1395: Li, C., Mao, S., van den Bosch, F.~C., Yang, X., Jing, Y.~P., Pasquali, A., 
1396: \& Mo, H.~J.\ 2007, \apjl, 662, L71 
1397: 
1398: \bibitem[{{Ghigna} {et~al.}(1998){Ghigna}, {Moore}, {Governato}, {Lake},
1399:   {Quinn}, \& {Stadel}}]{1998MNRAS.300..146G}
1400: {Ghigna}, S., {Moore}, B., {Governato}, F., {Lake}, G., {Quinn}, T., \&
1401:   {Stadel}, J. 1998, \mnras, 300, 146
1402: 
1403: \bibitem[{{Hahn} {et~al.}(2007){Hahn}, {Carollo}, {Porciani}, \&
1404:   {Dekel}}]{2007arXiv0704.2595H}
1405: {Hahn}, O., {Carollo}, C.~M., {Porciani}, C., \& {Dekel}, A. 2007, arXiv:0704.2595v1 [astro-ph]
1406: 
1407: 
1408: \bibitem[{{Hawley} \& {Peebles}(1975)}]{1975AJ.....80..477H}
1409: {Hawley}, D.~L. \& {Peebles}, P.~J.~E. 1975, \aj, 80, 477
1410: 
1411: \bibitem[{{Heavens} {et~al.}(2000){Heavens}, {Refregier}, \&
1412:   {Heymans}}]{2000MNRAS.319..649H}
1413: {Heavens}, A., {Refregier}, A., \& {Heymans}, C. 2000, \mnras, 319, 649
1414: 
1415: \bibitem[{{Heymans} \& {Heavens}(2003)}]{2003MNRAS.339..711H}
1416: {Heymans}, C. \& {Heavens}, A. 2003, \mnras, 339, 711
1417: 
1418: \bibitem[{{Holmberg}(1969)}]{1969ArA.....5..305H}
1419: {Holmberg}, E. 1969, Arkiv for Astronomi, 5, 305
1420: 
1421: \bibitem[{{Hopkins} {et~al.}(2005){Hopkins}, {Bahcall}, \&
1422:   {Bode}}]{2005ApJ...618....1H}
1423: {Hopkins}, P.~F., {Bahcall}, N.~A., \& {Bode}, P. 2005, \apj, 618, 1
1424: 
1425: \bibitem[{{Jing}(2002)}]{2002MNRAS.335L..89J}
1426: {Jing}, Y.~P. 2002, \mnras, 335, L89
1427: 
1428: \bibitem[{{Jing} \& {Suto}(2002)}]{2002ApJ...574..538J}
1429: {Jing}, Y.~P. \& {Suto}, Y. 2002, \apj, 574, 538
1430: 
1431: \bibitem[{{Kang} {et~al.}(2005){Kang}, {Mao}, {Gao}, \&
1432:   {Jing}}]{2005A&A...437..383K}
1433: {Kang}, X., {Mao}, S., {Gao}, L., \& {Jing}, Y.~P. 2005, \aap, 437, 383
1434: 
1435: \bibitem[Kang et al.(2007)]{2007MNRAS.378.1531K} Kang, X., van den Bosch, 
1436: F.~C., Yang, X., Mao, S., Mo, H.~J., Li, C., \& Jing, Y.~P.\ 2007, \mnras, 
1437: 378, 1531 
1438: 
1439: \bibitem[{{Kasun} \& {Evrard}(2005)}]{2005ApJ...629..781K}
1440: {Kasun}, S.~F. \& {Evrard}, A.~E. 2005, \apj, 629, 781
1441: 
1442: \bibitem[{{King} \& {Schneider}(2002)}]{2002A&A...396..411K}
1443: {King}, L. \& {Schneider}, P. 2002, \aap, 396, 411
1444: 
1445: \bibitem[Koch \& Grebel(2006)]{2006AJ....131.1405K} Koch, A., \& Grebel, 
1446: E.~K.\ 2006, \aj, 131, 1405 
1447: 
1448: \bibitem[Knebe et al.(2004)]{2004ApJ...603....7K} Knebe, A., Gill, 
1449: S.~P.~D., Gibson, B.~K., Lewis, G.~F., Ibata, R.~A., \& Dopita, M.~A.\ 
1450: 2004, \apj, 603, 7 
1451: 
1452: \bibitem[{{Kochanek}(2002)}]{2002sgdh.conf...62K}
1453: {Kochanek}, C.~S. 2002, in ''The Shapes of Galaxies and Their
1454:   Dark Matter Halos'', Eds. P. Natarajan (Singapore: World Scientific)
1455:   p. 62
1456: 
1457: \bibitem[{{Kroupa} {et~al.}(2005){Kroupa}, {Theis}, \&
1458:   {Boily}}]{2005A&A...431..517K}
1459: {Kroupa}, P., {Theis}, C., \& {Boily}, C.~M. 2005, \aap, 431, 517
1460: 
1461: \bibitem[{{Kuhlen} {et~al.}(2007){Kuhlen}, {Diemand}, \&
1462:   {Madau}}]{2007arXiv0705.2037K}
1463: {Kuhlen}, M., {Diemand}, J., \& {Madau}, P. 2007, arXiv:0705.2037v2 [astro-ph] 
1464: 
1465: \bibitem[{{Lee} {et~al.}(2005){Lee}, {Kang}, \& {Jing}}]{2005ApJ...629L...5L}
1466: {Lee}, J., {Kang}, X., \& {Jing}, Y.~P. 2005, \apjl, 629, L5
1467: 
1468: \bibitem[{{Libeskind} {et~al.}(2005){Libeskind}, {Frenk}, {Cole}, {Helly},
1469:   {Jenkins}, {Navarro}, \& {Power}}]{2005MNRAS.363..146L}
1470: {Libeskind}, N.~I., {Frenk}, C.~S., {Cole}, S., {Helly}, J.~C., {Jenkins}, A.,
1471:   {Navarro}, J.~F., \& {Power}, C. 2005, \mnras, 363, 146
1472: 
1473: \bibitem[Libeskind et al.(2007)]{2007MNRAS.374...16L} Libeskind, N.~I., 
1474: Cole, S., Frenk, C.~S., Okamoto, T., \& Jenkins, A.\ 2007, \mnras, 374, 16 
1475: 
1476: \bibitem[{{Lynden-Bell}(1982)}]{1982Obs...102..202L}
1477: {Lynden-Bell}, D. 1982, The Observatory, 102, 202
1478: 
1479: \bibitem[{{Macci{\`o}} {et~al.}(2006){Macci{\`o}}, {Moore}, {Stadel}, \&
1480:   {Diemand}}]{2006MNRAS.366.1529M}
1481: {Macci{\`o}}, A.~V., {Moore}, B., {Stadel}, J., \& {Diemand}, J. 2006, \mnras,
1482:   366, 1529
1483: 
1484: \bibitem[{{MacGillivray} {et~al.}(1982){MacGillivray}, {Dodd}, {McNally}, \&
1485:   {Corwin}}]{1982MNRAS.198..605M}
1486: {MacGillivray}, H.~T., {Dodd}, R.~J., {McNally}, B.~V., \& {Corwin}, Jr., H.~G.
1487:   1982, \mnras, 198, 605
1488: 
1489: \bibitem[{{Majewski}(1994)}]{1994ApJ...431L..17M}
1490: {Majewski}, S.~R. 1994, \apjl, 431, L17
1491: 
1492: \bibitem[{{Onuora} \& {Thomas}(2000)}]{2000MNRAS.319..614O}
1493: {Onuora}, L.~I. \& {Thomas}, P.~A. 2000, \mnras, 319, 614
1494: 
1495: \bibitem[{{Pen} {et~al.}(2000){Pen}, {Lee}, \& {Seljak}}]{2000ApJ...543L.107P}
1496: {Pen}, U.-L., {Lee}, J., \& {Seljak}, U. 2000, \apjl, 543, L107
1497: 
1498: \bibitem[{{Pereira} \& {Kuhn}(2005)}]{2005ApJ...627L..21P}
1499: {Pereira}, M.~J. \& {Kuhn}, J.~R. 2005, \apjl, 627, L21
1500: 
1501: \bibitem[Pereira et al.(2007)]{2007arXiv0707.1702P} Pereira, M.~J., Bryan, 
1502: G.~L., \& Gill, S.~P.~D.\ 2007, arXiv:0707.1702 
1503: 
1504: \bibitem[{{Plionis}(1994)}]{1994ApJS...95..401P}
1505: {Plionis}, M. 1994, \apjs, 95, 401
1506: 
1507: \bibitem[{{Plionis} \& {Basilakos}(2002)}]{2002MNRAS.329L..47P}
1508: {Plionis}, M. \& {Basilakos}, S. 2002, \mnras, 329, L47
1509: 
1510: \bibitem[{{Plionis} {et~al.}(2003){Plionis}, {Benoist}, {Maurogordato},
1511:   {Ferrari}, \& {Basilakos}}]{2003ApJ...594..144P}
1512: {Plionis}, M., {Benoist}, C., {Maurogordato}, S., {Ferrari}, C., \&
1513:   {Basilakos}, S. 2003, \apj, 594, 144
1514: 
1515: \bibitem[{{Porciani} {et~al.}(2002){Porciani}, {Dekel}, \&
1516:   {Hoffman}}]{2002MNRAS.332..339P}
1517: {Porciani}, C., {Dekel}, A., \& {Hoffman}, Y. 2002, \mnras, 332, 339
1518: 
1519: \bibitem[{{Rhee} \& {Latour}(1991)}]{1991A&A...243...38R}
1520: {Rhee}, G.~F.~R.~N. \& {Latour}, H.~J. 1991, \aap, 243, 38
1521: 
1522: \bibitem[{{Sharp} {et~al.}(1979){Sharp}, {Lin}, \&
1523:   {White}}]{1979MNRAS.187..287S}
1524: {Sharp}, N.~A., {Lin}, D.~N.~C., \& {White}, S.~D.~M. 1979, \mnras, 187, 287
1525: 
1526: \bibitem[{{Stadel}(2001)}]{2001PhDT........21S}
1527: {Stadel}, J.~G. 2001, PhD thesis, unpublished (University of Washington)
1528: 
1529: \bibitem[{{Statler}(1987)}]{1987ApJ...321..113S}
1530: {Statler}, T.~S. 1987, \apj, 321, 113
1531: 
1532: \bibitem[{{Struble}(1990)}]{1990AJ.....99..743S}
1533: {Struble}, M.~F. 1990, \aj, 99, 743
1534: 
1535: \bibitem[{{Tormen}(1997)}]{1997MNRAS.290..411T}
1536: {Tormen}, G. 1997, \mnras, 290, 411
1537: 
1538: \bibitem[{{Ulmer} {et~al.}(1989){Ulmer}, {McMillan}, \&
1539:   {Kowalski}}]{1989ApJ...338..711U}
1540: {Ulmer}, M.~P., {McMillan}, S.~L.~W., \& {Kowalski}, M.~P. 1989, \apj, 338, 711
1541: 
1542: \bibitem[{{van den Bosch} {et~al.}(1999){van den Bosch}, {Lewis}, {Lake}, \&
1543:   {Stadel}}]{1999ApJ...515...50V}
1544: {van den Bosch}, F.~C., {Lewis}, G.~F., {Lake}, G., \& {Stadel}, J. 1999, \apj,
1545:   515, 50
1546: 
1547: \bibitem[{{Wang} {et~al.}(2005){Wang}, {Jing}, {Mao}, \&
1548:   {Kang}}]{2005MNRAS.364..424W}
1549: {Wang}, H.~Y., {Jing}, Y.~P., {Mao}, S., \& {Kang}, X. 2005, \mnras, 364, 424
1550: 
1551: \bibitem[{{West}(1989)}]{1989ApJ...344..535W}
1552: {West}, M.~J. 1989, \apj, 344, 535
1553: 
1554: \bibitem[{{Yang} {et~al.}(2005){Yang}, {Mo}, {van den Bosch}, \&
1555:   {Jing}}]{2005MNRAS.356.1293Y}
1556: {Yang}, X., {Mo}, H.~J., {van den Bosch}, F.~C., \& {Jing}, Y.~P. 2005, \mnras,
1557:   356, 1293
1558: 
1559: \bibitem[{{Yang} {et~al.}(2006){Yang}, {van den Bosch}, {Mo}, {Mao}, {Kang},
1560:   {Weinmann}, {Guo}, \& {Jing}}]{2006MNRAS.369.1293Y}
1561: {Yang}, X., {van den Bosch}, F.~C., {Mo}, H.~J., {Mao}, S., {Kang}, X.,
1562:   {Weinmann}, S.~M., {Guo}, Y., \& {Jing}, Y.~P. 2006, \mnras, 369, 1293
1563: 
1564: \bibitem[{{York} {et~al.}(2000)}]{2000AJ....120.1579Y}
1565: {York, D.~G., et~al.} 2000, \aj, 120, 1579
1566: 
1567: \bibitem[{{Zaritsky} {et~al.}(1997){Zaritsky}, {Smith}, {Frenk}, \&
1568:   {White}}]{1997ApJ...478L..53Z}
1569: {Zaritsky}, D., {Smith}, R., {Frenk}, C.~S., \& {White}, S.~D.~M. 1997, \apjl,
1570:   478, L53+
1571: 
1572: \bibitem[{{Zentner} {et~al.}(2005){Zentner}, {Kravtsov}, {Gnedin}, \&
1573:   {Klypin}}]{2005ApJ...629..219Z}
1574: {Zentner}, A.~R., {Kravtsov}, A.~V., {Gnedin}, O.~Y., \& {Klypin}, A.~A. 2005,
1575:   \apj, 629, 219
1576: 
1577: \end{thebibliography}
1578: 
1579: %%
1580: \end{document}
1581: % LocalWords:  ApJ Faltenbacher al Jing Li Shude Mo Pasquali Institut fuer SK
1582: % LocalWords:  Astrophysik Nandan Jodrell Macclesfield DL UK nigstuhl virial MW
1583: % LocalWords:  GCS overdense onigstuhl BCGs BCG dFGRS SDSS Holmberg subhalo pre
1584: % LocalWords:  subhalos triaxial analyse CDM Zeldovich PPPM LCDMa FoF GCSs eps
1585: % LocalWords:  overdensity rms NFW ij ki kj arXiv resimulation infall peri apo
1586: % LocalWords:  labelled virialized NSFC KJCX YW HJM AISR IIS natexlab Adelman
1587: % LocalWords:  Agustsson Brainerd ph Allgood Primack Kravtsov Wechsler Altay
1588: % LocalWords:  Colberg Arag Calvo Weygaert der Hulst Azzaro Patiri Prada Bailin
1589: % LocalWords:  Zentner Norberg Zaritsky Basilakos Plionis Yepes Gottl ber Damen
1590: % LocalWords:  Turchaninov Binggeli Binney Natarajan Cappellari Zeeuw Emsellem
1591: % LocalWords:  Falc Barroso Krajnovi Kuntschner McDermid Peletier Sarzi
1592: