0801.4453/ms.tex
1: %\documentclass[12pt,preprint]{aastex}
2: 
3: %% manuscript produces a one-column, double-spaced document:
4: % \documentclass[manuscript]{aastex}
5: 
6: %% preprint2 produces a double-column, single-spaced document:
7: % \documentclass[preprint2]{aastex}
8: 
9:  \documentstyle[emulateapj,apjfonts,psfig]{article}
10: 
11: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
12: %                         OWN MACROS
13: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
14: \newcommand{\vdag}{(v)^\dagger}
15: \newcommand{\myemail}{skywalker@galaxy.far.far.away}
16: \newcommand{\apm}{AP$^3$M}
17: \newcommand{\LCDM}{$\Lambda$CDM}
18: \newcommand{\LWDM}{$\Lambda$WDM}
19: \newcommand{\hkpc}{$h^{-1}{\ }{\rm kpc}$}
20: \newcommand{\hMpc}{$h^{-1}{\ }{\rm Mpc}$}
21: \newcommand{\hMsun}{$h^{-1}{\ }{\rm M_{\odot}}$}
22: \newcommand{\kms}{${\rm{\ }km{\ }s^{-1}}$}
23: \newcommand{\nbody}{$N$-body}
24: \newcommand{\Dvir}{$\Delta_{\rm vir}$}
25: \newcommand{\Rvir}{$R_{\rm vir}$}
26: \newcommand{\Rhalo}{$R$}
27: \newcommand{\Mvir}{$M_{\rm vir}$}
28: \newcommand{\Mhalo}{$M$}
29: \newcommand{\rtrunc}{$r_{\rm trunc}$}
30: \newcommand{\rtidal}{$r_{\rm tidal}$}
31: \newcommand{\zform}{$z_{\rm form}$}
32: \newcommand{\Lsat}{$\vec{L}_{\rm sat}$}
33: \newcommand{\Lhost}{$\vec{L}_{\rm host}$}
34: \newcommand{\Ehost}{$\vec{E}_{1, \rm host}$}
35: \newcommand{\Rapo}{$\vec{R}_{\rm sat}^{\rm apo}$}
36: \newcommand{\Lapo}{$\vec{L}_{\rm sat}^{\rm apo}$}
37: \newcommand{\vecx}{\vec{x}}
38: \newcommand{\vecp}{\vec{p}}
39: \newcommand{\vecr}{\vec{r}}
40: \newcommand{\vecF}{\vec{F}}
41: \newcommand{\vecg}{\vec{g}}
42: \newcommand{\ddx}{\ddot{\vec{x}}}
43: \newcommand{\ddr}{\ddot{{r}}}
44: \newcommand{\dx}{\dot{\vec{x}}}
45: \newcommand{\vecnabla}{\vec{\nabla}}
46: \newcommand{\dvecx}{\dot{\vec{x}}}
47: \newcommand{\ddvecx}{\ddot{\vec{x}}}
48: \newcommand{\ddvecr}{\ddot{\vec{r}}}
49: \newcommand{\REF}{\textbf{REFS}}
50: \newcommand{\Table}[1]{Table~\ref{#1}}
51: \newcommand{\Sec}[1]{Section~\ref{#1}}
52: \newcommand{\Eq}[1]{Eq.~(\ref{#1})}
53: \newcommand{\Fig}[1]{Fig.~\ref{#1}}
54: \newcommand{\mlapm}{\texttt{MLAPM}}
55: \newcommand{\mhf}{\texttt{MHF}}
56: \newcommand{\mht}{\texttt{MHT}}
57: \newcommand{\ea}{et~al.~}                            % \ea      =  et al.
58: \newcommand{\AN}[3]     {\mbox{AN~\textbf{#1},~#2~(#3)}}
59: \newcommand{\AAA}[3]    {\mbox{A\&A~\textbf{#1},~#2~(#3)}}
60: \newcommand{\AAR}[3]    {\mbox{A\&A~Rev.~\textbf{#1},~#2~(#3)}}
61: \newcommand{\AAS}[3]    {\mbox{A\&A~Suppl.~\textbf{#1},~#2~(#3)}}
62: \newcommand{\ApJ}[3]    {\mbox{ApJ~\textbf{#1},~#2~(#3)}}
63: \newcommand{\ApJS}[3]   {\mbox{ApJ~Suppl.~\textbf{#1},~#2~(#3)}}
64: \newcommand{\ApJL}[3]   {\mbox{ApJ~Lett.~\textbf{#1},~#2~(#3)}}
65: \newcommand{\ARAA}[3]   {\mbox{Ann.~Rev.~A~\&~A~\textbf{#1},~#2~(#3)}}
66: \newcommand{\AJ}[3]     {\mbox{Astron.~J.~\textbf{#1},~#2~(#3)}}
67: \newcommand{\MNRAS}[3]  {\mbox{MNRAS~\textbf{#1},~#2~(#3)}}
68: \newcommand{\Nature}[3] {\mbox{Nature~\textbf{#1},~#2~(#3)}}
69: \newcommand{\NewA}[3]   {\mbox{NewA~\textbf{#1},~#2~(#3)}}
70: \newcommand{\Science}[3]{\mbox{Science~\textbf{#1},~#2~(#3)}}
71: \newcommand{\PhRevL}[3] {\mbox{Phys.~Rev.~Lett.~\textbf{#1},~#2~(#3)}}
72: \newcommand{\PhRevD}[3] {\mbox{Phys.~Rev.~\textbf{D#1},~#2~(#3)}}
73: \newcommand{\astroph}[1]{\mbox{\texttt{astro-ph/#1}}}
74: 
75: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
76: %                         OWN MACROS
77: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
78: 
79: \begin{document}
80: 
81: \title{On the Correlation between Spin Parameter and Halo Mass}
82: 
83: \author{Alexander Knebe\altaffilmark{1}, 
84:         Chris Power\altaffilmark{2}}
85: 
86: \altaffiltext{1}{Astrophysikalisches Institut Potsdam,
87: 				An der Sternwarte 16,
88: 				14482 Potsdam,
89: 				Germany}
90: \altaffiltext{2}{Centre for Astrophysics \& Supercomputing,
91: 				Swinburne University,
92: 				Mail H31, PO Box 218,
93: 				Hawthorn, VIC 3122,
94: 				Australia}
95: 
96: 
97: \begin{abstract}
98:   We report on a correlation between virial mass $M$ and spin parameter
99:   $\lambda$ for dark matter halos forming at redshifts $z \gtrsim 10$.
100:   We find that the spin parameter decreases with increasing halo mass.
101:   Interestingly, our analysis indicates that halos forming at later
102:   times do not exhibit such a strong correlation, in agreement with 
103:   the findings of previous studies. We briefly discuss the implications 
104:   of this correlation for galaxy formation at high redshifts and the galaxy
105:   population we observe today.
106: \end{abstract}
107: 
108: \keywords{galaxies: formation --- cosmology: theory --- cosmology:
109:   early Universe --- methods: numerical}
110: 
111: \section{Introduction} \label{sec:introduction}
112: %%%%%%%%%%%%%%%%%%%%%%%%%%
113: 
114: The physical mechanism by which galaxies acquire their angular momentum
115: is an important problem that has been the subject of investigation for
116: nearly sixty years (Hoyle 1949). This reflects the fundamental role played
117: by angular momentum of galactic material in defining the size and 
118: shapes of galaxies (e.g. Fall \& Efstathiou 1981). Yet despite its
119: physical significance, a precise and accurate understanding of the
120: origin of galactic angular momentum remains one of the missing pieces 
121: in the galaxy formation puzzle.
122: 
123: A fundamental assumption in current galaxy formation models is that
124: galaxies form when gas cools and condenses within the potential wells
125: of dark matter halos (White \& Rees 1978). Consequently it is probable that 
126: the angular momentum of the galaxy will be linked to the angular
127: momentum of its dark matter halo (e.g. Fall \& Efstathiou 1980; Mo, Mao
128: \& White 1998; Zavala, Okamoto \& Frenk 2007). Within the context
129: of hierarchical structure formation models, the angular momentum growth 
130: of a dark matter proto-halo is driven by gravitational tidal torquing during 
131: the early stages (i.e. the linear regime) of its assembly. This ``Tidal Torque
132: Theory'' has been explored in detail; it is a well-developed analytic 
133: theory (e.g. Peebles 1969, Doroshkevich 1979, White 1984) and its 
134: predictions are in good agreement with the results of cosmological 
135: $N$-body simulations (e.g. Barnes \& Efstathiou 1987; Warren~\ea 1992; 
136: Sugerman, Summers \& Kamionkowski 2000; Porciani, Dekel \& Hoffman
137: 2002). However, once the proto-halo has passed through maximum
138: expansion and the collapse has become non-linear, tidal torquing
139: no longer provides an adequate description of the evolution of the 
140: angular momentum (White 1984), which tends to decrease with time. 
141: During this phase it is likely that merger and accretion 
142: events play an increasingly important role in determining both the
143: magnitude and direction of the angular momentum of a galaxy (e.g. Bailin \&
144: Steinmetz 2005). Indeed, a number of studies have argued that mergers
145: and accretion events are the primary determinants of the angular
146: momenta of galaxies at the present day (Gardner 2001; Maller, Dekel~\&
147: Somerville 2002; Vitvitska~\ea 2002).\\
148: 
149: It is common practice to quantify the angular momentum of a dark matter 
150: halo by the dimensionless ``classical'' spin parameter (Peebles 1969),
151: \begin{equation}
152:   \label{eq:lambda}
153:   \lambda = \frac{J \sqrt{|E|}}{GM^{5/2}},
154: \end{equation}
155: where $J$ is the magnitude of the angular momentum of material within
156: the virial radius, $M$ is the virial mass, and $E$ is the total energy 
157: of the system. It has been shown that halos that have suffered a recent 
158: major merger will tend to have a higher spin parameter $\lambda$ than 
159: the average (e.g. Hetznecker \& Burkert 2006; Power, Knebe \& Knollmann 
160: 2008). Therefore one could argue that within the framework of hierarchical
161: structure formation that higher mass halos should have larger spin 
162: parameters \emph{on average} than less massive systems because they 
163: have assembled a larger fraction of their mass (by merging) more
164: recently. 
165: 
166: However, if we consider only halos in virial equilibrium, should we
167: expect to see a correlation between halo mass and spin? One might
168: na\"ively expect that more massive systems will have had their maximum
169: expansion more recently and so these systems will have been tidally
170: torqued for longer than systems that had their maximum expansion at
171: earlier times. This suggests that spin should \emph{increase} with
172: timing of maximum expansion and therefore halo mass. However, one
173: finds at best a weak correlation between mass and spin for equilibrium
174: halos at $z$=0 (e.g. Cole~\& Lacey 1996; Maccio et al. 2007; Bett et
175: al. 2007, hereafter B07), and the correlation is for spin to
176: \emph{decrease} with increasing halo mass, contrary to our na\"ive
177: expectation.
178: 
179: In this paper, we report on a (weak) correlation between spin and
180: mass for equilibrium halos at redshift $z$=10. The trend is for 
181: higher-mass halos to have smaller spins, and is qualitatively
182: similar to the one reported by B07 for the halo
183: population at $z$=0. We present the main evidence in support of this
184: correlation in Section~\ref{sec:correlation} and we consider its
185: implications for galaxy formation in Section~\ref{sec:conclusions}.
186: 
187: \section{The Simulations} \label{sec:simulations}
188: %%%%%%%%%%%%%%%%%%%%%%%%%
189: 
190: %\input{tab1.tex}
191: 
192: For the simulations presented in this paper we have adopted the
193: cosmology as given by Spergel~\ea (2003) ($\Omega_0=0.3$,
194: $\Omega_{\Lambda}=0.7$, $\sigma_8=0.9$, and $H_0=70$km/sec/Mpc). Each
195: run employed $N=256^3$ particles and differed in simulation box-size
196: $L_{\rm box}$, which leads to the particle mass $m_p$ differing
197: between runs -- $m_p=\rho_{\rm crit} \Omega_0 (L_{\rm box}/N)^3$,
198: where $\rho_{\rm crit}=3H_0^2/8\pi\,G$. This allows us to probe a
199: range of halo masses at redshift $z$=10. The primary parameters of
200: these simulations are summarized in \Table{tab:simu}.
201: 
202: Halos in all runs have been identified using the MPI parallelized
203: version of the \texttt{AHF} halo finder\footnote{\texttt{AHF} is
204:   already freely available from
205:   \texttt{http://www.aip.de/People/aknebe}}
206: (\texttt{AMIGA}'s-Halo-Finder), which is based on the \texttt{MHF}
207: halo finder of Gill, Knebe \& Gibson (2004). For each halo we compute
208: the virial radius \Rhalo, defined as the radius at which the mean
209: interior density is \Dvir\ times the background density of the
210: Universe at that redshift. This leads to the following definition for
211: the virial mass \Mhalo:
212: 
213: \begin{equation} \label{eq:Mvir}
214:   M = \frac{4\pi}{3} \Delta_{\rm vir} \Omega \rho_{\rm crit} R^3 \ .
215: \end{equation}
216: 
217: \noindent Note that \Dvir\ is a function of redshift and amounts to
218: $\Delta_{\rm vir} \approx 210$ at redshift $z$=10 , $\Delta_{\rm vir}
219: \approx 230$ at $z$=1, and the ``usual'' $\Delta_{\rm vir} \approx
220: 340$ at $z$=0 (cf. Gross 1997). \Table{tab:simu} summarises the total
221: number of halos ($N_{\rm halos}$) recovered by \texttt{AHF}, while
222: $z_{\rm final}$ gives the redshift that the simulation has been
223: evolved to.
224: 
225: We add that we ran five realisations of B20 to redshift $z$=10 in
226: order to have a statistically significant sample of halos in that
227: particular model. However, we also note that the fitting parameters 
228: presented in the following Section are robust in the sense that they 
229: do not depend on whether we stack the halos from those five runs or 
230: use them individually.
231: 
232: 
233: \section{The Halo Sample} \label{sec:virialisation}
234: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
235: We show in a companion paper (Power, Knebe \& Knollmann 2008) that a
236: substantial fraction of the halo population at high redshift is not in
237: virial equilibrium. Because we wish to examine the spin
238: distribution of equilibrium halos, it is important to account for
239: unrelaxed systems when investigating correlations between spin and
240: halo mass. For example, it has been shown that the spin can increase
241: sharply in the aftermath of mergers with mass ratios as modest as 5:1
242: (e.g. Hetznecker \& Burkert 2006), and that the degree to which a halo
243: is in dynamical relaxation is as important as recent merging history in its
244: influence on spin (D'Onghia \& Navarro 2007). To ensure
245: that halos in our sample are in virial equilibrium, we compute the
246: virial ratio for each halo, which we define as
247: 
248: \begin{equation}
249:   \label{eq:virial}
250:   Q = \frac{2T+S_p}{U} + 1 \ .
251: \end{equation}
252: 
253: \noindent Here $T$ represents the kinetic energy, $U$ the potential energy,
254: and $S_p$ the surface pressure of a given halo of mass \Mhalo. By
255: including $S_p$, we can account for the effect of infalling material on
256: the dynamical state of the halo. Each of these quantities are evaluated
257: using all gravitationally bound particles, and we adopt the formula of 
258: Shaw~\ea (2006) for the surface pressure term~$S_p$ (cf. equations.(4)-(6) in 
259: their study).
260: 
261: In Figure~\ref{fig:QspMass} we show that the relation between halo
262: mass and $Q$ can vary with mass. This is apparent at redshift $z$=1, 
263: where we find a trend for more massive halos to be less virialised. 
264: In contrast, high redshift halos are less virialised on average 
265: (as indicated by the increased average $\langle Q \rangle\approx-0.3$), 
266: but we find no apparent trend with mass.
267: 
268: Why is there a mass dependence at $z$=1 but not at $z$=10? There are
269: two factors. The first is that high redshift halos ``see'' an effective 
270: slope of the initial power spectrum of $n_{\rm eff} \approx -3$, and so 
271: the time at which a particular mass scale starts to collapse is
272: relatively insensitive to mass. Therefore we do not expect to find a
273: strong correlation between virial state $Q$ and mass. We have 
274: checked this halo populations drawn from the simulations of scale-free 
275: cosmologies of Knollmann, Power \& Knebe (2008) and our interpretation 
276: is consistent with the correlations we find in these runs.
277: The second is that the typical collapsing mass $M^{*}$ at $z$=10 is
278: small -- of order $10^3 h^{-1} \rm M_{\odot}$ -- and because we resolve 
279: mass scales that have collapsed more or less simultaneously, we see a 
280: population that has yet to relax. At $z$=1, the typical collapsing mass 
281: is much larger -- of order $10^{11} h^{-1} \rm M_{\odot}$ -- and so we 
282: resolve a population of halos whose mass accretion histories are more diverse. 
283: The most massive systems tend to be ones that have formed most
284: recently, and are therefore the least dynamically relaxed.
285: 
286: %\input{tab2.tex}
287: 
288: We have used the following relation between $Q$ and $M$ to classify
289: dynamically relaxed and unrelaxed systems:
290: 
291: \begin{equation} \label{eq:QspMass}
292: \begin{array}{rcll}
293: Q_{\rm allowed} & \propto & M^{-0.015}         & \mbox{, for $z=1$}  \\ 
294: Q_{\rm allowed} & \propto & \mbox{const.}    & \mbox{, for $z=10$} \\ 
295: \end{array}
296: \end{equation}
297: 
298: \noindent We allow the $Q$ values of halos in our sample to deviate from these
299: scaling relations by not more than 
300: 
301: \begin{equation} \label{eq:Qlim}
302: Q_{\rm allowed}-Q_{\rm lim} \le Q \le Q_{\rm allowed}+Q_{\rm lim}
303: \end{equation}
304: 
305: \noindent with $Q_{\rm lim} = 0.15$ (indicated by the dashed lines in
306: \Fig{fig:QspMass}). Furthermore, we consider only halos that contain
307: at least $N_{\rm min}=600$ particles within their virial radius to
308: ensure that we are not influenced by particle discreteness. Interestingly, 
309: when computing spin, the tighest restriction on particle number comes not 
310: from the calculation of angular momentum but from the calculation of the 
311: potential energy. By comparing analytic solutions with Monte Carlo 
312: realisations of Navarro, Frenk \& White (1997) haloes, we find that at least
313: 600 particles are required if the energy is to be computed to better than 
314: $10\%$.
315: 
316: \section{The Spin-Mass Correlation} \label{sec:correlation}
317: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
318: Calculating the total energy $E$ of a halo is computationally expensive, and so
319: computing $\lambda$ using \Eq{eq:lambda} is also expensive. This prompted 
320: Bullock et al. (2001) to introduce a modified spin parameter
321: 
322: \begin{equation} \label{eq:spinTsp}
323:   \lambda' = \frac{J}{\sqrt{2}MVR} \ ,
324: \end{equation}
325: 
326: \noindent where $V=\sqrt{GM/R}$ measured the circular velocity at the
327: virial radius $R$ and $J$ represents the absolute value of the angular
328: momentum. We follow Bullock et al. and compute spin using
329: equation~(\ref{eq:spinTsp}).
330: 
331: In Figure~\ref{fig:SpinMass}, we investigate the correlation between
332: halo spin $\lambda'$ and mass $M$. We show only halos that fulfill our
333: selection criteria (individual dots) and bin the data in five mass
334: bins equally spaced in log-space between $M_{\rm min}$ and $M_{\rm
335:   max}$ of the considered halos at the respective redshift. The values
336: plotted as histograms thereby represent the weighted mean of all spin
337: parameters in the respective mass range where the weight is inversely
338: proportional to the error estimate
339: 
340: \begin{equation}
341:  \frac{\sigma_J}{J} = \frac{0.2}{\lambda' \sqrt{N}}
342: \end{equation}
343: 
344: \noindent
345: for the spin parameter of a halo consisting of $N$ particles as
346: derived in Bullock et al. (2001) (cf. equation~(7) in that study). The
347: error bars indicate the standard deviation of the spin parameter
348: values in the bin from the weighted mean.\footnote{We like to note in
349:   passing that we also performed all of the analysis and stability
350:   checks using the median and the scatter about the median in each
351:   bin. The results remain unaffected and we therefore decide to only
352:   list them for the weighted means.}
353: 
354: The best fitting power-laws to these histograms reveal that
355: 
356: \begin{equation}\label{eq:logfit}
357: \begin{array}{lcll}
358: \displaystyle \lambda'  & \propto & M^{\alpha} \\ 
359: \end{array}
360: \end{equation}
361: 
362: \noindent with
363: 
364: \begin{equation}\label{eq:logslopes}
365: \begin{array}{lcll}
366: \displaystyle \alpha  & = & -0.002 \pm 0.149 & \mbox{\rm ,\ for $z=1$ } \\
367: \displaystyle \alpha  & = & -0.059 \pm 0.171 & \mbox{\rm ,\ for $z=10$ } .\\ 
368: \end{array}
369: \end{equation}
370: 
371: 
372: This indicates that there is a \emph{weak} correlation at high
373: redshifts for spin to decrease with increasing mass, albeit stronger than the
374: one at $z$=1. We compute Spearman rank correlation coefficients at $z$=10 (1) 
375: and find $R_s=-0.137 (0.06)$.
376: 
377: As an alternative approach, we fit a lognormal function to each of
378: our halo samples at $z$=10 and $z$=1,
379: 
380: \begin{equation} \label{lognormal}
381:  P(\lambda') = \displaystyle \frac{1}{\lambda' \sqrt{2\pi} \ \sigma_0}
382:               \exp \left( {-\frac{\ln^2 (\lambda'/\lambda'_0)}{2 \sigma_0^2}} \right).
383: \end{equation}
384: 
385: \noindent The resulting curves are presented in \Fig{fig:Pspin}
386: whereas the best-fit parameters, median values for $\lambda'_{\rm med}
387: = {\rm median}(\lambda')$ and median halo masses $M_{\rm med}$ are
388: given in \Table{tab:lambdaTsp}. Inspection of the best-fit parameters
389: confirm that the median spin declines as we move from less massive to
390: more massive objects at high redshift.
391: 
392: \section{Stability of Results} \label{sec:stability}
393: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
394: Because of the weak nature of the measured correlation it is
395: vitally important to check its credibility by performing a statistical
396: analysis. To this extent we investigate the sensitivity of the
397: logarithmic slope $\alpha$ with respect to a number of parameters that
398: enter into its determination, namely the number of bins $N_{\rm bins}$ 
399: used for the histograms; the virialisation criterion
400: parametrized via $Q_{\rm lim}$; and the minimum number of particles
401: $N_{\rm min}$ within a halo's virial radius. Note that we vary one
402: parameter at a time, keeping the others at their fixed ``standard'' values.  
403: The results are presented in \Table{tab:Nbins}, \Table{tab:Qlim}, and
404: \ref{tab:Nmin}.
405: 
406: We find that bin number has practically no effect on the
407: slope. Similarly we find that varying the virialisation criterion
408: $Q_{\rm lim}$ has little effect on the slope of the relation between
409: mass and spin, regardless of redshift (\Table{tab:Qlim}). In contrast,
410: we find that the minimum number of particles within a halo's virial
411: radius has a strong and systematic effect on the result at $z$=1 -- as
412: $N_{\rm min}$ increases, we find that the logarithmic slope becomes
413: shallower. This does not appear to be true at high redshift, although
414: as we go to earlier times we find that the number of massive haloes
415: becomes progressively smaller and our determination of $\alpha$
416: becomes increasingly unreliable.
417: 
418: These tests lead us to believe that our results are both stable and reliable, 
419: and our main result holds: \emph{the correlation between spin paramater 
420: $\lambda'$ and halo mass $M$ is one order of magnitude larger at redshift
421: $z$=10 than at $z$=1.}\\
422: 
423: As a further test of the credibility of the correlation we measure between 
424: halo mass and spin, we use the criteria of three other studies to select our
425:  halo sample. These are:
426: 
427: \begin{itemize}
428: 
429: \item Maccio et al. (2007) criteria:
430:   
431:   \begin{itemize}
432:   \item $N_{\rm min}=250$
433:   \item $x_{\rm off} < 0.04$
434:   \item $\rho_{\rm rms} < 0.4$
435:   \end{itemize}
436: 
437: \item Bett et al. (2007) criteria:
438:   
439:   \begin{itemize}
440:   \item $N_{\rm min}=300$
441:   \item $Q_{\rm lim}=0.5$
442:   \end{itemize}
443: 
444: \item Neto et al. (2007) criteria:
445:   
446:   \begin{itemize}
447:   \item $N_{\rm min}=600$
448:   \item $x_{\rm off} < 0.07$
449:   \item $f_{\rm sub} < 0.1$
450:   \end{itemize}
451: 
452: \end{itemize}
453: 
454: Here $N_{\rm min}$ is again the minimum number of particles in a halo,
455: $Q_{\rm lim}$ the virial limit as defined in \Eq{eq:Qlim}, $x_{\rm
456:   off}$ measures the distance between the most bound particle and the
457: centre of mass in units of the virial radius, $\rho_{\rm rms}$ is an
458: indicator of how well the density profile of the halo can be fitted by
459: a Navarro, Frenk~\& White (1997) profile (cf. Eq.~(2) in Maccio et
460: al. 2007), and $f_{\rm sub}$ is the fraction of mass in subhalos. The
461: resulting power law slopes $\alpha$ (cf. \Eq{eq:logfit}) for $z=1$ and
462: $z=10$ are presented in \Table{tab:EtAl}. Again we note that there
463: appears to be a much stronger correlation between $\lambda'$ and $M$
464: at higher redshift. While the relation is consistent with zero at
465: $z$=1 (as confirmed by Maccio et al. 2007) spin and mass are
466: correlated at $z$=10. How this result relates to the Bett et al. (2007) 
467: result -- who find a weak correlation at $z$=0 -- will be discussed in the
468: following Section.
469: 
470: 
471: \section{Conclusions and Discussion} \label{sec:conclusions}
472: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
473: We have performed a careful investigation of the relation between
474: virial mass and dimensionless spin parameter for dark matter halos
475: forming at high redshifts $z \gtrsim 10$ in a $\Lambda$CDM cosmology.
476: The result of our study, which is based on a series of cosmological
477: $N$-body simulations in which box size was varied while keeping
478: particle number fixed, indicates that there is a \emph{weak}
479: correlation between mass and spin at $z$=10, such that the spin
480: decreases with increasing mass. If there is a correlation
481: at $z$=1, we argue that it is significantly weaker than the one we find
482: at $z$=10; this is in qualitative agreement with the findings of
483: previous studies that focused on lower redshifts 
484: (Maccio~\ea 2007, Shaw~\ea 2005, Lemson~\& Kauffmann 1999).\\ 
485: 
486: Interestingly, B07 find a weak correlation between median spin and
487: halo mass at $z$=0 in the Millennium Simulation (Springel et
488: al. 2005), in the same spirit as the one presented here for $z$=10:
489: lower mass halos tend to have higher spins. However, as we show, the
490: correlation between halo mass and spin is weaker at $z$=1 than at
491: $z$=10, whereas the correlation reported in B07 for halos at $z$=0 is
492: much stronger than the one we find at $z$=1. This is not what one
493: would expect, and so it is important to try and understand the source
494: of the difference between our result at $z$=1 and the B07 result at
495: $z$=0. B07 fitted a 3$^{\rm rd}$-order polynomial to the median spins
496: of halos in the mass range $3\times 10^{11} \lesssim M/(h^{-1}
497: M_{\odot}) \lesssim 3\times 10^{14}$ at $z$=0. The form of this
498: polynomial is extremely sensitive to the precise values of the
499: best-fit parameters (Bett, private communication) and it is not
500: straightforward to extrapolate its behaviour outside of the given mass
501: range and redshift. We derive our estimates of the power-law exponents
502: from the spin distribution with respect to halo mass at $z$=1. Our
503: halos lie in the mass range $3\times 10^{9} \lesssim M/(h^{-1}
504: M_{\odot}) \lesssim 5\times 10^{12}$. B07 base their median spins upon
505: $\sim 1.5$ million halos with correspondingly small errors, and note
506: that the weak nature of the trend of spin with mass makes it hard to
507: detect. This suggests to us that the correlation between mass and spin
508: at $z$=10 is remarkably strong rather than the correlation at $z$=1
509: being too weak!
510: 
511: When studying correlations between halo mass and spin, great care must
512: be taken in defining the halo sample. In particular, we find that mass
513: resolution (i.e. the number of particles with which a halo is
514: resolved) and the degree of virialisation of a halo can have a
515: significant effect on the strength of the correlation (at least at
516: $z$=1, cf. \Table{tab:Nmin}). This -- at least -- is in good agreement
517: with the findings of B07.\\
518: 
519: We note that Power \& Knebe (2006) demonstrated that the size of 
520: simulation box can lead to a suppression of angular momentum in 
521: smaller boxes, due to the absence of longer wavelength perturbations 
522: in the initial conditions. This will lead to a bias in our estimate of
523: $\lambda$ (approximately a $\sim 10\%$ effect) but we have verified
524: that the spin distributions we obtain from a B20 run truncated on
525: scales larger than the longest wavelength perturbation modelled in the
526: B1 run produces results that are consistent. Indeed, we would expect
527: the correlation to be strengthened if the B1 spins were corrected for
528: box size effects.\\
529: 
530: It is interesting to speculate on the consequences of this correlation for
531: galaxy formation at high redshifts and the galaxy population we
532: observe today. In the standard picture of galaxy formation, gas cools
533: on to dark matter halos and is shock heated to the virial temperature
534: of the halo. The angular momentum of the gas and the dark matter should
535: (initially) be similar because they are subject to the same tidal
536: field. As the innermost densest parts of the gaseous halo cool, they
537: will settle into a gaseous disk with a scale length determined by the
538: specific angular momentum of the gas, which we would expect to be
539: related to the angular momentum of the halo (e.g. Zavala, Okamoto \&
540: Frenk 2007).
541: 
542: If more massive halos at high redshifts show a tendency to have smaller spin
543: parameters, the gas disks will have lower specific angular momenta and 
544: therefore will be more centrally concentrated. If star formation rate 
545: correlates with surface density, then we might expect the 
546: star formation rate to be enhanced in more massive halos. Because massive halos
547: tend to form preferentially in high density, highly clustered environments in 
548: which the merger rate also tends to be enhanced, then we might expect star
549: formation to proceed more rapidly and at earlier times in these halos.
550: Might this explain the effect of ``downsizing'' (e.g. Cowie et al. 1996), 
551: the successive shifting of star formation from high- to low-mass galaxies 
552: with decreasing redshift? We shall pursue this in a more quantitative manner 
553: in future work.
554: 
555: \acknowledgments
556: %%%%%%%%%%%%%%%%
557: AK would like to thank Swinburne University for its hospitality where
558: this work was initiated. AK further acknowledges funding through the
559: Emmy Noether programme of the DFG (KN 755/1). CP acknowledges funding
560: through the ARC Discovery Projected funded ``Commonwealth Cosmology
561: Initiative'', grant DP 0665574. The simulations presented in this
562: paper were carried out on the Beowulf cluster at the Centre for
563: Astrophysics~\& Supercomputing, Swinburne University as well as the
564: Sanssouci cluster at the Astrophysikalisches Institut Potsdam.
565: 
566: \begin{thebibliography}{}
567: %%%%%%%%%%%%%%%%%%%%%%%%%
568: 
569: \bibitem[Barnes87]{Barnes87}
570:         {Barnes, J.; Efstathiou, G.; 1987, ApJ 319, 575}
571: 
572: \bibitem[Bett07]{Bett07}
573:         {Bett, P.; Eke, V.; Frenk, C.S.; Jenkins, A.; Helly, A.; Navarro, J.; 2007, \astroph{0608607}}
574: 
575: \bibitem[Bullock01]{Bullock01}
576:         {Bullock, J. S.; Dekel, A.; Kolatt, T. S.; Kravtsov, A. V.; Klypin, A. A.; Porciani, C.; Primack, J. R.; 2001, ApJ 555, 240}
577: 
578: \bibitem[Cole96]{Cole96}
579:         {Cole, S.; Lacey, C.; 1996, MNRAS 281, 726}
580: 
581: \bibitem[Cowie96]{Cowie96}
582:         {Cowie L.L., Songaila A., Hu E.M., Choen J.G., 1996, AJ 112, 839}
583: 
584: \bibitem[Doroshkevich79]{Doroshkevich79}
585:         {Doroshkevich, 1979}
586: 
587: \bibitem[Fall80]{Fall80}
588:         {Fall~S.~M. \& Efstathiou~G. 1980, MNRAS, 193, 189}
589: 
590: \bibitem[Gao07]{Gao07}
591:         {Gao, L.; White, S.D.M.; 2007, MNRAS 377, L5}
592: 
593: \bibitem[Gardner01]{Gardner01}
594:         {Gardner, J.P.; 2001, ApJ 557, 616}
595: 
596: \bibitem[Gill04]{Gill04}
597:         {Gill, S.P.D.; Knebe, A.; Gibson, B.K.; 2004, MNRAS 351, 399}
598: 
599: \bibitem[Gottloeber07]{Gottloeber07}
600:         {Gottloeber S., Yepes G., 2007, ApJ 664, 117}
601: 
602: \bibitem[Gross97]{Gross97}
603:         {Gross, M.A.K., 1997, PhD thesis Univ. California, Santa Cruz}
604: 
605: \bibitem[Hetznecker06]{Hetznecker06}
606:         {Hetznecker, H.; Burkert, A.; 2006, MNRAS 370, 1905}
607: 
608: \bibitem[Knollmann08]{Knollmann08}
609:         {Knollmann S., Power C., Knebe A., 2008, MNRAS accepted}
610: 
611: \bibitem[Lemson99]{Lemson99}
612:         {Lemson, G.; Kauffmann, G.; 1999, MNRAS 302, 111}
613: 
614: \bibitem[Maccio07]{Maccio07}
615:         {Maccio, A.V.; Dutton, A.A.; van den Bosch, F.C.; Moore, B.; Potter, D.; Stadel, J.; 2007, MNRAS 378, 55}
616: 
617: \bibitem[Maller02]{Maller02}
618:         {Maller, A.H.; Dekel, A.; Somerville, R.; 2002, MNRAS 329, 423}
619: 
620: \bibitem[Mo98]{Mo98}
621:         {Mo, H.J.; Mao, S.; White, S.D.M.; 1998, MNRAS 295, 319}
622: 
623: \bibitem[NFW97]{NFW97}
624:         {Navarro J.F., Frenk C.S., White S.D.M., 1997, ApJ 490, 493}
625: 
626: \bibitem[Neto07]{Neto07}
627:         {Neto A.F., et al., \astroph{0706.2919}}
628: 
629: \bibitem[Peebles69]{Peebles69}
630:         {Peebles J.; 1969}
631: 
632: \bibitem[Power08]{Power08}
633:         {Power, C.B.; Knebe, A.; Knollmann, S.; 2008, MNRAS submitted}
634: 
635: \bibitem[Porciani02]{Porciani02}
636:         {Porciani; Dekel; Hoffman; 2002}
637: 
638: \bibitem[Shaw06]{Shaw06}
639:         {Shaw, L.D.; Weller, J.; Ostriker, J.P.; Bode, P.; 2006, ApJ 646, 815}
640: 
641: \bibitem[Spergel03]{Spergel03}
642:         {Spergel, D.; et al.; 2003}
643: 
644: \bibitem[Springel05]{Springel05}
645:         {Springel, V.; et al.; 2005, Nature 435, 629}
646: 
647: \bibitem[Sugarman00]{Sugerman00}
648:         {Sugerman; Summers; Kamionkowski; 2000}
649: 
650: \bibitem[Vitvitska02]{Vitvitska02}
651:         {Vitvitska, M.; Klypin, A.A.; Kravtsov, A.V.; Wechsler, R.H.; Primack, J.R.; Bullock, J.S.; 2002, ApJ 581, 799}
652: 
653: \bibitem[WhiteRees78]{WhiteRees78}
654:         {White, S.D.M. \& Rees, M. 1978, MNRAS, 183, 341}
655: 
656: \bibitem[White84]{White84}
657:         {White, S.D.M.; 1984}
658: 
659: \bibitem[Zavala07]{Zavala07}
660: 	{Zavala~J., Okamoto~T. \& Frenk.~C.~S.  \astroph{0710.2901}}
661: 
662: \end{thebibliography}
663: 
664: \clearpage
665: 
666: \begin{deluxetable}{ccccccccc}
667: \tablecaption{Summary of the cosmological simulations and number of halos. \label{tab:simu}}
668: \tablewidth{0pt}
669: \tablehead{ 
670: \colhead{run}                             & 
671: \colhead{$L_{\rm box}$ [\hMpc]}                 & 
672: \colhead{$m_p$ [\hMsun]}                   & 
673: \colhead{$z_{\rm final}$}                    & 
674: \colhead{$N_{\rm halos}^{z=1}$}     &
675: \colhead{$N_{\rm halos}^{z=10}$}  &
676: \colhead{$N_{\rm relaxed\ halos}^{z=1}$}     &
677: \colhead{$N_{\rm relaxed\ halos}^{z=10}$} 
678: }
679: \startdata
680:  B01 &  1\ \ \ & 4.9 $\times 10^3$ & 10 &    ---   &  8780  &  ---   & 286  \\
681:  B02 &  2.5    & 7.8 $\times 10^4$ & 10 &    ---   &  7991   &   ---   & 201  \\
682:  B05 &  5\ \ \ & 6.2 $\times 10^5$ & 1  &   16917  &  6532 &  832    & 109 \\
683:  B10 &  10     & 4.9 $\times 10^6$ & 1  &   18589  &  4360  &  949    & 37 \\
684:  B20 &  20     & 4.0 $\times 10^7$ & 1  &   20514  &  10947 &  995    & 27  \\
685: B100 &  100    & 4.9 $\times 10^9$ & 1  &   24696  & ---    &  978    & ---\\
686: \enddata
687: \end{deluxetable}
688: 
689: 
690: \begin{deluxetable}{ccccccccccc}
691: \tablecaption{Fitting parameters for $P(\lambda)$. \label{tab:lambdaTsp}}
692: \tablewidth{0pt}
693: \tablehead{ 
694: \colhead{run}                        & 
695: \colhead{$\lambda_0'$}                & 
696: \colhead{$\sigma$}                   & 
697: \colhead{$\lambda_{\rm med}'$}         & 
698: \colhead{$M_{\rm med}$ [\hMsun]}      &
699: &
700: \colhead{run}                        & 
701: \colhead{$\lambda_0'$}                & 
702: \colhead{$\sigma$}                   & 
703: \colhead{$\lambda_{\rm med}'$}         & 
704: \colhead{$M_{\rm med}$ [\hMsun]}      
705: }
706: \startdata
707: \multicolumn{5}{c}{$z$=1}  & & \multicolumn{5}{c}{$z$=10} \\
708: \hline
709:  \multicolumn{5}{c}{---}                & & B01 & 0.042 & 0.538 & 0.041 & 5.95e+06 \\
710:  \multicolumn{5}{c}{---}                & & B02 & 0.040 & 0.539 & 0.037 & 8.19e+07 \\
711: B05  & 0.033 & 0.521 & 0.031 & 7.92e+08 & & B05 & 0.036 & 0.516 & 0.035 & 5.63e+08 \\
712: B10  & 0.037 & 0.527 & 0.034 & 6.15e+09 & & B10 & 0.033 & 0.540 & 0.029 & 4.14e+09 \\
713: B20  & 0.038 & 0.543 & 0.036 & 5.40e+10 & & B20 & 0.030 & 0.251 & 0.027 & 3.33e+10 \\
714: B100 & 0.037 & 0.535 & 0.035 & 5.25e+12 & &       \multicolumn{5}{c}{---}       \\
715: \enddata
716: \end{deluxetable}
717: 
718: 
719: \begin{deluxetable}{ccccc}
720: \tablecaption{Variation of $\alpha$ with $N_{\rm bins}$ ($Q_{\rm lim}=0.15, N_{\rm min}=600$). \label{tab:Nbins}}
721: \tablewidth{0pt}
722: \tablehead{ 
723: \colhead{$N_{\rm bins}$}     & 
724: \colhead{$\alpha_{z=1} \pm \sigma_\alpha$}   & 
725: \colhead{$\alpha_{z=10} \pm \sigma_\alpha$}
726: }
727: \startdata
728: 4            & -0.001 $\pm$ 0.169 & -0.061 $\pm$ 0.191  \\
729: 5            & -0.002 $\pm$ 0.149 & -0.059 $\pm$ 0.171 \\
730: 6            & -0.005 $\pm$ 0.137 & -0.053 $\pm$ 0.155  \\
731: 7            & -0.006 $\pm$ 0.128 & -0.056 $\pm$ 0.144  \\
732: 8            & -0.003 $\pm$ 0.120 & -0.069 $\pm$ 0.132  \\
733: \enddata
734: \end{deluxetable}
735: 
736: \begin{deluxetable}{ccccc}
737: \tablecaption{Variation of $\alpha$ with $Q_{\rm lim}$ ($N_{\rm bins}=5, N_{\rm min}=600$). \label{tab:Qlim}}
738: \tablewidth{0pt}
739: \tablehead{ 
740: \colhead{$Q_{\rm lim}$}     & 
741: \colhead{$\alpha_{z=1} \pm \sigma_\alpha$}   & 
742: \colhead{$\alpha_{z=10} \pm \sigma_\alpha$}  &
743: \colhead{$N_{\rm relaxed\ halos}^{z=1}$}   & 
744: \colhead{$N_{\rm relaxed\ halos}^{z=10}$}
745: }
746: \startdata
747: 0.05 & -0.007 $\pm$ 0.148 & -0.058 $\pm$ 0.153 & 2123 & 264 \\
748: 0.10 & -0.004 $\pm$ 0.148 & -0.062 $\pm$ 0.168 & 3365 & 497 \\
749: 0.15 & -0.002 $\pm$ 0.149 & -0.059 $\pm$ 0.171 & 3811 & 660 \\
750: 0.20 & -0.005 $\pm$ 0.151 & -0.052 $\pm$ 0.173 & 3994 & 739 \\
751: 0.25 & -0.002 $\pm$ 0.153 & -0.052 $\pm$ 0.173 & 4165 & 774 \\
752: \enddata
753: \end{deluxetable}
754: 
755: \begin{deluxetable}{ccccc}
756: \tablecaption{Variation of $\alpha$ with $N_{\rm min}$ ($Q_{\rm lim}=0.15, N_{\rm bins}=5$). \label{tab:Nmin}}
757: \tablewidth{0pt}
758: \tablehead{ 
759: \colhead{$N_{\rm min}$}     & 
760: \colhead{$\alpha_{z=1} \pm \sigma_\alpha$}   & 
761: \colhead{$\alpha_{z=10} \pm \sigma_\alpha$}  &
762: \colhead{$N_{\rm relaxed\ halos}^{z=1}$}   & 
763: \colhead{$N_{\rm relaxed\ halos}^{z=10}$}
764: }
765: \startdata
766: 100  &  0.003 $\pm$ 0.129 & -0.037 $\pm$ 0.158 & 19707 & 5478 \\
767: 200  &  0.002 $\pm$ 0.137 & -0.038 $\pm$ 0.167 & 10809 & 2534 \\
768: 300  &  0.002 $\pm$ 0.142 & -0.035 $\pm$ 0.174 & 7336  & 1526 \\
769: 600  & -0.002 $\pm$ 0.149 & -0.059 $\pm$ 0.171 & 3811  &  660 \\
770: 1000 & -0.004 $\pm$ 0.155 & -0.058 $\pm$ 0.196 & 2303  &  343 \\
771: 2000 & -0.011 $\pm$ 0.164 & -0.048 $\pm$ 0.186 & 1154  &  120 \\
772: \enddata
773: \end{deluxetable}
774: 
775: \begin{deluxetable}{ccccc}
776: \tablecaption{Applying different virialisation criterion ($N_{\rm bins}=5$). \label{tab:EtAl}}
777: \tablewidth{0pt}
778: \tablehead{ 
779: \colhead{criterion}   & 
780: \colhead{$\alpha_{z=1}$}   & 
781: \colhead{$\alpha_{z=10}$}  &
782: \colhead{$N_{\rm relaxed\ halos}^{z=1}$}   & 
783: \colhead{$N_{\rm relaxed\ halos}^{z=10}$}
784: }
785: \startdata
786: Neto et al. (2007)   &  0.000 $\pm$ 0.149 & -0.041 $\pm$ 0.178 & 3486 & 429 \\
787: Maccio et al. (2007) &  0.001 $\pm$ 0.146 & -0.040 $\pm$ 0.184 & 4275 & 512  \\
788: Bett et al. (2007)   &  0.003 $\pm$ 0.146 & -0.040 $\pm$ 0.175 & 8582 & 1955 \\
789: \enddata
790: \end{deluxetable}
791: 
792: 
793: 
794: \clearpage
795: 
796: \begin{figure}
797:   \plottwo{f1a.eps}{f1b.eps}
798:   \caption{Relation between virial parameter $Q$
799:     (cf. equation~\ref{eq:virial}) and halo mass.  The solid lines
800:     represent the adopted virialisation criteria as given by
801:     \Eq{eq:QspMass}. Note that we already applied the mass cut of 600
802:     particles per halo for this plot and hence the number of halos appearing
803:     does not agree with the number given in \Table{tab:simu}.}\label{fig:QspMass}
804:   \end{figure}
805: 
806: \begin{figure}
807: \plottwo{f2a.eps}{f2b.eps}
808: \caption{Correlation of spin parameter $\lambda$ with mass $M$. The
809:   binned data (histograms) has been fitted to a power laws (dashed
810:   line, cf. \Eq{eq:logslopes}).
811:          \label{fig:SpinMass}}
812: \end{figure}
813: 
814: \begin{figure}
815: \plottwo{f3a.eps}{f3b.eps}
816: \caption{Lognormal distributions of the spin parameter $\lambda'$. 
817:          \label{fig:Pspin}}
818: \end{figure}
819: 
820: 
821: 
822: %\clearpage
823: 
824: 
825: %% The following command ends your manuscript. LaTeX will ignore any text
826: %% that appears after it.
827: 
828: \end{document}
829: 
830: %%
831: %% End of file `sample.tex'.
832: