1: %%% 1. ``\draft'' creates single column and double spaces format.
2: %%% 2. If you comment out ``\draft'', the output will be double column
3: %%% and single space.
4: %%% 3. For cross-references, the use of \label/\ref/\cite and the
5: %%% thebibliography environment is strongly recommended
6: %%% 4. Do NOT use \def/\renewcommand.
7: %%% 5. Do NOT redifine commands provided by PASJ00.cls.
8: \documentclass{pasj00}
9: \draft
10:
11: \begin{document}
12: \SetRunningHead{M.ando et al.}{LBGs at $z\sim5$: Rest-frame UV Spectra II}
13: \Received{2007/01/29}%{yyyy/mm/dd}
14: \Accepted{2007/05/07}%{yyyy/mm/dd}
15:
16: \title{Lyman Break Galaxies at $z\sim5$: Rest-frame UV Spectra II \thanks{\rm Based on data collected at Subaru Telescope,
17: which is operated by the National Astronomical Observatory of Japan.}}
18:
19: %%% begin:list of authors
20: \author{Masataka \textsc{Ando},
21: %\thanks{Example: Present Address is xxxxxxxxxx}
22: Kouji \textsc{Ohta}}
23: \affil{Department of Astronomy, Kyoto University,\\
24: Oiwake-cho, Kirashirakawa, Sakyo-ku, Kyoto, 606-8502}
25: \email{andoh@kusastro.kyoto-u.ac.jp, ohta@kusastro.kyoto-u.ac.jp}
26: \author{Ikuru \textsc{Iwata}}
27: \affil{Okayama Astrophysical Observatory, National Astronomical Observatory of Japan,\\
28: Kamogata, Okayama 719-0232}\email{iwata@oao.nao.ac.jp}
29: \author{Masayuki \textsc{Akiyama}, Kentaro \textsc {Aoki},
30: and Naoyuki \textsc{Tamura}}
31: \affil{Subaru Telescope, National Astronomical Observatory of Japan, \\
32: 650 North A'ohoku Place, Hilo, Hawaii 96720, USA}
33: \email{akiyama@subaru.naoj.org, kaoki@subaru.naoj.org, naoyuki@subaru.naoj.org}
34: %%% end:list of authors
35:
36: %%% Please use the following style in case that sorting by
37: %%% affilation is impossible.
38: %
39: % \author{%
40: % D-Firstname \textsc{D-Familyname}\altaffilmark{1}
41: % E-Firstname \textsc{E-Familyname}\altaffilmark{1,2}
42: % and
43: % F-Firstname \textsc{F-Familyname}\altaffilmark{2}}
44: % \altaffiltext{1}{Address of Institute}
45: % \email{ddddd@xxx.xxx.xx.xx}
46: % \email{eeeee@xxx.xxx.xx.xx}
47: % \altaffiltext{2}{Address of Institute}
48:
49: %% `\KeyWords{}' always has to be placed before `\maketitle'.
50: \KeyWords{galaxies: evolution --- galaxies: formation --- galaxies:
51: high-redshift}
52: %Do NOT move this preamble from here!
53:
54: \maketitle
55:
56: \begin{abstract}
57: We present the results of spectroscopy of
58: Lyman Break Galaxies (LBGs) at $z\sim5$ in the J0053+1234 field
59: with the Faint Object Camera and Spectrograph on the Subaru telescope.
60: Among 5 bright candidates with $z' < 25.0$ mag, 2 objects are confirmed to
61: be at $z\sim5$ from their Ly$\alpha$ emission and the continuum depression
62: shortward of Ly$\alpha$.
63: The EWs of Ly$\alpha$ emission of the 2 LBGs are not so strong
64: to be detected as
65: Ly$\alpha$ emitters, and one of them shows strong low-ionized
66: interstellar (LIS) metal absorption lines.
67: Two faint objects with $z' \geq 25.0$ mag are also
68: confirmed to be at $z\sim5$, and their spectra show strong Ly$\alpha$
69: emission in contrast to the bright ones.
70: These results suggest
71: a deficiency of strong Ly$\alpha$ emission in bright LBGs at
72: $z\sim5$, which has been discussed in our previous paper.
73: Combined with our previous spectra of LBGs at $z\sim5$
74: obtained around the Hubble Deep Field-North (HDF-N),
75: we made a composite spectrum of UV luminous
76: ($M_{1400} \leq -21.5$ mag) LBGs at $z\sim5$.
77: The resultant spectrum shows a weak Ly$\alpha$ emission and
78: strong LIS absorptions which suggests that the bright LBGs at $z\sim5$
79: have chemically evolved at least to $\sim$0.1 solar metallicity.
80: For a part of our sample in the HDF-N region, we obtained near-to-mid infrared
81: data, which constraint stellar masses of these objects.
82: With the stellar mass and the metallicity estimated
83: from LIS absorptions,
84: the metallicities of the LBGs at $z\sim5$ tend to be lower
85: than those of the galaxies with the same stellar mass at $z \lesssim 2$,
86: although the uncertainty is very large.
87: \end{abstract}
88:
89: \section{Introduction}
90:
91: Spectroscopic studies of high-redshift ($z>2\sim3$)
92: galaxies have been advanced recently.
93: One of the largest spectroscopic samples of high-z galaxies is
94: obtained by follow-up spectroscopy for Lyman Break galaxies (LBGs: e.g.,
95: \cite{Ste03}) which are selected by using multi-band photometric data.
96: So far, about a thousand of optical spectra have been obtained each for
97: LBGs at $z\sim3$ (e.g., \cite{Shap03})
98: and star-forming galaxies at $z\sim2$
99: (BM/BX objects: e.g., \cite{Ste04}).
100: Intensive spectroscopic studies based on magnitude-limited samples
101: have also identified several hundreds of galaxies at $z\sim2-4$
102: (e.g. FORS Deep Field (FDF) survey: \cite{Noll04}, VIRMOS-VLT Deep
103: Survey: \cite{Lefe05}).
104:
105: These studies revealed spectroscopic features of galaxies
106: at $z\sim2-3$ in their rest-frame UV wavelength region, and
107: deepened our understanding of their nature.
108: \citet{Shap03} found relations between the
109: strength of the Ly$\alpha$ equivalent width (EW) and other spectroscopic
110: features (e.g., EW of the low-ionized interstellar (LIS) metal
111: absorption lines, UV continuum slope, velocity offset between
112: Ly$\alpha$ and LIS absorptions) for the LBGs at $z\sim3$,
113: which sheds light on properties of an interstellar matter
114: such as a gas outflow, a velocity dispersion of the gas,
115: and a covering fraction of dust.
116: With a part of FDF data, \citet{Meh02} found that EW of the CIV
117: $\lambda\lambda 1548, 1550$ absorption line decreases from $z\sim1.4$
118: to $\sim3.4$.
119: The decrease in the EW of CIV is confirmed with the full sample of FDF,
120: and it seems to continue to $z\sim4$ \citep{Noll04}.
121: Since the EW of CIV is an indicator of the metallicity
122: in local starburst galaxies (e.g., \cite{Hek98}),
123: the result suggests a chemical evolution of galaxies in this
124: redshift range \citep{Meh02},
125: although the correlation between the metallicity and
126: the EW of CIV has not been calibrated at high-redshift
127: and thus is not ensured.
128: Near-infrared (NIR) spectra of galaxies at $z\sim2-3$ have also been
129: obtained and give information about kinematics and
130: metallicity of the galaxies.
131: The velocity offset between nebula emission lines and
132: interstellar absorption lines is generally seen in the LBGs, suggesting
133: the presence of the gas outflow (e.g., \cite{Pet01}).
134: From the emissions lines, gas metallicity in the galaxies can be derived
135: by using metallicity estimators (the $R_{23}$ and the N2 index);
136: for example, \citet{Tep00}, \citet{Kob00}, and \citet{Pet01}
137: derived metallicity of LBGs at $z\sim3$ which ranges typically from
138: 0.1$-$0.9 solar value.
139: \citet{Shap04} showed that majority of the seven NIR bright
140: ($K_{s,Vega}$ $\leq 20$ mag)
141: BX galaxies have near the solar metallicity, and
142: \citet{Erb06} investigated the extended sample
143: mainly with $K_{s,Vega}$ $< 21.5$ mag
144: and found the median metallicity of 87 galaxies at $z\sim2$ to be about
145: 0.5$-$0.6 solar value.
146:
147: Recently, surveys of galaxies at $z\gtrsim5$ have been carried out
148: (e.g.
149: \cite{Iwa03,Lehn03,Sta03,Dick04,Ouchi04a,Shima05,Yan05,Bou06a,Yoshida06}),
150: and challenges of finding galaxies even at
151: $z\sim7-10$ have also started
152: (e.g., \cite{Pello04,Kneib04,Bou05,Bou06b,Mann07}).
153: However, the progress of optical follow-up spectroscopy for galaxies at
154: $z\gtrsim4-5$ is much slower than that for galaxies at $z=2\sim3$
155: because objects are getting fainter, and characteristic features of
156: UV spectrum of
157: star-forming galaxies (e.g., Ly$\alpha$, LIS absorptions)
158: redshift to the wavelength region where night sky emissions
159: are very strong.
160: Although attempts of spectroscopic observations of galaxies at
161: $z\gtrsim5$ were made
162: (e.g., \cite{Spi98,Dey98,Wey98,Lehn03,Sta04}),
163: most of the galaxies were identified only with a Ly$\alpha$ emission
164: because of the low signal-to-noise ratio of their continuum light.
165:
166: We also started an optical spectroscopy of LBGs at $z\sim5$.
167: Targets were selected from our surveys for LBGs at $z\sim5$
168: (\cite{Iwa03}, 2007) which consist of
169: the deep and wide $V, I_C$, and $z'$-band imaging survey with
170: the Subaru/Suprime-Cam in the two target fields: the region including
171: the Hubble Deep Field-North (the HDF-N region) and the J0053+1234 region.
172: Our survey fields are quite wide (1290 arcmin$^2$ in total),
173: and we obtain 228 bright objects with $z'_{AB}<25.0$ mag
174: ($L\gtrsim L^{\ast}$ in UV luminosity function (UVLF)
175: of LBGs at $z\sim5$).
176: We reported an initial result of the spectroscopy for a part of
177: the objects in the HDF-N region (\cite{Ando04}: hereafter Paper I).
178: Thanks to the object brightness and good observing conditions,
179: we detected continuum features of the LBGs at $z\sim5$
180: and identified the galaxies to be at $z\sim5$
181: without Ly$\alpha$ emission line.
182: The most striking result was that there is no or a weak
183: (EW$_{\rm rest}<$ 10\AA ) Ly$\alpha$ emission line in 7 bright LBGs
184: identified.
185: Inspired by the result, the deficiency of the strong Ly$\alpha$
186: emission in the bright LBGs at $z\sim5$
187: was investigated by using combined data of Paper I,
188: our new results obtained in the J0053+1234 region,
189: and those from the literature;
190: the deficiency is generally seen in the LBGs at $z\sim5$
191: and even at $z\sim6$ \citep{Ando06}.
192: If the quenched Ly$\alpha$ emission is due to the dust, this deficiency
193: suggests that bright LBGs have larger amount of dust and higher
194: metallicity than faint ones.
195: This hypothesis suggests that bright LBGs experienced star formation
196: earlier than that of faint ones: differential evolution of the
197: high-redshift galaxies with their luminosity,
198: although there are other possible origins of the deficiency
199: (e.g., a velocity structure in/around the galaxy).
200:
201: In this paper, we present the data of spectroscopic observation of
202: LBGs at $z\sim5$ in the J0053+1234 region which were used
203: in our previous paper \citep{Ando06}.
204: In Section 2, we describe a sample selection, observation, and
205: data reduction.
206: The features of obtained spectra of the LBGs at $z=4-5$ in the J0053+1234
207: region are shown in Section 3.
208: Combining this result with our previous results (Paper I),
209: we described spectroscopic properties of bright LBGs at $z\sim5$ and
210: metallicity against stellar mass in the section.
211: Throughout this paper, we adopt flat $\Lambda$ cosmology, $\Omega_M=0.3$,
212: $\Omega_{\Lambda}=0.7$, and $H_0=70$ km s$^{-1}$Mpc$^{-1}$.
213: The magnitude system is based on AB magnitude \citep{Oke83}.
214:
215: \section{Observation and Data Reduction}
216: The photometric sample of LBGs at $z\sim5$ in the J0053+1234 region
217: was selected based on a deep and wide broad-band ($V, I_C,$ and $z'$)
218: imaging survey by using Subaru/Suprime-Cam \citep{Iye04,Miya}:
219: a total of 114 (236) candidates with $z' <$ 25.0 (25.5) mag in an effective
220: survey area ($\sim$800 arcmin$^2$).
221: Details of imaging observations and the color selection are described in
222: \citet{Iwa07}.
223: In order to detect continuum and absorption feature,
224: we selected the bright ($z' < 25.0$) LBG candidates at $z\sim5$
225: as the main spectroscopic targets.
226: Because the entire survey field was too wide to obtain spectra for the whole
227: sample of these LBG candidates with a limited field of view ($6'\phi$)
228: of the Faint Object Camera and Spectrograph
229: (FOCAS: \cite{kashi})
230: on the Subaru telescope and limited observing time,
231: we selected a target MOS field to contain the bright LBGs as many as possible.
232: When a slit of a bright target overlapped with a slit of another bright one,
233: we chose the object with a higher central surface brightness.
234: Finally, we designed one FOCAS mask which covers 5 bright LBGs.
235: The mask also covered 8 fainter ($z' \geq 25.0$ mag) LBGs
236: which are in the gaps of the slits of main targets on the mask.
237:
238: Spectroscopic observations with the Subaru/FOCAS were made on 2004
239: September 14
240: under a clear condition, and a seeing during the observing run
241: was typically $\sim 0.^{\prime\prime}$8 (FWHM).
242: We used a grism of 300 lines mm$^{-1}$ blazed at 7500 \AA\
243: and the SO58 order-cut filter (the same setting as Paper I),
244: which gave wavelength coverage from 5800\AA\ to 10000\AA\
245: (depending on a slit position on the mask) with a pixel scale of 1.34\AA.
246: The slit lengths were typically 10$^{''}$, and the slit widths were
247: fixed to be 0.$^{''}$8, giving a spectral resolution of R$\sim$700
248: which was measured with night sky emission lines.
249: An exposure time of each frame was 0.5 hours, and a total effective
250: exposure time was 6 hours.
251:
252: The data were reduced by using IRAF\footnote{Image
253: Reduction and Analysis Facility, distributed by National Optical
254: Astronomical Observatories, which are operated by the Association of
255: Universities for Research in Astronomy, Inc., under cooperative
256: agreement with the National Science Foundation.}
257: with the same manner as Paper I except for the re-subtraction of the
258: night sky emissions.
259: We subtracted bias from object frames by using
260: an overscan region of each frame
261: and averaged bias frame in order to remove the characteristic
262: bias-pattern of FOCAS CCDs.
263: Flat-fielding was made with normalized average of dome-flat images.
264: The wavelength calibration was made by using night-sky emission lines.
265: The RMS error of the wavelength calibration was $\sim0.7$\AA.
266: After the sky subtraction,
267: the spectrum of each object was carefully aligned and averaged.
268: Because some objects show too low continuum S/N to subtract sky emission
269: in each frame,
270: we carried out sky subtraction after combining spectra for these objects.
271: Five pixels were binned toward wavelength direction in order to improve
272: S/N, and one-dimensional spectra were obtained
273: with APALL task of IRAF.
274: Since our objects were faint, we did not fix the aperture width
275: for the extraction
276: but we determined it by eye for each object to trace it well.
277: The flux calibration and the sensitivity correction were made for the final
278: spectra by using standard stars of Wolf1346 and PG0823+546.
279: The correction for the atmospheric A and B bands was also made by tracing
280: those absorption features in the spectra of standard stars.
281: For some of the objects, the sky subtraction was not well achieved
282: due to the low-S/N of the continuum or the presence of the bad column of
283: the CCDs.
284: Thus we re-subtracted the scaled night sky emissions from the
285: one-dimensional spectrum of such objects.
286:
287: \section{Results and Discussions}
288:
289: \subsection{Redshift Determination}
290:
291: Among 5 bright targets, we identified 2 objects to be LBGs at $z=4-5$,
292: and the resultant spectra are shown in Figure 1 (No.1: top left, No.2:
293: top right).
294: LBG identifications were mainly made by the presence of
295: the continuum depression shortward of the redshifted
296: Ly$\alpha$ due to the inter-galactic HI absorption
297: and the Ly$\alpha$ emission line near the Ly$\alpha$ depression.
298: In addition, the Ly$\alpha$ emission line of each object shows an asymmetric
299: shape which is considered to be characteristic feature of LBGs and Ly$\alpha$
300: emitters (LAEs) at high-redshift.
301: Note that emission-like features in the spectrum of object No.1 at
302: $\sim$6300\AA\ and in the wavelength region longer than 7300\AA\
303: are the residuals of the night sky emissions since the sky subtraction was
304: not well achieved due to the bad column in the CCD.
305: For object No.2, we identified some LIS absorption lines
306: (SiII $\lambda$1260, OI+SiII $\lambda$1303, and CII $\lambda$1335)
307: at almost the same redshift of the Ly$\alpha$
308: which are characteristic features of nearby star-forming galaxies
309: (e.g., \cite{Hek98}) and LBGs at $z\sim3-5$
310: (e.g., \cite{Ste96a,Shap03,frye}; Paper I; \cite{Iwa05}),
311: supporting the identifications of high redshift star-forming galaxies.
312: Redshifts of these two bright LBGs were determined from the peak of
313: Ly$\alpha$ emission:
314: 4.797 and 4.267 for object No.1 and No.2, respectively.
315: The error of the redshift determination was $\sim0.005$.
316: The rest of the bright sample were not identified to be at $z\sim5$
317: because of their low S/N of the spectra.
318: These objects tend to have lower central surface
319: brightness compared with the confirmed objects.
320:
321: In addition to the bright targets,
322: we identified 2 objects among faint bonus targets with
323: $z'\geq25.0$ mag to be LBGs at $z=4-5$.
324: Spectra of them are shown in Figure 1 (No.3: bottom
325: left, and No.4: bottom right).
326: These objects show a strong (EW$_{\rm obs}\sim220$\AA\ and $\sim$420\AA\
327: for object No.3 and No.4, respectively) single emission line;
328: peak of the emission corresponds to the redshift of 4.391 and 4.491
329: for object No.3 and No.4, respectively if this emission is Ly$\alpha$.
330: A clear asymmetry of the emission line profile
331: and possible continuum depression shortward of the emission
332: are seen in the spectrum of object No.4,
333: which supports this object is really at $z=4-5$.
334: Because the redshift determinations for these objects were based only on
335: the single emission line, we examine the possibility that they are
336: foreground emission galaxies.
337: For example, if object No.3 (No.4) is an [OII] $\lambda3727$
338: emitter at $z=0.76$ (0.79),
339: they should show the redshifted H$\beta$ emission line at $\sim$8550(8710)\AA\
340: and the [OIII]$\lambda$5007 emission line
341: at $\sim$8810(8970)\AA\ with EW$_{\rm obs}$ of $\sim$70(130)\AA\ and
342: EW$_{\rm obs}$ of $\sim$110(210)\AA, respectively by assuming the median of
343: the observed line ratios of star-forming galaxies at $z=0.3\sim1.0$
344: (0.3 and 0.5, respectively: \cite{Kob04}).
345: As shown in Figure 1, emission-like
346: features are seen around at
347: a part of the expected wavelengths in both spectra.
348: However, these features are residual night-sky emissions, and
349: there is no other significant
350: (EW$_{\rm obs}\gtrsim100$\AA) emission lines at the expected wavelengths.
351: Thus these strong emissions are considered to be the Ly$\alpha$ emission
352: from LBGs at the target redshift.
353: We can not completely rule out the possibility that they are
354: foreground emission galaxies by only this analysis,
355: because some of the expected lines of foregrounds are in the wavelength
356: region of severe night sky emissions, and the line ratios have wide
357: object-to-object variances \citep{Kob04}.
358: But the $V-I_C$ colors of these objects (1.98 mag and 1.85 mag)
359: are too red to regard them as star-forming galaxies at $z<1$;
360: we calculated the SEDs of star-forming galaxies at $z<1$ by using
361: the population synthesis code P\'{E}GASE version 2 \citep{pegase}
362: with Salpeter IMF \citep{Sal55} and by
363: assuming constant star formation with an age of 100 Myr,
364: which gives their $V-I_C$ colors in observed-frame
365: of $-0.4-0.1$ mag and $0.2-0.8$ mag for $E(B-V)=0.0$ mag and
366: $E(B-V)=0.4$ mag, respectively with the
367: dust extinction curve by \citet{Cal00}.
368: Thus we consider both of the objects are at $z=4-5$.
369:
370: \begin{figure}
371: \begin{center}
372: \rotatebox{270}{\resizebox{6.5cm}{!}{\FigureFile(70mm,50mm){f1a.ps}}}
373: \hspace{0.5cm}
374: \rotatebox{270}{\resizebox{6.5cm}{!}{\FigureFile(70mm,50mm){f1b.ps}}}\\
375: \vspace{1.5cm}
376: \rotatebox{270}{\resizebox{6.5cm}{!}{\FigureFile(70mm,50mm){f1c.ps}}}
377: \hspace{0.5cm}
378: \rotatebox{270}{\resizebox{6.5cm}{!}{\FigureFile(70mm,50mm){f1d.ps}}}\\
379: \caption{Spectra of LBGs at $z=4-5$ in the J0053+1234 region.
380: {\it Top}: Object No.1 ({\it left}) and No.2 ({\it
381: right}). {\it Bottom}: Object No.3 ({\it left}) and No.4 ({\it right}).
382: These spectra are smoothed with a boxcar over 3-pixel.
383: Positions of line
384: features seen in LBGs at $z=3$ and local starburst galaxies
385: (e.g., \cite{Shap03,Hek98}) are
386: shown with vertical dashed lines.
387: A scaled sky spectrum is shown in a lower part of each panel, and the
388: atmospheric A-band and B-band absorptions are shown as vertical hatched
389: regions.}
390: \end{center}
391: \end{figure}
392:
393:
394: \subsection{Color Selection and Redshift Distribution of LBGs at $z\sim5$}
395:
396: Figure 2 shows the positions of observed objects in the $V - I_C$
397: and $I_C - z'$ two-color diagram from the present results and Paper I.
398: Photometric properties of spectroscopically identified LBGs at $z=4-5$
399: in the J0053+1234 region are summarized in Table 1.
400: Filled circles show LBGs at $z=4-5$ in this study: 2 bright objects
401: as large circles and 2 faint objects as small ones.
402: Filled squares show bright LBGs at $z\sim5$ in our previous results
403: (Paper I).
404: Open triangles and open-inverse triangles show bright LBG candidates
405: unidentified in our present observation and the previous results,
406: respectively.
407: So far, no bright spectroscopic targets (total of 22 LBG candidates)
408: have been identified as foreground galaxies, though a half of them
409: were not identified due to the low-S/N of their spectra.
410: In order to examine our color selection criteria further, we also show
411: the positions of the Galactic M stars which were identified in our
412: previous observations as filled pentagons in Figure 2.
413: Our selection
414: criteria for LBGs at $z\sim5$ are reasonable to exclude interlopers,
415: though more large and systematic spectroscopic observations are desirable.
416:
417: \begin{table}
418: \begin{center}
419: \caption{Photometric properties of the LBGs identified to be at $z\sim5$ in the J0053+1234 region.\label{tbl-1}}
420: \begin{tabular}{lccccccc}
421: \hline\hline
422: No. (ID) & RA(J2000) & DEC(J2000) &
423: $z'$\footnotemark[$*$] & $V-I_C$\footnotemark[$\dagger$] &
424: $I_C-z'$\footnotemark[$\dagger$] \\
425: %EW(LIS)\footnotemark[$\|$] &
426: %$\Delta V$(km s$^{-1}$)\footnotemark[$\#$] \\
427: \hline
428: 1 (106426) & 00:52:21.34 & +12:32:35.3 & 24.07 & 2.59 & 0.23 \\
429: 2 (104115) & 00:52:43.27 & +12:32:08.2 & 24.22 & 1.75 & 0.16 \\
430: 3 (091813) & 00:52:39.88 & +12:29:44.1 & 25.03 & 1.98 & 0.06 \\
431: 4 (093014) & 00:52:37.37 & +12:29:58.7 & 25.26 & $>$1.85 & 0.16 \\
432: \hline
433: %%%\multicolumn{8}{@{}l@{}}{\hbox to 0pt{\parbox{85mm}
434: \multicolumn{6}{@{}l@{}}{\hbox to 0pt{\parbox{160mm}{\footnotesize
435: \par\noindent
436: %Notes. --\\
437: %\noident
438: \footnotemark[$*$] $z'$ magnitude. MAG\_AUTO from SExtractor \citep{Bet}
439: is adopted. \\
440: %%%\noident
441: \footnotemark[$\dagger$] These values are measured with a
442: $1^{\prime\prime}.6$ aperture. \\
443: %%%\noident
444: }\hss}}
445: \end{tabular}
446: \end{center}
447: \end{table}
448:
449: \begin{figure}
450: \begin{center}
451: \FigureFile(160mm,100mm){fig2.eps}
452: \end{center}
453: \caption{
454: Positions of identified objects in the $I_C-z'$ and $V-I_C$ two-color
455: diagram.
456: Our color selection criteria for LBGs at $z\sim5$ (\cite{Iwa03}; 2007)
457: are indicated by thick lines.
458: The objects confirmed to be at $z\sim5$ in this study are shown as
459: filled circles with each identification number in Table 1.
460: The large circles show 2 bright objects, and small ones show 2 faint
461: objects.
462: Filled squares represent 7 bright LBGs at $z\sim5$ from Paper I.
463: Open triangles and open-inverse triangles show bright LBG
464: candidates unidentified in our present observation and the previous
465: results (Paper I), respectively.
466: Filled pentagons show objects identified to be foreground
467: objects (Galactic M stars) from Paper I.
468: A dashed (dot-dashed) line represents a color track of a model LBG
469: spectrum with $E(B-V)=0.0$ mag ($E(B-V)=0.4$ mag) from \citet{Iwa07}.
470: Small crosses are plotted on the tracks for $z \geq 4.0$ with an
471: interval of 0.1.
472: Model SEDs of LBGs are generated by using the population synthesis code
473: P\'{E}GASE version 2 \citep{pegase} with Salpeter IMF \citep{Sal55} and by
474: assuming constant star formation with an age of 100 Myr.
475: IGM absorptions at each redshift are calculated by using the analytic
476: formula by \citet{Inoue05}.
477: Calzetti's dust extinction curve \citep{Cal00} is assumed
478: to make model spectra with $E(B-V)=0.4$ mag.
479: A dotted line refers to a color track of an elliptical galaxy \citep{cww}.
480: Small open squares are plotted on the track with a redshift interval of 0.5.
481: Small open pentagons indicate the colors of A0 -- M9
482: stars calculated based on the library by \citet{Pick}.
483: }
484: \end{figure}
485:
486: The redshift distribution of the spectroscopically confirmed LBGs at $z\sim5$
487: in this study and Paper I is shown in Figure 3.
488: In Figure 3, we plot the expected redshift distribution of our
489: sample of LBGs at $z\sim5$ as a solid line normalized at $z=4.7$.
490: The distribution is calculated based on the expected detection rates
491: of the LBGs against the apparent magnitude and the redshift for each
492: survey field (figure 6 by \citet{Iwa07}).
493: However, the detection rates were derived through simulations
494: in the two-color diagram to estimate the completeness of the photometric
495: catalog, and they do not include the difference
496: of survey volumes for each redshift bin
497: and the difference of the number density at the corresponding UV
498: magnitude for an apparent magnitude at each redshift bin.
499: In order to derive the expected distribution and compare it with
500: the observed redshift distribution of the spectroscopic sample,
501: we corrected the detection rates
502: by adopting the redshift bin size of $\Delta z=0.2$
503: and assuming the UVLF of the $z\sim5$ LBGs \citep{Iwa07}.
504: Since almost all our spectroscopic sample covers $z'$ magnitude brighter
505: than 25.5 mag, we calculated the expected redshift distributions for
506: three magnitude ranges of 24.0$-$24.5 mag, 24.5$-$25.0 mag, and
507: 25.0$-$25.5 mag for each field, and took the average of them.
508: The distribution of spectroscopically confirmed LBGs is broadly
509: consistent with the expected distribution, but
510: there seems to be a deficiency of the LBGs at around
511: $z\sim5.0$.
512: Although the sample size is still small and the deficiency may
513: not be significant,
514: this may be due to the presence of severe night sky emission lines
515: which prevent us from detecting the continuum depression
516: and/or Ly$\alpha$ emission.
517: The night sky emission lines whose
518: wavelengths are converted to the redshifts corresponding to
519: the redshift of Ly$\alpha$ are also shown with dotted line
520: in the figure; the sky emission lines are severe for the
521: redshifts larger than 4.9.
522:
523: As described in section 3.1, spectroscopically identified LBGs in
524: the J0053 region have redshifts of $4.3-4.8$ with a mean of 4.5,
525: which is slightly lower than that of the objects in Paper I
526: with redshifts of $4.5-5.2$ (mean value of 4.7).
527: In fact, as seen in Figure 2,
528: the 4 objects in the J0053 region are bluer in
529: the $V-I_C$ color than the 7 objects in Paper I.
530: The mean $V-I_C$ color of spectroscopic sample is 2.62 mag and 2.04 mag
531: for the objects in the HDF-N region and in the J0053 region, respectively.
532: However the difference may be due to the variation among small
533: numbers of objects, and we should have more objects with spectroscopic
534: redshifts to examine this point further.
535:
536: \begin{figure}
537: \begin{center}
538: \FigureFile(140mm,80mm){fig3.eps}
539: \end{center}
540: \caption{Redshift distribution of LBGs at $z\sim5$ from this study and
541: our previous spectroscopic results (Paper I).
542: The night sky emission lines whose wavelengths are converted to the
543: redshifts corresponding to the redshift of Ly$\alpha$
544: are also shown with dotted line.
545: Expected redshift distribution normalized at $z=4.7$ for LBGs at $z\sim5$
546: is also plotted as solid line (right vertical axis,
547: see text for the details).
548: }
549: \end{figure}
550:
551: \subsection{Spectra of LBGs at $z\sim5$ in the J0053+1234 region}
552:
553: As shown in Figure 1, spectra of bright LBGs in the J0053+1234
554: region show Ly$\alpha$ and LIS absorptions (SiII $\lambda$1260, OI+SiII
555: $\lambda$1303, and CII $\lambda$1335) which are typical line features of
556: star-forming galaxies at low and high redshifts.
557: The Ly$\alpha$ EW$_{\rm rest}$s of object No.1 and No.2 are 14$\pm3$\AA\
558: and 21$\pm4$\AA , respectively, and their observed fluxes are
559: 2.5$\times 10^{-17}$ erg s$^{-1}$ cm$^{-2}$ and 2.4$\times 10^{-17}$
560: erg s$^{-1}$ cm$^{-2}$, respectively
561: corresponding to the luminosities of 5.8$\times 10^{42}$ erg s$^{-1}$ and
562: 4.3$\times10^{42}$ erg s$^{-1}$, respectively.
563: These values of Ly$\alpha$ do not include absorption component and are not
564: corrected for the IGM absorption.
565: The errors of EW values in this paper include
566: the uncertainty in determination of the continuum level
567: and the RMS error of the continuum.
568: The values of EW of Ly$\alpha$ are larger than the average value of
569: our previous results for 7 bright LBGs (2.5$\pm$3.7\AA , Paper I),
570: but they are not so large to be detected as definitive LAEs.
571: Typical Ly$\alpha$ EW$_{\rm rest}$ value for selecting LAEs is
572: 15$\sim$20\AA .
573: In contrast to the result for the bright targets,
574: the Ly$\alpha$ emissions of the faint LBGs are strong;
575: EW$_{\rm rest}$ of object No.3 and No.4 are 40$\pm12$\AA\ and
576: 77$\pm23$\AA , respectively.
577: The observed fluxes of Ly$\alpha$ are
578: 3.7$\times 10^{-17}$ erg s$^{-1}$ cm$^{-2}$ and 8.3$\times 10^{-17}$
579: erg s$^{-1}$ cm$^{-2}$, respectively corresponding to the luminosities of
580: 7.1$\times 10^{42}$ erg s$^{-1}$ and 1.7$\times10^{43}$ erg s$^{-1}$,
581: respectively.
582: We summarize the observed quantities in Table 2.
583:
584: The bright LBGs at $z\sim5$ tend to show no or weak Ly$\alpha$ emission;
585: average EW$_{\rm rest}$ of Ly$\alpha$ emission of the total of 9 bright
586: LBGs is $\sim5.8\pm7.2$\AA , while
587: strong Ly$\alpha$ emission is seen among the faint LBGs.
588: Because we can not deny the presence of faint LBGs with no or weak
589: Ly$\alpha$ emission from the current data, these results suggest,
590: at least, a deficiency of strong Ly$\alpha$ emission in bright LBGs at
591: $z\sim5$ and might suggest the
592: UV luminosity dependence of the Ly$\alpha$ emission.
593: Combining with other spectroscopic results from the literature,
594: we found that the deficiency is generally seen in LBGs at $z\sim5$
595: and possibly in LBGs at $z\sim6$ \citep{Ando06}.
596: The trend is also seen in the galaxies at $z=4\sim6$
597: (Vanzella et al. 2006; 2007).
598: Possible origins of the deficiency of strong Ly$\alpha$ emission
599: are discussed by \citet{Ando06}.
600: The deficiency or the UV luminosity dependence of the Ly$\alpha$
601: are also seen in the recent results of spectroscopy of LBGs at $z\sim3$;
602: \citet{Shap06} showed that
603: UV luminous ($M_{UV}\sim-21.6$ mag) LBGs at $z\sim3$ tend to show
604: weaker Ly$\alpha$ emission than the total LBG sample of \citet{Shap03}.
605: From the re-binned sample of \citet{Shap03} with their UV luminosity,
606: \citet{Keel06} showed that Ly$\alpha$ emission is
607: weaker in the UV luminous ($M_{UV}< -21.0$ mag) LBGs than in fainter LBGs.
608:
609: For object No.2, we identified LIS absorption lines
610: (SiII $\lambda$1260, OI+SiII $\lambda$1303, and CII $\lambda$1335)
611: at almost the same redshift as the Ly$\alpha$ emission.
612: We also identified SiIV $\lambda$1394 (EW$_{\rm rest}=-2.0\pm1.1$ \AA )
613: \footnote{We resolved SiIV $\lambda\lambda$1394,1403 doublet,
614: but $\lambda$1403
615: component was not detected significantly (comparable to 1$\sigma$),
616: which is consistent with the observed ratio for LBGs at $z\sim3$
617: \citep{Shap03}.}
618: and CIV $\lambda\lambda$1548,1551 (EW$_{\rm rest}=-4.9\pm1.4$\AA )
619: which are also seen in the spectra of LBGs at $\sim3$
620: (e.g., \cite{Shap03}).
621: The average EW of the three LIS absorption lines is $-4.2\pm1.4$\AA\
622: which is larger than the average value of our previous result
623: ($-2.8\pm1.2$\AA\ : Paper I).
624: The peak of Ly$\alpha$ emission line is redshifted by $280 \pm 210$
625: km s$^{-1}$ relative to the average of the LIS lines.
626: The peak-to-valley velocity offset which may be related to
627: a large scale gas outflow was also reported in LBGs at
628: $z=3\sim5$ (e.g., \cite{Shap03, frye}; Paper I).
629: The offset value of object No.2 is smaller than typical value
630: of LBGs at $z\sim3$ ($500-600$ km s$^{-1}$: \cite{Shap03})
631: and that of LBGs at $z\sim5$ ($\sim600$ km s$^{-1}$: Paper I).
632: For object No.1, we could not identify significant absorption
633: features due to the residuals of the sky subtraction.
634:
635: \begin{table}
636: \begin{center}
637: \caption{Spectroscopic properties of the LBGs identified to be at $z=4-5$ in the J0053+1234 region.\label{tbl-2}}
638: \begin{tabular}{lccccccc}
639: \hline\hline
640: No. (ID) & Redshift\footnotemark[$*$] &
641: EW(Ly$\alpha$)\footnotemark[$\dagger$] &
642: $F_{\rm Ly\alpha}$\footnotemark[$\ddagger$] &
643: $L_{\rm Ly\alpha}$\footnotemark[$\S$] \\
644: %EW(LIS)\footnotemark[$\|$] &
645: %$\Delta V$(km s$^{-1}$)\footnotemark[$\#$] \\
646: \hline
647: 1 (106426) & 4.797 & 14$\pm3$ & 2.5$\times 10^{-17}$ & 5.8$\times 10^{42}$ \\
648: 2 (104115) & 4.267 & 21$\pm4$ & 2.4$\times 10^{-17}$ & 4.3$\times 10^{42}$ \\
649: 3 (091813) & 4.391 & 40$\pm12$ & 3.7$\times 10^{-17}$ & 7.1$\times 10^{42}$ \\
650: 4 (093014) & 4.491 & 77$\pm23$ & 8.3$\times 10^{-17}$ & 1.7$\times 10^{43}$ \\
651: \hline
652: %%%\multicolumn{8}{@{}l@{}}{\hbox to 0pt{\parbox{85mm}
653: \multicolumn{5}{@{}l@{}}{\hbox to 0pt{\parbox{110mm}{\footnotesize
654: \par\noindent
655: %Notes. --\\
656: %\noident
657: \footnotemark[$*$] Redshifts are determined from the peak of
658: Ly$\alpha$ emission.
659: The error of the redshift determination is $\sim0.005$. \\
660: %%%\noident
661: \footnotemark[$\dagger$] Rest-frame equivalent width of Ly$\alpha$ emission in
662: units of angstroms, not including Ly$\alpha$ absorption.
663: EWs are taken to be positive for emission lines.
664: The error is estimated from
665: the uncertainty of the continuum level determination and the RMS error of
666: the continuum. \\
667: %%%\noident
668: \footnotemark[$\ddagger$] Observed flux of the Ly$\alpha$ emission line in
669: units of erg s$^{-1}$ cm$^{-2}$. \\
670: %%%\noident
671: \footnotemark[$\S$] Luminosity of the Ly$\alpha$ emission line
672: in units of erg s$^{-1}$.
673: The slit correction is not applied to the values. \\
674: %%%\noident
675: }\hss}}
676: \end{tabular}
677: \end{center}
678: \end{table}
679:
680: \subsection{Composite Spectrum of Luminous LBGs at $z\sim5$}
681:
682: In order to overcome the low-S/N of each spectrum and investigate
683: typical spectroscopic properties of luminous LBGs at $z\sim5$, we produced a
684: composite spectrum of 8 LBGs at $z\sim5$ with $M_{1400}<-21.5$ mag
685: by averaging the spectra obtained in this study and our previous study.
686: We derived the absolute UV magnitude from the observed broad band
687: magnitude to the rest-frame 1400\AA\ magnitude assuming a continuum
688: slope $\beta$ ($f_{\lambda}\propto\lambda^{\beta}$) of $-1$
689: which is a typical value for LBGs at $z\sim3$ \citep{Shap03}.
690: Before co-adding the spectra, all wavelengths were converted to the
691: rest-frame scale using individual redshifts, and each flux density
692: was normalized at 1250\AA .
693:
694: Figure 4 shows the resultant composite spectrum re-binned to
695: common dispersion of 1\AA\ pixel$^{-1}$.
696: Note that the emission-like features except for the Ly$\alpha$
697: (e.g., around 1275\AA , 1425\AA )
698: come from the residuals of sky emission lines,
699: and these features are not real.
700: The continuum depression shortward of Ly$\alpha$ by IGM is clearly seen in
701: the composite spectrum.
702: A depression factor $D_A$ \citep{Oke82} for the composite spectrum
703: is $\sim$0.6, which agrees with the value of
704: the composite spectrum of LBGs at $z\sim5$ (Paper I) and the values of
705: QSOs at $z\sim5$ ($0.1-0.7$: e.g., \cite{Son02,Son04}).
706: As expected from the previous subsection, EW$_{\rm rest}$ of the
707: Ly$\alpha$ emission is not large (7.6$\pm0.8$\AA ), and
708: LIS absorptions are also seen clearly.
709: The EWs of LIS absorptions are $-2.7\pm0.36$\AA , $-1.8\pm0.30$\AA ,
710: and $-1.5\pm0.29$\AA\
711: for SiII $\lambda$1260, OI+SiII $\lambda$1303, and CII $\lambda$1335,
712: respectively,
713: which suggests that bright LBGs have been chemically polluted to
714: some extent by metal at $z\sim5$.
715: The SiIV $\lambda\lambda$1394,1403 is also seen in the composite
716: spectrum, and the EW$_{\rm rest}$s of
717: the SiIV $\lambda1394$ and the SiIV $\lambda1403$
718: are $-2.0\pm0.32$ \AA\ and $-1.0\pm0.26$ \AA , respectively.
719:
720: The EW${_{\rm rest}}$ of Ly$\alpha$ is smaller than that of the composite
721: spectrum of LBGs at $z\sim3$ (15.1\AA : \cite{Shap03});
722: the EW values of LIS absorptions are comparable to or somewhat larger
723: than those of $z\sim3$
724: ($-1.7$\AA , $-2.3$\AA , and $-1.5$\AA\
725: for SiII $\lambda$1260, OI+SiII $\lambda$1303, and CII $\lambda$1335,
726: respectively).
727: However, the sample luminosity range of our LBGs at $z\sim5$
728: ($z' \lesssim 24.8$ mag corresponding to $M_{1400} \lesssim -21.5$ mag)
729: is different from that of LBGs at $z\sim3$
730: (mainly $R=23.5-25.5$ mag corresponding to $M_{1400}=-21.8\sim-19.8$ mag).
731: As described in the previous subsection, we found the deficiency of the
732: strong Ly$\alpha$ emission in luminous LBGs at $z\sim5$ and pointed out
733: the possible presence of the UV luminosity dependence on the spectroscopic
734: features.
735: Thus we can not discuss the evolution of the spectroscopic feature of LBGs
736: from a simple comparison of the composite spectra.
737:
738: \begin{figure}
739: \begin{center}
740: \rotatebox{270}{\resizebox{12cm}{!}{\FigureFile(140mm,80mm){f4.ps}}}
741: \end{center}
742: \caption{
743: Composite spectrum of 8 luminous ($M_{1400}<-21.5$ mag) LBGs at
744: $z\sim5$ from this study and Paper I.
745: Note that most of emission-like features (e.g. around 1275\AA , 1425\AA )
746: come from the residuals of sky emission lines.
747: }
748: \end{figure}
749:
750: Although the low S/N around the wavelength range
751: longer than 1400\AA\ does not allow us to estimate metallicity
752: by using 1425
753: index (e.g., \cite{Rix}) and CIV index (e.g., \cite{Meh02}),
754: we roughly estimate mean metallicity of the 8 luminous LBGs at $z\sim5$
755: from LIS absorption lines of the composite spectrum assuming the
756: relation between the mean EW of the three LIS absorptions
757: (SiII $\lambda$1260, OI+SiII $\lambda$1303, and CII $\lambda$1335)
758: and a metallicity calibrated with local star-forming galaxies \citep{Hek98}.
759: Since the sample by \citet{Hek98} does not cover the low metallicity
760: (i.e., weak LIS absorptions) region, we extrapolated the EW(LIS)-metallicity
761: relation in the middle-to-high (12+log(O/H)$\gtrsim$8.0)
762: metallicity region to the lower metallicity region.
763: The estimated metallicity of 12+log(O/H) is about 7.6$\pm0.6$
764: corresponding to $\sim$0.1 (0.3$-$0.02) solar value (8.66: \cite{Asp04}).
765: The errors come from the uncertainty of the EW measurement as well as
766: that of the empirical relation between the EW(LIS) and metallicity.
767: It should be worth noting that the LIS lines are saturated,
768: and EW(LIS) reflects not only the metallicity but also the velocity
769: structure and covering fraction of the gas (e.g., \cite{Hek98,Shap03,Noll04}).
770: \citet{Erb06} show a composite rest-frame UV spectrum
771: of objects at $z\sim2$ with a mean stellar mass of $7\times10^{10}M_{\odot}$
772: and mean metallicity of 12+log(O/H)$\sim8.55$
773: derived from N2 index method.
774: The EW(LIS) in the spectrum is $\sim2-3$\AA\
775: corresponding to 12+log(O/H) of $\sim7.6-8.2$
776: if we use the local relation by \citet{Hek98}.
777: The result suggests that the local EW(LIS)-metallicity relation
778: may systematically underestimate the metallicity of galaxies
779: at high redshift.
780:
781: \subsection{A Hint for Chemical Evolution of massive LBGs at $z\sim5$?}
782:
783: We derived the individual metallicity for 8 LBGs at $z\sim5$ with the
784: LIS absorption in this study and Paper I with the same manner mentioned above.
785: For the 3 objects in the HDF-N region among the 8 LBGs,
786: Subaru/CISCO $K'$ band data and Spitzer/IRAC mid-IR data are available,
787: and we can estimate their stellar masses by
788: fitting their rest-frame UV to optical SEDs to the model SEDs.
789: (Details of NIR and MIR data of our LBGs at $z\sim5$
790: will be described elsewhere.)
791: The metallicities (12+log(O/H)) of
792: 104268 (No.1 of Paper I), 144200 (No.2 of Paper I),
793: and 148198 (No.8 of Paper I)
794: are 7.7$\pm 0.8$, 7.7$\pm 0.8$, and 8.7$\pm 0.9$, respectively.
795: For the stellar mass estimation, we adopted the same manner used in
796: \citet{Iwa05}.
797: We used the P\'{E}GASE version 2 population synthesis model \citep{pegase}
798: with Salpeter IMF \citep{Sal55}, constant star formation history,
799: an the Calzetti's dust extinction curve \citep{Cal00} to make model spectra.
800: IGM absorptions were calculated for the given redshift
801: by using the analytic formula by \citet{Inoue05}.
802: Then the observed SEDs were fit to model SEDs, and the best fit model was
803: obtained by minimizing $\chi^2$.
804: Resultant stellar masses (best fit value with 68\% confidence level)
805: of the objects 104268, 144200, and 148198 are
806: 5.82$^{+6.95}_{-3.93}\times10^{10}M_{\odot}$,
807: 3.33$^{+0.94}_{-2.92}\times10^{10}M_{\odot}$, and
808: 1.33$^{+0.67}_{-0.90}\times10^{11}M_{\odot}$, respectively.
809:
810: In Figure 5, open circles represent the stellar masses and the
811: metallicities of the three luminous LBGs at $z\sim5$, and
812: filled circle represents the average value of them.
813: Pluses, diamonds, and triangles show the data for galaxies at
814: $z\sim0.1$, $z\sim0.7$ and $z\sim2$, respectively \citep{Tre,Sav,Erb06},
815: which show the presence of mass-metallicity relation at each
816: redshift and its evolution from $z\sim2$ to $z\sim0$
817: \footnote{The metallicity of galaxies at $z\sim2$ \citep{Erb06} was
818: derived from N2 index (e.g., \cite{Pet04}) which is different from
819: methods used by \citet{Tre} and \citet{Sav} and may have a problems in
820: calibration (e.g., \cite{Erb06})
821: which may partly be a cause for the offset.}.
822: Although the uncertainties are quite large (and may contain some
823: systematics),
824: and our way of metallicity
825: estimation is different from those for galaxies at other redshifts,
826: the metallicities of the LBGs at $z\sim5$ tend to be lower than those of the
827: galaxies with the same stellar mass at $z\lesssim2$
828: \footnote{The stellar masses by \citet{Erb06} were derived by using
829: Chabrier IMF.
830: Thus our data points should be $\sim$1.8 times smaller (e.g., \cite{Erb06})
831: in Figure 5 in order to compare them with the data by \citet{Erb06}.}.
832:
833: \begin{figure}
834: \begin{center}
835: \FigureFile(140mm,80mm){fig5.eps}
836: \end{center}
837: \caption{Stellar mass-metallicity relation for galaxies at various redshifts.
838: Open circles represent our results of 3 luminous LBGs at $z\sim5$, and
839: filled circle represents the average value of them.
840: Pluses, diamonds, and triangles show the results for galaxies at
841: $z\sim0.1$, $z\sim0.7$ and $z\sim2$, respectively \citep{Tre,Sav,Erb06}.
842: }
843: \end{figure}
844:
845: \bigskip
846:
847: We are grateful to the FOCAS team, especially Youichi Ohyama,
848: and the staffs of Subaru telescope for their supports during our
849: observation.
850: We also thank Kiyoto Yabe and Marcin Sawicki
851: for useful discussions in deriving stellar masses.
852: We appreciate the anonymous referee for the comments which improved
853: this paper.
854: The preparation of the observation was in part carried out on "sb"
855: computer system
856: operated by the Astronomical Data Analysis Center (ADAC) and Subaru
857: Telescope of the National Astronomical Observatory of Japan.
858: MA is supported by a Research Fellowship of the Japan Society for the
859: Promotion of Science for Young Scientists.
860: KO is supported by a Grant-in-Aid for Scientific Research from
861: Japan Society for the Promotion of Science (17540216).
862: This work is supported by the Grant-in-Aid for the 21st Century COE
863: "Center for Diversity and Universality in Physics" from the MEXT of Japan.
864:
865: \begin{thebibliography}{}
866: \bibitem[Asplund et al.(2004)]{Asp04}Asplund, M., Grevesse, N., Sauval, A. J., Allende Prieto, C., \& Kiselman, D. 2004, \aap, 417, 751
867: \bibitem[Ando et al.(2004)]{Ando04}Ando, M., Ohta, K., Iwata, I., Watanabe, C., Tamura, N., Akiyama, M., \& Aoki, K. 2004, \apj, 610, 635 (Paper I)
868: \bibitem[Ando et al.(2006)]{Ando06}Ando, M., Ohta, K., Iwata, I., Akiyama, M., Aoki, K., \& Tamura, N. 2006, \apj, 645, L9
869: \bibitem[Bertin and Arnouts(1996)]{Bet}Bertin, E., \& Arnouts, S. 1996, \aaps, 117, 393
870: \bibitem[Bouwens et al.(2005)]{Bou05}Bouwens, R. J., Illingworth, G.D., Thompson, R. I., \& Franx, M. 2005, \apj, 624, L5
871: \bibitem[Bouwens et al.(2006)]{Bou06a}Bouwens, R. J., Illingworth, G.D., Blakeslee, J. P., \& Franx, M. 2006, \apj, 653, 53
872: \bibitem[Bouwens \& Illingworth(2006)]{Bou06b}Bouwens, R. J., \& Illingworth, G.D. 2006, \nat, 443, 189
873: \bibitem[Calzetti et al.(2000)]{Cal00}Calzetti, D., Armus, L., Bohlin, R. C., Kinney, A. L., Koornneef, J., \& Storchi-Bergmann, T. 2000, \apj, 533, 682
874: \bibitem[Coleman et al.(1980)]{cww}Coleman, G. D., Wu, C.-C., \& Weedman, D. W. 1980, \apjs, 43, 393
875: \bibitem[Dey et al.(1998)]{Dey98}Dey, A., Spinrad, H., Stern, D., Graham, J. R., \& Chaffee, F. H. 1998, \apj, 498, L93
876: \bibitem[Dickinson et al.(2004)]{Dick04}Dickinson, M., et al. 2004, \apj, 600, L99
877: \bibitem[Dow-Hygelund et al.(2005)]{Dow05}Dow-Hygelund, C. C., et al. 2005, \apj, 630, L137
878: \bibitem[Erb et al.(2006)]{Erb06}Erb, D. K., Shapley, A. E., Pettini, M., Steidel, C. C., Reddy, N. A., \& Adelberger, K. L. 2006, \apj, 644, 813
879: \bibitem[Fioc \& Rocca-Volmerange(1997)]{pegase}Fioc, M., \& Rocca-Volmerange, B. 1997, \aap, 326, 950
880: \bibitem[Frye et al.(2002)]{frye}Frye, B., Broadhurst, T., \& Benitez, N. 2002, \apj, 568, 558
881: \bibitem[Heckman et al.(1998)]{Hek98}Heckman, T. M., Robert, C.,
882: Leitherer, C., Garnett, D. R., \& van der
883: Rydt, F. 1998, \apj, 503, 646
884: \bibitem[Inoue et al.(2005)]{Inoue05}Inoue, A. K., Iwata, I., Deharveng,
885: J.-M., \& Burgarella, D. 2005, \aap, 435,
886: 471
887: \bibitem[Iwata et al.(2003)]{Iwa03}Iwata, I., Ohta, K., Tamura, N.,
888: Ando, M., Wada, S., Watanabe, C., Akiyama,
889: M., \& Aoki, K. 2003, \pasj, 55, 415
890: \bibitem[Iwata, Inoue \& Burgarella(2005)]{Iwa05}Iwata, I., Inoue,
891: A. K., \& Burgarella, D. 2005, \aap, 440,
892: 881
893: \bibitem[Iwata et al.(2007)]{Iwa07}Iwata, I., Ohta, K., Tamura, N.,
894: Akiyama, M., Aoki, K., Ando, M., Kiuchi,
895: G., \& Sawicki, M. 2007, \mnras, 376, 1557
896: \bibitem[Iye et al.(2004)]{Iye04}Iye, M., et al. 2004, \pasj, 56, 381
897: \bibitem[Kashikawa et al.(2002)]{kashi}Kashikawa, N., et al. 2002, \pasj, 54, 819
898: \bibitem[Keel(2006)]{Keel06}Keel, W. C. 2006, \aj, 131, 2755
899: \bibitem[Kneib et al.(2004)]{Kneib04}Kneib, J.-P., Ellis, R. S., Santos, M. R., \& Richard, J. 2004, \apj, 607, 697
900: \bibitem[Kobulnicky \& Koo(2000)]{Kob00}Kobulnicky, H. A., \& Koo, D. C. 2000, \apj, 545, 712
901: \bibitem[Kobulnicky \& Kewley(2004)]{Kob04}Kobulnicky, H. A., \& Kewley, L. J. 2004, \apj, 617, 261
902: \bibitem[Le F\`{e}vre et al.(2005)]{Lefe05}Le F\`{e}vre, O., et al. 2005, \aap, 439, L845
903: \bibitem[Lehnert \& Bremer(2003)]{Lehn03}Lehnert, M. D., \& Bremer, M. 2003, \apj, 593, 630
904: \bibitem[Mannucci et al.(2007)]{Mann07}Mannucci, F., Buttery, H., Maiolino, R., Marconi, A., \& Pozzetti, L. 2007, \aap, 461, 423
905: \bibitem[Mehlert et al.(2002)]{Meh02}Mehlert, D., et al. 2002, \aap, 393, 809
906: \bibitem[Miyazaki et al.(2002)]{Miya}Miyazaki, S., et al. 2002, \pasj, 54, 833
907: \bibitem[Noll et al.(2004)]{Noll04}Noll, S., et al. 2004, \aap, 418, 885
908: \bibitem[Oke \& Korycansky(1982)]{Oke82}Oke, J. B., \& Korycansky, D. G. 1982, \apj, 255, 11
909: \bibitem[Oke \& Gunn(1983)]{Oke83}Oke, J. B., \& Gunn, J. E. 1983, \apj, 266, 713
910: \bibitem[Ouchi et al.(2004)]{Ouchi04a}Ouchi, M., et al. 2004, \apj, 611, 660
911: \bibitem[Pello et al.(2004)]{Pello04}Pello, R., Schaerer, D., Richard, J., Le Borgne, J.-F., \& Kneib, J.-P. 2004, \aap, 416, L35
912: \bibitem[Pettini et al.(2001)]{Pet01}Pettini, M., Shapley, A. E., Steidel, C. C., Cuby, J. G., Dickinson, M., Moorwood, A. F. M., Adelberger, K. L., \& Giavalisco, M. 2001, \apj, 554, 981
913: \bibitem[Pettini \& Pagel(2004)]{Pet04}Pettini, M., \& Pagel, B. E. J. 2004, \mnras, 348, L59
914: \bibitem[Pickles(1998)]{Pick}Pickles, A.J. 1998, \pasp, 110, 863
915: \bibitem[Rix et al.(2003)]{Rix}Rix, S. A., Pettini, M., Leitherer, C.,
916: Bresolin, F., Kudritzki, R-P., \& Steidel,
917: C. C. 2004, \apj, 615, 98
918: \bibitem[Salpeter(1955)]{Sal55}Salpeter, E. E. 1955, \apj, 121, 161
919: \bibitem[Savaglio et al.(2005)]{Sav}Savaglio, S., et al. 2005, \apj, 635, 260
920: \bibitem[Shapley et al.(2003)]{Shap03}Shapley, A. E., Steidel, C. C., Pettini, M., \& Adelberger, K. L. 2003, \apj, 588, 65
921: \bibitem[Shapley et al.(2004)]{Shap04}Shapley, A. E., Erb, D. K., Pettini, M., Steidel, C. C., \& Adelberger, K. L. 2004, \apj, 612, 108
922: \bibitem[Shapley et al.(2006)]{Shap06}Shapley, A. E., Steidel, C. C.,
923: Pettini, M, Adelberger, K. L., \& Erb,
924: D. K. 2006, \apj, 651, 688
925: \bibitem[Shimasaku et al.(2005)]{Shima05}Shimasaku, K., Ouchi, M.,
926: Furusawa, H., Yoshida, M., Kashikawa,
927: N., \& Okamura, S. 2005, \pasj, 57, 447
928: \bibitem[Songaila \& Cowie(2002)]{Son02}Songaila, A., \& Cowie, L. L. 2002, \aj, 123, 2183
929: \bibitem[Songaila(2004)]{Son04}Songaila, A. 2004, \aj, 127, 2598
930: \bibitem[Spinrad et al.(1998)]{Spi98}Spinrad, H., et al. 1998, \aj, 116, 2617
931: \bibitem[Stanway et al.(2003)]{Sta03}Stanway, E. R., Bunker, A. J., \& McMahon, R. G. 2003, \mnras, 342, 439
932: \bibitem[Stanway et al.(2004)]{Sta04}Stanway, E., Bunker, A., McMahon, R., Ellis, R. S., Treu, T., \& McCarthy, P. J. 2004, \apj, 607, 704
933: \bibitem[Steidel et al.(1996)]{Ste96a}Steidel, C. C., Giavalisco, M., Dickinson, M., \& Adelberger, K. L. 1996, \aj, 112, 352
934: \bibitem[Steidel et al.(2003)]{Ste03}Steidel, C. C., Adelberger, K. L., Shapley, A. E., Pettini, M., Dickinson, M., \& Giavalisco, M. 2003, \apj, 592, 728
935: \bibitem[Steidel et al.(2004)]{Ste04}Steidel, C. C., Shapley, A. E., Pettini, M., Adelberger, K. L., Erb, D. K., Reddy, N. A., \& Hunt, M. P. 2004, \apj, 604, 534
936: \bibitem[Teplitz et al.(2000)]{Tep00}Teplitz, H. I., et al. 2000, \apj, 533, L65
937: \bibitem[Tremonti et al.(2004)]{Tre}Tremonti, C. A., et al. 2004, \apj, 613, 898
938: \bibitem[Vanzella et al.(2006)]{VLT}Vanzella, E., et al. 2006, \aap, 454, 423
939: \bibitem[Vanzella et al.(2007)]{Van07}Vanzella, E., et al. 2007,
940: Proceedings for the
941: conference "At the Edge of the Universe:
942: Latest results from the deepest
943: astronomical surveys", (astro-ph/0612182)
944: \bibitem[Weymann et al.(1998)]{Wey98}Weymann, R. J., Stern, D., Bunker, A., Spinrad, H., Chaffee, F. H., Thompson, R. I., \& Storrie-Lombardi, L. J. 1998, \apj, 505, L95
945: \bibitem[Yan et al.(2005)]{Yan05}Yan, H., et al. 2005, \apj, 634, 109
946: \bibitem[Yoshida et al.(2006)]{Yoshida06}Yoshida, M., et al.
947: 2006, \apj, 653, 988
948: \end{thebibliography}
949:
950: \end{document}
951: