1: %%
2: %% This is a sample manuscript marked up using the
3: %% AASTeX v5.x LaTeX 2e macros.
4:
5: \documentclass[12pt,preprint]{aastex}
6:
7: %% manuscript produces a one-column, double-spaced document:
8:
9: %\documentclass[manuscript]{aastex}
10:
11: %% preprint2 produces a double-column, single-spaced document:
12:
13: %% \documentclass[preprint2]{aastex}
14:
15: %% Sometimes a paper's abstract is too long to fit on the
16: %% title page in preprint2 mode. When that is the case,
17: %% use the longabstract style option.
18:
19: %% \documentclass[preprint2,longabstract]{aastex}
20: \usepackage{graphicx}
21: %\usepackage{epstopdf}
22: \usepackage{psfig}
23: %% If you want to create your own macros, you can do so
24: %% using \newcommand. Your macros should appear before
25: %% the \begin{document} command.
26: %%
27: %% If you are submitting to a journal that translates manuscripts
28: %% into SGML, you need to follow certain guidelines when preparing
29: %% your macros. See the AASTeX v5.x Author Guide
30: %% for information.
31:
32: \newcommand{\vdag}{(v)^\dagger}
33: \newcommand{\myemail}{ccarilli@nrao.edu}
34: %\newcommand{\etal}{\mbox{et al.\}}
35:
36: \newcommand{\etal}{\mbox{et~al.}}
37: \newcommand{\Msol}{\mbox{$M_{\odot}$}}
38: \newcommand{\msun}{\mbox{$M_{\odot}$}}
39: \newcommand{\Msun}{\mbox{${\bf M_{\odot} }$}}
40: \newcommand{\Lsol}{\mbox{$L_{\odot}$}}
41: \newcommand{\lsun}{\mbox{$L_{\odot}$}}
42: \newcommand{\ergsec}{\mbox{erg s$^{-1}$}}
43: \newcommand{\beam}{\mbox{beam$^{-1}$}}
44: \newcommand{\Mdyn}{\mbox{$M_{\rm dyn}$}}
45: \newcommand{\Mgas}{\mbox{$M_{\rm gas}$}}
46: \newcommand{\HH}{\mbox{H$_2$}}
47: \newcommand{\Mgasfive}{\mbox{$M_{\rm gas}(r<500{\rm pc})$ }}
48: \newcommand{\Halpha}{\mbox{H$\alpha$} }
49: \newcommand{\tm}[1]{\tablenotemark{#1}}
50: \def\deg {{\ifmmode^\circ\else$^\circ$\fi}} %%% Overwrites TeX \deg
51: \def\Ho {{$H_{0}$} }
52: \def\qo {{$q_{0}$} }
53: \def\kmsMpc {{\ km\ s$^{-1}$\ Mpc$^{-1}$} }
54:
55: %% You can insert a short comment on the title page using the command below.
56:
57: \slugcomment{Resubmitted to COSMOS ApJ Special Issue October 16, 2006}
58:
59: %% If you wish, you may supply running head information, although
60: %% this information may be modified by the editorial offices.
61: %% The left head contains a list of authors,
62: %% usually a maximum of three (otherwise use et al.). The right
63: %% head is a modified title of up to roughly 44 characters.
64: %% Running heads will not print in the manuscript style.
65:
66: \shorttitle{Radio and millimeter properties of $z \sim 5.7$
67: Ly$\alpha$ emitters in the COSMOS field}
68: \shortauthors{Carilli et al.}
69:
70: %% This is the end of the preamble. Indicate the beginning of the
71: %% paper itself with \begin{document}.
72:
73: \begin{document}
74:
75:
76: \title{Radio and millimeter properties of $z \sim 5.7$
77: Ly$\alpha$ emitters in the COSMOS field:
78: limits on radio AGN, submm galaxies, and dust obscuration}
79:
80: \author{
81: C.L. Carilli\altaffilmark{1},
82: T. Murayama\altaffilmark{2},
83: R. Wang\altaffilmark{1},
84: E. Schinnerer\altaffilmark{3},
85: Y. Taniguchi\altaffilmark{2},
86: V. Smol{\v c}i{\' c}\altaffilmark{3},
87: F. Bertoldi\altaffilmark{4},
88: M. Ajiki\altaffilmark{2},
89: T. Nagao\altaffilmark{5},
90: S. S. Sasaki\altaffilmark{6},
91: Y. Shioya\altaffilmark{2},
92: J.~E.~Aguirre\altaffilmark{7,8}
93: A. W. Blain\altaffilmark{9},
94: N. Scoville\altaffilmark{9},
95: D. B. Sanders\altaffilmark{10}
96: }
97:
98:
99: \altaffiltext{$\star$}{Based on data collected at : the Subaru
100: Telescope, which is operated by the National Astronomical Observatory
101: of Japan; the National Radio Astronomy Observatory which is a facility
102: of the National Science Foundation operated under cooperative
103: agreement by Associated Universities, Inc; the IRAM 30m telescope; the
104: Caltech Submm Observatory}
105:
106: %
107: \altaffiltext{1}{National Radio Astronomy Observatory, P.O. Box 0, Socorro, NM
108: 87801}
109: %
110: \altaffiltext{2}{Physics Department, Graduate School of Science \&
111: Engineering, Ehime University, 2-5 Bunkyo-cho, Matsuyama
112: 790-8577, Japan}
113: %
114: \altaffiltext{3}{Max Planck Institut f\"ur Astronomie, K\"onigstuhl
115: 17, Heidelberg, D-69117, Germany}
116: %
117: \altaffiltext{4}{Dept. Astronomy, Bonn University, Bonn, Germany, D53121}
118: %
119: \altaffiltext{5}{National Astronomical Observatory of Japan,
120: 2-21-1 Osawa, Mitaka, Tokyo 181-8588, Japan}
121: %
122: \altaffiltext{6}{Astronomical Institute, Graduate School of Science,
123: Tohoku University, Aramaki, Aoba, Sendai 980-8578, Japan}
124: %
125: \altaffiltext{7}{Jansky Fellow, National Radio Astronomy Observatory}
126: %
127: \altaffiltext{8}{Center for Astrophysics and Space Astronomy and
128: Department of Astronomy and Planetary Sciences, University of
129: Colorado, 593 UCB, Boulder, CO 80303-0593}
130: %
131: \altaffiltext{9}{California Institute of Technology, MC 105-24, 1200 East
132: California Boulevard, Pasadena, CA 91125}
133: %
134: \altaffiltext{10}{Institute for Astronomy, 2680 Woodlawn Dr.,
135: University of Hawaii, Honolulu, Hawaii, 96822}
136: %
137:
138:
139: \begin{abstract}
140:
141: We present observations at 1.4 and 250 GHz of the $z\sim 5.7$
142: Ly$\alpha$ emitters (LAE) in the COSMOS field found by Murayama et
143: al.. At 1.4 GHz there are 99 LAEs in the lower noise regions of the
144: radio field. We do not detect any individual source down to 3$\sigma$
145: limits of $\sim 30\mu$Jy beam$^{-1}$ at 1.4 GHz, nor do we detect a
146: source in a stacking analysis, to a 2$\sigma$ limit of $2.5\mu$Jy
147: beam$^{-1}$. At 250 GHz we do not detect any of the 10 LAEs that are
148: located within the central regions of the COSMOS field covered by
149: MAMBO ($20' \times 20'$) to a typical 2$\sigma$ limit of $S_{250} <
150: 2$mJy. The radio data imply that there are no low luminosity radio
151: AGN with $L_{1.4} > 6\times 10^{24}$ W Hz$^{-1}$ in the LAE
152: sample. The radio and millimeter observations also rule out any highly
153: obscured, extreme starbursts in the sample, ie. any galaxies with
154: massive star formation rates $> 1500$ M$_\odot$ year$^{-1}$ in the
155: full sample (based on the radio data), or 500 M$_\odot$ year$^{-1}$
156: for the 10\% of the LAE sample that fall in the central MAMBO
157: field. The stacking analysis implies an upper limit to the mean
158: massive star formation rate of $\sim 100$ M$_\odot$ year$^{-1}$.
159:
160: \end{abstract}
161:
162: \keywords{galaxies: formation --- galaxies: evolution ---
163: galaxies: radio, submm, IR --- surveys}
164:
165:
166: \section{Introduction}
167:
168: Numerous studies have demonstrated the power of discovering high
169: redshift star forming galaxies using narrow band filters centered on
170: the Ly$\alpha$ line (Hu et al. 2002, 2004, Kodaira et al. 2003, Rhoads
171: et al. 2003, Malhotra \& Rhoads 2004, Tran et al. 2004, Kurk et
172: al. 2004, Santos et al. 2004, Martin \& Sawicki 2004, Taniguchi et
173: al. 2005; Iye et al. 2006). Indeed, the majority of galaxies known at
174: $z \sim 6$ have been discovered in this way. Finding galaxies at
175: these extreme redshifts is of paramount importance since the recent
176: discovery of Gunn-Peterson absorption by a partially neutral IGM
177: toward the highest $z$ QSOs ($z \sim 6$; Fan et al. 2006) -- a
178: signature of cosmic reionization. Reionization is a key benchmark in
179: cosmic structure formation, indicating the formation of the first
180: luminous objects (Fan, Carilli, Keating 2006).
181:
182: The Cosmic Evolution Survey (COSMOS), covering 2 \sq\deg, is designed
183: to probe the evolution of galaxies, AGN and dark matter in the context
184: of their cosmic environment. The COSMOS/HST field has extensive
185: supporting observations, ranging from the radio through the X-ray
186: (Scoville et al. 2006). Part of this program entails a SUBARU narrow
187: band survey of the full field centered on Ly$\alpha$ at $z \sim 5.7$
188: (Murayama et al. 2006). This survey has revealed a large sample of
189: galaxies at $z \sim 5.7$, with 110 candidate galaxies.
190:
191: Observations of the COSMOS field have been done at 1.5$"$ resolution
192: (FWHM) at 1.4 GHz down to an rms level between 8 and 10$\mu$Jy
193: beam$^{-1}$ (Schinnerer et al. 2006). Observations have also been
194: done at 250 GHz at a resolution of 10.6$''$ of the inner $20'\times
195: 20'$ of the COSMOS field using MAMBO at the IRAM 30m telescope to a
196: rms level of 0.9 mJy (Bertoldi et al 2006), and a somewhat larger,
197: shallower field ($30.6'\times 30.6'$) at a resolution of 31$''$ using
198: BOLOCAM at the Caltech Submm Observatory (Aguirre et al. 2006) to an
199: rms level of 1.9mJy.
200:
201: In this paper we use the data from the VLA, MAMBO, and BOLOCAM of the
202: COSMOS field to constrain the centimeter and millimeter properties of
203: the $z=5.7$ LAEs. This study represents the deepest radio continuum
204: study of high-$z$ LAEs, over the largest area, as well as the most
205: extensive study of these sources at millimeter wavelengths to date.
206: These data allow us to set limits on any low luminosity radio AGN, as
207: well as on the number of highly dust-obscured starburst galaxies, in
208: the LAE sample at $z=5.7$. We perform a stacking analysis to set a
209: limit to the mean UV obscuration of high-$z$ LAEs.
210:
211: \section{The sample and the radio and millimeter observations}
212:
213: \subsection{The sample}
214:
215: The sample is taken from the narrow band Ly$\alpha$ survey of Murayama
216: et al. (2006) centered on a redshift of $z=5.7 \pm 0.05$. They cover
217: the full COSMOS field, implying a comoving volume of $1.5\times10^6$
218: Mpc$^{3}$. They select sources that are detected in the narrow band
219: NB816 filter at NB816 $< 25.1$ mag, are undetected in shorter
220: wavelength broad band filters, and have NB-to-broad band near-IR
221: colors that imply Ly-$\alpha$ (observed) equivalent widths, $EW_{obs}
222: > 120 \AA$ (corresponding to rest frame $EW_{rest} = EW_{obs}/(1+z) >
223: 18\AA$).
224:
225: They find 110 candidate LAEs, and they estimate that the
226: contamination rate by low $z$ objects is $< 14\%$. Thirty seven
227: sources are also detected in longer wavelength filters ($z'$),
228: corresponding to rest frame UV emission (1250\AA).
229: All of the sources are small, $< 0.5"$, and a few ($\sim 5\%$) show
230: evidence for 2 or 3 compact components. No large ($\ge 10$'s kpc)
231: 'Ly-$\alpha$ blob' sources are detected (Steidel et al. 2000;
232: Matsuda et al. 2004).
233:
234:
235:
236: \subsection{The VLA observations}
237:
238: We have searched for radio emission from the LAEs in the COSMOS
239: field using the data presented in Schinnerer et al. (2006). At each
240: position we determine the flux density, and the rms noise in the
241: region. The relative astrometric accuracy between the radio and
242: optical images is better than 0.2$"$ (Aussel et al. 2006), while for a
243: 3$\sigma$ detection the positional uncertainty is given roughly by:
244: FWHM/SNR $\sim 0.5''$. We have searched for radio sources
245: within 0.6$''$ radius of the LAE optical position.
246: We exclude from the analysis 11 LAEs in higher noise regions
247: of the field, such as close to a bright continuum source, or
248: near the edge of the field, leaving a sample of 99 sources total,
249: and 33 with UV continuum detections.
250:
251: We do not detect any source $>3\sigma$ at 1.4 GHz within 0.6$''$ of
252: any LAE in the sample of 99. The typical 3$\sigma$ limit is 30$\mu$Jy
253: beam$^{-1}$ at 1.4GHz. Note that, for the full sample of 99 sources,
254: we expect 0.05 chance coincidences within 0.6$''$ at the level of
255: 30$\mu$Jy beam$^{-1}$, based on faint radio source counts (Fomalont et
256: al. 2006, in prep).
257:
258: We have also performed a radio stacking analysis of the sources,
259: summing images centered on the positions of the LAEs, weighted
260: by the rms in each subfield. Stacking all the LAEs, we do
261: not detect a source at the LAE position to a $2\sigma$ limit of
262: 2.5$\mu$Jy beam$^{-1}$. If we only stack the UV-detected sources (33
263: sources), we find a $2\sigma$ limit of $4 \mu$Jy beam$^{-1}$ (Figure
264: 1).
265:
266: One LAE, J10000.51+014940.1, has a marginal ($2.7\sigma$) 1.4 GHz
267: source of $27\pm 10\mu$Jy located just $0.2''$ from the optical
268: position (Figure 2). If real, the implied luminosity density at a rest
269: frame frequency of 1.4 GHz is $L_{1.4} = 6\times 10^{24}$ W
270: Hz$^{-1}$, assuming a spectral index of --0.75. We do not consider
271: this a firm detection, but deeper radio imaging would be very
272: interesting for this source.
273:
274: We note that in the study of the GOODS North field, Ajiki et al. (2006)
275: found 10 LAEs at $z \sim 5.7$ using a similar technique as was employed
276: for the COSMOS field. Comparing these sources to the deep radio survey
277: of Richards (2000), we again find that no LAE has a radio counterpart
278: to a 5$\sigma$ detection limit of $40\mu$Jy.
279:
280:
281: \subsection{The MAMBO and BOLOCAM observations}
282:
283: Ten of the LAEs are located within the $20'\times 20'$ field imaged
284: with MAMBO at 250 GHz (Bertoldi et al. 2006). Given a FWHM of
285: 10.6$''$, we searched for MAMBO counterparts $> 3\sigma$, within
286: 3.5$''$ of an LAE. None of the LAEs have a MAMBO counterpart, to a
287: typical 3$\sigma$ upper limit of $S_{250} < 3$ mJy. A stacking
288: analysis leads to a $2\sigma$ limit to the mean 250 GHz flux density
289: of $S_{250} < 0.7$mJy. For reference, based on submm galaxy source
290: counts (Bertoldi et al. 2006), we expect 0.01 chance coincidences
291: within 3.5$''$ with $S_{250} \ge 3$mJy for the 10 LAEs located in the
292: MAMBO-COSMOS field.
293:
294: There is one LAE, J100040.22+021903.8, that has a potential MAMBO
295: source located 5$"$ south of the LAE position, with a flux density of
296: $S_{250} = 3.2 \pm 0.91$ mJy (Figure 3). The BOLOCAM image shows a
297: value of $1.7\pm 1.9$ mJy at this position. The radio image shows a
298: surface brightness of $15\pm10\mu$Jy beam$^{-1}$ at the LAE position.
299: This LAE is not detected in the UV continuum, and the total star
300: formation rate based on the Ly$\alpha$ luminosity is only 6 M$_\odot$
301: year$^{-1}$ (uncorrected for obscuration). Given the positional
302: offset, and the relatively low significance of the MAMBO detection, we
303: cannot claim either reality of the MAMBO source, or an association of
304: the LAE and the (marginal) MAMBO source. If real, the implied FIR
305: luminosity is: $L_{FIR} = 1.1\times10^{13}$ L$_\odot$, and the
306: predicted radio flux density at 1.4 GHz is 10$\mu$Jy based on a star
307: forming galaxy template (Carilli \& Yun 2000). Deeper observations at
308: 250 and 1.4 GHz are required to check the reality of this source.
309:
310: We also searched the wider, shallower BOLOCAM field for counterparts
311: to the LAEs. There are 12 LAEs in the BOLOCAM field (10 are common to
312: the MAMBO field), and again, no source is detected with BOLOCAM to
313: a typical 3$\sigma$ limit of 5.7mJy. A stacking analysis provides a
314: 2$\sigma$ limit of 1 mJy.
315:
316: \section{Discussion}
317:
318: We do not detect any individual source to a typical 3$\sigma$ limit of
319: 30$\mu$Jy beam$^{-1}$ at 1.4 GHz. A limit of 30$\mu$Jy beam$^{-1}$ at
320: an observing frequency of 1.4 GHz implies a limit to the radio
321: luminosity at an emitted frequency of 1.4 GHz of $L_{1.4} <
322: 6\times10^{24}$ W Hz$^{-1}$, assuming a spectral index of $-0.75$.
323: For comparison, the nearby Fanaroff-Riley class I (ie. low luminosity)
324: radio galaxy M87 has $L_{1.4} = 9\times10^{24}$ W Hz$^{-1}$. The lack
325: of radio AGN in the LAE sample is not surprising, since Taniguchi et
326: al. (2006, in prep) show that the narrow band search technique selects
327: against broad line QSOs, for which the emission lines are typically
328: broader than the filter. Likewise, Hu et al. (1998) and Keel et
329: al. (1999) find a relatively low fraction (between 17\% and 40\%) of
330: narrow line AGN in lower $z$ LAE samples, while Shapley et al. (2003)
331: find only 3\% of the Ly-break galaxies at $z \sim 3$ show optical
332: emission line spectra consistent with an AGN. Overall, our
333: non-detection of even a low luminosity radio AGN in any of the 99
334: COSMOS LAEs is broadly consistent with the conclusion that the narrow
335: band Ly$\alpha$ search technique preferentially selects for
336: star-forming galaxies.
337:
338: We should point out that, in their extensive study of a sample of NB
339: selected LAEs at $z \sim 4.5$, Malhotra \& Rhoads (2002) found a
340: surprising fraction of the sources ($\sim 60\%$) had Ly-$\alpha$
341: $EW_{rest} > 240\AA$. They state that such large EW's cannot arise
342: through normal star formation, requiring either: (i) a narrow-line
343: AGN, (ii) a top-heavy IMF, or (iii) low metalicities. The largest
344: measured $EW_{rest}$ in the COSMOS LAE sample is 103\AA, however, the
345: majority of sources are not detected in the UV, and hence only lower
346: limits on the $EW$ values can be set (see Section 2.1). Murayama et
347: al. (2006) discuss the $EW$ distribution for the COSMOS sample in more
348: detail.
349:
350: The radio luminosity limit also corresponds to a massive ($>
351: 5$M$_\odot$) star formation rate $\sim 1500$ M$_\odot$ year$^{-1}$
352: (Condon 1992). Hence, we can rule-out any highly dust obscured,
353: 'hyperluminous' infrared starburst galaxy. Such a source would
354: correspond to a bright 'submm' galaxy with a 250 GHz flux density of
355: $\sim 9$mJy, assuming that the local far-IR--radio correlation
356: continues to apply to redshift $z\simeq 6$ (eg. Blain et al. 2001;
357: Bertoldi et al. 2006; Carilli \& Yun 1999, 2000). The MAMBO image of
358: the inner 20$'$ of the COSMOS field pushes this limit down to 3mJy
359: (500 M$_\odot$ year$^{-1}$), at least for the 10\% of the LAE sample
360: that fall within this area.
361:
362: The mean total (0.1 to 100 M$_\odot$) star formation rate for all the
363: sources based on the Ly-$\alpha$ luminosity is $\sim 8$ M$_\odot$
364: year$^{-1}$ (Murayama et al. 2006). A similar number is found for the
365: star formation rates derived from the Ly-$\alpha$ luminosity for the
366: UV-detected subsample. For comparison, the star formation rates
367: derived from UV luminosities are systematically higher, with the mean
368: total star formation rate derived from the UV luminosities for the
369: UV-detected sources $\sim 12$ M$_\odot$ year$^{-1}$. The implied
370: massive star formation rates (5 to 100 M$_\odot$), assuming a
371: Salpeter IMF, are a factor 5.6 smaller, or 1.4 and 2.1 M$_\odot$
372: year$^{-1}$, respectively. The difference between the Ly-$\alpha$
373: derived and UV luminosity derived star formation rates is discussed
374: in Murayama et al. (2006), and likely relates to extra attenuation of
375: the Ly-$\alpha$ line due to associated Ly-$\alpha$ absorption. Note
376: that none of these values have been corrected for dust extinction.
377:
378: From the radio stacking analysis, we derive a (2$\sigma$) upper limit
379: to the mean massive star formation rate of 81 M$_\odot$ year$^{-1}$
380: for all the LAEs, and 130 M$_\odot$ year$^{-1}$ for just the
381: UV-detected sources. These radio limits to the star formation rate
382: are independent of the dust content. Hence, the upper limit to the
383: mean obscuration of the LAE galaxies in either the UV continuum or the
384: Ly-$\alpha$ line, is about factor of 60. For comparison, the typical
385: Ly-break galaxy is thought to have its UV emission attenuated by a
386: factor $\sim 5$ due to intrinsic dust (Steidel et al. 1999), and the
387: mean obscuration for galaxies selected using the Ly-$\alpha$ narrow
388: band technique is thought to be even smaller (Shapely et
389: al. 2003). Hence, while our study represents the most sensitive,
390: widest field radio and mm study of high-$z$ LAEs to date, it also
391: accentuates the relatively poor limits that can be reached in the
392: radio and mm for star forming galaxies at the highest redshifts, when
393: compared to studies using the Ly$\alpha$ line.
394:
395: The main result of this work is to rule-out the existence of any
396: highly obscured massive starburst, or low luminosity radio AGN in the
397: COSMOS LAE sample. Clearly, to push down to normal star forming
398: galaxies will require the one to two orders of magnitude improvement
399: in sensitivity afforded by the up-coming Expanded Very Large Array,
400: and the Atacama Large mm Array (ALMA).
401:
402:
403: \acknowledgments
404:
405: CC acknowledges partial support from the Max-Planck Society and the
406: Alexander von Humboldt Foundation through the Max-Planck
407: Forshungspreise 2005. We thank the referee for very useful comments.
408: The HST COSMOS Treasury program was supported through NASA grant
409: HST-GO-09822. We gratefully acknowledge the contributions of the
410: entire COSMOS collaboration. More information on the COSMOS survey is
411: available at {\bf \url{http://www.astro.caltech.edu/cosmos}}.
412:
413:
414: \begin{thebibliography}{}
415:
416: \bibitem[]{}Ajiki, M., Mobasher, B., Taniguchi, Y., Shioya, Y. etal.
417: 2006, ApJ, 638, 596
418:
419: \bibitem[]{}Aguirre, J. et al. 2006, ApJ, in press
420:
421: \bibitem[]{}Aussel, H. et al. 2006, ApJ, in press
422:
423: \bibitem[]{}Bertoldi, F. et al. 2006, ApJ, in press
424:
425: \bibitem[]{} Blain, A., Smail, I., Ivison, R., Kneib, J.-P.,
426: Frayer, D. 2002, Phys. Rep., 369, 1
427:
428: \bibitem[]{}Carilli, C.L., Yun, M.S. 1999, ApJ,
429: 513, L13
430:
431: \bibitem[]{}Carilli, C.L., Yun, M.S. 2000, ApJ, 530, 618
432:
433: \bibitem[]{}Condon, J. 1992, ARAA, 30, 575
434:
435: \bibitem[]{fan06a} Fan, X., Strauss, M., Becker, R., et al. 2006a, AJ,
436: 132, 117
437:
438: \bibitem[Fan et al. 2006]{fan06b} Fan, X., Carilli, C.L., Keating,
439: B. 2006, ARAA, 44, 413
440:
441: \bibitem[]{}Finoguenov, A. et al. 2006, ApJ, in press
442:
443: \bibitem[Hu et al. 2004]{2004AJ....127..563H} Hu, E.~M., Cowie,
444: L.~L., Capak, P., McMahon, R.~G., Hayashino, T., \& Komiyama, Y.\
445: 2004, \aj, 127: 563-575
446:
447: \bibitem[]{}Hu, E., Cowie, L., McMahon, R. 1998, ApJ, 502, L99
448:
449: \bibitem[]{}Iye, M. Ota, M., Kashikawa, N. et al. 2006, Nature, 443, 186
450:
451: \bibitem[]{}Keel, W., Cohen, S., Windhorst, R., Waddington, I. 1999,
452: AJ, 118, 2547
453:
454: \bibitem[Kodaira et al. 2003]{2003PASJ...55L..17K} Kodaira, K.,
455: Taniguchi, Y., Kashikawa, et al. 2003, \pasj, 55: L17
456:
457: \bibitem[Kurk et al. 2004]{2004A&A...422L..13K} Kurk, J.~D., Cimatti,
458: A., di Serego Alighieri, S., Vernet, J., Daddi, E., Ferrara, A., \&
459: Ciardi, B.\ 2004, \aap, 422: L13-L17
460:
461: \bibitem[Malhotra \& Rhoads 2004]{2004ApJ...617L...5M} Malhotra, S.,
462: \& Rhoads, J.~E.\ 2004, \apjl, 617: L5-L8
463:
464: \bibitem[Malhotra \& Rhoads 2002]{2002ApJ...565L...71M} Malhotra, S.,
465: \& Rhoads, J.~E.\ 2002, \apjl, 565: L71
466:
467: \bibitem[Martin \& Sawicki 2004]{2004ApJ...603..414M} Martin, C.~L.,
468: \& Sawicki, M.\ 2004, \apj, 603: 414-424
469:
470: \bibitem[]{}Matsuda, Y., Yamada, T., Hayashino, T., Tamura, H.,
471: Yamauchi, R. et al. 2004, AJ, 128, 569
472:
473: \bibitem[]{}Murayama, M. et al. 2006, ApJ, in press
474:
475: \bibitem[Rhoads et al. 2003]{2003AJ....125.1006R} Rhoads, J.~E.,
476: Dey, A., Malhotra, S., et al. 2003, \aj, 125: 1006-1013
477:
478: \bibitem[]{}Richards, E. 2000, \apj, 533, 611
479:
480: \bibitem[Santos et al. 2004]{2004ApJ...606..683S} Santos, M.~R.,
481: Ellis, R.~S., Kneib, J.-P., Richard, J., \& Kuijken, K.\ 2004, \apj,
482: 606: 683-701
483:
484: \bibitem[Shapely03]{Shapely03} Shapley, A.,
485: Steidel, C., Pettini, M., Adelberger, K. 2003, \apj,
486: 588, 65
487:
488: \bibitem[]{}Schinnerer, E. et al. 2006, ApJ, in press
489:
490: \bibitem[]{}Scoville, N. et al. 2006, ApJ, in press
491:
492: \bibitem[]{}Shapley, A., Steidel, C., Pettini, M.,
493: Adelberger, K. 2003, ApJ, 588, 65
494:
495: \bibitem[]{}Steidel, C., Adelberger, K., Giavalisco, M., Dickinson,
496: M., Pettini, M. 1999, ApJ, 519, 1
497:
498: \bibitem[]{}Steidel, C., Adelberger, K., Shapley, A., Pettini, M.,
499: Dickinson, M., Giavalisco, M. 2000, ApJ, 532, 170
500:
501: \bibitem[Taniguchi et al. 2005]{2005PASJ...57..165T} Taniguchi, Y.,
502: Ajiki, M., Nagao, T. et al. 2005, \pasj, 57: 165
503:
504: \bibitem[Tran et al. 2004]{2004ApJ...612L..89T}
505: Tran, K.-V.~H., Lilly, S.~J., Crampton, D., \& Brodwin, M.\ 2004,
506: \apjl, 612: L89
507:
508: \end{thebibliography}
509:
510: \clearpage\newpage
511:
512: \noindent{\bf Note Added in Proof}
513:
514: Subsequent to the acceptance of the paper: "Radio and submm
515: observations of LAEs in the Cosmos field," by Carilli et al., an
516: additional nine LAEs were discovered on further investigation of the
517: Subaru images, making for a total of 119 LAEs in the Cosmos field
518: (Murayama et al. 2006). We have searched the radio and submm Cosmos
519: images for counterparts, and do not detect any source to similar
520: limits as those presented for the original sample of 110 LAEs. The
521: stacking analysis is effectively unaltered by these new sources.
522:
523: There is one source in the new sample of nine (J095825.26+022651.32 =
524: source 4 in the final LAE catalog; Murayama et al. 2006) which
525: projects within 5$"$ north-west of a strong radio hot spot. This
526: radio hot spot is at the end of one of the radio lobes of an
527: arcminute-sized luminous radio galaxy. The (likely) optical
528: identification of the radio host galaxy is an early-type galaxy with a
529: photometric redshift of $1.1\pm 0.2$, situated in a cluster of
530: galaxies at this redshift identified by Finoguenov et al. (2006). We
531: feel the projected proximity of the LAE and the radio hot spot is most
532: likely just a coincidence, although it is possible that gravitational
533: lensing by the cluster may magnify the LAE, or that the detected
534: excess in the NB816 filter is due to broad [OII] 372.7nm nebular
535: emission at $z=1.19$, associated with shocked gas preceding the radio
536: hot spot. Spectroscopy of this object is needed to test these
537: possibilities.
538:
539: \clearpage
540: \newpage
541:
542: \begin{figure}[ht]
543: \psfig{figure=f1.ps,width=5in}
544: \caption{The stacked 1.4 GHz image of the
545: LAEs in the COSMOS field (99 sources). The rms noise level is 1.25 $\mu$Jy
546: beam$^{-1}$. The cross marks the stacking position, centered
547: on the LAE positions (the absolute coordinates axes are arbitrary).
548: The contour levels are: -6, -4, -2, 2, 4, 6 $\mu$Jy beam$^{-1}$,
549: and the beam has FWHM = 1.5$"$.
550: }
551: \label{}
552: \end{figure}
553:
554: \clearpage\newpage
555:
556: \begin{figure}[ht]
557: \psfig{figure=f2.ps,width=5in}
558: \caption{The VLA 1.4 GHZ image of the field centered on
559: the $z=5.7$ LAE J10000.51+014940.1 in Ajiki et al. (2006).
560: The cross marks the position of the LAE.
561: The contour levels are: -27, -18, -9, 9, 18,
562: 27 $\mu$Jy beam$^{-1}$, and the beam has FWHM = 1.5$"$.
563: }
564: \label{}
565: \end{figure}
566:
567: \clearpage\newpage
568:
569: \begin{figure}[ht]
570: \psfig{figure=f3.ps,width=5in,angle=-90}
571: \caption{The MAMBO 250 GHz image of the field centered on
572: the $z=5.7$ LAE J100040.22+021903.8 in Ajiki et al. (2006).
573: The cross marks the position of the LAE, and the
574: size of the cross corresponds to the FWHM of the
575: MAMBO beam. The contour levels are: -3, -2, -1, 1, 2, 3 mJy
576: beam$^{-1}$, and the beam has FWHM = 10.6$''$.
577: }
578: \label{}
579: \end{figure}
580:
581: \end{document}
582:
583: