1: \documentclass{emulateapj}
2: \usepackage{apjfonts}
3: \newcommand{\etal}{{et al}\/.}
4: \newcommand{\hh}{^{\rm h}}
5: \newcommand{\mm}{^{\rm m}}
6: \begin{document}
7: \slugcomment{accepted for publication in ApJ Letters}
8: \shorttitle{Gemini imaging of a z=10 candidate}
9: \shortauthors{M. N. Bremer \etal}
10: \title{Gemini H-band imaging of the field of a z=10 candidate}
11: \author{M.N. Bremer}
12: \affil{Department of Physics, University of Bristol, Tyndall Avenue,
13: Bristol BS8 1TL, UK}
14: \author{Joseph B. Jensen}
15: \affil{Gemini Observatory, 950 N. Cherry Ave., Tucson, AZ 85719}
16: \author{M. D. Lehnert, N.M. F\"{o}rster Schreiber}
17: \affil{Max-Planck-Institut f\"ur extraterrestrische Physik,
18: Giessenbachstra\ss e, 85748 Garching bei M\"{u}nchen, Germany}
19: \and
20: \author{Laura Douglas}
21: \affil{Department of Physics, University of Bristol, Tyndall Avenue,
22: Bristol BS8 1TL, UK}
23:
24:
25: \begin{abstract}
26:
27: We present a deep $H$-band image of the field of a candidate $z{=}10$
28: galaxy magnified by the foreground ($z{=}0.25$) cluster Abell 1835. The
29: image was obtained with NIRI on Gemini North to better constrain the
30: photometry and investigate the morphology of the source. The image is
31: approximately one magnitude deeper and has better spatial resolution
32: (seeing was 0.4-0.5 arcsec) than the existing $H$-band image obtained
33: with ISAAC on the VLT by \cite{pello04}. The object is not detected
34: in our new data. Given the published photometry ($H_{AB}$=25.0), we
35: would have expected it to have been detected at more than $\sim7\sigma$
36: in a 1.4 arcsec diameter aperture. We obtain a limit of $H_{AB}$$>$26.0
37: (3$\sigma$) for the object. A major part of the evidence that this object
38: is at $z{=}10$ was the presence of a strong continuum break between the $J$
39: and $H$ band, attributed to absorption of all continuum shortward
40: of 1216 \AA~ in the rest-frame of the object. Our $H-$band non-detection
41: substantially reduces the magnitude of any break and therefore weakens the
42: case that this object is at $z{=}10$. Without a clear continuum break,
43: the identification of an emission line at 1.33745$\mu$m as Ly$\alpha$
44: at $z\approx10$ is less likely. We show that the width and flux of
45: this line are consistent with an alternative emission line such as
46: [OIII]$\lambda$5007 from an intermediate redshift HII/dwarf galaxy.
47:
48: \end{abstract} \keywords{cosmology: observations - early universe -
49: galaxies: distances and redshifts - galaxies: evolution - galaxies:
50: formation}
51:
52: \maketitle
53: \section{Introduction}
54: \label{intro}
55: Eight-meter class telescopes such as Gemini, VLT and Keck have opened
56: up the Universe at $z>5$ for detailed study. The first $z>5$ galaxy
57: was discovered in 1998 \citep{Dey98}. In the past two years the number
58: of confirmed and candidate $z>5$ galaxies has grown substantially,
59: \citep[see {\it e.g.,}][]{L03,bremer04, S04a, S04b, bunker03, ajiki03,
60: rhoads03, bouwens04, hu04}. These have been discovered using a variety
61: of techniques including slitless spectroscopy, narrow-band imaging
62: surveys for Ly$\alpha$ emitters, and broad-band photometry followed by
63: spectroscopy to identify UV-bright Lyman dropouts.
64:
65: Until recently, most searches for distant galaxies have concentrated
66: on redshifts out to $z\sim6.6$. Beyond 7000 \AA~ the sky emission is
67: increasingly dominated by OH emission bands, making spectroscopic
68: identification of Ly$\alpha$ emission more and more difficult, except
69: in the gaps between the OH bands. As a consequence the redshift
70: distribution of $z>5$ galaxies with spectroscopic redshifts is
71: non-uniform, mostly reflecting the wavelength distribution of these
72: gaps. Nevertheless, with 8m telescopes it is entirely possible to
73: unambiguously identify galaxies at $z\sim5-6$, \cite[see e.g., the
74: spectra in][]{L03, ando04}. Beyond 9000 \AA~ the sky shows a higher
75: density of bright telluric lines and a brighter continuum. As the
76: typical $z\sim5-6$ galaxy has a continuum magnitude of AB$>25$ and a
77: roughly flat intrinsic f$_{\nu}$ spectrum (zero color in AB), this
78: makes identifying candidates at even higher redshifts increasingly
79: difficult.
80:
81: Nevertheless, surveys for even more distant galaxies are
82: particularly important for our understanding of reionization. The
83: likely presence of Gunn-Peterson troughs in the spectra of $z>6$
84: SDSS quasars \citep{becker01, djorgovski01, fan02} indicate that the
85: reionization of hydrogen in the IGM ended at $z\ga6$. Reionization may
86: have begun much earlier; analysis of Wilkinson Microwave Anisotropy
87: Probe (WMAP) first-year temperature and polarization data indicates
88: possible substantial reionization at $z>10$ \citep{kogut03}. As the
89: UV emission from early star-forming galaxies is thought to be a major
90: (and possibly dominant) source of ionizing photons, the evolution in the
91: number and luminosity of such galaxies is likely to be directly linked
92: to the reionization history of the IGM. Based on work carried out thus
93: far to $z\sim 6$ it appears that the Universal UV luminosity and star
94: formation density declines with increasing redshift beyond $z\sim 3$
95: \citep[e.g.,][]{L03, bunker04}. If this trend continues beyond $z=6$,
96: it is difficult to envisage star formation in moderate mass galaxies
97: being responsible for ionizing the Universe at $z>6$ \citep[see
98: e.g.,][]{ricotti04}.
99:
100: One technique developed to overcome the difficulty of detecting even more
101: distant galaxies is the use of gravitational lensing by an intervening
102: galaxy cluster to boost their apparent brightness. This boost can be
103: as much as a factor of 10-100 if the galaxy happens to lie on a critical
104: line. \cite{santos04} proved the utility of this technique out to $z=5.6$
105: and \cite{kneib04} discovered a probable lensed $z\sim 7$ Lyman break galaxy
106: behind Abell 2218.
107:
108: Recently, \cite{pello04} identified a probable highly magnified galaxy
109: at $z{=}10$ lying on a critical line of the cluster Abell 1835. Using
110: broad-band optical imaging from HST and CFHT along with near-IR imaging
111: and spectroscopy from ISAAC/VLT, they presented evidence for a redshift
112: of 10. The object was not detected in $V,R,I$ optical bands and was
113: formally only detected at more than 4$\sigma$ in $H$ ($H_{AB}=25.00\pm
114: 0.25$) and about 3$\sigma$ in $K$ ($K_{s,AB}=25.51\pm 0.36$). The $J-$band
115: detection quoted by \citeauthor{pello04}was $J_{AB}=26.8\pm 1$. $J-$band
116: spectroscopy covered the wavelength range 1.162 to 1.399 $\mu$m and showed
117: an emission line at 1.33745 $\mu$m, detected in two separate central
118: wavelength settings of the spectrograph, with a combined significance
119: level of $4-5\sigma$.
120:
121: This photometry and spectroscopy led \citeauthor{pello04}to argue that
122: the emission line is most likely Ly$\alpha$ at $z{=}10.0175$. A key
123: part of this case was the shape of the spectral energy distribution
124: measured from the imaging data. The object was undetected in the optical,
125: showed a large break between the $J$ and $H$ bands, and had a blue
126: $H-K_s$ color ($H_{AB}-K_{s,AB}<0$). This spectral energy distribution
127: is consistent with a young stellar population observed at $z\sim10$,
128: with light shortward of 1216\AA~ in the rest-frame heavily absorbed by
129: intervening neutral hydrogen. This is consistent with the detected line
130: being Ly$\alpha$ at $z{=}10.0175$.
131:
132: However, given the low signal-to-noise detection, the extremely
133: low flux of the emission line and the limited wavelength covered by
134: the spectroscopy, other emission line identifications are plausible
135: (e.g., [OIII]$\lambda$5007, one component of the [OII]$\lambda$3727
136: doublet, H$\alpha$). Although \citeauthor{pello04} claim the source
137: is on a $z=10$ critical line for Abell 1835, which could be used as
138: a compelling evidence of its extreme redshift, there is no published
139: detailed analysis of the accuracy of the mass model. The critical lines
140: for other redshifts are close to those at $z=10$, so uncertainty in the
141: mass model can lead to the source falling on or near a critical line
142: of a different redshift. This uncertainty is reflected in the range of
143: magnifications given in \cite{pello04}. With all of the uncertainties
144: in the other pieces of evidence that the source is at z=10,
145: the strength of the $H-$band detection is key to this interpretation;
146: without a strong break between $J$ and $H$, there is no other compelling
147: evidence that the line should be identified as Ly$\alpha$.
148:
149: Therefore, given the crucial role of the $H-$band imaging in assigning a
150: redshift to this object, we here report on deep $H-$band images of the
151: field of the $z{=}10$ candidate obtained with the Near-Infrared Imager
152: (NIRI) on the Frederick C. Gillett Gemini Telescope (Gemini-North).
153: These images were obtained both to better constrain the $H-$band photometry
154: (the original detection was not highly significant) and to investigate
155: the morphology of the source with images taken under the excellent seeing
156: conditions that are often attainable at Gemini-North.
157:
158: \section{Data and analysis}
159:
160: \subsection{Observations and Data Reduction}
161:
162: NIRI observations of Abell 1835 were carried out using Director's
163: Discretionary Time on the nights of 30 May and 6 Jun 2004 UT. NIRI is
164: an infrared imager with 0.117 arcsec pixels and $2 \times 2$
165: arcmin$^2$ field of view \citep{hodapp03}. We obtained a total of
166: 22,635 s integration over 2 nights in the $H$ filter centered at 1.65
167: \micron\ (1.49 to 1.78 \micron). The weather was photometric and the
168: seeing was good for both nights. On the first night, the image
169: quality of the final image was 0.40 arcsec FWHM for 11,340 s of
170: exposure. The second night was less good, and the overall image with
171: 11,295 s of total exposure had 0.53 arcsec FWHM. The images were
172: centered on the location of the $z{=}10$ candidate, rather than the
173: center of the cluster (as was the \citeauthor{pello04} data), to allow
174: better sky subtraction.
175:
176: Individual 45 s exposures were reduced using the Gemini IRAF package.
177: Each frame was sky-subtracted and flattened using flat field
178: images produced by the Gemini calibration unit. A unique sky image was
179: derived for each individual exposure by averaging the nearest 8 to 11
180: frames taken within 6 min, with stars and galaxies masked out. Since the
181: $z{=}10$ candidate is located in a relatively low density region of the
182: cluster, residual background variations due to incomplete masking
183: of extended galaxies are not severe. Individual sky-subtracted and
184: flat-fielded frames were then registered using integer shifts, averaged
185: without any weighting, and cosmic rays were rejected. Data from the
186: two nights were reduced separately, and later combined to produce the
187: final image. The image quality in the combined image is 0.47 arcsec FWHM.
188: Figure 1 shows the region around the candidate object.
189:
190: Two UKIRT faint standard stars \citep{hawarden01} were observed and
191: reduced in the same way as the cluster data to determine the photometric
192: calibration. The calibration determined for these two nights agreed to
193: within 0.01 mag with the extinction-corrected zero point over the two
194: month period prior to our observations. We obtained a zero point at
195: H-band of 24.146 (compared to the NIRI mean of H=24.14 measured over
196: many months). Our measured average sky brightnesses were H=14.15 and
197: 14.27 mag per square arcsec for the two nights (typical for Mauna Kea).
198: The sky variations throughout both nights were a smooth function of
199: airmass, with the vast majority of data frames having values within 0.1
200: magnitudes of the above averages. In addition, NIRI uses Hall effect
201: sensors to identify which filter is in the telescope beam. These identify
202: in a unique and absolute way a given filter making it impossible to have
203: an inappropriate filter in the beam. The zero points and sky brightnesses
204: are in excellent agreement for NIRI in the H-band and are inconsistent
205: with any filter other than H.
206:
207: For comparison, we retrieved the raw ISAAC/VLT data used by
208: \citeauthor{pello04}from the ESO data archive and re-reduced their
209: $H$-band data. The resulting image is also shown in Figure 1 with the
210: same display parameters as the NIRI/Gemini image. The final ISAAC
211: $H-$band image was made from the same $H-$band data as presented in
212: \citeauthor{pello04}as judged by comparing the total integration time
213: listed in their paper with the total integration time of data we
214: reduced (i.e., all the data from ESO programme ID 70.A-0355).
215:
216: \subsection{Photometry}
217:
218: To assess our ability to reliably detect objects with $H_{\rm
219: AB}\,{=}\,25.0$ \citep[the magnitude of the $z=10$ candidate measured
220: by][]{pello04}, we added twenty-eight artificial stars with $H_{\rm
221: AB}\,{=}\,25.0$ and 0.47 arcsec FWHM to the final Gemini image, three
222: of which are shown in Figure~2. The mean recovered magnitude for these
223: objects was $H_{\rm AB}\,{=}\,25.035$, with 1$\sigma$ uncertainties
224: on each measurement typically 0.18 magnitudes. The aperture used for
225: photometry was a circle of diameter 3 times the seeing disk (a diameter of
226: 1.4 arcsec). The good agreement between input and recovered magnitudes
227: indicate that no aperture correction is necessary. The artificial
228: sources are similar in brightness to the four faint objects marked in
229: Figure~2 and suggest what the \citeauthor{pello04} object should look
230: like if it were present in our image. All of the objects were easily
231: detected in the NIRI image (see Figure~2).
232:
233: Despite this, our Gemini $H$-band image shows no sign of an object at
234: the location of the $z{=}10$ galaxy candidate. Given the measured sky
235: noise and the results of our artificial star analysis, a point source with
236: $H_{\rm AB}\,{=}\,25.0$ would have been detected with a $S/N\,{\sim}\,7-8$
237: in a photometric aperture three times the seeing FWHM (1.4 arcseconds
238: diameter). We randomly placed twenty-five 1.4 arcsec diameter apertures
239: on sky regions within the target area shown in Figure~1 in order to
240: determine detection limits for the candidate. The distribution of residual
241: flux in these apertures gave a $3-\sigma$ limit of $H_{\rm AB} >26.30$,
242: a result which agrees with the measured uncertainties on the magnitudes
243: of the artificial sources.
244:
245: We also measured the distribution of the individual pixel counts about
246: the mode of the sky value for the entire frame, after excising pixels
247: containing flux from identified objects. This more conservative approach
248: takes into account systematic variations in sky level due to subtracting
249: the individual unique sky images, which are inevitably influenced by
250: small amounts of flux from objects not fully masked out when creating
251: the images. The resulting $3-\sigma$ limit of $H_{\rm AB} >26.03$
252: differs from the aperture determination by 0.27 magnitudes, indicating
253: that these systematic uncertainties are minor. Although, as noted above,
254: these systematic uncertainties are likely to be less of an issue in
255: the region of the candidate than for the frame as a whole, for the rest
256: of this paper we use the more conservative limit of $H_{\rm AB} >26.0$
257: for the flux at the position of the $z{=}10$ candidate.
258:
259: Several objects detected in the region of our image shown in Figure~2 have
260: magnitudes comparable to or somewhat fainter than \citeauthor{pello04}'s
261: $H$-band magnitude for the candidate $z{=}10$ galaxy. Four of these
262: objects lie within 15 arcsec of the $z{=}10$ candidate and provide
263: a robust indication of the depth of the image. Object brightnesses
264: were determined using the IRAF apphot aperture photometry package.
265: All four have measured $S/N\,{\gtrsim}\,5$ within an aperture 1.4 arcsec
266: in diameter. The positions of these four comparison objects are shown
267: in Figure~2, and their brightnesses are listed in Table~1.
268:
269: \section{Discussion}
270: \label{discuss}
271:
272: The lack of an $H-$band detection in our Gemini North NIRI data at a
273: level significantly fainter than the detection by \cite{pello04} is
274: puzzling. Our observations are deeper (by about one magnitude), and
275: have better spatial resolution and sampling. It is possible that the
276: object is time-variable or transient, in which case non-concurrent
277: multi-band photometry does not constrain the redshift, or has a large
278: proper motion ({\it i.e.}, is a solar system object: the ecliptic
279: latitude of the source is about 14 degrees) given the range of dates
280: over which the ESO data were obtained and the time elapsed between the
281: Gemini and VLT observations.
282:
283: In any event, our non-detection in $H$ greatly weakens the argument
284: based on the large break between the optical, $J$, and $H$ bands which
285: supported the claim that the line detected in the spectroscopy is most
286: likely Ly$\alpha$ at $z=10.0$. Given the photometry in $J$ reported
287: by \citeauthor{pello04}, there is no formal continuum detection in
288: the optical or $J$. The photometry by \citeauthor{pello04}indicates
289: the object is only marginally detected in $H$ ($\sim$4$\sigma$;
290: $H_{AB}=25.00\pm0.25$) and $K$ (3$\sigma$; $K_{s,AB}=25.51\pm0.36$). With
291: our new $H-$band data ($H_{AB}>26.0$), the broad-band photometry can no
292: longer be said to constrain the redshift of this object at all.
293:
294: This leaves the $\sim$4$\sigma$ detection of the emission line from
295: the spectroscopy. Recently \cite{Weatherley04} have thrown doubt on
296: the reality of this line. Taken together with our work, the reality
297: of any source at this position has to be strongly questioned. However,
298: whereas our work is based on an independent data set, \cite{Weatherley04}
299: reanalyse the \citeauthor{pello04} data. It is possible that small
300: differences in reduction may lead to contentious results, especially
301: given the faintness and low signal-to-noise of the detection claimed by
302: \cite{pello04}. Consequently, the following discussion assumes that the
303: line is real. But even if it is not, the discussion is directly relevant
304: to all searches for high redshift line emitters.
305:
306: It is still possible that this line is Ly$\alpha$ at $z{=}10$, but with
307: only this level of detection and no corroboration from photometry,
308: the case is weakened considerably and perhaps this is not the most
309: likely interpretation. Other lines such as [OII], [OIII], H$\alpha$
310: at redshifts between $0.77<z<2.75$ were detectable given the wavelength
311: range used in the spectroscopy. The $H-K_s$ color can no longer be
312: used as evidence for an intrinsically blue and hence extremely young
313: high redshift galaxy, given our sensitive upper limit.
314:
315: Luminosity distributions for high redshift objects in \cite{SLBGs4}
316: indicate that there should be several tens of low continuum luminosity
317: galaxies per square arcminute between $0.77<z<2.7$ with continuum
318: fluxes below the optical flux limits quoted in \cite{pello04}. Indeed,
319: \cite{richard03} discovered a faint $z=1.7$ galaxy using the same imaging
320: and spectroscopic data as used by \citeauthor{pello04} This object
321: was identified in continuum longward of the $I$ band from imaging and
322: in three emission lines from spectroscopy. A comparable galaxy several
323: times fainter in both continuum and lines would have been undetected
324: in the photometry and would have only been detected in a single narrow
325: line. Thus it remains possible that the \citeauthor{pello04}object is
326: a galaxy of this type, almost regardless of which of the above emission
327: lines is the true identification.
328:
329: HII dwarf galaxies in the local universe and at moderate redshifts
330: show a correlation between H$\beta$ luminosity and emission line
331: widths \citep{melnick00}. \citeauthor{pello04}measure a line flux of
332: $\sim$4$\times$10$^{-18}$ ergs s$^{-1}$ cm$^{-2}$ and an upper limit to
333: the line width of about 60 km s$^{-1}$. If we assume the galaxy to be at
334: redshift appropriate for the most likely alternative line identifications
335: such as [OII]$\lambda$3726, [OII]$\lambda$3729, [OIII]$\lambda$5007
336: or H$\alpha$, a reasonable range of ratio of H$\beta$ to these other
337: lines for dwarf galaxies, and a small amount of magnification due to the
338: intervening cluster, we find that the emission line plausibly lies along
339: the correlation of \cite{melnick00} for line width versus H$\beta$ line
340: luminosity. As an example, suppose that the actual line identification
341: is [OIII]$\lambda$5007 at a redshift of 1.67, leading to a luminosity
342: of 40.9 ergs s$^{-1}$ in the log assuming reasonable cosmological
343: parameters (H$_0$=70 km s$^{-1}$ Mpc$^{-1}$, $\Omega_{matter}$=0.3
344: and $\Omega_{\Lambda}$=0.7). For HII/dwarf galaxies, the ratio of
345: [OIII]$\lambda$5007 to H$\beta$ ranges from about 2 to up to 10,
346: consequently the H$\beta$ luminosity would be between 39.9 to 40.6
347: ergs s$^{-1}$ in the log. For the published line width, this range in
348: luminosity falls on the correlation between line luminosity and width
349: given by \citeauthor{melnick00} for HII/dwarf emission line galaxies,
350: showing that identifying this line as an optical emission line from a
351: moderate redshift HII/dwarf galaxy is plausible.
352:
353: \acknowledgments
354:
355: We wish to thank the pair of anonymous referees for their insightful and
356: encouraging comments, and Roser Pell\'{o}, Daniel Schaerer, and Dan Stern
357: for their helpful comments. These results are based on observations
358: obtained at the Gemini Observatory, which is operated by the Association
359: of Universities for Research in Astronomy, Inc., under a cooperative
360: agreement with the NSF on behalf of the Gemini partnership: the National
361: Science Foundation (United States), the Particle Physics and Astronomy
362: Research Council (United Kingdom), the National Research Council (Canada),
363: CONICYT (Chile), the Australian Research Council (Australia), CNPq
364: (Brazil) and CONICET (Argentina). These Gemini observations were supported
365: by Director's Discretionary Time. The data retrieved from the ESO data
366: archive were from observations made with the Paranal Observatory under
367: programme ID 70.A-0355(C). LD acknowledges receipt of a PPARC studentship.
368:
369: \begin{thebibliography}{}
370:
371: \bibitem[Ajiki et al. (2003)]{ajiki03}Ajiki, M. et al. 2003, \aj,
372: 126, 2091
373:
374: \bibitem[Ando et al. (2004)]{ando04}Ando, M., Ohta, K., Iwata, I.,
375: Watanabe, C., Tamura, N., Akiyama, M., \& Aoki, K. 2004, astro-ph/0404227
376:
377: \bibitem[Becker et al. (2001)]{becker01}Becker, R. H. et al. 2001, \aj,
378: 122, 2850
379:
380: \bibitem[Bouwens et al. (2004)]{bouwens04}Bouwens, R. J et al. 2004,
381: \apj, 606, 25
382:
383: \bibitem[Bremer et al. (2004)]{bremer04}Bremer, M. N., Lehnert, M. D.,
384: Waddington, I., Hardcastle, M. J., Boyce, P. J., \& Phillipps, S. 2004,
385: \mnras, 347, 7
386:
387: \bibitem[Bunker et al. (2003)]{bunker03}Bunker, A. J., Stanway, E. R.,
388: Ellis, R. S., McMahon, R. G., \& McCarthy, P. J.
389:
390: \bibitem[Bunker et al. (2004)]{bunker04}Bunker, A. J., Stanway, E. R.,
391: Ellis, R. S., \& McMahon, R. G. 2004, astro-ph/0403223
392:
393: \bibitem[Dey et al. (1998)]{Dey98}Dey, A., Spinrad, H., Stern, D.,
394: Graham, J. R., \& Chaffee, F. H. 1998, \apj,498, L93
395:
396: \bibitem[Djorgovski et al. (2001)]{djorgovski01}Djorgovski, S. G., Castro, S., Stern, D., \& Mahabal, A. A. 2001, \apj, 560, 5
397:
398: \bibitem[Fan et al. (2002)]{fan02}Fan et al. 2002, \aj, 123, 1247
399:
400: \bibitem[Hawarden et al. (2001)]{hawarden01}Hawarden, T. G., Leggett,
401: S. K., Letawsky, M. B., Ballantyne, D. R., \& Casali, M. M. 2001, MNRAS,
402: 325, 563
403:
404: \bibitem[Hodapp et al. (2003)]{hodapp03}Hodapp, K. W., Jensen, J. B., Irwin, E.
405: M., Yamada, H. Chung, R., Fletcher, K., Robertson, L., Hora, Joseph L.,
406: Simons, D. A., Mays, W., Nolan, R., Bec, Matthieu, Merrill, M., \& Fowler,
407: A. M. 2003, \pasp, 115, 1388
408:
409: \bibitem[Hu et al. (2004)]{hu04}Hu, E. M., Cowie, L. L., Capak, P.,
410: McMahon, Richard G., Hayashino, T., \& Komiyama, Y. 2004, \aj, 127, 563
411:
412: \bibitem[Kneib et al. (2004)]{kneib04}Kneib, J.-P., Ellis, R. S., Santos,
413: M. R., \& Richard, J. 2004, \apj, 607, 697
414:
415: \bibitem[Kogut et al.(2003)]{kogut03} Kogut, A., et al.\ 2003,
416: \apjs, 148, 161
417:
418: \bibitem[Lehnert \& Bremer (2003)]{L03} Lehnert, M. D. \& Bremer, M. N.
419: 2003, \apj, 593, 630
420:
421: \bibitem[Melnick, Terlevich, \& Terlevich (2000)]{melnick00}Melnick,
422: J., Terlevich, R., \& Terlevich, E. 2000, \mnras, 311, 629
423:
424: \bibitem[Pell\'{o} et al. (2004)]{pello04} Pelló, R., Schaerer, D., Richard,
425: J., Le Borgne, J.-F., \& Kneib, J.-P. 2004, \aa, 416, 35
426:
427: \bibitem[Ricotti, et al. (2004)]{ricotti04}Ricotti, M., Haehnelt, M. G.,
428: Pettini, M., \& Rees, M. J.
429:
430: \bibitem[Rhoads et al. (2003)]{rhoads03}Rhoads, J. E. et al. 2003, \aj,
431: 125, 1006
432:
433: \bibitem[Richard et al. (2003)]{richard03} Richard, J., Schaerer, D.,
434: Pelló, R., Le Borgne, J.-F., \& Kneib, J.-P. 2003, \aa, 412, 57
435:
436: \bibitem[Santos et al. (2004)]{santos04}Santos, M. R., Ellis, R. S.,
437: Kneib, J.-P., Richard, J., \& Kuijken, K. 2004, \apj, 606, 683
438:
439: \bibitem[Stanway et al. (2004a)]{S04a} Stanway, E. R., Bunker, A. J.,
440: McMahon, R. G., Ellis, R. S., Treu, T., \& McCarthy, P. J. 2004, \apj,
441: 607, 704
442:
443: \bibitem[Stanway et al. (2004b)]{S04b}Stanway, E. R. et al. 2004, \apj,
444: 604, 13
445:
446: \bibitem[Steidel et al. (1999)]{SLBGs4}Steidel, C. C., Adelberger, K. L.,
447: Giavalisco, M., Dickinson, M., \& Pettini, M. 1999, \apj, 519, 1
448:
449: \bibitem[Weatherley et al. (2004)]{Weatherley04}Weatherley, S.J., Warren, S.J., Babbedge, T.S.R. 2004, \mnras, submitted, astro-ph/0407150
450:
451: \end{thebibliography}
452:
453: \begin{deluxetable}{cc}
454: \tablewidth{150pt}
455: \tablecolumns{2}
456: \tablecaption{Aperture Photometry of Objects in the Abell 1835 Field}
457: \tablehead{
458: \colhead{Object} &
459: \colhead{$H_{\rm AB}$}
460: \\
461: \colhead{ID\tablenotemark{a}}&
462: \colhead{(AB mag)}
463: }
464: \startdata
465: Cand & $>26.0$\tablenotemark{b} \\
466: 1 & $25.26 \pm 0.21$ \\
467: 2 & $25.23 \pm 0.21$ \\
468: 3 & $24.70 \pm 0.14$ \\
469: 4 & $25.17 \pm 0.20$ \\
470: Art 1& $25.04 \pm 0.18$ \\
471: Art 2& $25.00 \pm 0.17$ \\
472: Art 3& $25.06 \pm 0.18$ \\
473:
474: \enddata \tablenotetext{a}{Artificial objects all have input $H_{\rm
475: AB}\,{=}\,25.0$.} \tablenotetext{b}{3-$\sigma$ upper limit}
476: \tablecomments{All magnitudes and limits measured in 1.4 arcsec
477: diameter apertures. Uncertainties were determined from measurements of
478: rms pixel-to-pixel variation in the sky level directly around each
479: aperture and across the frame as a whole, the quoted value being the
480: larger of these two in each case.}
481:
482: \end{deluxetable}
483:
484: \clearpage
485:
486: \begin{figure}
487: \includegraphics[scale=0.9]{f1.eps}
488: \caption{Deep $H$-band NIRI image of the region around the $z{=}10$
489: candidate object (left) and our reconstruction of the ISAAC/VLT $H$-band
490: image (right). Both images are 18 by 22 arcsec, the NIRI image has
491: 0.47 arcsec FWHM image quality, and both images are displayed with
492: the same linear range and stretch with respect to their noise levels.
493: The circles indicate the location of the candidate object, and are 1.4
494: arcsec in diameter. A comparison of the aperture (1.4 and 3.0 arcsec in
495: diameter) magnitudes of 15 relatively bright (H$_{AB}$$\approx$19-23)
496: objects near the position of the candidate object (estimated using
497: the average zero-point from the ESO web pages for ISAAC for the month
498: around the time of the observations) suggests that the photometry between
499: the ISAAC and NIRI images are consistent with an average difference of
500: $<$H$_{AB,ISAAC}$-H$_{AB,NIRI}$$>$=$-$0.09$\pm$0.10 magnitudes, with no
501: dependence on the $J-H$ color of an object.}
502: \end{figure}
503:
504: \begin{figure}
505: \epsscale{0.7}
506: \plotone{f2.eps}
507: \caption{The same Gemini data as in Figure 1, this time with
508: positions of four other sources near the candidate object with $H_{\rm
509: AB}\,{\gtrsim}\,25$. These are barely visible in the VLT data shown
510: in Figure 1. In addition, the objects labelled ``Art'' are artificial
511: sources with $H_{\rm AB}\,{=}\,25.0$. The three artificial sources are
512: an illustrative subset of 28 artificial point sources used to assess the
513: detectability of objects with this brightness in our image. The photometry
514: for sources 1-4, and these three artificial sources is given in Table 1.}
515: \end{figure}
516:
517: \end{document}
518: