1:
2: \documentclass[12pt,preprint]{aastex}
3: %% \documentclass[preprint2]{aastex}
4: %% \documentclass[preprint2,longabstract]{aastex}
5:
6: %\usepackage{amssymb}
7: %\usepackage{amsmath}
8:
9: %% If you want to create your own macros, you can do so
10: %% using \newcommand. Your macros should appear before
11: %% the \begin{document} command.
12:
13:
14: \shorttitle{PopIII Stars' Remnants}
15: \shortauthors{Trenti, Santos \& Stiavelli}
16:
17: \begin{document}
18:
19: %% LaTeX will automatically break titles if they run longer than
20: %% one line. However, you may use \\ to force a line break if
21: %% you desire.
22:
23: \title{Where can we really find the First Stars' Remnants today?}
24:
25:
26:
27: %% Use \author, \affil, and the \and command to format
28: %% author and affiliation information.
29: %% Note that \email has replaced the old \authoremail command
30: %% from AASTeX v4.0. You can use \email to mark an email address
31: %% anywhere in the paper, not just in the front matter.
32: %% As in the title, use \\ to force line breaks.
33:
34: %\author{The Three of Us}
35:
36: \author{M. Trenti, M.~R. Santos, and M. Stiavelli}
37: \affil{Space Telescope Science Institute, 3700 San Martin Drive Baltimore MD 21218 USA}
38: \email{trenti@stsci.edu; msantos@stsci.edu; mstiavel@stsci.edu}
39:
40: %% Notice that each of these authors has alternate affiliations, which
41: %% are identified by the \altaffilmark after each name. Specify alternate
42: %% affiliation information with \altaffiltext, with one command per each
43: %% affiliation.
44:
45:
46:
47:
48: %-------------------------------------------------%
49: \begin{abstract}
50:
51: A number of recent numerical investigations concluded that
52: the remnants of rare structures formed at very high redshift, such as
53: the very first stars and bright redshift $z\approx 6$ QSOs, are
54: preferentially located at the center of the most massive galaxy clusters
55: at redshift $z=0$. In this paper we readdress this question using a
56: combination of cosmological simulations of structure formation and
57: extended Press-Schechter formalism and we show that the typical
58: remnants of Population III stars are instead more likely to be found
59: in a group environment, that is in dark matter halos of mass $\lesssim
60: 2 \cdot 10^{13}~h^{-1} \mathrm{M_{\sun}}$. Similarly, the descendants of the
61: brightest $z \approx 6$ QSOs are expected to be in medium-sized
62: clusters (mass of a few $10^{14}~h^{-1} \mathrm{M_{\sun}}$), rather than in the
63: most massive superclusters ($M>10^{15}~h^{-1} \mathrm{M_{\sun}}$) found within
64: the typical 1~Gpc$^3$ cosmic volume where a bright $z\approx 6$ QSO
65: lives. The origin of past claims that the most massive clusters
66: preferentially host these remnants is rooted in the numerical method
67: used to initialize their numerical simulations:
68: Only a small region of the cosmological volume of interest was
69: simulated with sufficient resolution to identify low-mass halos at
70: early times, and this region was chosen to host the most massive halo
71: in the cosmological volume at late times. The conclusion that the
72: earliest structures formed in the entire cosmological volume evolve
73: into the most massive halo at late times was thus arrived at by
74: construction. We demonstrate that, to the contrary, the first
75: structures to form in a cosmological region evolve into relatively
76: typical objects at later times. We propose alternative numerical
77: methods for simulating the earliest structures in cosmological
78: volumes.
79:
80: \end{abstract}
81:
82: \keywords{cosmology: theory - galaxies: high-redshift - early universe
83: - methods: N-body simulations}
84:
85:
86: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
87: \section{Introduction}
88:
89: Rare dark matter halos at high redshifts host interesting
90: astrophysical objects, especially before or at the end of the
91: reionization epoch. One example is given by the very first Population~III (PopIII)
92: stars formed in the universe at $z \gtrsim 40$, which started the metal
93: enrichment of the interstellar medium and the reionization process
94: \citep{abel02,san02,bromm04,naoz06}, and possibly also produced
95: intermediate mass black hole seeds that grow to become
96: super-massive black holes ($M_{BH}> 10^9 M_{\sun}$) within the first
97: billion year after the Big Bang \citep{vol03}. Another example are
98: bright $z \approx 6$ QSOs \citep{fan04}, considered to be hosted in the
99: most massive dark matter halos at that time \citep{MR05}. The
100: luminosity of such an object is powered through accretion onto a
101: supermassive black hole \citep[e.g. see][]{hop06}, which may be the
102: descendant of one of the first generation of PopIII stars formed in
103: the universe, at $z \gtrsim 40$ (\citealt{mad01}; see also \citealt{ts07a}).
104:
105: Many numerical and theoretical investigations have aimed at
106: characterizing the properties of both PopIII stars
107: \citep[e.g. see][]{abel02,bromm04,ciar,ree05,gao05,osh07} and of high
108: redshift QSOs \citep[e.g. see][]{hop06,MR05,dimatteo05,li07}. However,
109: the rarity of these structures makes a fully self-consistent treatment
110: even of the formation of the underlying dark matter halos
111: typically outside the current capabilities of a
112: standard cosmological simulation, as the dynamic range that needs to
113: be resolved is too large. The main limitation at high mass resolution is the box size,
114: $l_{box}$: The enforcement of periodic boundary conditions
115: limits the power spectrum of density fluctuations to modes with
116: wavelengths shorter than $l_{box}$. This results in a severe bias on
117: the measured number density of rare, massive halos, especially
118: before the epoch of the reionization of the universe when the abundance
119: of galaxies derived from numerical simulations can be underestimated
120: by up to an order of magnitude \citep{bar04}. While the number density
121: of rare objects can be estimated in the context of
122: Press-Schechter-type modeling \citep{lacey93,mo_white96,she99,bar04},
123: studying the details of the formation histories of the first galaxies
124: and QSOs, possibly including hydrodynamics and radiative feedback
125: processes, requires high-resolution numerical simulation.
126:
127: Thus, given a simulation box that is large enough to contain one or
128: more of the high-redshift halos of interest, the problem is to
129: identify a sub-region that contains one of them for high-resolution
130: simulation. In passing we note that this challenge is different from
131: that of developing a simulation code that is able to adaptively
132: increase the resolution when following the gas collapse that leads to
133: the formation of the first stars. This has been successfully
134: implemented with adaptive mesh refinement codes (for example ENZO -
135: \citealt{bry98}). \citet{gao05} proposed a method to identify rare
136: structures formed at very high redshift by recursively resimulating
137: successively smaller, nested, sub-regions of a simulation box at
138: progressively higher resolutions. Specifically, their method
139: resimulates the region of the box centered on the most massive dark
140: matter halo, reidentified at higher redshift and lower mass within the
141: sub-region at each resimulation step, until the desired resolution is
142: achieved in a small fraction of the entire simulation box. This
143: procedure is well-motivated by the fact that number density of massive
144: halos is increased in regions of large-scale over-density
145: \citep{bar04}, and thus structure formation in the sub-region on small
146: scales at early times is accelerated by sitting at the top of an
147: over-density, an over-density that is known to exist because it collapsed
148: into the massive halo identified in the previous resimulation
149: step. The \citet{gao05} method extends high-resolution resimulation
150: techniques used previously \citep[e.g. see ][]{nav94,whi00}, and has
151: been used recently by several authors to study the first stars and
152: QSOs \citep{gao05,ree05,gao07,li07}.
153:
154: The recursive resimulation method for volume selection is very
155: effective at identifying a region with a very high-redshift halo when
156: compared to random selection of a region of equal volume within the
157: parent simulation box, but still it does not lead to the rarest
158: high-redshift halos in the volume. This is due to the stochastic
159: nature of the formation and growth of dark matter halos
160: \citep[e.g. see][]{PS,bond,she99}. The most massive dark matter halo
161: in a given volume at some early time does not in general evolve into
162: the most massive halo at a later time. The probability of the most
163: massive halo to evolve into the most massive halo, quantified
164: following \citet{lacey93}, decreases as time passes and the
165: characteristic mass scale increases. As we demonstrate, the
166: probability falls far below unity for the evolution of the first stars
167: and QSOs to the present time. The stochastic nature of growth
168: histories of dark matter halos is addressed and discussed in
169: \citet{gao05}, but unfortunately some of the subsequent studies using
170: their method do not take into account this stochasticity and assume a
171: very strong correlation between the location of the richest clusters
172: at $z=0$ and that of rare halos at very high redshift.
173:
174: In this paper we clarify this issue by presenting some basic, but
175: often overlooked, results of Gaussian random fields in extended
176: Press-Schechter theory to quantify the locations at later times of the
177: first dark matter halos to form in a simulation box. We compare these
178: analytical and numerical results to those published using recursive
179: resimulation and highlight the benefits and limitations of that
180: method. In addition, we propose an alternative method, based on
181: analysis of the density field in the initial conditions, to select the
182: sub-region of a simulation box that contains some of the earliest
183: structure in the box without being biased toward the regions hosting
184: the largest halos at $z=0$. The paper is organized as follows. In
185: Sec.~\ref{sec:QSOs} we investigate where the descendants of $z=6$ QSOs
186: are today, while in Sec.~\ref{sec:PopIII} we extend the result to the
187: first Population III stars in the Universe. In Sec.~\ref{sec:Ic} we
188: discuss the implications of our results for the recursive mesh
189: refinement method and propose a viable
190: alternative. Sec.~\ref{sec:conc} summarizes and concludes.
191:
192: \section{Bright $z\approx 6$ QSOs: where are they today?} \label{sec:QSOs}
193:
194: As a first step toward characterizing the assembly history of rare,
195: massive, dark matter halos, we consider the link between $z \approx 6$
196: QSOs (considered to be hosted in the largest protoclusters at that
197: time, e.g. see \citealt{MR05}) and the largest clusters at $z=0$. For
198: this we use the publicly available merger trees constructed upon the
199: Millennium Run \citep{MR05,lemson}. This cosmological simulation has
200: more than $10^{10}$ dark matter particles within a volume of
201: $500^3~h^{-3}\mathrm{Mpc}^3$ and has been run using a concordance
202: $\Lambda CDM$ cosmology based on WMAP year 1 results
203: \citep{WMAP1}\footnote{Note that a change in the cosmological
204: parameters, and in particular on the value of $\sigma_8$, the root
205: mean squared mass fluctuation in a sphere of radius
206: $8~h^{-1}\mathrm{Mpc}$ extrapolated to $z=0$ using linear theory, does
207: influence the details of our results, such as the expected mass of a
208: QSO host halo, but not the essence of the relation between dark matter
209: halos at different redshifts.}.
210:
211: From the Millennium Catalog\footnote{http://www.mpa-garching.mpg.de/millennium} we select all dark matter halos with more
212: than $10^3$ particles (corresponding to a minimum halo mass of $8.6
213: \cdot 10^{11}~h^{-1} \mathrm{M_{\sun}}$) in the $z=0$ snapshot and all
214: halos with more than $100$ particles in the $z=6.18$ snapshot. Within
215: these catalogs we also identify the descendants at $z=0$ of the
216: $z=6.18$ halos and plot them in Fig.~\ref{fig:MR}. The figure
217: immediately shows that the descendants of the most massive halos at
218: $z>6$ live in a variety of environments at $z=0$. Some of the most
219: massive halos at $z=6.18$ have accreted relatively little mass by
220: $z=0$ (a factor of a few of their initial mass), while others have
221: increased their masses by more than hundred times, reaching
222: $M>10^{15}~h^{-1} \mathrm{M_{\sun}}$. The scatter is substantial. In
223: particular the descendant of the most massive $z=6.18$ halo in this
224: simulation has a mass of only $2.2 \cdot 10^{14}~h^{-1}
225: \mathrm{M_{\sun}}$ at $z=0$, to be compared with the largest cluster
226: in the box, which has $M=3 \cdot 10^{15}~h^{-1}
227: \mathrm{M_{\sun}}$. Fig.~\ref{fig:MR} also quantifies the relation
228: between progenitor and descendant halos in terms of the dimensionless
229: variable $\nu = \delta_{c}^2 / \sigma^2(M)$ used in extended Press-Schechter
230: modeling. Here $\sigma^2(M)$ is the variance of density
231: fluctuations over a scale that contains a mass $M$ and $\delta_c(z)$
232: is the critical value of an overdensity in linear theory at redshift
233: $z$.
234:
235: A complementary picture is given by considering the mass distribution
236: of the most massive $z=6.18$ progenitor for the $10000$ largest $z=0$
237: halos, that is with a mass $M \gtrsim 4 \cdot 10^{13} h^{-1}
238: \mathrm{M_{\sun}}$ (Fig.~\ref{fig:MR1}). Again a considerable scatter
239: is present in the plot, with only a modest correlation between
240: descendant and progenitor mass. These results reflect the fact that
241: there are additional contributions to the density fluctuation power
242: spectrum over a scale $M_1$ compared to a scale $M_2>M_1$. This can be
243: also illustrated by random walks generated using a $\Lambda CDM$ power
244: spectrum (see Fig.~\ref{fig:random_walk}). The typical fluctuations at
245: small scales greatly exceed those at large scale, so that a walk with
246: excess power at small scales often has a rather typical amplitude
247: at large scales, and vice versa.
248:
249: A quantitative model for this behavior is available in the context of
250: extended Press-Schechter theory, for example by Eqs. 2.15 and 2.16 in
251: \citet{lacey93}. Using this formalism one can compute the probability
252: that a dark matter halo of mass $M_1$ at redshift $z_1$ evolves into a
253: halo of mass $M>M_2$ at $z_2<z_1$ [$P(M>M_2,z_2|M_1,z_1)$]. This is
254: shown in Fig.~\ref{fig:EPS}, where we plot the contour lines of $P$
255: for a $M_1=5 \cdot 10^{12}~h^{-1} \mathrm{M_{\sun}}$ halo at
256: $z_1=6.18$. The median of the distribution at $z=0$ is $M=4 \cdot
257: 10^{14}~h^{-1} \mathrm{M_{\sun}}$ and at the 68\% confidence level
258: interval is $M \in [1.8:8.8] \cdot 10^{14}~h^{-1}
259: \mathrm{M_{\sun}}$. In this respect the Millennium Run is typical as
260: it lies within the $1 \sigma$ interval. From Fig.~\ref{fig:EPS} one
261: can also immediately see that it is relatively improbable for the
262: descendant of the most massive halo at $z=6.18$ to be the most massive
263: halo at $z=0$. In fact there is only a probability $p<2\%$ that the
264: mass of the descendant halo is above $2 \cdot 10^{15}~h^{-1}
265: \mathrm{M_{\sun}}$. The Millennium Run has a volume large enough to
266: contain $z=0$ halos more massive than $2 \cdot 10^{15}~h^{-1}
267: \mathrm{M_{\sun}}$ (in fact there are two of them), thus it is
268: expected that in less than $2\%$ of Millennium-like realizations the
269: most massive $z \approx 6$ halo is the progenitor of the most massive
270: $z=0$ halo. As a reference we provide the number density contours from
271: the \citet{she99} mass function, integrated to obtain the number of
272: objects above that mass at that redshift in the volume of the
273: Millennium Run (red dotted lines in Fig.~\ref{fig:EPS}): when $z
274: \lesssim 2.5$, the upper 2$\sigma$ confidence level contour lies at a
275: lower mass than the contour for the number density ($n=1$) associated
276: with the most massive dark matter halo in the volume. This mean that
277: the typical mass of the descendants progressively shifts toward that
278: of more common halos as the redshift decreases, as it is also
279: immediately visible when looking at the redshift evolution of the
280: value of $\nu$ for the median descendant, which decreases
281: progressively (small panel of fig.~\ref{fig:EPS}).
282:
283: \section{The very first PopIII stars: where are they today?} \label{sec:PopIII}
284:
285: A similar scenario holds for the relation between the location of the
286: very first PopIII stars, formed in dark matter halos with $M \approx
287: 10^6 \mathrm{M_{\sun}}$ at $z>40$, and that of dark matter halos at
288: $z=0$. Here there is even less correlation than in the case of $z=6$
289: QSO halos (see Sec.~\ref{sec:QSOs}) because the halo mass dynamic
290: range involved is larger and thus there is an additional contribution
291: to the scatter from modes at small scales in the density fluctuation
292: power spectrum. Using a simulation volume of $\approx 1~\mathrm{Gpc}^3$,
293: \citet{ts07a} showed that the remnants of the very first PopIII stars
294: formed in dark matter halos of mass $10^6~h^{-1} \mathrm{M_{\sun}}$ at
295: $z>40$ end up at $z=0$ in dark matter halos with a median mass of $3
296: \cdot 10^{13}~h^{-1} \mathrm{M_{\sun}}$, about two order of magnitude
297: smaller than the largest halo in the simulation box. Here we confirm
298: and extend their result using extended Press-Schechter modeling.
299:
300: Using Eq. 2.15 and 2.16 of \citet{lacey93}, we compute at different
301: redshifts the probability distribution for the mass of the descendant
302: of a $10^6~h^{-1} \mathrm{M_{\sun}}$ dark matter halo formed at
303: $z_1=40$. The results are shown in Fig.~\ref{fig:EPS_PopIII}, where we
304: plot the median descendant mass versus the redshift $z_2$ (black
305: line), the $1 \sigma$ confidence level contours (blue lines) and $2
306: \sigma$ confidence level contours (green lines). These analytic
307: results agree well with the numerical simulations in \citet{ts07a} and
308: confirm that at $z=0$ the typical remnant of one of the very first
309: PopIII halos does not live in a supercluster, but is rather hiding in
310: a more common group environment. This is, e.g., contrary to the
311: conclusions ( ``The very oldest stars should be found today in the
312: central regions of rich galaxy clusters'') that \citet{whi00} drew
313: from the resimulation of a massive galaxy cluster. Similarly, at $z=6$
314: the remnant of a typical $z \geq 40$ PopIII star does not live in the
315: largest dark matter halos of that time, but rather has seeded a dark
316: matter halo of mass $3-4 \cdot 10^{10}~h^{-1} \mathrm{M_{\sun}}$,
317: typical for the faint $z\approx 6$ galaxies observed in deep surveys
318: such as the Hubble Space Telescope Ultra Deep Field \citep{ts08}.
319:
320: %%%%%%%% Discussion about more common PopIII stars
321: These results have been obtained for very rare PopIII stars formed at
322: $z>40$. Extended Press-Schechter modeling predicts that the remnants
323: of more common PopIII stars formed at lower redshift live in
324: yet lower mass dark matter halos. For example a $10^6~h^{-1} \mathrm{M_{\sun}}$ dark matter halo
325: formed at $z=20$ has a $z=0$ descendant with median mass $4 \cdot
326: 10^{12}~h^{-1} \mathrm{M_{\sun}}$. More massive, rarer, PopIII halos, with virial
327: temperature above $10^4$ K, can cool by neutral hydrogen rather than
328: molecular hydrogen so that their formation is less sensitive to
329: radiative feedback from other stars \citep[e.g. see][]{bromm04}. These
330: halos have a typical mass of $\approx 10^{8}~h^{-1} \mathrm{M_{\sun}}$ and if
331: they form at $z=20$, then their typical descendants are similar to
332: those of a $z=40$ $10^6 h^{-1} \mathrm{M_{\sun}}$ dark halo (see
333: Fig.~\ref{fig:EPS_PopIII}).
334:
335: \section{Generation of Initial Conditions around Rare High-Redshift Halos}\label{sec:Ic}
336:
337: Given the large scatter in the assembly histories of dark matter halos
338: is there an optimal way to select a region of a simulation box
339: centered around a rare high-redshift halo under the constraint of
340: limited computational resources?
341:
342: \subsection{Analysis of the recursive resimulation method}
343:
344: The method introduced by \citet{gao05} certainly presents a very
345: competitive advantage over a random selection of an equal subvolume
346: though it identifies a halo with an atypical
347: accretion history, biased toward having an above average merging rate
348: and living in an environment that tends to have an overdensity of
349: nearby halos \citep{bar04}.
350: %This fact must be properly taken into account if the
351: %properties of the resimulation are then used to infer average
352: %properties of the galaxy population, especially if baryon physics
353: %and/or radiative feedback is added to the resimulation.
354: To better quantify the properties of halos identified by recursive
355: resimulation, we use the Monte Carlo method presented in
356: \citet{ts07a}, which is based on the identification of virialized dark
357: matter halos as density peaks in the linear density field. This
358: approach predicts well and without introducing systematic biases the
359: location and virialization redshift of the very first dark matter
360: halos when compared to the full non-linear dynamics of the simulation,
361: even though there are some statistical fluctuations on a halo-to-halo
362: basis \citep{bond96a,mes07}. For the comparison with \citet{gao05} we
363: adopt the following parameters from their paper: (i) $\Omega_M=0.3$,
364: $\Omega_{\lambda}=0.7$, $\Omega_b = 0.04$, $h=0.7$, $\sigma_8=0.9$,
365: spectral index $n_s=1$; (ii) parent box size edge $l_{box}=479$ Mpc
366: $h^{-1}$; (iii) largest dark matter halo in the box at $z=0$ of mass
367: $M_1=8.1 \cdot 10^{14} h^{-1} \mathrm{M_{\sun}}$; (iv) final
368: high-resolution simulation sphere of $1.25 h^{-1}$~Mpc (with volume
369: $V_5 = 8.18 h^{-3}$~Mpc$^3$); (v) the most massive halo in the final
370: resimulation region R5 is $M_{halo_{R5}} = 1.2 \cdot 10^5 h^{-1}
371: \mathrm{M_{\sun}}$ at $z=48.84$. Based on these assumptions, our
372: analysis yields the following results:
373: \begin{itemize}
374:
375: \item The number density of dark matter halos of mass $M \geq
376: M_{halo_{R5}}$ is $n(M>M_{halo_{R5}}) = 0.17 h^3 $Mpc$^{-3}$
377: (Sheth-Tormen mass function) or $n(M>M_{halo_{R5}}) = 0.0025 h^3
378: $Mpc$^{-3}$ (Press-Schechter mass function) at $z=48.84$, so these
379: halos are not particularly rare. Specifically, in a random region of
380: volume $V_5$ at $z=48.84$, the expectation value for the number of
381: halos more massive than $M_{halo_{R5}}$ is greater than unity ($n(M>
382: M_{halo_{R5}}) \cdot V_5 = 1.39$) when using the \citet{she99} formula.
383: However, due to the high bias of the halos, $b \approx 16$, the actual
384: fraction of empty volumes as given by our Monte Carlo code is
385: $98.7\%$. Thus the \citet{gao05} method is a substantial improvement
386: over random selection of a volume $V_5$ within the parent box.
387:
388: \item The first halo of mass $M_{halo_{R5}} = 1.2 \cdot 10^5 h^{-1}
389: \mathrm{ M_{\sun}}$ in the \emph{whole} box virializes at $z>62$ at
390: 99\% of confidence level. The possibility that the most massive halo
391: identified by \citet{gao05} in their final resimulated volume is the
392: most massive of the whole box at that redshift is ruled out at a
393: confidence level greater than $1-10^{-7}$ (confidence level limited by
394: the precision of the numerical integration).
395:
396: \item The first halo of mass $M_{halo_{R5}}$ that ends up at $z=0$ in
397: a dark matter halo of mass $M_1=8.1 \cdot 10^{14} h^{-1} \mathrm{
398: M_{\sun}}$ is formed at $z>50.8$ at 99.9\% of confidence level. Thus
399: it is unlikely that the \citet{gao05} halo of mass $M_{halo_{R5}}$ at
400: $z=48.84$ is the most massive progenitor of the cluster that will
401: contain it at $z=0$.
402:
403: \end{itemize}
404:
405: From this analysis it is indeed confirmed that recursive refinement
406: works well to select a sub-region for resimulation that host a rare
407: high-z dark matter halo, but still this halo is not one of the very
408: first of its kind in the box. If the interest is primarily in the
409: local physics around one of these rare structures, a viable
410: alternative is the constrained realization method
411: \citep{hf91,bert01}. A constrained realization does not require a
412: hierarchy of resimulations, but rather introduces the overdensity in
413: the initial conditions by construction, so that the realized initial
414: conditions do not carry information about its rarity or its typical
415: surrounding environment.
416:
417: \subsection{Unbiased selection of rare halos: the density field method}
418:
419: If one is instead interested in selecting one of the very first dark
420: matter halos in a given simulation volume, with a particular interest
421: in representative halos, that is halos with unbiased accretion
422: histories, we propose instead to select the final region of interest
423: from a density field at uniform resolution. When the dynamic range of
424: the resimulation is not too large, it is of course possible to
425: identify the region to be refined directly from the low-resolution
426: dark matter halo catalogs. If this is not possible, then the
427: resimulation volume can be identified from a high resolution, but
428: uniform, refinement of the density field. The idea is the following:
429: %%%%%%%%%%%%%%%
430: \begin{enumerate}
431:
432: \item A high-resolution density field is generated over the whole box.
433: The mass of a field cell is that of the high-$z$ dark
434: matter halo of interest (e.g. $10^6 h^{-1} \mathrm{ M_{\sun}} $ for a PopIII
435: halo). A technical, but important, detail is that a top-hat
436: filter in frequency space must be used when generating this field, as this
437: guarantees that the mass function obtained from peak analysis matches
438: that of \citet{PS}.
439:
440: \item The highest peak in the high-resolution field is
441: identified. This is the location where one of the first halos at the
442: mass scale of the density field is formed within the \emph{whole} simulation
443: box.
444:
445: \item A region around that peak can now be selected for resimulation,
446: with additional fluctuations added over the high resolution density
447: field, for example by using the GRAFIC2 refinement method of
448: \citet{bert01}. Outside the selected region the high resolution field
449: can be degraded if necessary to generate more massive particles.
450: \end{enumerate}
451: %%%%%%%%%%%%%%
452:
453: This method does not necessarily require overwhelming computational
454: resources. For example, to select the largest $z=6$ dark matter halo
455: in a volume relevant for the study of bright high-$z$ QSOs, a mass
456: resolution of $\approx 5 \times 10^{11} h^{-1} \mathrm{ M_{\sun}} $ is
457: more than sufficient, which translates into a $N = 512^3$ density
458: field grid for a cosmic volume of $\approx (1 \mathrm{Gpc}/h)^3$. Such
459: a grid only requires 512MB of RAM for storing. Even identifying the
460: positions of the first PopIII stars in a large volume is not
461: unrealistic. A $N=2048^3$ density grid requires only 32GB of RAM and
462: provides a mass resolution of $1.2 \times 10^5 h^{-1}
463: \mathrm{ M_{\sun}} \equiv M_{halo_{R5}}$ (the first halo in
464: \citealt{gao05}) over a volume of $(23.74 h^{-1}
465: \mathrm{Mpc})^3$. Given the number density of such halos---$n \approx
466: 0.11 h^3 \mathrm{Mpc}^{-3}$---about 1471 will have formed on average
467: in the box by $z=48.84$. In fact, using the Monte Carlo code of
468: \citet{ts07a} we predict that there is a probability $p< 10^{-3}$ of not
469: forming one of these halos before $z=48.84$. The median redshift of
470: formation for the first halo of mass $M_{halo_{R5}}$ is $z=51.6$. With
471: larger resources one can generate a $4096^3$ density grid over a
472: volume of $(100 \mathrm{Mpc}/h)^3$, obtaining a mass resolution of $10^6 h^{-1}
473: \mathrm{ M_{\sun}} $. This requires about 256GB of RAM, easily
474: available on a parallel computing cluster of moderate size. With a
475: top-end supercomputer one can use more than $10^{12}$ cells, with a
476: dynamic range that permits identification of smaller mass halos in the
477: same volume or an increase the volume of the parent box for the same
478: assumed PopIII halo mass.
479:
480: Our proposed density field method has also another advantage: As we
481: generate a uniform, very high-resolution density grid over the whole
482: box, aliasing errors in the zoom refinement procedure are less severe
483: than if one follows the \citet{gao05} recursive refinement method (see
484: \citealt{bert01} for a detailed analysis of the errors introduced when
485: large zooming factors are used).
486:
487: \subsection{Preliminary testing of the density field method}\label{sec:valid}
488:
489: Detailed analysis and applications of our method to select
490: high-redshift halos is deferred to a follow-up paper. Here we present
491: a preliminary testing to assess its performance. For this we consider
492: a $N=512^3$ cosmological simulation in a box of edge $l_{box}=512
493: h^{-1} \mathrm{Mpc}$ (details on the simulation are discussed in
494: \citealt{ts07a}). We construct the halo catalog for a snapshot at
495: $z=5$ using the HOP halo finder \citep{eis98}, finding that the most
496: massive halo has a mass of $1.25 \cdot 10^{13} h^{-1}
497: \mathrm{M_{\sun}}$ ($187$ particles). The density field used to
498: generate the initial conditions for this $N=512^3$ run is then (i)
499: downgraded to a $N_{down}=128^3$ grid (where one low resolution cell
500: corresponds to 64 of the original grid cells) using the \citet{hf91}
501: constrained realization method and (ii) evolved in linear theory to
502: $z=5$. The highest density peaks in the low-resolution field are then
503: identified and their location compared with that of the dark matter
504: halos identified from the N-body simulation at full resolution. Within
505: the ten highest peaks (with linear overdensities from $\delta =1.84$
506: to $\delta = 1.75$), six of them are associated with dark matter halos
507: with more than $64$ particles, including all the top three
508: overdensities. The top overdensity is matched to the fourth most
509: massive halo of the snapshot, with $141$ particles (versus the $187$ of
510: the most massive). Of the remaining unmatched peaks, two are
511: associated with halos with less than 64 particles and two appear to be
512: still in the process of virialization, so the central density of their
513: groups in the N-body run does not qualify them as halos. The most
514: massive halo in the box is missed by the density field analysis
515: because its particles are spread over several adjacent cells in the
516: downgraded density field. This is an intrinsic limitation of our
517: method rooted in the use of a fixed grid in the position space, which
518: is bound to miss some of the the dark matter halos that are off-center
519: with respect to the spatial location of the grid cells. However this
520: only leads to miss a random fraction of the rare halos, without an
521: environmental bias as in the \citet{gao05} method.
522:
523: \section{Conclusion and Discussion}\label{sec:conc}
524:
525: By means of both analysis of numerical simulations and of extended
526: Press-Schechter modeling we investigated the relation between the most
527: massive dark matter halos at different redshifts. The main conclusion
528: of this work is that---contrary to expectations from many recent works
529: (e.g. see \citealt{MR05,ree05,gao05,gao07,li07})---the most massive
530: halo at a redshift $z_1>z_2$ does not necessarily evolve into the most
531: massive at $z=z_2$. This is a robust conclusion that can be naturally
532: understood in the context of growth of dark matter density
533: perturbations, for example by constructing merger trees through the
534: \citet{lacey93} model. Rare high-redshift objects, such as the
535: remnants of the first PopIII stars and QSOs, are not hosted at $z=0$
536: in the most massive halos, but rather live in a variety of
537: environments. For example, the typical $z>40$ PopIII star remnant
538: lives in a dark matter halo that at $z=0$ has a mass of $\approx 2
539: \cdot 10^{13} h^{-1} \mathrm{M_{\sun}}$, typical for a galaxy group,
540: and not within rich clusters as claimed by \citet{whi00}. Similarly
541: the descendant of a typical $z\approx 6$ QSO is not located within the
542: most massive clusters at $z=0$ as assumed by \citet{li07}.
543:
544: These conclusions have important consequences on the application of
545: the recursive simulation method introduced by \citet{gao05} to
546: identify high-redshift rare halos based on progressive refinement of
547: regions centered around the most massive $z=0$ cluster. In fact, while
548: the recursive method is indeed effective at identifying a sub-region
549: of the simulation with earlier-than-average structure formation, it
550: finds neither the earliest structures in the box, which are dominated
551: by small-scale density fluctuations, nor typical early structure, as
552: it preferentially identifies objects located in the regions with the
553: highest bias.
554:
555: These limitations may have only a minor effect when the goal is to
556: investigate the formation of one rare Population III halo in the
557: simulation box, as it is done for example in \citet{gao05} and in
558: \citet{ree05}. However in different physical scenarios it is important
559: to correctly estimate the rarity of the halo simulated and to assess
560: how typical their growth histories are. This is critical if additional
561: physics beyond gravitational interactions is included, such as star
562: formation and radiative feedback. One such example is the formation of
563: rare high-redshift QSOs: \citet{li07} use the \citet{gao05} method to
564: identify at $z\approx 6.5$ ``the most massive halo in a $\approx$ 3
565: Gpc$^3$ volume'' and then conclude that the QSO formed in this halo
566: reproduces the properties of observed QSOs with the same number
567: density. From our analysis in Sec.~\ref{sec:QSOs} it is clear that the
568: halo identified by \citet{li07} as progenitor of the largest $z=0$
569: cluster is not likely to be the most massive at $z \approx 6.5$. Thus
570: other similar or more massive halos are expected to be present at
571: $z \approx 6.5$ in their $\approx$ 3 Gpc$^3$ simulation volume: in principle
572: any of these halos could host a bright QSOs, with important
573: consequences for the comparison between observed and simulated QSO
574: number densities. In addition, when the goal is to study the
575: environment in which QSOs live, selecting host halos with the
576: resimulation method introduces systematic effects in the results
577: difficult to quantify and correct for, because these halos would have
578: above-average growth (and merging) histories.
579:
580: To avoid selecting only the halos with atypical accretion histories
581: and in an attempt to improve over the identification of some of the
582: rarest high-redshift halos in a box, we propose instead to select the
583: initial conditions for high-resolution resimulation based on the
584: analysis of the linear density field at uniform resolution. The
585: method, described in Sec.~\ref{sec:Ic}, identifies subregions of the
586: simulation box with high-redshift halos as those with the highest
587: peaks in the density field, requiring a mass resolution in the field
588: comparable to that of the mass of the halos one wishes to select. The
589: applicability of the method is thus limited only by the size of the
590: largest density grid that can be accommodated in the available memory,
591: otherwise requiring only a modest amount of computing time compared to the
592: \citet{gao05} method. This is because our method bypasses the need of
593: a series of N-body simulations to be carried out in addition to the
594: final run.
595:
596: Unfortunately, our method does not guarantee identification of \emph{the
597: first} halo on the desired mass scale, as highlighted by some preliminary
598: testing we presented in Sec.~\ref{sec:valid}. This seems still an
599: elusive goal. When the density field is defined over a fixed grid,
600: there is not a perfect match between the halo catalog constructed from
601: the density field and that obtained by increasing the resolution of
602: the field and then following the full non-linear dynamics with an
603: N-body simulation. An extensive validation of our linear density field
604: initial conditions generation and its application to the formation of the first
605: bright QSOs will be discussed in a subsequent paper.
606:
607:
608: \acknowledgements
609:
610: This work was supported in part by NASA grants JWST IDS NAG5-12458 and
611: HST-GO10632. We thank the referee for useful comments and
612: suggestions. We are grateful to Gerard Lemson and to the Millennium
613: Run collaboration for allowing us to use their merger tree catalogs.
614:
615: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
616: \begin{thebibliography}{}
617:
618: \bibitem[Abel et al.(2002)]{abel02} {Abel}, T., {Bryan}, G.~L. and {Norman}, M.~L. 2002, Science, 295, 93
619:
620: \bibitem[Bertschinger(2001)]{bert01} Bertschinger E. 2001, \apj, 137, 1
621:
622: \bibitem[Barkana \& Loeb(2004)]{bar04} Barkana, R. and Loeb A. 2004, \apj, 609, 474
623:
624: \bibitem[Bond et al.(1991)]{bond} Bond, J. R., Cole, S., Efstathiou, G. and Kaiser, N. 1991, \apj, 379, 440
625:
626: \bibitem[Bond \& Myers(1996)]{bond96a} Bond, J.~R. and Myers, S.~T. 1996, \apjs, 103, 1
627:
628: \bibitem[Bryan \& Norman(1998)]{bry98} Bryan G.L. and Norman, M.L. 1998, \apj, 495, 80
629:
630: \bibitem[Bromm \& Larson(2004)]{bromm04} {{Bromm}, V. and {Larson}, R.~B.} 2004, \araa, 42, 79
631:
632: \bibitem[Casas-Miranda et al.(2002)]{cas02} {Casas-Miranda}, R., {Mo}, H.~J., {Sheth}, R.~K. and {Boerner}, G. 2002, \mnras, 333, 730
633:
634: \bibitem[Ciardi \& Ferrara(2005)]{ciar} Ciardi, B. and Ferrara, A. 2005, Space Sci. Rev., 116, 625
635:
636: \bibitem[Di Matteo et al.(2005)]{dimatteo05} {{Di Matteo}, T.,
637: {Springel}, V. and {Hernquist}, L.} 2005, \nat, 433, 604
638:
639: \bibitem[Eisenstein \& Hut(1998)]{eis98} Eisenstein, D.~J. and Hut, P. 1998, \apj, 498, 137
640:
641: \bibitem[Eisenstein \& Hu(1999)]{eis99} Eisenstein, D.~J. and Hu, W. 1999, \apj, 511, 5
642:
643: \bibitem[Fan et al.(2004)]{fan04} {{Fan}, X. et al.} 2004, \aj, 128,
644: 515
645:
646: \bibitem[Gao et al.(2005)]{gao05} Gao, L., White, S. D. M.,
647: Jenkins, A., Frenk, C. S. and Springel, V. 2005, \mnras, 363, 379
648:
649: \bibitem[Gao et al.(2007)]{gao07} Gao, L., Yoshida, N., Abel,
650: T., Frenk, C.S., Jenkins A. and Springel, V. 2007, \mnras, 378, 449
651:
652: \bibitem[Hoffman \& Ribak(1991)]{hf91} Hoffman, Y. and Ribak, E. 1991, \apj, 380, L5
653:
654: \bibitem[Hopkins et al.(2005)]{hop06} {Hopkins}, P.~F., {Hernquist}, L., {Martini}, P., {Cox}, T.~J.,
655: {Robertson}, B., {Di Matteo}, T. and {Springel}, V. 2005, \apjl, 625, 71
656:
657: \bibitem[Lacey \& Cole(1993)]{lacey93} Lacey, C. and Cole, S. 1993, \mnras, 262, 627
658:
659: \bibitem[Lemson(2006)]{lemson} Lemson G. and the Virgo Consortium, 2006, astro-ph/0608019
660:
661: \bibitem[Li et al.(2007)]{li07} Li, Y., Hernquist L., Robertson B.,
662: Cox T.~J., Hopkins, P.~F., Springel, V., Gao, L., Di Matteo, T.,
663: Zentner A.~R., Jenkins, A. and Yoshida N. 2007, \apj, 665, 187
664:
665: \bibitem[Madau \& Rees(2001)]{mad01} Madau, P. and Rees, M. 2001, \apjl, 551, 27
666:
667: \bibitem[Mesinger \& Furlanetto(2007)]{mes07} Mesinger, A. and Furlanetto, S. 2007, \apj, 669, 663
668:
669: \bibitem[Mo \& White(1996)]{mo_white96} Mo, H. J.and White, S. D. M. 1996, \mnras, 282, 347
670:
671: \bibitem[Mu{\~n}oz \& Loeb(2008)]{mun08} Mu{\~n}oz, J.~A. and Loeb, A. 2008, \mnras, in press, arXiv:0711.0467
672:
673: \bibitem[Naoz et al.(2006)]{naoz06} Naoz, S., Noter, S. and Barkana R. 2006, \mnras, 373, 98
674:
675: \bibitem[Navarro \& White(1994)]{nav94} Navarro, J.~F. and White, S.~D.~M. 1994, \mnras, 267, 401
676:
677: \bibitem[O'Shea \& Norman(2007)]{osh07} O'Shea, B.~W. and Norman, M.~L. 2007, \apj, 654, 66
678:
679: \bibitem[Press \& Schechter(1974)]{PS} Press, W.~H. and Schechter, P. 1974, \apj, 187, 425
680:
681: \bibitem[Reed et al.(2005)]{ree05} {Reed}, D.~S., {Bower}, R., {Frenk}, C.~S. and {Gao}, L.,
682: {Jenkins}, A., {Theuns}, T. and {White}, S.~D.~M. 2005, \mnras, 363, 393
683:
684: \bibitem[Santos et al.(2002)]{san02} Santos, M.~R., Bromm V. and Kamionkowski M. 2002, \mnras, 336, 1082
685:
686: \bibitem[Sheth \& Tormen(1999)]{she99} Sheth, R.~K. and Tormen G. 1999,
687: \mnras, 308, 119
688:
689: \bibitem[Spergel et al.(2003)]{WMAP1} Spergel, D.~N. et al. 2003, \apjs, 148, 175
690:
691: \bibitem[Springel et al.(2005)]{MR05} {Springel}, V. et al. 2005,
692: \nat, 435, 629
693:
694: \bibitem[Volonteri et al.(2003)]{vol03} {{Volonteri}, M., {Haardt}, F. and {Madau}, P.} 2003, \apj, 582, 559
695:
696: \bibitem[Trenti \& Stiavelli(2007)]{ts07a} Trenti M. and Stiavelli, M. 2007, \apj, 667, 38
697: \bibitem[Trenti \& Stiavelli(2008)]{ts08} Trenti M. and Stiavelli, M. 2008, \apj, 676, 767
698:
699: \bibitem[White \& Springel(2000)]{whi00} {White}, S.~D.~M. and
700: {Springel}, V. 2000, in ``The First Stars'',{Weiss}, A. and {Abel},
701: T.~G. and {Hill}, V. eds., Springer-Verlag.
702:
703: \end{thebibliography}
704:
705: \clearpage
706:
707: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
708: \begin{figure}
709: \plotone{f1.eps} \caption{Mass at $z=0$ of the descendants
710: of the most massive dark matter halos at $z=6.18$ in the Millennium
711: Run \citep{MR05}. The shaded area within the region with the highest
712: density of points represents for a given $z=6.18$ mass the 68\%
713: confidenence level interval for the mass of the $z=0$ descendant,
714: while the blue central line is the median of the distribution. Some
715: of the most massive halos at $z>6$ have accreted only relatively
716: little mass and a considerable scatter is present. The upper and
717: right axes in the plot translate the halo mass into the
718: dimensionless variable $\nu = \delta_c^2/ \sigma^2(M)$, used in the extended
719: Press-Schechter formalism.}\label{fig:MR}
720: \end{figure}
721:
722: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
723: \begin{figure}
724: \plotone{f2.eps} \caption{Mass at redshift $z=6.18$ of the most
725: massive progenitor of the most massive halos at $z=0$ for the
726: Millennium Run \citep{MR05}. $\nu$ values are given as in
727: Fig.~\ref{fig:MR}. The few points with $m_{z=6.18} = 8.5 \cdot 10^9
728: h^{-1} \mathrm{M_{\sun}}$ are associated to those $z=0$ halos that
729: do not have a $z=6.18$ progenitor identified in the simulation (that
730: is its below the minimum halo mass of $8.5 \cdot 10^{10} h^{-1}
731: \mathrm{M_{\sun}}$ --- corresponding to $100$ particles). Given the
732: considerable scatter in these plots, adaptive refinement of the most
733: massive $z=0$ halo in a simulation would not necessarily lead to the
734: most massive $z \approx 6$ halo of that box. }\label{fig:MR1}
735: \end{figure}
736:
737: \clearpage
738:
739:
740: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
741: \begin{figure}
742: \plotone{f3.eps}\caption{Random walks generated using a Millennium run
743: cosmology. The green shaded area represents for any value of
744: $\sigma^2(M)$ --- evaluated at $z=0$ --- the area enclosing 68\% of
745: the walks and thus has amplitude $\pm \sigma(M)$. The red and blue
746: walks have been selected from $10^3$ random realizations as (i) red:
747: the walk with the largest overdensity at $M=3 \cdot 10^5 h^{-1}
748: \mathrm{M_{\sun}}$ and (ii) blue: the walk with the largest
749: overdensity at $M=2 \cdot 10^{15} h^{-1} \mathrm{M_{\sun}}$.
750: }\label{fig:random_walk}
751: \end{figure}
752:
753: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
754:
755:
756: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
757: \begin{figure}
758: \plotone{f4.eps} \caption{Main panel: Contours for the probability
759: that a halo of mass $M_1=5 \cdot 10^{12} h^{-1} \mathrm{M_{\sun}}$
760: at redshift $z_1=6.18$ is more massive than $M_2$ at redshift
761: $z_2<z_1$ [$P(M>M_2,z_2|M_1,z_1)$]. The black line is the median of
762: the distribution, the blue lines the $1 \sigma$ confidence levels
763: and the green lines the $2 \sigma$ confidence levels. Red dotted
764: lines: contour lines for the number density of dark matter halos in
765: a $10^8 (\mathrm{Mpc}/h)^3$ volume (1,10,100,1000,10000 halos from
766: top to bottom), obtained using the Sheth \& Tormen mass
767: function. Small inset: Evolution of the main panel median line
768: represented in terms of the extended Press-Schechter variable
769: $\nu$. From this inset it is immediately clear that the
770: descendant of a rare density peak progressively becomes more
771: common.}\label{fig:EPS}
772: \end{figure}
773:
774: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
775:
776:
777: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
778: \begin{figure}
779: \plotone{f5.eps} \caption{As in fig.~\ref{fig:EPS} but for a
780: dark matter halo of mass $M_1= 10^{6} h^{-1} \mathrm{M_{\sun}} $
781: formed at $z_1=40$.}\label{fig:EPS_PopIII}
782: \end{figure}
783:
784: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
785:
786:
787:
788: \end{document}
789: