1: % sample pages for Proceedings IAU Colloquium document class
2: % (based on v1.0 cca2esam.tex)
3: % v1.04 released 17 May 2004 by TechBooks
4: %% small changes and additions made by KAvdH/IAU 4 June 2004
5: % Copyright (2004) International Astronomical Union
6: \NeedsTeXFormat{LaTeX2e}
7: \documentclass{iauc}[12pt]
8: \renewcommand{\topfraction}{.85}
9: \renewcommand{\bottomfraction}{.7}
10: \renewcommand{\textfraction}{.15}
11: \renewcommand{\floatpagefraction}{.66}
12: \usepackage{graphicx}
13: \pubyear{2005}
14: \volume{199}
15: \pagerange{1--}
16: \setcounter{page}{1}
17: \jname{Probing Galaxies through Quasar Absorption Lines}
18: \editors{P. R. Williams, C. Shu, and B. M\'{e}nard, eds.}
19: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20: \title[Spectroscopic Studies of \boldmath{$z \sim5.7$} and $z\sim6.5$ Galaxy Samples] %% %%
21: {Spectroscopic Studies of \boldmath{$z \sim5.7$} and $z\sim6.5$ Galaxies: Implications for Reionization}
22:
23: \author[Hu et al.] %% give here short author list %%
24: {Esther M. Hu, Lennox L. Cowie, Peter Capak%
25: \thanks{Present address: Department of Astronomy, California Institute of Technology, MS 105-24, 1201 E. California Blvd., Pasadena, CA, 91125, USA.},
26: \and Yuko Kakazu}
27:
28: \affiliation{Institute for Astronomy, University of Hawaii,
29: 2680 Woodlawn Drive, Honolulu, HI 96822, USA}%\break
30: %{email: %hu@ifa.hawaii.edu,{\thinspace}cowie@ifa.hawaii.edu,{\thinspace}capak@astro.caltech.%edu,{\thinspace}kakazu@ifa.hawaii.edu}}%\\[\affilskip]
31:
32: \begin{document}
33:
34: \maketitle
35:
36: \begin{abstract}
37: The recent development of large, complete samples which identify
38: high-redshift galaxies at $z\sim5.7$ and $z\sim 6.5$ from deep,
39: wide-field surveys provides detailed information on the earliest
40: galaxies, their numbers, spatial and kinematic distributions,
41: and implications for early reionization of the IGM. In this
42: contribution we present results of spectroscopic studies of $z\sim
43: 5.7$ and $z\sim6.5$ galaxies identified from our deep, Lyman alpha
44: narrowband and multicolor surveys conducted with the SuprimeCam
45: mosaic CCD camera on the 8.3-m Subaru telescope and observed
46: with the DEIMOS multi-object spectrograph on Keck. The luminosity
47: function of the $z\sim6.5$ galaxies is shown to be similar to the
48: luminosity function of the $z\sim 5.7$ galaxy samples, suggesting
49: that a substantial star-forming population is already in place at
50: $z\sim 6.5$. Comparisons of both individual and stacked spectra of
51: galaxies in these two samples show that the Lyman alpha emission
52: profiles, equivalent widths, and continuum break strengths do not
53: substantially change over this redshift interval. The wide-field
54: nature of the surveys also permits mapping the large-scale
55: distribution of the high-redshift galaxies in spatial structures
56: extending across individual SuprimeCam fields ($\sim 60$ Mpc).
57: Field-to-field variations in the number of objects at $z\sim 6.5$
58: may shortly be able to place constraints on the porosity of the
59: reionization boundary.
60:
61: \keywords{cosmology: observations --- early universe --- galaxies: distances
62: and redshifts --- galaxies: evolution --- galaxies: formation --- galaxies: high-redshift}
63:
64:
65: %\keywords{Keyword1, keyword2, keyword3, etc.}
66: %% add here a maximum of 10 keywords, to be taken form the file <Keywords.txt>.
67:
68: \end{abstract}
69: \firstsection % if your document starts with a section,
70: % remove some space above using this command.
71: \section{Introduction}
72: Substantial progress has been made over the last 5 years in
73: high-redshift galaxy studies (see \cite{spinrad04} for a review).
74: The significant aspects of recent work carried out since that review
75: are the rapid growth in the numbers of galaxies that populate
76: the high-redshift end of the samples ($z\sim6$ and above) and
77: the development of large, spectroscopically complete samples from
78: wide-area surveys for a number of discrete fields (e.g., \cite{lum5},
79: \cite{tani}, \cite{hdfz5}). These give a handle on the problem
80: of cosmic variance, and also allow us to build up the large samples
81: required for luminosity function studies. Spectroscopic completeness
82: makes it possible to estimate the star-forming population at high
83: redshift, and its evolution. Detailed examination and comparison
84: of the strength and profile shape of the Ly$\alpha$ emission line
85: may provide insights to possible changes in the underlying stellar
86: population, or in the neutral hydrogen content of the surrounding
87: intragalactic medium.
88:
89: \section{The Hawaii Wide-Field Narrowband Surveys}
90: The most successful method of identifying galaxies at redshifts
91: beyond $z\sim 5.5$, where galaxy continuua are faint against a
92: strong nightsky background, is by using the redshifted Ly$\alpha$
93: emission line. The present surveys were were driven by the
94: development of wide-field, red-sensitive instruments on the large
95: telescopes: the half-degree FOV SuprimeCam mosaic CCD imager
96: (Miyazaki et al.\ 2002) on the 8.3-m Subaru Telescope, and the
97: wide-field ($16.7'\times 5'$) DEIMOS multi-slit spectrograph
98: and imager on the Keck II 10-m telescope. Narrowband imaging is
99: used in combination with deep multicolor $BVRIz'$ and moderate
100: infrared $K'$ imaging over the SuprimeCam fields to identify
101: high-redshift galaxies with strong Ly$\alpha$ emission and
102: continuum color breaks.
103:
104: Current investigations use $\sim 120$\AA\
105: bandpass filters centered at 8150\AA\ and 9130\AA\ (Ly$\alpha$
106: at $z\sim 5.7$ and $z\sim 6.5$) for the narrowband studies, with
107: follow-up DEIMOS spectra (3.6\AA\ resolution, $R\sim2700$) at Keck.
108: Survey fields are chosen from half a dozen extensively studied
109: fields: SSA22, GOODS-N, Lockman Hole NW, SSA13, SSA17, and A370.
110: Completed studies at $z\sim 5.7$ for SSA22 (\cite{lum5}) show a
111: high success rate in identifying and confirming candidates (19
112: confirmed of 22 observed), as well as success in discriminating
113: against red stars and identifying/confirming T dwarf candidates
114: (\cite{kakazu}). A second $z\sim5.7$ study has been completed for
115: the region including the HDF-N and GOODS-N fields (\cite{hdfz5}).
116: At present 62 confirmed redshifts at $z\sim5.7$ are in hand for
117: 3 fields, with 14 confirmed $z\sim6.5$ redshifts (extending to
118: $z=6.74$) for 3 fields.
119:
120: \begin{figure}[tp]
121: \includegraphics[width=5.2in]{hu01.ps}
122: \caption{({\it Left panel\/}) shows the Hubble Deep Field (HDF-N)
123: imaged with the WFPC2 camera on HST through the F814W continuum
124: filter. ({\it Right panel\/}) shows the same region imaged from
125: the ground through a $120\,$\AA-wide narrowband NB816 filter (for
126: Ly$\alpha$ emission at $z\sim 5.7$) using the SuprimeCam mosaic
127: CCD camera on Subaru. The narrowband images in the survey are
128: comparable in depth to the HDF, and each field center surveys a
129: region 70 times larger than the area shown here, with good image
130: quality. Wide area coverage and image depth are required to identify
131: and study the high-redshift samples.}\label{fig:hdf_nb_comp}
132: \end{figure}
133: At magnitudes brighter than $AB\sim25.5$ the number density of such
134: high-redshift galaxies is only a few hundred per square degree and
135: the distribution is highly correlated on sub-degree scales (e.g.,
136: Capak 2004). A combination of depth and area coverage is needed
137: for these studies. Figure~\ref{fig:hdf_nb_comp} shows the depth
138: and image quality of the survey's SuprimeCam narrowband exposures
139: compared to deep HDF continuum images. No $z\sim5.7$ galaxies
140: are found in the survey within the HDF-N. Over the larger
141: ACS GOODS-N field (still a sub-region of the SuprimeCam
142: field) only 6 $z\sim5.7$ galaxies were found. And none of the
143: $z\sim6.5$ sources were found within the region of ACS GOODS-N.
144:
145:
146: \section{Luminosity Functions}
147: With the high-redshift samples in hand, we can construct luminosity
148: functions at $z\sim5.7$ and $z\sim6.5$, and compare these to
149: results at lower redshifts. The most direct comparison is to use
150: the UV continuum luminosities of the Ly$\alpha$-selected samples and
151: compare these with luminosity functions of the Lyman-break galaxies
152: (\cite{stei1999}, \cite{stei2000}) at redshifts $z\sim3$ and $z\sim4$
153: (Figure~\ref{fig:uvlum}). This avoids problems of estimating
154: possible dust contamination on the emission line.
155: Here the main uncertainty is how to
156: correct the raw counts for the fractional population with strong
157: Ly$\alpha$ emission. If the estimate of \cite{stei2000}
158: based on the LBGs at $z\sim3-4$ with strong Ly$\alpha$ is used
159: to assume that the Ly$\alpha$ emitters represent only 20\% of
160: the star-forming galaxy population at $z\sim5.7$, then the raw
161: numbers for Ly$\alpha$-selected galaxies ({\it filled boxes\/})
162: are corrected upwards to the {\it open boxes\/} that overlie the
163: $z\sim3$ and $z\sim4$ Lyman break galaxy UV continuum luminosity
164: functions, implying there has not been a substantial decrease in
165: the star formation rate out to redshift $z\sim5.7$.
166:
167: \begin{figure}%\hskip1.6in
168: \centering
169: \includegraphics[width=4.2in,scale =0.9]{hu02.ps}
170: \caption{The UV continuum luminosity function of identified
171: $z\sim 5.7$ galaxies ({\it boxes\/}) compared with the UV continuum
172: luminosity functions at $z\sim3$ ({\it diamonds}) and at $z\sim 4$
173: reported by \cite{stei1999}. Measured points are shown as {\it
174: solid boxes} with $1\sigma$ Poisson uncertainties based on the
175: number of objects in each bin; the open boxes that match closely
176: to the lower redshift luminosity functions show assumed values if
177: emitters pick out 20\% of the LBG sample as \cite{stei2000} find
178: to be the case at $z\sim3$.}\label{fig:uvlum}
179: \end{figure}
180:
181: For the $z\sim6.5$ sample the longest wavelength deep continuum band
182: ($z'$ around 9200 \AA) spans the wavelength of Ly$\alpha$ emission, making
183: it difficult to estimate the line-free ultraviolet continuum.
184: Here, we compare the Ly$\alpha$ luminosity function with
185: the luminosity function of the emission line obtained in narrowband samples
186: at $z\sim3.4$ (\cite{cow98}, \cite{smitty}).
187: The results are shown in Figure~\ref{fig:lalum}. The $z\sim 3.4$ points
188: are plotted with {\it open boxes\/} and the $z\sim 5.7$ points are shown
189: with {\it filled boxes\/}. The vertical dashed line indicates the limits of
190: the sample, which does not reach the flat part of the $z\sim5.7$
191: luminosity function. The {\it filled box and upper limit drawn with heavy bars\/}
192: represent the points for the $z\sim6.5$ luminosity function and overlie the
193: $z\sim5.7$ luminosity function data. The points
194: are consistent with no marked change in the luminosity function
195: between $z\sim5.7$ and $z\sim6.5$ as suggested by \cite{mal}, who used
196: a likelihood analysis to merge disparate data samples with varying
197: degrees of incompleteness to compare the two redshift samples.
198: The advantage of the present results lies in the larger and
199: homogeneous samples, and the completeness of identifications.
200:
201: \begin{figure}%\hskip1.6in
202: \centering
203: \includegraphics[width=4.2in,scale =0.9]{hu03.ps}
204: \caption{The Ly$\alpha$ luminosity function of identified
205: $z\sim 5.7$ galaxies ({\it filled boxes\/}) compared with the Ly$\alpha$
206: luminosity functions at $z\sim3.4$ ({\it open boxes})
207: (\cite{cow98}, \cite{smitty}). Measured points are shown with $1\sigma$ Poisson uncertainties based on the
208: number of objects in each bin. The vertical dashed line indicates the limits of
209: the $z\sim5.7$ survey, which does not extend to the faint end of the $z\sim3.4$
210: Ly$\alpha$ luminosity function. The points and upper limits for the $z\sim6.5$
211: luminosity function are shown with the thick bars, and are consistent with the
212: $z\sim5.7$ luminosity function, but are based on only 14 galaxies.}\label{fig:lalum}
213: \end{figure}
214:
215: \section{Ly$\alpha$ Emission Line Profiles}
216: If reionization occurred at redshifts lower than
217: $z=6.5$ we should see a major change in the equivalent widths
218: of the Lyman alpha emission line between the $z=6.5$ sample and
219: the $z=5.7$ sample since the damping wings of the neutral gas
220: should scatter much of the redward wing of the Lyman alpha
221: lines at the higher redshifts where the intergalactic gas
222: is neutral. We can examine the emission-line
223: profiles in the two samples. Figure~\ref{fig:z5z6comp} shows that
224: the Ly$\alpha$ profiles, which have been stacked to improve
225: our sensitivity, are extremely similar. Both profiles show the
226: strong blue asymmetry due to the Ly$\alpha$ forest that is
227: characteristic of the very high-redshift galaxies. The profiles have
228: similar equivalent widths and shapes.
229:
230: Does this provide clear evidence that the epoch of reionization must lie
231: at higher redshifts? Not necessarily. For Ly$\alpha$-emitting galaxies
232: that are sufficiently luminous, self-clearing can complicate their effectiveness
233: as a probe of the surrounding neutral IGM (e.g., \cite{haiman}).
234: Even for lower luminosity objects, clustered distributions or neighboring
235: objects may ionize the local region. The $z\sim 6.5$ samples are
236: still relatively small, and show a surprising amount of field-to-field
237: variation, with 10 of the identified systems coming from the A370 field --
238: coincidentally, the first field with a discovered $z>6$ galaxy (\cite{z6}).
239: Preferred (low density) lines of sight could be ionized, and there
240: can be selection biases in our $z\sim 6.5$ sample.
241: The properties could reflect large-scale structure effects
242: or a porous reionization boundary in both the profile and
243: luminosity function analyses.
244:
245:
246: \begin{figure}\hskip1.6in
247: \centering
248: \includegraphics[width=3.3in,angle=90,scale =0.9]{hu04.ps}
249: \vspace{-0.25in}
250: \caption{ Stacked profiles of identified $z\sim 6.5$ galaxies
251: and $z\sim5.7$ galaxies. The dotted profile shows the spectral
252: resolution from the night sky lines, and the steep blue fall-off
253: due to absorption by neutral hydrogen is clearly evident in the
254: asymmetric line profile. The profiles are extremely similar in
255: equivalent width (56\AA\ and 60\AA) and shape. This may suggest
256: that the reionization epoch is at higher redshift, since we might
257: hope to see a change in the profiles due to the expected increase
258: in scattering redward of Ly$\alpha$ from hydrogen damping wings
259: due to the surrounding neutral medium if we sample galaxies above
260: and below the reionization redshift (suggested at $z\sim 6.2$).}\label{fig:z5z6comp}
261: \end{figure}
262:
263: \section{Large-Scale Structure}
264: The $z\sim5.7$ spectroscopic samples (\cite{lum5}, \cite{hdfz5}) with near-complete coverage in the SSA22 and HDF-N SuprimeCam
265: fields provided the first quantitative demonstration of the existence of structured kinematic and redshift structures on large scales ($\sim 60$ Mpc co-moving
266: distance) at these redshifts. The long diagonal filament in {\it right panel\/}
267: of Figure~\ref{fig:structure} makes up a distinct redshift system with Ly$\alpha$
268: emission falling in the bottom quarter of the filter. Such features are also seen in
269: maps of the HDF-N. Similar structured distribution
270: at $z\sim 5.7$ has recently been reported by \cite{ouchi} for the Subaru XMM Deep
271: Field.
272:
273: \section{Conclusions}
274: We are now able to obtain large samples of $z\sim5.7$ and $z\sim6.5$ galaxies.
275: The continuum and Ly$\alpha$ luminosity functions seem to be similar to those
276: at lower redshifts, indicating a strong contribution to to the ionization of the
277: IGM from star-formation at these redshifts. Both the Ly$\alpha$ luminosity function and
278: the Ly$alpha$ line shapes are similar at $z\sim5.7$ and $z\sim6.5$. Large scale
279: structure revealed in the distribution of these high-redshift galaxies leads us to the
280: exciting conclusion that we may
281: be able to make 3-dimensional maps of the cosmic web at these redshifts!
282:
283: \begin{figure}[ht]
284: \includegraphics[angle=90, width=2.3in]{hu05.ps}\hskip 0.3in
285: \includegraphics[width=2.3in]{hu06.ps}
286: \caption{({\it Left})The $z\sim6.55$ emitters in A370 display a filamentary
287: extent. The box
288: indicates a $z\sim 6.74$ galaxy, currently, the highest confirmed
289: redshift. ({\it Right}) Redshift $z\sim5.7$ structure in the expanded SSA22 field.
290: Small crosses (circled region) show candidates not yet spectroscopically observed
291: along the extended filament.
292: }\label{fig:structure}
293: \end{figure}
294: \begin{acknowledgments}
295: This research was supported by NSF grants AST 00-71208 to E.\,M.\,H.
296: and AST99-84816 to L.\,L.\,C., and by NASA grant GO-7266.01-96A from Space Telescope Science Institute, which is operated by AURA, Inc. under NASA contract NAS 5-26555. We are grateful to the staff of the
297: Keck and Subaru Telescopes for supporting these observations.
298: \end{acknowledgments}
299:
300: \begin{thebibliography}{}
301:
302: \bibitem[Capak \etal\ (2004)]{capak}
303: {Capak, P. \etal} 2004,
304: \textit{AJ} 127, 180
305: \bibitem[Cowie \& Hu (1998)]{cow98}
306: {Cowie, L.L. \& Hu, E.M.} 1998,
307: \textit{AJ} 115, 1319
308: \bibitem[Haiman (2002)]{haiman}
309: {Haiman, Z.} 2002,
310: \textit{ApJ} (Letters) 576, L1
311: \bibitem[Hu \etal\ (2005)]{hdfz5}
312: {Hu, E.M., Cowie, L.L., \& Capak, P.} 2005,
313: \textit{AJ} submitted
314: \bibitem[Hu \etal\ (2004)]{lum5}
315: {Hu, E.M., Cowie, L.L., Capak, P., Hayashino, T., \& Komiyama, Y.} 2004,
316: \textit{AJ} 127, 563
317: \bibitem[Hu \etal\ (1999)]{smitty}
318: {Hu, E.M., Cowie, L.L., \& McMahon, R.G.} 1999
319: \textit{ApJ} (Letters) 502, L99
320: \bibitem[Hu \etal\ (2002)]{z6}
321: {Hu, E.M., Cowie, L.L., McMahon, R.G., Capak, P., Iwamuro, F., Kneib,
322: J.-P. Maihara, T., \& Motoharo, K.} 2002,
323: \textit{ApJ} (Letters) 568, L75; Erratum, \textit{ApJ} (Letters) 576, L99
324: \bibitem[Kakazu \etal\ in preparation]{kakazu}
325: {Kakazu, Y. \etal\ } 2005,
326: in preparation
327: \bibitem[Malhotra \& Rhoads (2004)]{mal}
328: {Malkhotra \& Rhoads} 2004,
329: \textit{ApJ} (Letters) 617, L5
330: \bibitem[Ouchi \etal\ (2005)]{ouchi}
331: {Ouchi \etal\ } 2005,
332: \textit{ApJ} (Letters) 620, L1
333: \bibitem[Spinrad (2004)]{spinrad04}
334: {Spinrad, H.} 2004,
335: in: J.W. Mason (ed.),
336: \textit{Astrophysics Update},
337: (Berlin: Springer-Verlag and Chichester, UK: Praxis
338: Publishing), p.\ 155, (astro-ph/0308411)
339: \bibitem[Steidel \etal\ (1999)]{stei1999}
340: {Steidel, C.C., Adelberger, K.L., Giavalisco, M., Dickinson, M., \&
341: Pettini, M.} 1999,
342: \textit{ApJ} 519, 1
343: \bibitem[Steidel \etal\ (2000)]{stei2000}
344: {Steidel, C.C., Adelberger, K.L., Shapley, A.E., Pettini, M., Dickinson,
345: M., \&
346: Giavalisco, M.} 2000,
347: \textit{ApJ} 532, 170
348: \bibitem[Taniguchi \etal\ (2005)]{tani}
349: {Taniguchi \etal\ } 2005,
350: \textit{PASJ} 57, 165
351: \end{thebibliography}
352:
353: \end{document}
354: