1: %\documentclass{emulateapj}
2:
3: \documentclass[12pt,preprint]{aastex}
4:
5: %% manuscript produces a one-column, double-spaced document:
6:
7: % \documentclass[manuscript]{aastex}
8:
9: %% preprint2 produces a double-column, single-spaced document:
10:
11: % \documentclass[preprint2]{aastex}
12:
13: %\shorttitle{Collapsed Cores in Globular Clusters}
14: %\shortauthors{Djorgovski et al.}
15:
16: %% This is the end of the preamble. Indicate the beginning of the
17: %% paper itself with \begin{document}.
18:
19: \begin{document}
20:
21: \title{Low-Metallicity Gas Clouds in
22: a Galaxy Proto-Cluster at Redshift 2.38} %% As in the title, you can use \\ to force line breaks.
23:
24: \author{Paul J. Francis\altaffilmark{1}}
25: \affil{Research School of Astronomy and Astrophysics, the Australian
26: National University, Canberra 0200, Australia}
27: \email{pfrancis@mso.anu.edu.au}
28:
29: \altaffiltext{1}{Joint Appointment with the Department of Physics,
30: Faculty of Science, the Australian National University}
31:
32: \and
33:
34: \author{Gerard M. Williger}
35: \affil{Department of Physics and Astronomy, Johns Hopkins University,
36: 3701 San Martin Drive, Baltimore, MD 21218}
37: \email{williger@pha.jhu.edu}
38:
39:
40: \begin{abstract}
41:
42: We present high resolution spectroscopy of a QSO whose sight-line
43: passes through the halo of a pair of elliptical galaxies at redshift 2.38.
44: This pair of galaxies probably lies at the center of
45: a galaxy proto-cluster, and is embedded in a luminous extended Ly$\alpha$
46: nebula.
47:
48: The QSO sight-line intersects two small gas clouds within this halo.
49: These clouds have properties similar
50: to those of high velocity clouds (HVCs) seen in the halo of the Milky Way.
51: The gas is in a cool ($< 2 \times 10^4$K) and at least 20\% neutral phase,
52: with metallicities in the
53: range -3.0 $<$ [Fe/H] $<$ -1.1 and neutral hydrogen column densities of
54: $\sim 10^{19.5}{\rm cm}^{-2}$.
55:
56: The origin of these clouds is unclear. The presence of low metallicity gas
57: within this possible proto-cluster implies either that the intra-cluster
58: medium has not been enriched with metals at this redshift, or the
59: clouds are embedded within a hot, ionized, metal-rich gas phase.
60:
61: \end{abstract}
62:
63:
64: \keywords{galaxies: clusters: individual (2142-4420) --- galaxies: halos ---
65: quasars: absorption lines}
66:
67: \section{Introduction}
68:
69: There has been considerable recent work on investigating the inter-relationship
70: between gas (traced by QSO absorption lines) and galaxies at high
71: redshifts \citep[eg.][]{pas01,fra01,wil02,ade03,zha03}. A complex picture
72: is emerging, with gas tracing galaxies on larger scales but some
73: classes of absorption-line system avoiding galaxies on small scales.
74:
75: QSO absorption-line measurements such as these could potentially
76: shed light on the origin of the highly metal enriched intra-cluster medium
77: of galaxy clusters today. Many models have been proposed
78: \citep[eg.][]{ren97,chi00,mar00,agu01,del03} to produce this gas, but all have
79: problems, and there is currently no consensus.
80:
81: In this paper, we present high resolution spectra of a QSO that lies behind
82: a probable galaxy proto-cluster at redshift 2.38 (Fig~\ref{closemap}).
83: QSO LBQS 2139$-$4434 lies
84: at redshift 3.23. Its sight line passes only 22\arcsec\ (160 projected
85: proper kpc, for $H_0 = 70
86: {\rm km\ s}^{-1}{\rm Mpc}^{-1}$, $\Omega_{\rm matter} = 0.3$ and
87: $\Omega_{\Lambda} = 0.7$) from the extremely red object B1. \citet{fra01}
88: demonstrated that B1 is actually a pair of large elliptical galaxies at
89: redshift 2.38..
90:
91: B1 is embedded within a luminous, extended Ly$\alpha$ emitting
92: nebula or ``blob''. Such Ly$\alpha$ nebulae are thought to
93: be associated with massive dark halos, and may be the ancestors of
94: cooling flows today \citep{ste00,fra01}. It lies within a probable
95: galaxy proto-cluster \citep{fra93,fra96,fra97,fww01}, which is itself
96: embedded in a 80 Mpc scale ``Great Wall'' \citep{pal04,fra04}.
97:
98: This sight-line thus allows us to probe the gas in an unusual part of the
99: early universe: the halo of an cluster elliptical galaxy.
100:
101: \section{Observations and Reduction\label{obsred}}
102:
103: Our existing spectra of LBQS 2139$-$4434 \citep{fra93,fww01} were of too low
104: a resolution and
105: narrow a wavelength coverage to seriously constrain the physical properties of
106: the gas. We therefore re-observed it using the University College London Echelle
107: Spectrograph (UCLES) on the Anglo-Australian Telescope (AAT).
108: Observations were carried out on the nights of 2001 August 20 --- 23 and the
109: total usable integration time was 30,600 sec. \citet{dod02} independently
110: obtained high resolution spectra of this QSO,
111: using the UVES spectrograph. They kindly
112: allowed us to use their spectrum in this analysis.
113:
114: The data were reduced using standard procedures, and set to the vacuum
115: heliocentric frame. Absorption-lines
116: properties were measured interactively using the XVOIGT program \citep{mar95}.
117: We fit our UCLES spectrum, the UVES spectrum, and a lower resolution AAT
118: spectrum which gave better coverage at UV wavelengths \citep{fww01}, using
119: for each line whichever spectrum gave the most reliable data. Where a given
120: line was seen in multiple spectra, the derived parameters were always
121: consistent within the error bars. Parameters we derived from the UVES spectra
122: agree very well with those derived by \citet{dod02}.
123:
124: \section{Results}
125:
126: QSO LBQS 2139$-$4434 was previously known to show high equivalent width
127: Ly$\alpha$ absorption at the redshift of the elliptical galaxies making up
128: B1 (and of the cluster). We confirm this: the absorption centroid redshift
129: matches the \ion{C}{4} emission-line centroid of B1 to within
130: $200 {\rm km\ s}^{-1}$, the uncertainty being dominated by the width of
131: the emission lines.
132:
133: For the first time, we unequivocally detect metal-line absorption from
134: this absorption-line system (Fig~\ref{detplot}). No high ionization
135: absorption is seen: the tentative detection of \ion{C}{4} by
136: \citet{fww01} turns out, at higher resolution, to be caused by blends
137: of unrelated lines at other redshifts.
138:
139: The metal-line absorption is clearly resolved into two narrow components,
140: separated in velocity by $59 {\rm km\ s}^{-1}$: a lower redshift component
141: (component 1) at $z=2.3797$ and a higher redshift component (component 2)
142: at $z=2.3804$. Both components are narrow and
143: spectrally unresolved: we place upper limits on the Doppler $b$ parameter of
144: $< 6.0 {\rm km\ s}^{-1}$ in all transitions. For the \ion{Fe}{2}
145: transitions of component 1
146: we can place a tighter upper limit of $b < 5 {\rm km\ s}^{-1}$. There is no
147: evidence for velocity substructure within either of these components.
148:
149: If we assume that the neutral hydrogen is located in one or both of the
150: metal-line clouds, this allows us to break
151: the degeneracy (noted by \citet{fww01}) between velocity dispersion and
152: column density in our fits to the Ly$\alpha$ emission profile, and measure
153: the neutral hydrogen column density. The combined neutral atomic hydrogen column
154: density is $10^{19.7 \pm 0.1}{\rm cm}^{-2}$. We are unable to constrain how it
155: is divided up between the two components.
156:
157: There is also a strong degeneracy between the assumed velocity dispersion
158: $b$ and the inferred metal column densities. For component 1, we can place
159: a lower limit of $b>2 {\rm km\ s}^{-1}$ from the \ion{O}{1} line:
160: if $b$ were lower, then to get a good fit we would need such a high
161: column density that we would start to see damping wings, which are not
162: observed. We cannot place an observational lower limit on $b$ in component
163: 2, but physically the sound speed of even a cold neutral medium should be
164: at least $1 {\rm km\ s}^{-1}$, so we take that as our lower limit. In
165: Table~\ref{coldens}, we use these limits on $b$ to bracket the possible column
166: densities.
167:
168: We list upper limits on non-detected lines by assuming the same limiting
169: values of $b$, and seeing interactively how strong such an absorber could be
170: before the residuals go clearly into emission.
171:
172: \section{The Physical State of the Absorbing Gas}
173:
174: The strength of the low ionization species, such as \ion{O}{1} and
175: \ion{Si}{2}, compared with our upper limits on high ionization
176: species such as \ion{C}{4} and \ion{Si}{4} clearly indicate that the
177: absorbing gas is cool, and at least partially neutral.
178:
179: There are two possibilities for the temperature of these gas clouds:
180: we could have a warm neutral/partially ionised phase with
181: $T \sim 10^4$K or a cold neutral phase with $10^2 < T < 10^3$ K
182: \citep[eg.][]{sut93}.
183:
184: We modeled the first possibility (warm neutral/partially ionised phase)
185: using Gary
186: Ferland's CLOUDY code \citep{fer96}. For plausible choices of the
187: UV ionizing background \citep{haa96,sco00}, we can reproduce the
188: absorption-line columns measured with a cloud of density
189: $0.1 < N_H < 10 {\rm cm}^{-3}$, size 2 --- 200 pc and mass
190: 0.001 -- 10 $M_{\sun}$. If the density is towards the lower end of this range,
191: the hydrogen could be up to $\sim 80$\% ionised.
192:
193: The second possibility is almost impossible to model, due to the complex
194: molecular cooling processes, but cold neutral gas in our own galaxy does show
195: ionization states and column densities comparable to those we observe.
196:
197: Which of these models is correct? Our stringent upper limits on the velocity
198: dispersions of the two components provide some evidence in favor of the cold
199: model. The sound speed in the warm model is
200: $\sim 10 {\rm km\ s}^{-1}$, while in the cold model it would be only
201: 1 --- 3 ${\rm km\ s}^{-1}$. One would expect the internal
202: turbulent velocity dispersions to be comparable to the sound speed. The
203: rapid movement of these clouds through B1's dark matter halo should be
204: ample to excite and sustain such turbulent motions. Our upper limit
205: of $B < 6 {\rm km\ s}^{-1}$ is thus most consistent with a cool model.
206:
207: On the other hand, if the velocity dispersion really is low ($b \sim 2 {\rm km\
208: s}^{-1}$), we are forced to invoke extremely large \ion{O}{1} column
209: densities to fit the observed absorption line. This would imply
210: an extraordinary overabundance of oxygen compared to silicon: [O/Si] $> 1.5$
211: for $b \sim 2$. Larger values of $b$ bring the inferred column density of
212: \ion{O}{1} down to more reasonable values. A cold neutral phase with
213: supersonic turbulence would be an interesting possibility.
214:
215: If the gas were in the cold phase,
216: we might expect dust to be present. We can place upper
217: limits on the quantity of dust by noting, however, that the QSO has a UV
218: continuum slope indistinguishable from other QSOs in the Large Bright
219: QSO Sample \citep{fra91}. This allows us to place a rough upper limit on
220: the extinction in these clouds of $E(B-V)<0.1$.
221:
222: We also note (Fig~\ref{detplot}) that the \ion{O}{1} line touches zero flux.
223: Thus
224: the absorbing cloud must be large enough to completely obscure the background
225: QSO continuum emitting region.
226:
227: In conclusion, we are unable to discriminate between the two physical
228: models. We consider the oxygen overabundance argument somewhat stronger
229: than the velocity dispersion argument, but neither is conclusive. Higher
230: resolution spectra will be needed to resolve this puzzle.
231:
232: \section{Conclusions}
233:
234: Our line of sight though the halo of B1, and probably though the central
235: regions
236: of a galaxy proto-cluster, is thus intersecting two small gas
237: clouds. Their properties are very similar to those of high velocity clouds
238: (HVCs) in the halos of our own galaxy \citep[eg.][]{gib01,tri03}. Their
239: metallicities must be low: $-3.0 < $[Fe/H] $< -1.1$,
240: [C/H] $< -1.65$ and $-2.9 <$ [Si/H] $<-1.65$ (Table~\ref{coldens}, and allowing
241: for neutral hydrogen fractions as low as 20\%). The warmer the gas, the lower
242: our inferred metallicity. If
243: there is substantial molecular hydrogen, the metallicities would
244: be lower still.
245:
246: Given the continued controversy over the nature even of the HVCs in our own
247: galactic halo, it would be foolhardy to make dogmatic statements about
248: the origins of these distant clouds. Many of the same models that have been
249: proposed for our local HVCs (eg. tidally disrupted dwarf galaxies, in-falling
250: primordial gas, gas trapped in dark matter sub-halos, debris from galactic
251: fountains) could apply here.
252:
253: The metallicities we measure for these clouds are lower than the typical
254: intra-cluster medium (ICM) metallicities of galaxy clusters today. This implies
255: either that this enrichment of this cluster had not yet taken place, or
256: that the cold gas clouds we see are in some way isolated from the enriched
257: gas. It is possible, for example, that these clouds are embedded in a very hot,
258: ionized, X-ray emitting, metal-rich ICM. We would not detect such a hot phase in
259: absorption. We do not detect B1 in X-rays (Williger et al. 2004, in
260: preparation), but our current limits do not place tight constraints on the
261: existence of such an ICM.
262:
263: These observations may also shed light on the enigmatic Ly$\alpha$ blobs. While
264: our sight-line does not pass through the blob, it does probe the same dark
265: matter halo. It is quite possible that the Ly$\alpha$ emission region is made up
266: of small clouds similar to those we are probing in absorption. The emission
267: could be driven by photoionization and photo-evaporation of these clouds by
268: some concealed AGN, or could be produced by fast shocks created at cloud-cloud
269: collisions, as discussed by \citet{fra01}.
270:
271: \acknowledgments
272:
273: We would like to thank D'Odorico, Petitjean and Cristiani for allowing us to
274: use their UVES spectrum of QSO 2139$-$4434, and Ralph Sutherland for
275: many illuminating discussions. We would also like to thank all
276: the people who helped us get back to work at Mt Stromlo Observatory so soon
277: after the January 2003 bushfires.
278:
279: \begin{thebibliography}{}
280:
281: \bibitem[Adelberger et al.(2003)]{ade03} Adelberger, K.L., Steidel, C.C.,
282: Shapley, A.E. \& Pettini, M. 2003, \apj , 584, 45
283:
284: \bibitem[Aguirre et al.(2001)]{agu01} Aguirre, A., Hernquist, L.,
285: Schaye, J., Katz, N., Weinberg, D.H. \& Gardner, J. 2001, \apj , 561, 521
286:
287: \bibitem[Chiosi(2000)]{chi00} Chiosi, C. 2000, \aap , 364, 423
288:
289: \bibitem[De Lucia, Kauffmann \& White(2003)]{del03} De Lucia, G., Kauffmann, G.
290: \& White, S.D.M. 2003, \mnras , submitted (astro-ph/0310268)
291:
292: \bibitem[D'Odorico, Petitjean \& Cristiani(2002)]{dod02}
293: D'Odorico, V., Petitjean, P. \& Cristiani, S. 2002, \aap , 390, 13
294:
295: \bibitem[Ferland(1996)]{fer96} Ferland, G.J. 1996, ``Hazy, a Brief
296: Introduction to Cloudy'', University of Kentucky Department of Physics and
297: Astronomy Internal Report.
298:
299: \bibitem[Francis et al.(1991)]{fra91} Francis, P.J., Hewett, P.C., Foltz, C.B.,
300: Chaffee, F.H., Weymann, R.J. \& Morris, S.L. 1991, \apj , 373, 465
301:
302: \bibitem[Francis and Hewett(1993)]{fra93} Francis, P.J. \& Hewett, P.C. 1993,
303: \aj , 105, 1633
304:
305: \bibitem[Francis et al.(1996)]{fra96} Francis, P.J. et al. 1996, \apj , 457,
306: 490
307:
308: \bibitem[Francis et al.(2001)]{fra01} Francis, P.J. et al. 2001, \apj , 554,
309: 1001
310:
311: \bibitem[Francis et al.(2004)]{fra04} Francis, P.J., Palunas, P.,
312: Teplitz, H.I., Williger, G.M. \& Woodgate, B.E. 2004, in preparation.
313:
314: \bibitem[Francis, Wilson and Woodgate(2001)]{fww01}
315: Francis, P.J., Wilson, G.M. \& Woodgate, B.E. 2001, PASA, 18, 64
316:
317: \bibitem[Francis, Woodgate and Danks(1997)]{fra97}
318: Francis, P.J., Woodgate, B.E. \& Danks, A.C. 1997, \apjl , 482, L25
319:
320: \bibitem[Gibson et al.(2001)]{gib01} Gibson, B.K., Giroux, M.L., Penton, S.V.,
321: Stocke, J.T., Shull, J.M \& Tumlinson, J. 2001, \aj , 122, 3280
322:
323: \bibitem[Haardt \& Madau(1996)]{haa96} Haardt, F. \& Madau, P. 1996,
324: \apj , 461, 20
325:
326: \bibitem[Kauffman et al.(1999)]{kau99} Kauffmann, G.,
327: Colberg, J.M., Diaferio, A., \& White, S.D.M. 1999, \mnras ,
328: 303, 188
329:
330: \bibitem[Mar and Bailey(1995)]{mar95} Mar, D.P., \& Bailey, G. 1995,
331: PASA, 12, 239
332:
333: \bibitem[Martinelli, Matteucci and Colafrancesco(2000)]{mar00}
334: Martinelli, A., Matteucci, F. \& Colafrancesco, S. 2000, \aap , 354, 387
335:
336: \bibitem[Palunas et al.(2004)]{pal04} Palunas, P., Teplitz, H.,
337: Francis, P.J., Williger, G.M. \& Woodgate, B.E. 2004, in press
338:
339: \bibitem[Pascarelle et al.(2001)]{pas01} Pascarelle, S.M., Lanzetta, K.M.,
340: Chen, H-W., \& Webb, J.K. 2001, \apj , 560, 101
341:
342: \bibitem[Renzini(1997)]{ren97} Renzini, A. 1997, \apj , 488, 35
343:
344: \bibitem[Scott et al.(2000)]{sco00} Scott, J., Bechtold, J.,
345: Dobrzycki, A. \& Kulkarni, V.P. 2000, \apjs , 130, 67
346:
347: \bibitem[Steidel et al.(2000)]{ste00} Steidel, C.C., Adelberger, K.L.,
348: Shapley, A.E., Pettini, M., Dickinson, M., \& Giavalisco, M. 2000, \apj ,
349: 532, 170
350:
351: \bibitem[Sutherland \& Dopita(1993)]{sut93} Sutherland, R.S. \& Dopita,
352: M.A. 1993, \apjs , 88, 253
353:
354: \bibitem[Tripp et al.(2003)]{tri03} Tripp, T.M. 2003, \aj , 152, 3122
355:
356: \bibitem[Williger et al.(2002)]{wil02} Williger, G.M., Campusano, L.E.,
357: Clowes, R.G. \& Graham, M.J. 2002, \apj , 578, 708
358:
359: \bibitem[Zhan, Fang \& Burstein(2003)]{zha03} Zhan, H., Fang, L-Z.,
360: \& Burstein, D. 2003, \apjl , 592, 1
361:
362: \end{thebibliography}
363:
364:
365: \clearpage
366:
367: %% Use the figure environment and \plotone or \plottwo to include
368: %% figures and captions in your electronic submission.
369:
370:
371: \clearpage
372:
373:
374: \begin{figure}
375: \plotone{f1.eps}
376: \caption{
377: Close-up image of QSO 2139$-$4434, and the $z=2.38$ galaxy B1, which lies
378: 22\arcsec\ away. Greyscale is a NICMOS F160W image (rest-frame $B$-band),
379: while the contours show Ly$\alpha$ emission, measured with the Rutgers
380: Fabry-Perot on the CTIO Blanco Telescope.
381: \label{closemap}}
382: \end{figure}
383:
384: \begin{figure}
385: \plotone{f2.eps}
386: \caption{
387: UVES spectra of the nine best metal-line detections in
388: QSO 2139-4434. Velocities are relative to a nominal
389: redshift of 2.38.
390: \label{detplot}}
391: \end{figure}
392:
393: \clearpage
394:
395: \begin{deluxetable}{lcccc}
396: \tablewidth{0pt}
397: \tablecolumns{5}
398: \tablecaption{LBQS 2139$-$4434 Absorption-Line Column Densities \label{coldens}}
399: \tablehead{
400: \colhead{Ion} &
401: \multicolumn{4}{c}{Log(Column Density) (cm$^{-2}$)} \\
402: \colhead{ ~ } & \multicolumn{2}{c}{z=2.3797 Component} &
403: \multicolumn{2}{c}{z=2.3804 Component} \\
404:
405: \colhead{ ~ } & \colhead{$b=2{\rm km\ s}^{-1}$} &
406: \colhead{$b=5{\rm km\ s}^{-1}$} &
407: \colhead{$b=1{\rm km\ s}^{-1}$} &
408: \colhead{$b=6{\rm km\ s}^{-1}$} \\
409: }
410: \startdata
411:
412: \ion{H}{1} & \multicolumn{4}{c}{19.6 --- 19.8} \\
413: \ion{C}{1} & $<$12.2 & $<$12.2 & $<$12.8 & $<$12.8 \\
414: \ion{C}{2} & $<$14.5 & $<$15.5 & $<$14.4 & $<$13.9 \\
415: \ion{C}{4} & $<$12.4 & $<$12.5 & $<$12.5 & $<$12.5 \\
416: \ion{O}{1}
417: & 16.4 --- 17.0 & 14.2 --- 14.7 & 16.9 --- 17.3 & 14.0 --- 14.3 \\
418: \ion{Si}{1} & $<$11.7 & $<$11.9 & $<$11.6 & $<$11.9 \\
419: \ion{Si}{2}
420: & 13.0 --- 13.3 & 13.0 --- 13.1 & 12.4 --- 13.1 & 12.5 --- 12.9 \\
421: \ion{Si}{4} & $<$11.6 & $<$11.8 & $<$12.2 & $<$12.4 \\
422: \ion{S}{1} & $<$12.0 & $<$11.9 & $<$12.4 & $<$12.6 \\
423: \ion{Fe}{2}
424: & 13.8 --- 13.9 & 13.0 --- 13.1 & 13.0 --- 13.2 & 12.6 --- 12.8 \\
425:
426: \enddata
427: \end{deluxetable}
428:
429:
430:
431:
432: \end{document}
433:
434: