0710.0007/ms.tex
1: \documentclass[apj]{emulateapj}
2: 
3: %===========================================================================
4: % NPH defined commands
5: \newcommand{\ii}{$i'$}
6: \newcommand{\iband}{$i_{775}$}
7: \newcommand{\zz}{$z'$}
8: \newcommand{\zband}{$z_{850}$}
9: \newcommand{\uu}{$U$}
10: \newcommand{\bb}{$B$}
11: \newcommand{\vv}{$V$}
12: \newcommand{\jj}{$J$}
13: \newcommand{\iz}{\ensuremath{(i'\!-\!z')}}
14: \newcommand{\gi}{\ensuremath{(g\!-\!i')}}
15: \newcommand{\zj}{\ensuremath{(z'-J)}}
16: \newcommand{\cge}{$\gtrsim$}
17: \newcommand{\cle}{$\lesssim$}
18: \newcommand{\etal}{{et\thinspace al.}}
19: \newcommand{\cg}{c.\thinspace g.}
20: \newcommand{\Ang}{\AA\thinspace}
21: \newcommand{\SN}{$S/N$}
22: \newcommand{\Ho}{$H_{0}$}
23: \newcommand{\super}[1]{$^{#1}$}
24: \newcommand{\sub}[1]{$_{#1}$}
25: \newcommand{\tabref}[1]{Table~\ref{#1}}
26: \newcommand{\figref}[1]{Figure~\ref{#1}}
27: \newcommand{\secref}[1]{\S~\ref{#1}}
28: \newcommand\arcspt   {{$\buildrel{\prime\prime}\over .$}}
29: \newcommand\Msun     {{\ $M_{\odot}$} }
30: 
31: %===========================================================================
32: 
33: \begin{document}
34: 
35: \title{Surface  Brightness  Profiles of  Composite  Images of  Compact \\
36: Galaxies  at \boldmath{$\lowercase{z}\!\simeq\!4\!-\!6$}  in the  HUDF}
37: 
38: \shorttitle{Composite Surface Brightness Profiles at $z\!\simeq\!4\!-\!6$}
39: 
40: \author{N.  P.  Hathi\altaffilmark{1},  R. A.  Jansen\altaffilmark{2,1},
41: R.   A.   Windhorst\altaffilmark{2,1},   S.  H.   Cohen\altaffilmark{2}, \\
42: W.   C.  Keel\altaffilmark{3},   M.   R.  Corbin\altaffilmark{4}   and
43: R. E. Ryan Jr.\altaffilmark{1}}
44: 
45: \altaffiltext{1}{Department  of  Physics,  Arizona  State  University,
46: Tempe, AZ 85287-1504, USA}
47: 
48: \altaffiltext{2}{School of Earth  and Space Exploration, Arizona State
49: University, Tempe, AZ 85287-1404, USA}
50: 
51: \altaffiltext{3}{Department  of Physics  and Astronomy,  University of
52: Alabama, Box 870324, Tuscaloosa, AL 35487, USA}
53: 
54: \altaffiltext{4}{U. S. Naval Observatory, 10391 W. Naval Observatory
55: Road, Flagstaff, AZ 86001-8521, USA}
56: 
57: \email{Nimish.Hathi@asu.edu}
58: \shortauthors{Hathi et al}
59: 
60: %--------------------------------------------------
61: 
62: \begin{abstract}
63: 
64: The Hubble  Ultra Deep Field  (HUDF) contains a significant  number of
65: \bb-, \vv- and  \ii-band dropout objects, many of  which were recently
66: confirmed to  be young star-forming  galaxies at $z\!\simeq\!4\!-\!6$.
67: These galaxies are too  faint individually to accurately measure their
68: radial surface brightness profiles.   Their average light profiles are
69: potentially of  great interest,  since they may  contain clues  to the
70: time since  the onset of  significant galaxy assembly.   We separately
71: co-add \vv, \ii- and \zz-band  HUDF images of sets of $z\!\simeq\!4,5$
72: and $6$  objects, pre-selected to have nearly  identical compact sizes
73: and the  roundest shapes.  From these  stacked images, we  are able to
74: study the average(d) radial structure  of these objects at much higher
75: signal-to-noise ratio  than possible  for an individual  faint object.
76: Here we explore the reliability and usefulness of a stacking technique
77: of compact  objects at $z\!\simeq\!4\!-\!6$ in the  HUDF.  Our results
78: are:  (1) image  stacking provides  reliable and  reproducible average
79: surface  brightness profiles;  (2) the  shape of  the  average surface
80: brightness profile  shows that even  the faintest $z\!\simeq\!4\!-\!6$
81: objects are  \emph{resolved}; and  (3) if late-type  galaxies dominate
82: the  population  of  galaxies  at  $z\!\simeq\!4\!-\!6$,  as  previous
83: \emph{HST}  studies have  shown for  $z\!\lesssim\!4$, then  limits to
84: dynamical age  estimates for these galaxies from  their profile shapes
85: are comparable with  the SED ages obtained from  the broadband colors.
86: We  also present accurate  measurements of  the sky-background  in the
87: HUDF and its associated 1$\sigma$ uncertainties.
88: 
89: \end{abstract}
90: 
91: \keywords{galaxies: high-redshift --- galaxies: structure --- galaxies: formation}
92: 
93: %--------------------------------------------------
94: 
95: \section{Introduction}\label{introduction}
96: 
97: In  the last  decade,  ground  and space  based  observations of  high
98: redshift  galaxies  have  begun  to  outline  the  process  of  galaxy
99: assembly.   The details of  that process  at high  redshifts, however,
100: remain poorly constrained.  There  is increasing support for the model
101: of  galaxy formation,  in  which the  most  massive galaxies  assemble
102: earlier      than       their      less      massive      counterparts
103: \citep[e.g.][]{cowi96,guzm97,koda04,mcca04}.   A detailed  analysis of
104: the  `fossil record'  of  the current  stellar  populations in  nearby
105: galaxies   selected   from  the   \emph{Sloan   Digital  Sky   Survey}
106: \citep[SDSS;][]{york00} provides  strong evidence for  this downsizing
107: picture  \citep{heav04,pant07}.   The  increasing number  of  luminous
108: galaxies  spectroscopically   confirmed  to  be   at  $z\!\simeq\!6.5$
109: \citep[e.g.][]{hu02,  kodi03,  kurk04,  rhoa04,  ster05,  tani05},  or
110: \cle0.9 Gyr  after the Big  Bang, also supports this  general picture.
111: In an  alternate hierarchical scenario, arguments have  been made that
112: significant number  of low luminosity  dwarf galaxies were  present at
113: these times,  and were the main  contributor to finish  the process of
114: reionization   of  the  intergalactic   medium  \citep{yan04a,yan04b}.
115: However,  there  is  presently  little information  on  the  dynamical
116: structure of  these or  other galaxies at  $z\!\simeq\!6$.  It  is not
117: clear  whether  these  objects  represent isolated  disk  systems,  or
118: collapsing spheroids, mergers or other dynamically young objects.
119: 
120: \citet{ravi06}  used deep, multi-wavelength  images obtained  with the
121: \emph{Hubble Space Telescope} (\emph{HST}) Advanced Camera for Surveys
122: (ACS) as part  of the Great Observatories Origins  Deep Survey (GOODS)
123: to analyze 2-D surface brightness distributions of the brightest Lyman
124: Break Galaxies (LBGs)  at $2.5\!<\!z\!<\!5$.  They distinguish various
125: morphologies  based on the  S\'{e}rsic index  $n$, which  measures the
126: shape of  the azimuthally  averaged surface brightness  profile (where
127: $n$=1  for exponential  disks and  $n$=4  for a  de Vaucouleurs  law).
128: \citet{ravi06} find that 40\% of the LBGs have light profiles close to
129: exponential, as seen for disk  galaxies, and only $\sim$30\% have high
130: $n$,  as seen  in  nearby  spheroids.  They  also  find a  significant
131: fraction ($\sim$30\%) of galaxies with light profiles \emph{shallower}
132: than  exponential, which appear  to have  multiple cores  or disturbed
133: morphologies, suggestive  of close  pairs or on-going  galaxy mergers.
134: Distinction  between  these possible  morphologies  and, therefore,  a
135: better estimate of the formation  redshifts of the systems observed at
136: $z\!\simeq\!4\!-\!6$  in  particular,  is  important for  testing  the
137: galaxy assembly  picture, and for  the refinement of  galaxy formation
138: models.
139: 
140: One possible technique involves the radial surface brightness profiles
141: of  the most  massive objects  --- those  that will  likely  evolve to
142: become  the massive  elliptical galaxies,  which  we see  in place  at
143: redshifts $z\!\lesssim\!2$  \citep{driv98,vand03,vand04}.  This can be
144: analytically   understood  in  the   context  of   the  \citet{lynd67}
145: relaxation formalism and the numerical galaxy formation simulations of
146: \citet{vana82},  which  describe  collisionless collapse  and  violent
147: relaxation as the formation mechanism for elliptical galaxies.  As the
148: time-scale for  relaxation is shorter in  the inner than  in the outer
149: parts of a galaxy, convergence toward a $r^{1/n}$-profile will proceed
150: from  the  inside  to  progressively  larger  radii  at  later  times.
151: Moreover,  \citet{korm77} has  shown that  tidal perturbations  due to
152: neighbors can  cause the radial surface brightness  profile to deviate
153: from a  pure de~Vaucouleurs  profile in the  outer parts of  a galaxy.
154: This implies  that the radius where surface  brightness profiles start
155: to deviate significantly from  an $r^{1/n}$ profile \emph{might} serve
156: as a ``\emph{virial  clock}'' that traces the time  since the onset of
157: the last major  merger, accretion events or global  starburst in these
158: objects.
159: 
160: Image   stacking  methods   have  been   used  extensively   on  X-ray
161: \citep{nand02,bran01}  and radio  \citep{geor03,whit07} data  to study
162: the mean properties (e.g. flux, luminosity) of well-defined samples of
163: sources  that are  otherwise too  faint to  be  detected individually.
164: \citet{pasc96} applied  such a  stacking method to  a large  number of
165: optically very  faint, compact objects at $z\!=\!2.39$  to trace their
166: ``average''   structure.    This   approach   was  also   applied   by
167: \citet{zibe04} to  detect the presence  of faint stellar  halos around
168: disk  galaxies selected  from  the  SDSS.  An  attempt  to apply  this
169: technique  to  high  redshift   galaxies  in  the  Hubble  Deep  Field
170: \citep[HDF;][]{will96}  was  not  conclusive  (H.   Ferguson;  private
171: communication) due to the  poorer spatial sampling and shallower depth
172: of   the    HDF   compared   to   the   Hubble    Ultra   Deep   Field
173: \citep[HUDF;][]{beck06}.
174: 
175: In this paper, we use  the exceptional depth and fine spatial sampling
176: of the HUDF  to study the potential of  this image stacking technique,
177: and will estimate limits to dynamical ages of faint, young galaxies at
178: $z\!\simeq\!4\!-\!6$.  The HUDF  reaches $\sim$1.5~mag deeper than the
179: equivalent  HDF  exposure in  the  \ii-band,  and  has better  spatial
180: sampling than the HDF.  The  HUDF depth also allows us to characterize
181: the sky background very accurately, which is critical for successfully
182: using  a  stacking  method  to  measure  the  mean  surface-brightness
183: profiles for these faint young galaxies.
184: 
185: This  paper  is  organized  as follows:  In  \secref{observations}  we
186: summarize the HUDF observations, and in \secref{sample} we discuss the
187: selection   of    our   $z\!\simeq\!4,5$   and    $6$   samples.    In
188: \secref{analysis}  we  describe  our  data  analysis,  which  includes
189: accurately  measuring the 1$\sigma$  sky-subtraction error,  the image
190: stacking method to generate  mean surface-brightness profiles, and our
191: test of  its reliability.  In \secref{results} we  present and discuss
192: our  results in terms  of the  average surface-brightness  profiles of
193: $z\!\simeq\!4\!-\!6$ galaxies, and  in \secref{conclusion} we conclude
194: with a summary of our results.
195: 
196: Throughout  this paper we  refer to  the \emph{HST}/ACS  F435W, F606W,
197: F775W,  and F850LP  filters as  the \bb-,  \vv-, \ii-,  and \zz-bands,
198: respectively.  We assume a \emph{Wilkinson Microwave Anisotropy Probe}
199: (WMAP)  cosmology   of  $\Omega_m$=0.24,  $\Omega_{\Lambda}$=0.76  and
200: \Ho=73~km~s$^{-1}$~Mpc$^{-1}$, in  accord with the  most recent 3-year
201: WMAP results  of \citet{sper07}.  This  implies a current age  for the
202: Universe  of 13.65~Gyr.   All magnitudes  are given  in the  AB system
203: \citep{oke83}.
204: 
205: %--------------------------------------------------
206: 
207: \section{Observations}\label{observations}
208: 
209: The HUDF  contains \cge100 objects that are  \ii-band dropouts, making
210: them candidates for  galaxies at $z\!\simeq\!6$ \citep{bouw04, bouw06,
211: bunk04, yan04b}.   Similarly, there are  larger numbers of  objects in
212: the  HUDF  that are  \bb-band  dropouts  (415  in total)  or  \vv-band
213: dropouts  (265   in  total),  and  are  candidates   for  galaxies  at
214: $z\!\simeq\!4$  and $z\!\simeq\!5$, respectively.   \citet{beck06} and
215: \citet{bouw07} find  similar number of  \bb- and \vv-band  dropouts in
216: the  HUDF.  A significant  fraction of  these objects  to AB\cle27~mag
217: have  recently  been  spectroscopically  confirmed to  have  redshifts
218: $z\!\simeq\!4\!-\!6$ through  the detection of  Ly$\alpha$ emission or
219: identifying their  Lyman break  \citep{ malh05,dow07}. We  discuss our
220: detailed  drop-out   selection  criteria  below.   Despite  the  depth
221: (AB\cle29.5~mag)  of the  HUDF images,  however, these  objects appear
222: very  faint and with  little, if  any, discernible  structural detail.
223: Visual  inspection of all  these objects  shows their  morphologies to
224: divide into four broad  categories: symmetric, compact, elongated, and
225: amorphous.
226: 
227: %--------------------------------------------------
228: 
229: \section{Sample Selection}\label{sample}
230: 
231: We  construct  three separate  catalogs  for these  $z\!\simeq\!4,5,6$
232: galaxy candidates, selecting  only the \emph{isolated}, \emph{compact}
233: and \emph{symmetric} galaxies. We  exclude objects with obvious nearby
234: neighbors, to  avoid a bias  due to dynamically disturbed  objects and
235: complications    due   to   chance    superpositions.    \figref{fig1}
236: demonstrates  that  our  completeness  limit  for  $z\!\simeq\!4$  and
237: $z\!\simeq\!5$  objects  is  AB\cle29.3~mag,  and  for  $z\!\simeq\!6$
238: objects  it  is AB\cle29.0~mag.   Therefore,  all  three catalogs  are
239: complete  to  AB\cle29.0~mag,  which  is  equivalent  to  at  least  a
240: 10$\sigma$ detection  for objects that  are nearly point  sources. For
241: each   object  in   our  $z\!\simeq\!4,5,6$   samples,   we  extracted
242: 51$\times$51 pixel postage stamps  (which at 0\farcs03 pix$^{-1}$ span
243: $1\farcs53$ on a  side) from the HUDF \vv,  \ii\thinspace and \zz-band
244: images, respectively.  Each postage stamp was extracted  from the full
245: HUDF, such that the centroid of an object (usually coincident with the
246: brightest pixel) was at the center of that stamp.
247: 
248: %--------------------------------------------------
249: \begin{figure}
250: \epsscale{1.0}
251: \plotone{f1.eps}
252: \caption{The HUDF  number counts for all $z\!\simeq\!4,  5, 6$ objects
253: before  the sub-selection  of  compact isolated  $z\!\simeq\!4, 5,  6$
254: objects was  made.  The  vertical dotted line  shows the  magnitude to
255: which the number counts of  all these redshifts are complete. The area
256: of  the  HUDF  is  3.15$\times$10$^{-3}$  deg$^2$.}\label{fig1}
257: \end{figure}
258: %--------------------------------------------------
259: 
260: \subsection{\boldmath {The $z\!\simeq\!4$ and $z\!\simeq\!5$ Objects (\bb-, \vv-band dropouts)}}
261: 
262: We used the \ii-band  selected $BVi'z'$ HUDF catalog \citep{beck06} to
263: select  the  $z\!\simeq\!4$  and  $z\!\simeq\!5$  objects.   With  the
264: \texttt{HyperZ} code \citep{bolz00},  we computed photometric redshift
265: estimates, using the magnitudes and associated uncertainties tabulated
266: in   the   HUDF    catalog.    All   objects   with   3.5$\leq\!z_{\rm
267: phot}\!\leq$4.5 were assigned to the bin of $z\!\simeq\!4$ candidates,
268: and all  objects with 4.5$\leq\!z_{\rm  phot}\!\leq$5.5 to the  bin of
269: $z\!\simeq\!5$ candidates.
270: 
271: We  then   applied  color  criteria,  similar  to   those  adopted  by
272: \citet{giav04},   to   select  the   \bb   ($z\!\simeq\!4$)  and   \vv
273: ($z\!\simeq\!5$) dropout samples.  For \bb-band dropouts, we require:
274: \begin{displaymath}
275:       \left\{ \begin{array} {ll}
276:         (B-V) \ge 1.2 + 1.4 \times (V-z') \quad \hbox{mag} \\ 
277:         \hbox{and}\quad (B-V) \ge 1.2  \quad \hbox{mag} \\
278:         \hbox{and}\quad (V-z') \le 1.2 \quad \hbox{mag} 
279: 		\end{array} \right.
280: \end{displaymath} 
281: For \vv-band dropouts, the following color selection was applied:
282: \begin{displaymath}
283:       \left\{ \begin{array} {ll}
284: 	(V-i') > 1.5 + 0.9 \times (i'-z') \; \; \hbox{or} \; \; (V-i') > 2.0  \quad \hbox{mag} \\
285:         \hbox{and}\quad (V-i') \ge 1.2  \quad \hbox{mag} \\
286: 	\hbox{and}\quad (i'-z') \le 1.3 \quad \hbox{mag} 
287:                 \end{array} \right.
288: \end{displaymath}
289: We  note, that  only objects  satisfying \emph{both}  color \emph{and}
290: photometric   redshift  criteria   were  selected   in   our  samples.
291: \citet{vanz06}  using  VLT/FORS2  observed  $\sim$100 \bb-,  \vv-  and
292: \ii-band  dropout  objects in  the  Chandra  Deep  Field South  (CDFS)
293: selected based on above mentioned color criteria \citep{giav04}.  They
294: have spectroscopically confirmed $>$90\% of their high redshift galaxy
295: candidates.   Therefore, we expect  only a  small number  ($<$10\%) of
296: contaminants in  our sample  of dropouts.  One  or two objects  in our
297: final  sample  could be  such  contaminants,  but  because we  have  3
298: different  realizations  of  10  objects (3$\times$10),  each  showing
299: similar profiles, they do not appear to affect our results.
300: 
301: The $z\!\simeq\!4$  sample has  415 objects, while  the $z\!\simeq\!5$
302: sample has 265 objects.   In \figref{fig2}ab, we show the distribution
303: of the  FWHM and ellipticity, $\epsilon=(1-b/a)$, measured  in each of
304: the two samples  using \texttt{SExtractor} \citep{bert96}.  We further
305: constrained  our samples  by  imposing limits  on  compactness and  on
306: roundness of FWHM $\le 0\farcs3$  and $\epsilon \le 0.3$.  Again, this
307: is   to  minimize  the   probability  that   the  $z\!\simeq\!4\!-\!5$
308: candidates  are significantly dynamically  disturbed, and  to maximize
309: the probability of selecting  physically similar objects.  Our goal is
310: to find  the visibly most  symmetric, least disturbed systems  for the
311: current  study.    This  sub-selection  leaves  204   objects  in  the
312: $z\!\simeq\!4$ sample  and 102  objects in the  $z\!\simeq\!5$ sample.
313: Most of these  objects are faint, and are only a  few pixels across in
314: size, and,  hence, have larger uncertainties in  their measurements of
315: FWHM and ellipticity.  Therefore, we also checked our objects visually
316: to   eliminate  any   possibility  of   our  selected   objects  being
317: contaminated by  unrelated nearby objects, being  clearly extended, or
318: objects with complex morphologies.
319: 
320: %--------------------------------------------------
321: \begin{figure}
322: \epsscale{1.0}
323: \plotone{f2.eps}
324: \caption{Ellipticity,   ($1-b/a$),  versus   object   FWHM,  for   all
325: $z\!\simeq\!4$   {\bf   {(a)}},   $z\!\simeq\!5$   {\bf   {(b)}}   and
326: $z\!\simeq\!6$ {\bf {(c)}} objects selected in the HUDF.  Measurements
327: were  performed  in  \ii-band  for $z\!\simeq\!4$  and  $z\!\simeq\!5$
328: objects, while  we used the \zz-band for  $z\!\simeq\!6$ objects.  The
329: FWHM of a stellar image/PSF  is $\sim$3 pixels or 0\farcs09, indicated
330: by the leftmost hatched area  in each panel. Objects within the shaded
331: area meet  our additional  selection criteria on  roundness ($\epsilon 
332: \le   0.3$)    and   compactness   (FWHM   $\le    0\farcs3$   or   10
333: pixels).}\label{fig2}
334: \end{figure}
335: %--------------------------------------------------
336: 
337: \subsection{\boldmath {$z\!\simeq\!6$ Objects (\ii-band dropouts)}}
338: 
339: \citet{yan04b} found 108  possible 5.5$\leq\!z\!\leq$6.5 candidates in
340: the HUDF to $m_{AB}$($z_{850}$)=30.0~mag. \citet{bunk04} independently
341: found  the brightest  54  of these  108  $z\!\simeq\!6$ candidates  to
342: AB=28.5~mag.  Similarly, deep \emph{HST}/ACS grism spectra of the HUDF
343: \ii-band dropouts confirm \cge90\%  of these objects at AB\cle27.5 mag
344: to be  at $z\!\simeq\!6$ \citep{malh05,hath07}.  Using  the catalog of
345: \citet{yan04b},  we extracted  108 postage  stamps,  each 51$\times$51
346: pixels in size, from the HUDF \zz-band image.
347: 
348: Like  for  the $z\!\simeq\!4$  and  $z\!\simeq\!5$  objects, for  each
349: $z\!\simeq\!6$ object  we measured  its \zz-band FWHM  and ellipticity
350: using   \texttt{SExtractor}.    \figref{fig2}c   shows  the   measured
351: ellipticity  versus FWHM  for  all 108  $z\!\simeq\!6$ candidates.   A
352: smaller sample of 67 objects satisfies our constraints on the FWHM and
353: ellipticity.  Further visual inspection,  to make sure that our sample
354: has only isolated, compact and round objects, leaves 30 objects in our
355: $z\!\simeq\!6$  sample.  We  therefore  imposed a  sample  size of  30
356: objects also on the two lower redshift bins after visual inspection.
357: 
358: The  results  in  this  paper  are therefore  based  on  approximately
359: (30/415)$\sim$7\%,  (30/265)$\sim$11\%, and (30/108)$\sim$28\%  of the
360: total $z\!\simeq\!4,5$ and $6$ galaxy populations.
361: 
362: %--------------------------------------------------
363: 
364: \section{Results}\label{analysis}
365: 
366: \subsection{The HUDF Sky Surface-Brightness Level and its rms Variation}\label{skyerror}
367: 
368: For  the  present  work,  it  is \emph{critical}  that  we  accurately
369: characterize  the  sky-background, and  correctly  propagate the  true
370: 1$\sigma$ errors due to the subtraction of this sky-background. In the
371: following, we  will pursue  two complimentary approaches  to determine
372: the sky surface-brightness, and  compare the results. Here, we discuss
373: the \zz-band measurements in detail.
374: 
375: We first measured the sky-background in each of the 415 $z\!\simeq\!4$
376: object  stamps  (`local'  sky  measurements).   The  Interactive  Data
377: Language  (IDL\footnote{IDL  Website http://www.ittvis.com/index.asp})
378: procedure  \texttt{SKY/MMM.pro}\footnote{Part  of  the  IDL  Astronomy
379: User's Library,  see: http://idlastro.gsfc.nasa.gov/homepage.html} was
380: used to  measure the sky-background.   This procedure is  adapted from
381: the \texttt{DAOPHOT} \citep{stet87} routine of the same name and works
382: as follows.   First, the average and  sigma are obtained  from the sky
383: pixels.  Second,  these values are  used to eliminate outliers  with a
384: low  probability.   Third, the  values  are  then  recomputed and  the
385: process is repeated up to  20 iterations.  If there is a contamination
386: due to an object, then the contamination is estimated by comparing the
387: mean and median of the remaining sky pixels to get the true sky value.
388: The output of this procedure is the modal sky-level in the image.
389: 
390: \figref{fig3}c  shows a  histogram of  the \zz-band  modal  sky values
391: obtained from all  415 object stamps extracted from  the drizzled HUDF
392: images.   The 1$\sigma$  uncertainty in  the sky,  $\sigma_{\rm sky}$,
393: determined   from    a   Gaussian    fit   to   the    histogram,   is
394: $2.19\times10^{-5}$  electrons   sec$^{-1}$  in  the   \zz-band.   The
395: sky-background level  within the HUDF  was obtained from  the original
396: flat-fielded ACS images, because the final co-added HUDF data products
397: are sky-subtracted.   The header parameters MDRIZSKY  and EXPTIME were
398: used to  obtain the actually  observed sky-value. MDRIZSKY is  the sky
399: value  in   electrons  ($e^-$)  computed  by   the  MultiDrizzle  code
400: \citep{koek02}, while EXPTIME is the total exposure time for the image
401: in seconds, so that the average sky-value in the HUDF has the units of
402: $e^-$   sec$^{-1}$.   \figref{fig4}d  shows   the  histogram   of  the
403: sky-values  obtained from  288 HUDF  \zz-band  flat-fielded exposures.
404: The average  value of  the sky background,  I$_{\rm sky}$,  is 0.02051
405: $e^-$  sec$^{-1}$ pix$^{-1}$.   That  sky-value is  measured from  the
406: flat-fielded  individual  ACS images  with  pixel  sizes of  0\farcs05
407: pix$^{-1}$  and  hence, in  the  following  calculations, the  average
408: sky-value is  multiplied by  a factor of  (0.030/0.05)$^2$=0.60$^2$ to
409: obtain the corresponding average sky-value for the HUDF drizzled pixel
410: size of  0\farcs030 pix$^{-1}$.  Using  these values, we  estimate the
411: relative rms random sky-subtraction error as follows:
412: \[
413: \Sigma_{\rm ss,ran} = \frac{\sigma_{\rm sky,ran}}{{\rm I}_{\rm sky}} = \frac{2.19\times10^{-5}}{2.05\times10^{-2}\ \cdot\ 0.60^2} = 2.97\times10^{-3} 
414: \]
415: 
416: %--------------------------------------------------
417: \begin{figure}
418: \epsscale{1.0}
419: \plotone{f3.eps}
420: \caption{Distribution  of  the  modal  sky background  level  used  to
421: estimate the 1$\sigma$  uncertainty in that level, as  measured in the
422: 415  $z\!\simeq\!4$ \emph{object stamps}  extracted from  the drizzled
423: HUDF images  {\bf {(a)}}  for \vv-band, {\bf  {(b)}} for  \ii-band and
424: {\bf {(c)}} for  \zz-band.  The mean ($\mu$) and  the sigma ($\sigma$)
425: of the  best-fit Gaussian  to these distributions  are also  shown in
426: each panel.}\label{fig3}
427: \end{figure}
428: %--------------------------------------------------
429: \begin{figure}
430: \epsscale{1.0}
431: \plotone{f4.eps}
432: \caption{The \emph{actual sky values} measured using header parameters
433: MDRIZSKY and EXPTIME from  flat-fielded HUDF exposures.  {\bf (a)} for
434: \bb-band  using  112 exposures.   {\bf  (b)}  for  \vv-band using  112
435: exposures.  {\bf (c)} for \ii-band using 288 exposures.  {\bf (d)} for
436: \zz-band  using  288  exposures.   The  mean  ($\mu$)  and  the  sigma
437: ($\sigma$) of the best-fit  Gaussian to these distributions are shown
438: in each panel.}\label{fig4}
439: \end{figure}
440: %-------------------------------------------------- 
441: 
442: The measured average sky background level can then be expressed as the \zz-band
443: sky surface brightness as follows:
444: \begin{eqnarray*}
445: \mu_{z'} &=& 24.862 - 2.5 \cdot \log\left(\frac{0.0205\ \cdot\ 0.60^2}{0.030^2} \right) \\
446:          &=& 22.577 \pm 0.003 \ \hbox{mag arcsec}^{-2} \qquad\qquad\qquad\qquad\quad\null 
447: \end{eqnarray*}
448: where  24.862 is the  ACS/WFC \zz-band  AB zero-point,  and 0\farcs030
449: pixel$^{-1}$ is the drizzled pixel scale.  This is consistent with the
450: values  obtained  by  extrapolating  the on-orbit  $BVI$  sky  surface
451: brightness  of \citet{wind94,wind98} to  \zz, with  the sky-background
452: estimates from  the ACS  Instrument Handbook \citep{gonz05},  and with
453: the colors obtained by  convolving the filter transmission curves with
454: the  solar  spectrum.   \tabref{table1}  gives the  measured  electron
455: detection rate,  surface brightness and  colors of the  sky background
456: with  their  corresponding  errors  for  the HUDF  $BVi'z'$  bands  as
457: calculated from \figref{fig3}  and \figref{fig4}.  The contribution of
458: the zodiacal  background dominates the total  sky-background, which we
459: find to  be only $\sim$10\%  redder in (\vv--\ii) and  (\ii--\zz) than
460: the  Sun.   The  \zz-band  surface  brightness  corresponding  to  the
461: 1$\sigma$ sky-subtraction uncertainty is therefore:
462: \begin{eqnarray*}
463: \mu_{z'} - 2.5\cdot \log (\Sigma_{\rm ss,ran}) &=& 22.577 - 2.5\cdot \log (2.97\times10^{-3}) \\
464:                                               &=& 28.895\ \hbox{mag arcsec}^{-2}
465: \end{eqnarray*}
466: 
467: Next, we measure the sky-background from 415 `\emph{blank}' sky stamps
468: (51$\times$51  pixel) distributed  throughout the  HUDF  (`global' sky
469: measurements).   We measure  the  sky background  using  the same  IDL
470: algorithm as used above.
471: 
472: \figref{fig5}c shows the histogram  of the measured \zz-band modal sky
473: values.  A Gaussian  distribution was fit to this  histogram, giving a
474: sky-sigma of $2.00\times10^{-5}$  $e^-$ sec$^{-1}$.  The average value
475: of the  sky remains 0.02051 $e^-$  sec$^{-1}$ (\figref{fig4}d).  Using
476: these   values,   we   can   estimate  a   relative   rms   systematic
477: sky-subtraction error as follows:
478: \[
479: \Sigma_{\rm ss,sys} = \frac{\sigma_{\rm sky,sys}}{{\rm I}_{\rm sky}} = \frac{2.00\times10^{-5}}{2.05\times10^{-2}\ \cdot\ 0.60^2} = 2.71\times10^{-3}
480: \]
481: Since  the   \zz-band  sky  surface  brightness   remains  22.577  mag
482: arcsec$^{-2}$, this gives us  for the surface brightness corresponding
483: to the 1$\sigma$ sky subtraction uncertainty:
484: \begin{eqnarray*}
485: \mu_{z'} - 2.5\cdot \log (\Sigma_{\rm ss,sys}) &=& 22.577 - 2.5\cdot \log (2.71\times10^{-3}) \\ 
486:                                                &=& 28.995\ \hbox{mag arcsec}^{-2}
487: \end{eqnarray*}
488: 
489: %--------------------------------------------------
490: \begin{figure}
491: \epsscale{1.0}
492: \plotone{f5.eps}
493: \caption{Distribution  of  the  modal  sky background  level  used  to
494: estimate the 1$\sigma$  uncertainty in that level, as  measured in 415
495: \emph{`blank'  51$\times$51  pixel  sky  stamps}  extracted  from  the
496: drizzled HUDF  images {\bf (a)}  for \vv-band, {\bf (b)}  for \ii-band
497: and {\bf (c)} for \zz-band.  The mean ($\mu$) and the sigma ($\sigma$)
498: of to the best-fit Gaussian  to these distributions are shown in each
499: panel.}\label{fig5}
500: \end{figure}
501: %--------------------------------------------------
502: 
503: From  these two  complementary approaches,  we can  conclude  that all
504: surface     brightness    measurements    become     unreliable    for
505: surface-brightness    levels     fainter    than    28.95$\pm$0.05~mag
506: arcsec$^{-2}$ in the \zz-band. We have also experimented with slightly
507: larger cutouts (75$\times$75 pixels instead of 51$\times$51 pixels) to
508: estimate  the sky-subtraction  error.  We  find that  with  the larger
509: cutouts,  the  surface   brightness  corresponding  to  the  1$\sigma$
510: sky-subtraction  error is  $\sim$0.1--0.2  mag arcsec$^{-2}$  fainter.
511: For larger cutouts  we expect this surface brightness  to be $\sim$0.4
512: mag fainter  but we  find about 0.1--0.2  mag fainter.  This  might be
513: because of  residual systematic errors in the  HUDF images. Therefore,
514: we are  at the limit  of accurately measuring this  surface brightness
515: and hence, we  will here quote the conservative  brighter limit of the
516: surface  brightness corresponding  to  this 1$\sigma$  sky-subtraction
517: error.   Expected  contributions to  this  surface  brightness due  to
518: uncertainties in  the bias  level determinations, which  correspond to
519: $\sim$0.001  counts  sec$^{-1}$ for  typical  HUDF  exposures (A.   M.
520: Koekemoer; private communication), are less than 1\%.
521: 
522: \figref{fig5}  clearly  shows  that  the  distribution  of  the  modal
523: sky-values is  not as symmetric  around zero as in  \figref{fig3}, and
524: hence, the use of a `global' sky value for the HUDF is not as reliable
525: as `local'  sky measurements.   Therefore, for the  surface brightness
526: profiles and the following  discussion, we will adopt the \emph{local}
527: 1$\sigma$ random sky-subtraction error for all objects in our study.  
528: 
529: The average  modal sky values and  their 1$\sigma$ errors  in the \vv-
530: and  \ii-bands were  calculated in  exactly the  same way  as  for the
531: \zz-band, as shown in \figref{fig3},  4 and 5.  The resulting $BVi'z'$
532: sky  values and  the sky  surface-brightness levels  are all  given in
533: \tabref{table1}.
534: 
535: %--------------------------------------------------
536: \begin{deluxetable*}{cccccc}
537: \tabletypesize{\scriptsize}
538: \tablewidth{0pt}
539: \tablecaption{Measured sky values in $BVi'z'$ (filters) for the HUDF \label{table1}}
540: \tablenum{1}
541: \tablehead{\colhead{HUDF} & \colhead{Number of} & \colhead{Mean Sky Value$^a$} & \colhead{Sky SB$^c$} & 
542: \colhead{Sky Color$^c$} & \colhead{1$\sigma$ Sky-Subtraction} \\
543: \colhead{Filter} & \colhead{Exposures} & \colhead{($e^-$/s) and rms error$^b$} & \colhead{(AB mag arcsec$^{-\
544: 2}$)} & \colhead{(AB mag)} & \colhead{error (AB mag arcsec$^{-2}$)} }
545: 
546: \startdata
547: $B$ & 112 & 0.015909 $\pm$ 0.000065 & 23.664 $\pm$ 0.003 & ($B-V$)$_{\rm sky}$=0.800 & 29.85 $\pm$ 0.05 \\
548: $V$ & 112 & 0.070276 $\pm$ 0.000297 & 22.864 $\pm$ 0.002 & ($V-i'$)$_{\rm sky}$=0.222 & 30.15 $\pm$ 0.15 \\
549: $i'$ & 288 & 0.040075 $\pm$ 0.000088 & 22.642 $\pm$ 0.002 & ($i'-z'$)$_{\rm sky}$=0.065 & 29.77 $\pm$ 0.20 \\
550: $z'$ & 288 & 0.020511 $\pm$ 0.000047 & 22.577 $\pm$ 0.003 & ($V-z'$)$_{\rm sky}$=0.287 & 28.95 $\pm$ 0.05 \\
551: \enddata
552: 
553: \tablenotetext{a}{From \figref{fig4}}
554: \tablenotetext{b}{Error is standard deviation of the mean ($\sigma$/$\surd{N}$)}
555: \tablenotetext{c}{Sky surface brightness values and colors are consistent with the solar colors in AB mag of
556: (\vv-\ii)=0.19, (\vv-\zz)=0.21 and (\ii-\zz)=0.01 [except for bluest color (\bb-\vv)], and is dominated by
557: the zodiacal background.}
558: \end{deluxetable*}
559: %--------------------------------------------------
560: 
561: \subsection{Composite Images and Surface Brightness Profiles}
562: 
563: For  each  redshift   bin  ($z\!\simeq\!4,5,6$),  we  generated  three
564: ``stacked'' composite  images from subsets  of 10 postage  stamps that
565: were  selected as  follows.  After  placing  all 30  image stamps  per
566: redshift  bin into  a $30\times$  (51$\times$51) pixel  IDL  array, 10
567: stamps  were randomly  drawn without  selecting any  object  more than
568: once.   An output image  was generated,  in which  the values  at each
569: pixel are the  average of the corresponding pixels  in the 10 selected
570: input stamps. From  the remaining 20 stamps, we  again randomly select
571: 10, from which we generated  a second composite image, after which the
572: final 10  images were  averaged into the  third composite  image.  The
573: three composite  images per redshift bin are  therefore independent of
574: each  other.  In  none of  our realizations  did we  produce composite
575: images   that  were   essentially  unresolved.    Even   the  faintest
576: $z\!\simeq\!4\!-\!6$    galaxies    are    clearly   resolved.     The
577: $z\!\simeq\!4,5,6$ objects used to  generate the composite images have
578: an  apparent magnitude  range  of approximately  27.5$\pm$1.0 AB  mag.
579: Because the magnitude range is  relatively small and the S/N per pixel
580: is low  even in their central  pixel, we have given  all objects equal
581: weight.   To test  whether this  range  in magnitude  will affect  our
582: stacks and hence,  our profiles, we created 3  stacks depending on the
583: apparent magnitude, i.e. one stack  of the 10 brightest objects in the
584: sample, a second stack of the  10 next brightest objects in the sample
585: and a  third stack of the 10  faintest objects in the  sample. This is
586: summarized  in \figref{fig6}d. We  found that  the profiles  were very
587: similar except that  the profiles of the fainter  stacks fall-off more
588: quickly  at larger  radius compared  to the  profile of  the brightest
589: stack, but  the inner  profile and the  deviation in the  profiles are
590: clearly visible in all 3  stacks.  Therefore, we conclude that for our
591: range   in   apparent   magnitudes,   our  stacks/profiles   are   not
592: affected. Perhaps  most surprisingly, \figref{fig6}d  shows that $r_e$
593: value  of all  3  flux ranges  ($\sim$26.0--27.0, $\sim$27.0--28.0  \&
594: $\sim$28.0--29.0 mag)  are all  about the same  over $\sim$3-4  mag in
595: flux, so  the primary parameter  that distinguishes the  brighter from
596: the   fainter  $z\!\simeq\!6$  dropouts   is  their   central  surface
597: brightness (which thus also varies by $\sim$3--4 mag).
598: 
599: %--------------------------------------------------
600: \begin{figure}
601: \epsscale{1.0}
602: \plotone{f6.eps}
603: \caption{Composite surface  brightness profiles for  three independent
604: sets   of  10  objects   at  {\bf   (a)}  $z\!\simeq\!4$,   {\bf  (b)}
605: $z\!\simeq\!5$ and  {\bf (c)} $z\!\simeq\!6$,  respectively.  The thin
606: solid curve  represents the ACS \vv, \ii\thinspace  and \zz-band PSFs,
607: respectively, while  the horizontal dashed line  indicates the surface
608: brightness level corresponding to the 1$\sigma$ sky--subtraction error
609: in the HUDF images. The vertical dotted line marks the radius at which
610: the profile starts to  deviate significantly from the extrapolation of
611: the inner  $r^{1/n}$ profile observed  at smaller radii. Note  that at
612: $z\!\simeq\!6$, this  deviation is still  well above the  red \zz-band
613: PSF  halo   at  $r$\cge0\farcs30.   The   panel  {\bf  (d)}   shows  3
614: $z\!\simeq\!6$  composite profiles  (each with  a set  of  10 objects)
615: divided  by  apparent  magnitudes.   The brightest  composite  profile
616: (dotted)  has an average  \zz-band magnitude  of $\sim$26.8  mag.  The
617: next brightest composite profile  (short dash) has an average \zz-band
618: magnitude  of  $\sim$27.9  mag,  and the  faintest  composite  profile
619: (dot-dash)   has   an  average   \zz-band   magnitude  of   $\sim$28.9
620: mag.}\label{fig6}
621: \end{figure}
622: %-------------------------------------------------- 
623: 
624: We used the IRAF\footnote{IRAF (http://iraf.net) is distributed by the
625: National Optical  Astronomy Observatories,  which are operated  by the
626: Association  of Universities  for Research  in Astronomy,  Inc., under
627: cooperative   agreement  with   the   National  Science   Foundation.}
628: procedure \texttt{ELLIPSE} to fit surface brightness profiles shown in
629: \figref{fig6} to  each of the  three independent composite  images per
630: redshift bin.  We also computed a mean surface-brightness profile from
631: the  three composite  surface brightness  profiles generated  from the
632: three   independent   composite   images   for  each   redshift   bin.
633: \figref{fig7} shows  composite images for  $z\!\simeq\!4,5,6$ objects.
634: Here each  composite image  is a stack  of 30  objects.  
635: \figref{fig8} shows  the average surface brightness  profiles for each
636: of the  redshift intervals $z\!\simeq\!4,5,6$.  The  thin solid curves
637: in \figref{fig6}  and the  dot-dash curves in  \figref{fig8} represent
638: the  observed  ACS  \vv,   \ii\thinspace  and  \zz-band  Point  Spread
639: Functions  (PSFs),  while the  horizontal  dashed  lines indicate  the
640: surface    brightness   level    corresponding   to    the   1$\sigma$
641: sky--subtraction  error in  each of  the HUDF  images as  discussed in
642: \secref{skyerror}.  It  is important  to note that  we scaled  the ACS
643: PSFs to match the surface brightness  of the central data point in our
644: mean surface-brightness  profile, to  determine how extended  the mean
645: surface-brightness profile is with respect to the PSFs.
646: 
647: In  \figref{fig8},   we  fitted  all  possible   combinations  of  the
648: Sers\'{i}c  profiles (convolved  with  the ACS  PSF)  to the  observed
649: profiles  and using  $\chi^2$ minimization,  found the  best  fits for
650: galaxies at  $z\!\simeq\!4,5,6$.  The best fit  S\'{e}rsic index ($n$)
651: for  all  three profiles  ($z\!\simeq\!4,5,6$)  is $n\!<\!2$,  meaning
652: these galaxies  follow mostly exponential disk-type  profiles in their
653: central regions.  We find that  the observed profiles start to deviate
654: from  the  best-fit  profiles  at  $r\!\gtrsim$0\arcspt  27,  somewhat
655: depending on  the redshift.  From  \figref{fig8}, we also see  that in
656: each   of  $V$  ($z\!\simeq\!4$),   $i'$  ($z\!\simeq\!5$)   and  $z'$
657: ($z\!\simeq\!6$), the  PSF declines more rapidly with  radius than the
658: composite radial  surface brightness profile  for $r\!\gtrsim$0\arcspt
659: 27.  It is  therefore unlikely that the observed  `breaks' result from
660: the  halos   and  structure  of   the  ACS  PSFs.    Specifically,  at
661: $z\!\simeq\!6$ the  most significant deviations  in the light-profiles
662: are seen  at levels 1.5--2.0  mag above the  1$\sigma$ sky-subtraction
663: error,  and  well above  the  PSF wings.   Each  of  the mean  surface
664: brightness profiles display a  well-defined break, the radius of which
665: appears to  change somewhat with  redshift. The vertical  dotted lines
666: (in \figref{fig6} and \figref{fig8}) mark the radius at which the mean
667: surface brightness  profiles start  to deviate significantly  from the
668: extrapolation of the $r^{1/n}$ profile observed at smaller radii.
669: 
670: %--------------------------------------------------
671: \begin{figure}
672: \epsscale{1.0}
673: \plotone{f7.eps}
674: \caption{Composite images for  {\bf Left} $z\!\simeq\!4$, {\bf Center}
675: $z\!\simeq\!5$  and  {\bf Right}  $z\!\simeq\!6$  objects.  Here  each
676: composite image is a stack of 30 objects. Each stamp is $1\farcs53$ on
677: a side.}\label{fig7}
678: \end{figure}
679: %--------------------------------------------------
680: \begin{figure}
681: \epsscale{1.1}
682: \plottwo{f8a.eps}{f8b.eps}
683: \epsscale{0.5}
684: \plotone{f8c.eps}
685: \caption{Mean surface  brightness profiles with a  best fit S\'{e}rsic
686: profiles  for 30 composite  images at  {\bf (a)}  $z\!\simeq\!4$, {\bf 
687: (b)} $z\!\simeq\!5$  and {\bf (c)}  $z\!\simeq\!6$, respectively.  The
688: thin dot-dash curve represents the ACS \vv, \ii\thinspace and \zz-band
689: PSFs,  respectively, while  the horizontal  dashed line  indicates the
690: surface    brightness   level    corresponding   to    the   1$\sigma$
691: sky--subtraction error  in the HUDF  images. The vertical  dotted line
692: marks the radius at which  the profile starts to deviate significantly
693: from  the extrapolation  of the  inner $r^{1/n}$  profile  observed at
694: smaller  radii.  The  $n$  and  r$_c$  are  the  best  fit  S\'{e}rsic
695: parameters.  }\label{fig8}
696: \end{figure}
697: %--------------------------------------------------  
698: 
699: \subsection{Test of the Stacking Technique on Nearby Galaxies}
700: 
701: To test  the general  validity of the  stacking technique itself  on a
702: local galaxy sample, we used  surface photometry from the Nearby Field
703: Galaxy  Survey   \citep[NFGS:][]{jans00a,jans00b}.   The  NFGS  sample
704: contains 196 nearby galaxies,  that were objectively selected from the
705: CfA redshift catalog \citep[CfA\,I;][]{davi83,huch83} to span the full
706: range in absolute $B$  magnitude present in the CfA\,I ($-14.7\lesssim
707: \!M_B\!\lesssim -22.7$  mag).  The absolute  magnitude distribution in
708: the  NFGS sample  approximates  the local  galaxy luminosity  function
709: \citep[e.g.,][]{marz94},  while  the  distribution  over  Hubble  type
710: follows  the changing  mix of  morphological  types as  a function  of
711: luminosity in the local  galaxy population.  The NFGS sample \citep[as
712: detailed in][]{jans00a}  minimizes biases,  and yields a  sample that,
713: with  very  few  caveats,   is  representative  of  the  local  galaxy
714: population.   As part  of  the NFGS,  $UBR$  surface photometry,  both
715: integrated (global) and nuclear spectrophotometry, as well as internal
716: kinematics  were   obtained  \citep[see][]{jans01}.   Here,   we  will
717: concentrate on the $U$-band surface photometry, since it is closest in
718: wavelength    to    the    rest-frame    wavelengths    observed    at
719: $z\!\simeq\!4\!-\!6$.   Although,  ideally,  we  would want  a  filter
720: further into  the UV, \citet{tayl07} and \citet{wind02}  show that for
721: the majority  of late-type nearby galaxies, the  apparent structure of
722: galaxies does  not change dramatically once one  observes shortward of
723: the Balmer break. Early-type  galaxies, however, are a clear exception
724: to this, but these are  not believed to dominate the galaxy population
725: at $z\!\simeq\!4\!-\!6$, as discussed before.
726: 
727: \figref{fig9} shows  stacked profiles for  relatively luminous early-,
728: spiral-, and  late-type galaxies drawn from the  NFGS. Vertical dotted
729: lines indicate  the half-light radii  and their intersection  with the
730: profiles,  the  surface  brightness  at  that  radius.   Dashed  lines
731: indicate  exponential  fits to  the  outer  portion  of each  profile.
732: \figref{fig9}  also  shows   that  co-adding  profiles  for  disparate
733: morphological types and  for mid-type spiral galaxies with  a range in
734: bulge-to-disk ratios  can produce breaks in the  composite profile. No
735: such breaks are  seen when the profiles of  either early-type galaxies
736: (E,  S0) or late-type  galaxies (Sd--Irr)  are co-added.   This figure
737: shows  that, \emph{if}  galaxies at  $z\!\simeq\!4\!-\!6$  had similar
738: morphological types  as local galaxies,  then it would be  possible to
739: produce  a  break in  the  profiles  (as  shown in  \figref{fig6}  and
740: \figref{fig8}), merely  by mixing different types of  galaxies.  We do
741: not  believe  that  the  galaxy  populations  at  $z\!\simeq\!4\!-\!6$
742: morphologically resemble those at  low redshift.  Hence, for primarily
743: late-type galaxies, which dominate the faint blue galaxy population at
744: AB$\ge$24 mag  \citep{driv98}, and  which likely dominate  the fainter
745: end of the luminosity  function at $z\!\simeq\!4\!-\!6$ that we sample
746: here  \citep{yan04a,yan04b},  the image  stacking  is  likely a  valid
747: exercise.
748: 
749: %--------------------------------------------------
750: \begin{figure}
751: \epsscale{1.0}
752: \plotone{f9.eps}
753: \caption{Stacked  radial surface  brightness  profiles for  relatively
754: luminous early-, spiral- and  late-type nearby galaxies drawn from the
755: Nearby  Field  Galaxy  Survey \citep{jans00a,jans00b}.   The  vertical
756: dotted  line indicates the  half-light radius,  while the  dashed line
757: represents an exponential fit to  the outer portions of each composite
758: profile.  Co-adding profiles for disparate morphological types and for
759: spiral  galaxies with  a  range in  bulge-to-disk  ratios can  produce
760: breaks in the composite profile. No significant breaks are seen in the
761: \emph{outer} light  profiles, when  the profiles of  either early-type
762: galaxies    (E,   S0)    or   late-type    galaxies    (Sd--Irr)   are
763: co-added.}\label{fig9}
764: \end{figure}
765: %--------------------------------------------------
766: 
767: The primary goal of this section was to show that the profile stacking
768: technique  is  valid  and  can  be  used  to  get  meaningful  surface
769: brightness  profiles.  We  are not  comparing our  nearby  sample with
770: galaxies at  $z\!\simeq\!4\!-\!6$. These nearby  galaxies are unlikely
771: to be local analogues of  high redshift galaxies.  If we apply surface
772: brightness dimming to UV light-profiles of these nearby galaxies, they
773: would be mostly invisible to  \emph{HST}, and in some cases visible to
774: \emph{JWST} in  long integration \citep[see  e.g.][]{wind06}.  This is
775: another way of saying  that the $z\!\simeq\!4\!-\!6$ objects are truly
776: different from $z\!\simeq\!0$ objects.
777: 
778: %--------------------------------------------------
779: 
780: \section{Discussions}\label{results}
781: 
782: \figref{fig8} shows that the  mean surface brightness profiles deviate
783: significantly   from    an   inner   $r^{1/n}$    profile   at   radii
784: $r\!\gtrsim$0\arcspt  27--0\arcspt  35,   depending  somewhat  on  the
785: redshift bin.   These deviations appear real, with  the break/point of
786: departure located \cge  1.5--2~mag above the 1$\sigma$ sky-subtraction
787: error and above  the PSF-wings.  In the following,  we discuss several
788: possible explanations for the observed shapes of our composite surface
789: brightness profiles.
790: 
791: \subsection{Galaxies with Different Morphologies}
792: 
793: Our test  on nearby galaxies  (\figref{fig9}) shows that, if  we stack
794: many  galaxies with different  morphologies (early-type,  late-type or
795: spiral galaxies),  it is possible  to get a slope-change  (`break') in
796: the average surface brightness  profile. \citet{ravi06} find that 40\%
797: of the brighter LBGs at $2.5\!<\!z\!<\!5$ have light profiles close to
798: exponential, as seen for disk  galaxies, and only $\sim$30\% have high
799: $n$,  as seen  in  nearby  spheroids.  They  also  find a  significant
800: fraction ($\sim$30\%) of galaxies with light profiles \emph{shallower}
801: than  exponential, which appear  to have  multiple cores  or disturbed
802: morphologies, suggestive  of close  pairs or on-going  galaxy mergers.
803: Therefore,  if  galaxies at  $z\!\simeq\!4\!-\!6$  have  a variety  of
804: morphological types, then the  shape of the average surface brightness
805: profile that we  see may be due to the stacking  of different types of
806: galaxies.   Therefore, we find  that the  exponential and  the flatter
807: profiles  found by  \citet{ravi06} for  galaxies  at $2.5\!<\!z\!<\!5$
808: also apply to higher redshifts ($z\!\ge\!5$).
809: 
810: Also, we believe that it is  more likely that the high redshift, faint
811: galaxy population consists primarily  of small galaxies with late-type
812: morphologies   and   with  sub-L$^{*}$   luminosities,   as  seen   at
813: $z\!\simeq\!2\!-\!3$     \citep{driv95,driv98}.      So     if     the
814: $z\!\simeq\!4\!-\!6$  population consists of  such a  late-type galaxy
815: population, then the slope-change in  the light profiles is likely not
816: the result of co-adding images of objects with disparate morphological
817: types.
818: 
819: \subsection{Central Star Formation/Starburst}
820: 
821: \emph{HST} optical  images of galaxies  at $z\!\simeq\!4\!-\!6$ sample
822: their rest-frame UV ($\sim$1200 \Ang), where the contribution from the
823: actively  star-forming regions (very  young, massive  stars) dominates
824: the   UV-light.    \citet{hath07}   have   shown  that   galaxies   at
825: $z\!\simeq\!5\!-\!6$ are  high redshift starbursts  and these galaxies
826: have similar starburst intensity limit as local starbursting galaxies.
827: Therefore, it  is possible that galaxies  at $z\!\simeq\!4\!-\!6$ have
828: centrally concentrated star  formation or starburst.  This possibility
829: is  based  on three  key  assumptions: (1)  most  of  the galaxies  at
830: $z\!\simeq\!4,5,6$     are    intrinsically     later-type    galaxies
831: \citep{driv98,stei99}; (2)  the Spectral Energy  Distribution (SED) of
832: these galaxies at $z\!\simeq\!4,5,6$ are dominated by early A- to late
833: O-type stars, respectively;  and (3) there are no  old stars with ages
834: at  $z\!\simeq\!4\!-\!6$  greater  than  2-1 Gyr  in  WMAP  cosmology,
835: respectively.
836: 
837: \citet{hunt06}   studied  azimuthally   averaged   surface  photometry
838: profiles for large sample of nearby irregular galaxies. They find some
839: galaxies have double exponentials that  are steeper (and bluer) in the
840: inner  parts compared to  outer parts  of the  galaxy.  \citet{hunt06}
841: discuss that this  type of behavior is expected  in galaxies where the
842: centrally  concentrated  star  formation  or  starburst  steepens  the
843: surface brightness profiles  in the center. If that  is the case, then
844: one might expect a better correlation between the break in the surface
845: brightness profiles and changes  in color profiles. Unfortunately, for
846: our sample  of galaxies at  $z\!\simeq\!4\!-\!6$, we don't  have high-
847: resolution  restframe \uu\bb\vv  color information.   The  objects are
848: generally too  faint for \emph{Spitzer Space Telescope},  and hence we
849: cannot  confirm  or reject  this  possibility  for  the shape  of  our
850: composite surface brightness profiles.
851: 
852: \subsection{\boldmath {Limits to Dynamical Ages for $z\!\simeq\!4,5,6$ Objects}}\label{ages}
853: 
854: The  average compact $z\!\simeq\!4\!-\!6$  galaxy is  clearly extended
855: with respect  to the ACS PSFs  (\figref{fig8}), and is best  fit by an
856: exponential   profile   ($n\!<\!2$)  out   to   a   radius  of   about
857: $r\!\simeq$0\arcspt   35,   0\arcspt    31,   and   0\arcspt   27   at
858: $z\!\simeq\!4,5$ and $6$, respectively.  The apparent progression with
859: redshift is  noteworthy.  The  radius at which  the profile  starts to
860: deviate  from  $r^{1/n}$ (in  this  case  at  radius $r$\cge  0\arcspt
861: 35--0\arcspt 27) may be an  important constraint to the dynamical time
862: scale of the system,  as discussed in \secref{introduction}.  If this
863: argument is valid,  then we can estimate limits  to the dynamical ages
864: of $z\!\simeq\!4,5,6$ galaxies as follows.
865:   
866: In WMAP cosmology,  a radius of $r$\cge 0\arcspt  35 at $z\!\simeq\!4$
867: corresponds   to  $r$\cge   2.5   kpc.   The   dynamical  time   scale
868: \citep[e.g.,][]{binn87},  $\tau_{dyn}$,  goes  as  $\tau_{dyn}$  =  $C
869: r^{3/2} \!/\!\sqrt{G\,M}$, where the constant $C=\pi/2$. For a typical
870: dwarf  galaxy mass  range of  $\sim10^9\!-\!10^8$\Msun  inside $r$=2.5
871: kpc,  we  infer  that  the  limits  to  the  dynamical  age  would  be
872: $\tau_{dyn}\simeq$ 90--290  Myr, which is the lifespan  expected for a
873: late-type B-star.  This means that the last major merger that affected
874: this  surface brightness  profile  and that  triggered its  associated
875: starburst  may  have occurred  $\sim$0.20  Gyr before  $z\!\simeq\!4$,
876: ---assuming that the star-formation wasn't spontaneous, but associated
877: with some accretion or a merging event.
878: 
879: Table 2 shows  the break-radius and inferred limits  to dynamical ages
880: for the $z\!\simeq\!4\!-\!6$ objects.  At $z\!\simeq\!5$, we find that
881: the  limits   to  dynamical   age  at  the   break  radius   would  be
882: $\tau_{dyn}\simeq$ 70--210  Myr, which is the lifespan  expected for a
883: mid B-star,  while at $z\!\simeq\!6$,  $\tau_{dyn}\simeq$ 50--150 Myr,
884: which is the lifespan expected for a late O--early B-star.  This means
885: that  the last  major merger  that affected  these  surface brightness
886: profiles at  $z\!\simeq\!5$ and $6$ and that  triggered its associated
887: starburst  may  have occurred  $\sim$0.14  and  $\sim$0.10 Gyr  before
888: $z\!\simeq\!5$ and $6$, respectively.
889: 
890: %--------------------------------------------------
891: \begin{deluxetable}{cccc}
892: \tablewidth{0pt}
893: \tablecaption{Dynamical Ages for $z\!\simeq\!4-6$ objects in the HUDF\label{table2}}
894: \tablenum{2}
895: \tablehead{\colhead{Redshift} & \colhead{``Break'' Radius$^a$} & \colhead{``Break'' Radius$^b$} & 
896: \colhead{Dynamical Age$^c$} \\
897: \colhead{$z$} &      \colhead{(arcsec)} &   \colhead{(kpc)} &      \colhead{($\tau_{dyn}$)} }
898: 
899: \startdata
900: 4 & 0.35 & 2.5 & 0.09--0.29 Gyr \\
901: 5 & 0.31 & 2.0 & 0.07--0.21 Gyr \\
902: 6 & 0.27 & 1.6 & 0.05--0.15 Gyr \\
903: \enddata
904: 
905: \tablenotetext{a}{From composite surface brightness profiles (\figref{fig6} and \figref{fig8}).}
906: \tablenotetext{b}{Radius in kpc corresponding to radius in arcsec at given redshift.}
907: \tablenotetext{c}{If ``break radius'' interpreted as indicator of dynamical age.}
908: \end{deluxetable}
909: %--------------------------------------------------
910: 
911: The  dynamical time is  a lower  limit to  the actual  time available,
912: since it  assumes matter  starts from rest.   Any angular  momentum at
913: start will increase the available time.  The best-fit SED age from the
914: GOODS \emph{HST} and \emph{Spitzer} photometry on some of the brighter
915: of  these objects  --- using  \citet{bruz03} templates  --- is  in the
916: range  of about  $\sim$150--650  Myr \citep{yan05,eyle05,eyle07},  the
917: lower end  of which is consistent  with our limits  to their dynamical
918: age  estimates, while  the  somewhat  larger SED  ages  could also  be
919: affected by the onset of  the AGB in the stellar population increasing
920: the observed  \emph{Spitzer} fluxes and  hence possibly overestimating
921: ages \citep{mara05}.   Our age estimates  for $z\!\simeq\!4\!-\!6$ are
922: consistent  with the trend  of SED  ages suggested  for $z\!\simeq\!7$
923: \citep{labb06}.  It  is noteworthy that, given  the uncertainties, the
924: two  independent  age  estimates  are  consistent. If  our  limits  to
925: dynamical age  estimates for the  image \emph{stacks} are  thus valid,
926: they are consistent with the SED ages, and point to a consistent young
927: age for these objects.
928: 
929: Furthermore, the  presence of young, massive late  O--early B-stars at
930: $z\!\simeq\!6$ has implications for  the reionization of the universe.
931: From observations of the  appearance of complete Gunn-Peterson troughs
932: in the  spectra of $z\!\ga\!5.8$  quasars \citep{fan06}, we  know that
933: the epoch of reionization had ended by $z\!\simeq\!6$.  From the steep
934: ($\alpha$=--1.8)  faint-end  slope   of  the  luminosity  function  of
935: $z\!\simeq\!6$  galaxies, \citet{yan04a,yan04b}  concluded  that dwarf
936: galaxies,   and   not  quasars,   likely   finished  reionization   by
937: $z\!\simeq\!6$.   Should  the present  interpretation  of their  light
938: profiles  be correct,  then it  would appear  to add  support  to this
939: picture, in the  sense that such objects are  dominated by B-stars and
940: did  not   start  their  most  recent  major   starburst  long  before
941: $z\!\simeq\!6$.
942: 
943: \section{Summary}\label{conclusion}
944: 
945: We  used  the stacked  HUDF  images  to  analyze the  average  surface
946: brightness  profiles of  $z\!\simeq\!4\!-\!6$ galaxies.   Our analysis
947: shows  that even  the  faintest galaxies  at $z\!\simeq\!4\!-\!6$  are
948: resolved. This  may have implications  on the stellar density  and its
949: relation to the stellar density in present-day galaxies.  We also find
950: that  the average  surface  brightness profiles  display  breaks at  a
951: radius that progresses toward lower redshift from $r\simeq$0\arcspt 27
952: (1.6  kpc) at  $z\!\simeq\!6$  to $r\simeq$0\arcspt  35  (2.5 kpc)  at
953: $z\!\simeq\!4$.
954: 
955: The shape  of the  radial surface brightness  profile that  we observe
956: could  result  from a  mixture  of  different  morphological types  of
957: galaxies,  if  they  exist  at $z\!\simeq\!4\!-\!6$,  because  we  can
958: produce similar breaks in the  surface brightness profiles when we mix
959: different types of nearby  galaxies.  Alternatively, if these galaxies
960: are dominated by a central  starburst then they could show such double
961: exponential-type profiles, as discussed by \citet{hunt06}.  In a third
962: scenario,  if the  galaxies at  $z\!\simeq\!4-6$ are  truly  young and
963: mostly late-type, the  outer profiles seen in our  mean radial surface
964: brightness  profiles  at  $z\!\simeq\!4-6$  bear the  imprint  of  the
965: hierarchical build-up  process, and  are still dominated  by infalling
966: material, which is \emph{not} detectable in the individual HUDF images
967: of these  faint objects.  We  have estimated limits to  dynamical ages
968: from  the  break radius  at  $z\!\simeq\!4,  5,  6$, very  roughly  as
969: $\sim$0.20,  0.14  and 0.10  Gyr,  respectively,  and  those ages  are
970: similar   to   the   SED   ages   inferred   at   $z\!\simeq\!4\!-\!6$
971: \citep{yan05,eyle05,eyle07},  and consistent  with SED  ages suggested
972: for  $z\!\simeq\!7$ \citep{labb06}.  Hence,  at $z\!\simeq\!4,  5, 6$,
973: the last  major merger that  affected the surface  brightness profiles
974: that we observe, and that  triggered the observed star-burst, may have
975: occurred respectively  $\sim$0.20, 0.14 and 0.10 Gyr  earlier, or very
976: approximately at $z\!\simeq\!4.5, 5.5, 6.5$.  This would be consistent
977: with  the  hierarchical assembly  of  galaxies  and  with the  end  of
978: reionization,  since  it  would  imply  that  from  $z\!\simeq\!4$  to
979: $z\!\simeq\!6$,  the  SEDs  become  progressively  more  dominated  by
980: late-B--late-O stars.   This implies that the  sub-$L^*$ (i.e.  dwarf)
981: galaxies may have produced  sufficient numbers of energetic UV photons
982: to   complete   the  reionization   process   by  $z\!\simeq\!6$,   as
983: \citet{yan04a,yan04b} suggested.  It will  be imperative to study with
984: future    instruments    like    \emph{HST}/WFC3    and    \emph{JWST}
985: \citep{wind06,wind07} whether  the dominant stellar  population indeed
986: changes  from  late-O--early-B  at  $z\!\simeq\!6$ (i.e.   capable  of
987: reionizing) to  mid- to late-B  at $z\!\simeq\!4-5$ (i.e.   capable of
988: maintaining reionization),  and to what extent the  intrinsic sizes of
989: these faint objects will ultimately limit deep \emph{JWST} surveys.
990: 
991: %--------------------------------------------------
992: 
993: \acknowledgments This  work was partially  supported by HST  grants AR
994: 10298 and GO 9780 from the Space Telescope Science Institute, which is
995: operated by AURA  under NASA contract NAS 5-26555.   The authors thank
996: Deidre Hunter, Alan Dressler,  Henry Ferguson, Anton Koekemoer, Robert
997: Morgan for  their helpful  discussions. RAW acknowledges  support from
998: NASA JWST Interdisciplinary Scientist grant NAG5-12460 from GSFC, that
999: supported an investigation of the  implications of this work for JWST.
1000: We specially thank our referee, Dr.  Patrick McCarthy, for his helpful
1001: comments that have improved this paper.
1002: 
1003: Facilities: \facility{HST(ACS)}
1004: %--------------------------------------------------
1005: 
1006: \begin{thebibliography}{}
1007: \bibitem[Beckwith  \etal(2006)]{beck06} Beckwith,  S.,  Stiavelli, M.,
1008: Koekemoer, A. M., et al. 2006, AJ, 132, 1729
1009: \bibitem[Bertin  \&  Arnouts(1996)]{bert96}  Bertin, E.,  \&  Arnouts,
1010: S. 1996, A\&AS, 117, 393
1011: \bibitem[Binney \& Tremaine(1987)]{binn87} Binney, J. J., \& Tremaine,
1012: S. 1987, Galactic Dynamics (Princeton: Princeton Univ. Press)
1013: \bibitem[Bolzonella  \etal(2000)]{bolz00}  Bolzonella,  M.,  Miralles,
1014: J. M., \& Pell\'{o}, R. 2000, A\&A, 363, 476
1015: \bibitem[Bouwens \etal(2004)]{bouw04} Bouwens, R., Illingworth, G. D.,
1016: Thompson, R. I., et al. 2004, ApJ, 606, L25
1017: \bibitem[Bouwens  \etal(2006)]{bouw06}  Bouwens,  R. J.,  Illingworth,
1018: G.  D.,  Blakeslee,   J.  P.,  \&  Franx,  M.   2006,  ApJ,  653, 53
1019: \bibitem[Bouwens  \etal(2007)]{bouw07}  Bouwens,  R. J.,  Illingworth,
1020: G. D., Franx, M., \& Ford, H. 2007, ApJ, in press (astro-ph/0707.2080)
1021: \bibitem[Brandt  \etal(2001)]{bran01}  Brandt,  W. N.,  Hornschemeier,
1022: A.  E., Schneider, D.  P., Alexander,  D. M.,  Bauer, F.  E., Garmire,
1023: G. P., \& Vignali, C. 2001, ApJ, 558, L5
1024: \bibitem[Bruzual  \& Charlot(2003)]{bruz03}  Bruzual, G.,  \& Charlot,
1025: S. 2003, MNRAS, 344, 1000
1026: \bibitem[Bunker \&  Stanway(2004)]{bunk04} Bunker, A.  J., \& Stanway,
1027: E. R. 2004, (astro-ph/0407562)
1028: \bibitem[Cowie  \etal(1996)]{cowi96} Cowie, L.  L., Songaila,  A., Hu,
1029: E. M., \& Cohen, J. G. 1996, AJ, 112, 839
1030: \bibitem[Davis  \&  Peebles(1983)]{davi83}   Davis,  M.,  \&  Peebles,
1031: P. J. M. 1983, ApJ, 267, 465
1032: \bibitem[Dow-Hygelund     \etal(2007)]{dow07}     Dow-Hygelund,    C.,
1033: Holden,  B., Bouwens, R., et al. 2007, ApJ, 660, 47
1034: \bibitem[Driver \etal(1995)]{driv95} Driver,  S. P., Windhorst, R. A.,
1035: \& Griffiths, R. E. 1995, ApJ, 453, 48
1036: \bibitem[Driver  \etal(1998)]{driv98} Driver,  S.  P., Fernandez-Soto,
1037: A.,  Couch, W. J.,  Odewahn, S.  C., Windhorst,  R. A.,  Phillips, S.,
1038: Lanzetta, K., \& Yahil, A. 1998, ApJ, 496, L93
1039: \bibitem[Eyles  \etal(2005)]{eyle05}  Eyles,  L.  P., Bunker,  A.  J.,
1040: Stanway, E.  R., Lacy, M.,  Ellis R. S.,  \& Doherty, M.  2005, MNRAS,
1041: 364, 443
1042: \bibitem[Eyles \etal(2007)]{eyle07} Eyles, L. P., Bunker, A. J., Ellis
1043: R. S., Lacy,  M., Stanway, E. R., Stark, D., \&  Chiu, K. 2007, MNRAS,
1044: 374, 910
1045: \bibitem[Fan  \etal(2006)]{fan06}  Fan, X.,  Strauss,  M. A.,  Becker,
1046: R. H., et al.  2006, AJ, 132, 117
1047: \bibitem[Georgakakis  \etal(2003)]{geor03}  Georgakakis, A.,  Hopkins,
1048: A. M., Sullivan, M., Afonso, J., Georgantopoulos, I., Mobasher, B., \&
1049: Cram, L. E. 2003, MNRAS, 345, 939
1050: \bibitem[Giavalisco  \etal(2004)]{giav04}  Giavalisco, M.,  Dickinson,
1051: M., Ferguson, H. C., et al. 2004, ApJ, 600, L103
1052: \bibitem[Gonzaga \etal(2005)]{gonz05} Gonzaga,  S., et al. 2005, ``ACS
1053: Instrument Handbook", Version 6.0, (Baltimore:STScI)
1054: \bibitem[Guzman  \etal(1997)]{guzm97} Guzman,  R.,  Gallego, J.,  Koo,
1055: D. C., Phillips,  A. C., Lowenthal, J. D.,  Faber, S. M., Illingworth,
1056: G. D., \& Vogt, N. P. 1997 ApJ, 489, 559
1057: \bibitem[Hathi  \etal(2007)]{hath07} Hathi,  N. P.,  Malhotra,  S., \&
1058: Rhoads, J. E.  2007, ApJ, submitted (astro-ph/0709.0520)
1059: \bibitem[Heavens   \etal(2004)]{heav04}  Heavens,   A.,   Panter,  B.,
1060: Jimenez, R., \& Dunlop, J. 2004, Nature, 428, 625
1061: \bibitem[Hu  \etal(2002)]{hu02}  Hu, E.  M.,  Cowie,  L. L.,  McMahon,
1062: R. G., Capak, P., Iwamuro, F., Kneib, J.-P., Maihara, T., \& Motohara,
1063: K. 2002, ApJ, 568, L75
1064: \bibitem[Huchra \etal(1983)]{huch83} Huchra, J. P., Davis, M., Latham,
1065: D., \& Tonry, J. 1983, ApJS, 52, 89
1066: \bibitem[Hunter   \&  Elmegreen(2006)]{hunt06}   Hunter,  D.   A.,  \&
1067: Elmegreen, B. G. 2006, ApJS, 162, 49
1068: \bibitem[Jansen  \etal(2000a)]{jans00a}  Jansen,  R.  A.,  Franx,  M.,
1069: Fabricant, D., \& Caldwell, N. 2000a, ApJS, 126, 271
1070: \bibitem[Jansen \etal(2000b)]{jans00b}  Jansen, R. A.,  Fabricant, D.,
1071: Franx, M., \& Caldwell, N. 2000b, ApJS, 126, 331
1072: \bibitem[Jansen   \&  Kannappan(2001)]{jans01}   Jansen,   R.  A.   \&
1073: Kannappan, S. J. 2001, Ap\&SS, 276, 1151
1074: \bibitem[Kodaira  \etal(2003)]{kodi03}  Kodaira,  K.,  Taniguchi,  Y.,
1075: Kashikawa, N., et al. 2003, PASJ, 55, L17
1076: \bibitem[Kodama \etal(2004)]{koda04} Kodama,  T., Yamada, T., Akiyama,
1077: M., et al. 2004, MNRAS, 350, 1005
1078: \bibitem[Koekemoer  \etal(2002)]{koek02} Koekemoer,  A.  M., Fruchter,
1079: A. S., Hook, R. N., \&  Hack, W. 2002, The 2002 \emph{HST} Calibration
1080: Workshop,   ed.   S.   Arribas,   A.  Koekemoer,   and   B.   Whitmore
1081: (Baltimore:STScI), 337
1082: \bibitem[Kormendy(1977)]{korm77} Kormendy, J. 1977, ApJ, 218, 333
1083: \bibitem[Kurk \etal(2004)]{kurk04} Kurk, J. D., Cimatti, A., di Serego
1084: A, S., Vernet,  J., Daddi, E., Ferrara, A., \&  Ciardi, B. 2004, A\&A,
1085: 422, L13
1086: \bibitem[Labb\'{e}  \etal(2006)]{labb06} Labb\'{e},  I.,  Bouwens, R.,
1087: Illingworth, G. D., \& Franx, M. 2006, ApJ, 649, L67
1088: \bibitem[Lynden-Bell(1967)]{lynd67} Lynden-Bell,  D. 1967, MNRAS, 136,
1089: 101
1090: \bibitem[Maraston(2005)]{mara05} Maraston, C. 2005, MNRAS, 362, 799
1091: \bibitem[Malhotra  \etal(2005)]{malh05} Malhotra,  S., Rhoads,  J. E.,
1092: Pirzkal, N., et al. 2005, ApJ, 626, 666
1093: \bibitem[Marzke \etal(1994)]{marz94}  Marzke, R. O., Huchra,  J. P. \&
1094: Geller, M. J. 1994, ApJ, 428, 43
1095: \bibitem[McCarthy \etal(2004)]{mcca04} McCarthy, P. J. 2004, BAAS, 36,
1096: 1555
1097: \bibitem[Nandra  \etal(2002)]{nand02} Nandra,  K.,  Mushotzky, R.  F.,
1098: Arnaud,  K.,  Steidel, C.  C.,  Adelberger,  K.  L., Gardner,  J.  P.,
1099: Teplitz, H. I., \& Windhorst, R. A. 2002, ApJ, 576, 625
1100: \bibitem[Oke \&  Gunn(1983)]{oke83} Oke, J.  B., \& Gunn, J.  E. 1983,
1101: ApJ, 266, 713
1102: \bibitem[Panter \etal(2007)]{pant07} Panter, B., Jimenez, R., Heavens,
1103: A. F., \& Charlot, S. 2007, MNRAS, 378, 1550
1104: \bibitem[Pascarelle \etal(1996)]{pasc96} Pascarelle, S. M., Windhorst,
1105: R. A., Keel, W. C., \& Odewahn, S. C. 1996, Nature, 383, 45
1106: \bibitem[Ravindranath     \etal(2006)]{ravi06}    Ravindranath,    S.,
1107: Giavalisco,  M.,  Ferguson,  H.  C.,   et  al.  2006,  ApJ, 652, 963
1108: \bibitem[Rhoads \etal(2004)]{rhoa04}  Rhoads, J.  E.,  Xu, C., Dawson,
1109: S., et al. 2004, ApJ, 611, 59
1110: \bibitem[Spergel  \etal(2007)]{sper07}  Spergel,   D.  N.,  Bean,  R.,
1111: Dor\'{e}, O., et al. 2007, ApJS, 170, 377
1112: \bibitem[Steidel  \etal(1999)]{stei99}  Steidel,  C.  C.,  Adelberger,
1113: K. L., Giavalisco, M., Dickinson, M., \& Pettini, M. 1999, ApJ, 519, 1
1114: \bibitem[Stern  \etal(2005)]{ster05} Stern, D.,  Yost, S.  A., Eckart,
1115: M. E., Harrison,  F. A., Helfand, D. J.,  Djorgovski, S. G., Malhotra,
1116: S., \& Rhoads, J. E. 2005, ApJ, 619, 12
1117: \bibitem[Stetson(1987)]{stet87} Stetson, P. B. 1987, PASP, 99, 191
1118: \bibitem[Taniguchi  \etal(2005)]{tani05}  Taniguchi,  Y.,  Ajiki,  M.,
1119: Nagao, T., et al. 2005, PASJ, 57, 165
1120: \bibitem[Taylor \etal(2007)]{tayl07} Taylor, V.  A., Conselice, C. J.,
1121: Windhorst, R. A., \& Jansen, R. A. 2007, ApJ, 659, 162
1122: \bibitem[van Albada(1982)]{vana82} van Albada, T. S. 1982, MNRAS, 201,
1123: 939
1124: \bibitem[van   Dokkum  \etal(2003)]{vand03}   van   Dokkum,  P.    G.,
1125: F\"{o}rster, S., Natascha, M., et al. 2003, ApJ, 587, L83
1126: \bibitem[van  Dokkum \etal(2004)]{vand04} van  Dokkum, P.   G., Franx,
1127: M., F\"{o}rster, S., et al. 2004, ApJ, 611, 703
1128: \bibitem[Vanzella  \etal(2006)]{vanz06} Vanzella,  E.,  Cristiani, S.,
1129: Dickinson, M., et al. 2006, A\&A, 454, 423
1130: \bibitem[White  \etal(2007)]{whit07} White, R.   L., Helfand,  D.  J.,
1131: Becker,  R. H.,  Glikman, E.,  \& de  Vries, W.   2007, ApJ,  654, 99
1132: \bibitem[Williams \etal(1996)]{will96}  Williams, R. E.,  Blacker, B.,
1133: Dickinson, M., et al. 1996, AJ, 112, 1335
1134: \bibitem[Windhorst  \etal(1994)]{wind94}  Windhorst,  R.  A.,  Gordon,
1135: J.  M., Pascarelle,  S. M.,  Schmidtke, P.  C., Keel,  W.  C., Burkey,
1136: J. M., \& Dunlop, J. S. 1994, ApJ, 435, 577
1137: \bibitem[Windhorst \etal(1998)]{wind98} Windhorst, R. A., Keel, W. C.,
1138: \& Pascarelle, S. M. 1998, ApJ, 494, 27
1139: \bibitem[Windhorst  \etal(2002)]{wind02}  Windhorst,  R.  A.,  Taylor,
1140: V. A., Jansen, R. A., et al. 2002, ApJS, 143, 113
1141: \bibitem[Windhorst \etal(2006)]{wind06} Windhorst, R. A., Cohen, S. H., 
1142: Jansen, R. A., Conselice, C., \& Yan, H. 2006, NewAR, 50, 113
1143: \bibitem[Windhorst  \etal(2007)]{wind07}   Windhorst,  R.  A.,  Hathi,
1144: N. P.,  Cohen, S. H., \&  Jansen, R. A. 2007,  \emph{Advances in Space
1145: Research}, in press (astro-ph/0703171)
1146: \bibitem[Yan  \&  Windhorst(2004a)]{yan04a}   Yan,  H.  \&  Windhorst,
1147: R. 2004a, ApJ, 600, L1
1148: \bibitem[Yan  \&  Windhorst(2004b)]{yan04b}   Yan,  H.  \&  Windhorst,
1149: R. 2004b, ApJ, 612, L93
1150: \bibitem[Yan \etal(2005)]{yan05} Yan, H., Dickinson, M., Stern, D., et
1151: al. 2005, ApJ, 634, 109
1152: \bibitem[York \etal(2000)]{york00} York, D. G., et al. 2000, 120, 1579
1153: \bibitem[Zibetti \etal(2004)]{zibe04} Zibetti, S., White, S. D. M., \&
1154: Brinkmann, J.  2004, MNRAS, 347, 556
1155: \end{thebibliography}
1156: 
1157: %--------------------------------------------------  
1158: 
1159: \end{document}
1160: