1: \documentstyle[emulateapj,psfig,apjfonts]{article}
2:
3: \def\gs{\mathrel{\raise0.35ex\hbox{$\scriptstyle >$}\kern-0.6em
4: \lower0.40ex\hbox{{$\scriptstyle \sim$}}}}
5: \def\ls{\mathrel{\raise0.35ex\hbox{$\scriptstyle <$}\kern-0.6em
6: \lower0.40ex\hbox{{$\scriptstyle \sim$}}}}
7: \def\farcs{\hbox{$.\!\!^{\prime\prime}$}}
8:
9: \submitted{Received 2001 August 31; accepted 2001 September 24}
10:
11: \lefthead{Ivison et al.}
12: \righthead{A high-resolution imaging study of SMM\,J14011+0252}
13:
14: \begin{document}
15:
16: \title{Locating the starburst in the SCUBA galaxy, SMM\,J14011+0252}
17:
18: \author{R.\ J.\ Ivison\altaffilmark{1}, Ian Smail\altaffilmark{2},
19: D.\ T.\ Frayer\altaffilmark{3}, J.-P.\ Kneib\altaffilmark{4}
20: \& A.\ W.\ Blain\altaffilmark{5}}
21: \altaffiltext{1}{Astronomy Technology Centre, Royal Observatory,
22: Blackford Hill, Edinburgh EH9 3HJ, UK}
23: \altaffiltext{2}{Department of Physics, University of Durham, South
24: Road, Durham DH1 3LE, UK}
25: \altaffiltext{3}{{\em SIRTF} Science Centre, California Institute of
26: Technology, IPAC, MS\,220-06, Pasadena, CA 91125, USA}
27: \altaffiltext{4}{Observatoire Midi-Pyr\'en\'ees, UMR 5572,
28: 14 Avenue E.\ Belin, F-31400 Toulouse, France}
29: \altaffiltext{5}{Astronomy 105-24, California Institute of
30: Technology, Pasadena, CA 91125, USA}
31:
32: \setcounter{footnote}{5}
33:
34: \begin{abstract}
35: We present new, multi-wavelength, high-resolution imaging of the
36: luminous, submillimeter (submm) galaxy, SMM\,J14011+0252, an
37: interacting starburst at $z=2.56$. Our observations comprise optical
38: imaging from the {\it Hubble Space Telescope}, sensitive radio mapping
39: from the Very Large Array and CO observations from the Owens Valley
40: Radio Observatory and Berkeley-Illinois-Maryland Array. Aided by
41: well-constrained gravitational amplification, we use these new data to
42: map the distribution of gas and both obscured and unobscured
43: starlight. The maps show that the gas and star formation are extended
44: on scales of $\gs$10\,kpc, much larger than starbursts seen in local
45: ultraluminous galaxies, and larger than the rest-frame UV-bright
46: components of SMM\,J14011+0252, J1/J2. The most vigorous star
47: formation is marked by peaks in both the molecular gas and radio
48: emission, $\sim$1$''$ north of J1/J2, in the vicinity of J1n, an
49: apparent faint extension of J1. Using new sub-$0.5''$ $K$-band
50: imaging from UKIRT, we identify J1n as an extremely red object
51: (ERO). We suggest that while J1 and J2 are clearly associated with the
52: submm source, they are merely windows through the dust, or unobscured
53: companions to a large and otherwise opaque star-forming
54: system. Hence, their rest-frame UV properties are unlikely to be relevant
55: for understanding the detailed internal physics of the starburst.
56: \end{abstract}
57:
58: \keywords{cosmology: observations ---
59: galaxies: individual (SMM\,J14011+0252; SMM\,J14009+0252) ---
60: galaxies: evolution}
61:
62:
63: %
64: %
65: %
66: \section{Introduction}
67:
68: The resolution into individual sources of the majority of the optical,
69: submm and X-ray backgrounds (Madau \& Pozzetti 2000; Blain et al.\
70: 1999; Brandt et al.\ 2001) means that we can now investigate the
71: relationship and balance between obscured and unobscured star
72: formation and AGN activity at high redshifts. For example, the
73: limited overlap between X-ray and submm populations has been taken as
74: evidence that the bulk of the submm background arises from reprocessed
75: starlight (Fabian et al.\ 2000; Barger et al.\ 2001; Almaini et al.\
76: 2001). While the broad equality between the intensity of the optical
77: and far-IR background radiation has then been used to argue that
78: comparable amounts of star formation occur in obscured and unobscured
79: environments (Hauser et al.\ 1998). A major unresolved issue is the
80: relationship between the obscured and unobscured modes of star
81: formation, as represented by SCUBA-selected galaxies and Lyman-break
82: galaxies, respectively (Smail, Ivison, \& Blain 1997; Steidel,
83: Pettini, \& Hamilton 1995). Is unobscured activity simply the tip of
84: the iceberg, and can it be used to predict the obscured component
85: (Adelberger \& Steidel 2000), or do Lyman-break galaxies represent a
86: physically distinct population with no meaningful relationship with
87: galaxies selected in the submm (van der Werf et al.\ 2001; Meurer et
88: al.\ 2001; also see Bell et al.\ 2001)? This has important
89: implications for the relative contributions of optical- and
90: submm-selected galaxies to the history of star formation (Blain et
91: al.\ 1999b; Chapman et al.\ 2000; Peacock et al.\ 2001).
92:
93: \setcounter{figure}{0}
94: %
95: % Figure 1
96: %
97: \begin{figure*}[htb]
98: \centerline{\psfig{file=f1.eps,width=7.1in,angle=0}}
99: \vspace*{-2mm}
100: \caption{\footnotesize
101: Three views of SMM\,J14011+0252: a true-color $UR_{702}K$ image (using
102: $U$-band data from I00), the {\it HST} F702W frame and the UKIRT
103: $K$-band image. The true-color frame is $18''\times 18''$, while the
104: F702W/$K$-band views are zoomed to give a $6''\times 6''$ field to
105: better illustrate the internal structure of this galaxy. We identify
106: the two UV-bright components of SMM\,J14011+0252 on the F702W panel:
107: the morphologically complex J1 and the compact, blue component J2, as
108: well as the diffuse, very red component, J1n. We overlay a contour plot of the
109: seeing-matched F702W image on the $K$-band panel to contrast the
110: optical/IR morphologies (the contours are in 1-mag increments starting
111: at $\mu_R=24.0$\,mag\,arcsec$^{-2}$). North is up; east is left;
112: minor tickmarks denote 1$''$ increments.}
113: \end{figure*}
114:
115: To address this issue we need to study the properties of individual
116: submm galaxies. Unfortunately, progress has been constrained by their
117: faintness in the rest-frame UV and the resultant dearth of
118: spectroscopic redshifts and detailed morphologies for more than a
119: handful of systems (Ivison et al.\ 1998, 2000a --- I00). For the
120: moment all we have to work with are a few, perhaps atypical,
121: optically-bright galaxies with precise redshifts. There are also
122: concerns about the over-representation of AGN in the current samples
123: of spectroscopically-identified submm galaxies, reflecting a bias from
124: the relative ease of measuring redshifts for their strong-lined
125: spectra. Although star formation, rather than the AGN, is thought to
126: dominate the energetics of these systems, the presence of an AGN
127: clearly complicates any detailed analysis. We are therefore left with
128: only a single example from current surveys of the class of galaxy
129: which is expected to comprise the majority of the submm population:
130: the $z=2.56$, gas-rich, luminous starburst, SMM\,J14011+0252 (I00),
131: although even this appears unusually bright in the rest-frame UV.
132:
133: SMM\,J14011+0252, lying behind the core of the $z=0.25$ cluster
134: A\,1835, has an 850-$\mu$m flux of 6\,mJy, after correcting for
135: amplification by the foreground cluster lens (a mean factor of
136: $2.5\times$, dominated by amplification along position angle
137: 9$^{\circ}$), showing that this is an ultraluminous IR galaxy (ULIRG)
138: with a far-IR luminosity of $L_{\rm FIR}\sim 6\times
139: 10^{12}$\,L$_\odot$, a star-formation rate (SFR) of $\sim
140: 10^3$\,M$_\odot$\,yr$^{-1}$ and a large dust mass, $M_{\rm d} \sim
141: 10^8$--$10^9$\,M$_{\odot}$. I00 present detailed observations of this
142: submm-selected source, which they identified with an
143: interacting/merging pair of galaxies, J1 and J2, separated by $2.1''$,
144: with a combined absolute magnitude of $M_R =
145: -25.0$\footnote{Throughout we will assume $q_o=0.5$ and
146: $H_o=50$\,km\,s$^{-1}$\,Mpc$^{-1}$, giving a scale of $1''\equiv
147: 7.7$\,kpc at $z=2.56$. We quote all magnitudes and linear scales as
148: observed, rather than corrected for lens amplification, unless
149: otherwise stated.}. J2 contributes the bulk of the luminosity in the
150: UV, with the redder J1 dominating the near-IR. Optical and near-IR
151: spectra of J1 and J2 show no hint of AGN characteristics (I00) and the
152: system is undetected in hard X-ray observations with {\it Chandra}
153: (Fabian et al.\ 2000), supporting the contention that its luminosity
154: is predominantly produced by an intense starburst. The accurate
155: redshift for this system allowed Frayer et al.\ (1999) to detect
156: CO(3$\rightarrow$2) emission, confirming the presence of a large gas
157: reservoir, $M({\rm H}_2)\sim 10^{11}$\,M$_\odot$ (probably a lower
158: limit based on evidence of cold gas from CO(1$\rightarrow$0) emission
159: in another source, Papadopoulos et al.\ 2001).
160:
161: Although the observational dataset on SMM\,J14011+0252 is the most
162: comprehensive available for any submm galaxy, there is still
163: considerable freedom in interpreting the results. In particular, we
164: need to identify the location of the intense far-IR emission within
165: the system, and the attendant gas reservoir, to understand the exact
166: relationship between the UV and far-IR emission (Adelberger \& Steidel
167: 2000) and determine if they are coming from physically disjoint
168: components. Unfortunately, the published observations had
169: insufficient spatial resolution to answer this question. We have
170: therefore undertaken a campaign to image SMM\,J14011+0252 at high
171: resolution in the radio, millimeter (CO) and optical/near-IR wavebands
172: to investigate in detail the distribution of obscured and unobscured
173: star formation in this galaxy.
174:
175: We describe our new high-resolution images in \S2, analyse these in
176: \S3 and present our conclusions in \S4.
177:
178: %
179: %
180: %
181: \section{Observations and Reduction}
182:
183: \setcounter{figure}{1}
184: %
185: % Figure 2
186: %
187: \begin{figure*}[htb]
188: \centerline{\psfig{file=f2a.eps,width=3.2in,angle=0}
189: \hspace*{0.5in}
190: \psfig{file=f2b.eps,width=3.2in,angle=0}}
191: \vspace*{-3mm}
192: \caption{\footnotesize (a) Combined OVRO/BIMA CO image of
193: SMM\,J14011+0252, with a noise level of 0.8\,mJy\,beam$^{-1}$; (b) VLA
194: 1.4-GHz map, with a noise level of 12\,$\mu$Jy\,beam$^{-1}$. The CO
195: and radio emission are co-spatial, peaking $\sim$1$''$ north of
196: J1. The CO subtends $6.6''\pm 1.4''$ in the north-south direction, the
197: radio somewhat less. All contours are plotted at $-3.5$, $-2.5$, 2.5,
198: 3.5, 4.5, 5.5 $\times \sigma$; positions are relative to the position
199: of J1, shown as a greyscale; synthesized beams are shown.}
200: \end{figure*}
201:
202: \subsection{Optical and Near-IR Imaging}
203:
204: The {\it HST}\,\footnote{This paper is based upon observations
205: obtained with the NASA/ESA {\it Hubble Space Telescope} which is
206: operated by STScI for the Association of Universities for Research in
207: Astronomy, Inc., under NASA contract NAS5-26555.} imaging of
208: SMM\,J14011+0252, in the field of A\,1835, comprises three orbits in
209: the F702W filter, giving a combined integration time of 7.5\,ks. We
210: adopt the Vega-based {\it WFPC2} photometric system, $R_{702}$, from
211: Holtzman et al.\ (1995). The final WFPC2 frame (see Fig.~1) has an
212: effective resolution of 0.15$''$ and a 3-$\sigma$ detection limit
213: within a 2$''$--diameter photometry aperture of $R_{702}\sim26.6$. We
214: also exploit new, near-IR imaging of SMM\,J14011+0252. Using UFTI on
215: the UKIRT,\footnote{The United Kingdom Infrared Telescope (UKIRT) is
216: operated by the Joint Astronomy Centre on behalf of the Particle
217: Physics and Astronomy Research Council.} Smith et al.\ (2001) obtained
218: a 6.5-ks $K$-band exposure during 2001 April 4--7. The image
219: (Fig.~1c) has a 3-$\sigma$ limit of $K\sim 21.2$ and a FWHM of
220: 0.45$''$. The reduction and analysis of both the {\it HST} and UKIRT
221: data is described in full in Smith et al.\ (2001).
222:
223:
224: \subsection{Centimeter and Millimeter Mapping}
225:
226: The initial Owens Valley Millimeter Array (OVRO)\footnote{OVRO is
227: operated by the California Institute of Technology and is supported by
228: NSF grant AST 9981546.} CO observations taken of J1/J2 in 1998 showed
229: evidence for a possible extended component in the north-south
230: direction (Frayer et al.\ 1999). We subsequently obtained 78\,hr of
231: high-resolution CO observations at OVRO and the
232: Berkeley-Illinois-Maryland Array (BIMA)\footnote{The BIMA array is
233: operated by the Berkeley-Illinois-Maryland Association under funding
234: from the NSF.} to constrain the CO morphology and position of the
235: source (Table~1).
236:
237: %
238: % Table 1 (MM Log)
239: %
240: {\scriptsize
241: \begin{center}
242: \centerline{\sc Table 1}
243: \centerline{\sc Millimeter Observations of SMM\,J14011+0252}
244: \begin{tabular}{cccccc}
245: \hline\hline
246: \noalign{\smallskip}
247: Observing & Array & Baseline & Time on & RMS & $\theta_b^a$ \cr
248: Dates & & Lengths (m) & Source (h) & (mJy)& ($''$) \cr
249: \noalign{\smallskip}
250: \hline
251: \noalign{\smallskip}
252: 1998 Oct--Dec & OVRO(L+E) & 15--119 & 41.4 & 1.0 & $6.5\times 4.3$\cr
253: 1999 Dec & BIMA(A) & 80--1730 & 16.7 & 2.0 & $0.7\times 0.6$\cr
254: 2000 Feb--Mar & OVRO(H+U) & 35--483 & 19.8 & 1.8 & $2.4\times 1.9$\cr
255: \hline
256: \end{tabular}
257: \end{center}
258: \addtolength{\baselineskip}{-3pt}
259: $^a$ The rms and synthesized beam sizes ($\theta_b$) are for a
260: natural-weighted map averaged over the CO line width of
261: 320\,km\,$s^{-1}$. The data were taken in four OVRO configurations
262: (L, E, H, U) and the highest resolution BIMA configuration (A).
263:
264: }
265: \medskip
266:
267: Fig.~2a shows the integrated CO map of the combined OVRO and BIMA
268: data. In order to resolve the galaxy while maintaining good signal to
269: noise, we adopt a natural-weighting scheme with a UV taper of
270: 120\,$k\lambda$ using {\sc aips}. We achieved an rms noise level
271: 0.8\,mJy\,beam$^{-1}$ integrated over the CO emission-line width of
272: 320\,km\,s$^{-1}$, with a beam of $3.9''\times 3.9''$.
273:
274: The A\,1835 field was also mapped with the National Radio Astronomy
275: Observatory's (NRAO) Very Large Array (VLA),\footnote{NRAO is operated
276: by Associated Universities Inc., under a cooperative agreement with
277: the National Science Foundation.} during 1998 April and 2000 November.
278: After standard calibration and editing of the data and their
279: associated weights using {\sc aips}, the wide-field imaging task, {\sc
280: imagr}, was used to map the central $10'\times 10'$ field (see, e.g.\
281: Richards 2000). The resulting map (Fig~2b) has a noise level of
282: 12\,$\mu$Jy\,beam$^{-1}$ with 1.45$''$ resolution.
283:
284: \subsection{Astrometry}
285:
286: To reliably compare the positions of the components of
287: SMM\,J14011+0252 we have revisited the astrometry from I00 to tie the
288: optical and near-IR images directly to the radio coordinate frame. By
289: identifying bright radio sources within the {\it HST} field we align
290: the frames to a precision of $\sim$0.3$''$. We then measure the
291: position of J1 to be $\rm 14^h 01^m 04.^{\!\!s}95 \pm 0.^{\!\!s}02,
292: +02^{\circ}$ 52$'$ 24.$\!\!''$0 $\pm$ 0.3$''$ (J2000).
293:
294: The peak of the 1.4-GHz radio source lies at $\rm 14^h 01^m
295: 04.^{\!\!s}92, +02^{\circ}$ 52$'$ 24.$\!\!''$8 (J2000), with an
296: uncertainty of $\pm 0.2''$, coincident with the CO emission which
297: peaks at $\rm 14^h 01^m 04.^{\!\!s}93, +02^{\circ}$ 52$'$ 25.$\!\!''$0
298: (J2000), with an uncertainty of $\pm 0.7''$.
299:
300: %
301: %
302: %
303: \section{Analysis and Discussion}
304:
305: The {\it HST} imaging of SMM\,J14011+0252, aided by the gravitational
306: amplification from A\,1835, provides a sub-kpc scale view of the
307: components of this submm source in the rest-frame UV, $\sim
308: 2000$\,\AA. This shows that J1 is well resolved, with a half-light
309: diameter of $1.34'' \pm 0.08''$ ($3.3\pm 0.2$\,kpc in the source
310: plane). As with another submm galaxy, Lockman 850.1 (Lutz et al.\
311: 2001), its morphology is complex: the main component appears to be
312: relatively regular, with an extended envelope. Superimposed on this
313: are several bright knots, in two groups: one to the south-east and the
314: other to the north (which also appears in $K$). The northern group
315: points towards another bright, resolved blue knot, $2.5''$ away. This
316: knot defines the outer limit of a low-surface-brightness extension,
317: J1n, to the north of J1, which has an extent of 10\,kpc in the source
318: plane and which is much more prominent in $K$ than $R$ (Fig.~1). We
319: note, as did Ivison et al.\ (1998), that red plumes such as this may
320: be due to line emission: probably H\,$\alpha$ at 2.34\,$\mu$m (I00),
321: which could be mapped in the future using near-IR integral-field
322: spectroscopy.
323:
324: The second bright component, J2, is more luminous than J1 at
325: rest-frame wavelengths of $\ls 2000$\,\AA, but rapidly becomes fainter
326: in redder wavebands. J2 has a compact appearance, with a half-light
327: diameter of $0.50'' \pm 0.06''$. The morphology of J2 suggests it
328: could be an edge-on disk extending out to $\sim 0.5''$, with a
329: high-surface-brightness nucleus, though the prominence of this
330: extension may result from the higher north-south amplification.
331:
332: Photometry from the seeing-matched F702W and $K$-band frames gives:
333: J1, $(R_{702}-K)=3.29\pm 0.03$ and J2, $(R_{702}-K)=2.12\pm 0.05$,
334: both in apertures twice the size of their respective half-light
335: diameters. J1n is much redder than either of the bright components
336: with $(R_{702}-K)=5.03\pm 0.10$, very similar to the extreme colors
337: seen in other submm galaxies\footnote{SMM\,J14009+0252 is a 14.5-mJy
338: 850-$\mu$m source in the same field as SMM\,J14011+0252. It has a very
339: faint near-IR counterpart (J5) which was, and remains, undetected at
340: optical wavelengths. We can now place a 3--$\sigma$ limit on its
341: color of $(R_{702}-K)\geq 5.8$ in a $2''$-diameter aperture, meaning
342: that this galaxy is an ERO or a Class~0/I submm source (Ivison et al.\
343: 2000b). This brings the fraction of EROs in the Smail et al.\ sample
344: to $\ge$27 per cent ($\ge$44 per cent for the 4$\sigma$ sample).}
345: (Smail et al.\ 1999).
346:
347: As shown in Fig.~2, the peaks of the CO and radio emission in our
348: high-resolution maps lie $\sim$1$''$ north of J1 and $\sim$2$''$ east
349: of J2. Given the precision of our astrometry, these offsets are
350: significant and so we can rule out the starburst arising from either
351: of these components. The radio/CO emission are not coincident with the
352: ERO, J1n, but are more consistent with this component than with any
353: other. Thus the bulk of the luminosity, gas and dust appear be located
354: close to, or within, the extremely red component.
355:
356: I00 estimated the SFR for $\ge8$\,M$_{\odot}$ stars in
357: SMM\,J14011+0252, based on far-IR and radio fluxes, of $\sim
358: 10^3$\,M$_{\odot}$\,yr$^{-1}$. In contrast, the H$\alpha$ luminosity
359: indicates only $\sim 10^2$\,M$_{\odot}$\,yr$^{-1}$. They concluded
360: that the likely cause of this discrepency is significant dust
361: extinction, extending to rest-frame $\sim$6600\AA (observed $K$-band),
362: which obscures the most vigorously star forming regions within the
363: galaxy. This is consistent with the faint appearance of the
364: star-forming region in our $K$-band image and its complete absence in
365: the {\it HST} frame, as well as the conclusions of studies which
366: constrast the very different appearance of local starbursts in the
367: optical and the near-IR/radio. We suggest that the known optical
368: components, J1/J2, merely represent foreground star-forming knots, or
369: windows through the dust in a large and otherwise opaque star-forming
370: complex.
371:
372: Since the starburst is obscured even at $K$, we must turn to our
373: long-$\lambda$ observations to investigate its morphology. The
374: distributions of both the radio and CO emission are extended roughly
375: north--south, consistent with alignment of J1--J1n, with the emission
376: resolved along this axis. After deconvolving the beam, the molecular
377: gas traced by CO subtends $6.6''\pm1.4''$ along the major axis
378: (Fig.~2a). Assuming conservatively that the amplification is entirely
379: north-south, the molecular gas must cover at least 20\,kpc. The best
380: Gaussian fit to the 1.4-GHz emission gives $2.3'' \times 1.5''$ ({\sc
381: fwhm}), although a fainter emission extends to the south (Fig.~2b). We
382: stress that several nearby radio sources remain unresolved in our map,
383: so this extension is real. This strongly suggests that the far-IR
384: emission is powered by a starburst: the vast molecular gas reservoir
385: appears to be converted into stars and ultimately into SNe on a
386: galaxy-wide scale. The growing evidence of highly extended starbursts
387: in the submm galaxy population (Ivison et al., in prep) indicates that
388: physical conditions (gas densities, dust shielding, wind formation) in
389: these systems will be very different from those expected on the basis
390: of local, compact ($\le 1$\,kpc) ULIRGs.
391:
392: %
393: %
394: %
395: \section{Conclusions}
396:
397: We have presented new high-resolution optical, CO and radio
398: observations of the $z=2.56$ SCUBA-selected ULIRG, SMM\,J14011+0252.
399: Our new $K$-band imaging has allowed us to resolve an ERO component
400: within this system ($R_{702}-K=5.0$). Of the known optical/near-IR
401: components, this heavily dust-obscured component is closest to the
402: starburst, as probed by our high-resolution CO and radio maps.
403:
404: These observations are relevant for the issue of the relationship
405: between the Lyman-break and submm galaxy populations. In this paper
406: we have studied a galaxy which has been claimed to represent the
407: transition between these two populations, with UV properties that
408: confirm the reliability of large extinction corrections of
409: the type applied to Lyman-break galaxies (Adelberger \& Steidel 2000).
410: Our higher resolution data have demonstrated that the properties of
411: SMM\,J14011+0252 are much more complex than previously thought, with
412: the bulk of the luminosity coming from an extremely red component
413: which is not seen in rest-frame UV observations, or possibly from a
414: region which is blank even in the near-IR. While the optically bright
415: components, J1 and J2, are clearly associated with the submm source,
416: they appear as distinct regions within the system and their rest-frame
417: UV properties are therefore unlikely to provide a reliable indication
418: of the physical processes occuring within the highly obscured
419: starburst. J1, J2 and even J1n may merely mark windows through the
420: dust in an otherwise opaque star-forming complex. This calls into
421: question the validity of using UV-selected objects to trace the
422: evolution of obscured star formation at high redshifts.
423:
424: Nevertheless, in the absence of the less obscured components, J1 and
425: J2, this system would have been difficult to study in the detail
426: achieved here. The presence of UV-bright companions may prove to be
427: the only practical way to measure redshifts for submm galaxies for the
428: foreseeable future (see the discussion of J5 in I00; cf.\ Blain et
429: al.\ 2000; Townsend et al.\ 2001).
430:
431: New information on the luminous submm source, SMM\,J14009+0252,
432: previously suggested as a radio-loud AGN with ERO characteristics,
433: confirm that this galaxy is an ERO, raising the fraction of known
434: extremely red counterparts to almost half of the most robust parent
435: sample.
436:
437: High-redshift dusty galaxies such as SMM\,J14011+0252 and
438: SMM\,J14009+0252 account for $\gs 50$ per cent of the submm background
439: (Blain et al.\ 1999), leaving room for only a modest contribution
440: from the submm counterparts of UV-selected systems (Peacock et al.\
441: 2000). This is in sharp contrast to the situation in the local
442: Universe, where very luminous dusty starbursts contribute a minor
443: fraction of the total energy output, underlining the rapid evolution
444: which must occur in the ULIRG population (Smail et al.\ 1997).
445: Identifying the physical process responsible for this rapid increase
446: in star formation, as well as the reason for the much larger physical
447: scale of high-redshift starbursts compared to local examples, will
448: yield important insights into the formation of massive galaxies.
449:
450: %
451: %
452: %
453: \acknowledgments
454:
455: We thank Leo Blitz, Harald Ebeling, Frazer Owen, Nick Scoville, Graham
456: Smith, Jack Welch and Mel Wright for help and advice. IRS and JPK
457: acknowledge support from the Royal Society, the Leverhulme Trust and
458: CNRS.
459:
460: %
461: %
462: %
463: \begin{thebibliography}{}
464:
465: \bibitem[]{} Adelberger, K.\ L., \& Steidel, C.\ C.\ 2000, ApJ, 544,
466: 218 {}{}
467: \bibitem[]{} Almaini, O., et al.\ 2001, MNRAS, submitted (astro-ph/0108400){}{}
468: \bibitem[]{} Barger, A.\ J., Cowie, L.\ L., Steffen, A.\ T., Hornschemeier,
469: A.\ E., Brandt, W.\ N., \& Garmire, G.\ P.\ 2001, ApJ, submitted
470: (astro-ph/0107252){}{}
471: \bibitem[]{} Bell, E.\ F., Gordon, K.\ D., Kennicutt, R.\ C.\ \&
472: Zaritsky, D.\ 2001, ApJ, submitted (astro-ph/0108367){}{}
473: \bibitem[]{} Brandt, W.\ N., et al.\ 2001, AJ, 122, 1 {}{}
474: \bibitem[]{} Blain, A.\ W., Kneib, J.-P., Ivison, R.\ J., \& Smail, I.\
475: 1999, ApJ, 512, L87{}{}
476: \bibitem[]{} Blain, A.\ W., Frayer, D.\ T., Bock, J.\ J., \& Scoville, N.\ Z.\
477: 2000, MNRAS, 313, 559{}{}
478: \bibitem[]{} Chapman, S., et al.\ 2000, MNRAS, 319, 318 {}{}
479: \bibitem[]{} Fabian, A.\ C., et al.\ 2000, MNRAS, 315, L8{}{}
480: \bibitem[]{} Frayer, D.\ T., et al.\ 1999, ApJ, 514, L13{}{}
481: \bibitem[]{} Hauser, M.\ G., et al.\ 1998, ApJ, 508, 25{}{}
482: \bibitem[]{} Holtzman, J.\ A., et al.\ 1995, PASP, 107, 156{}{}
483: \bibitem[]{} Ivison, R.\ J., Smail, I., Le Borgne, J.-F., Blain, A.\ W., Kneib,
484: J.-P., B\'ezecourt, J., Kerr, T.\ H., \& Davies, J.\ K.\ 1998,
485: MNRAS, 298, 583{}{}
486: \bibitem[]{} Ivison, R.\ J., Dunlop, J.\ S., Smail, I., Dey, A., Liu,
487: M.\ C., \& Graham, J.\ R.\ 2000b, ApJ, 542, 27{}{}
488: \bibitem[]{} Ivison, R.\ J., Smail, I., Barger, A., Kneib, J.-P., Blain,
489: A.\ W., Owen, F.\ N., Kerr, T.\ H., \& Cowie, L.\ L.\ 2000a,
490: MNRAS, 315, 209 [I00]{}{}
491: \bibitem[]{} Lutz, D., et al.\ 2001, A\&A, in press (astro-ph/0108131){}{}
492: \bibitem[]{} Madau, P., \& Pozzetti, L.\ 2000, MNRAS, 312, 9 {}{}
493: \bibitem[]{} Meurer, G.\ R., Heckman, T.\ M., Seibert, M., Goldader, J.\ D.,
494: Calzetti, D., Sanders, D., \& Steidel, C.\ C.\ 2001, pre-print
495: (astro-ph/0011201){}{}
496: \bibitem[]{} Papadopoulos, P., Ivison, R., Carilli, C., \& Lewis G.\ 2001,
497: Nature, 409, 58{}{}
498: \bibitem[]{} Peacock, J.\ A., et al.\ 2001, MNRAS, 318, 535 {}{}
499: \bibitem[]{} Richards, E.\ 2000, ApJ, 533, 611{}{}
500: \bibitem[]{} Smail, I., Ivison, R.\ J., \& Blain, A.\ W.\ 1997, ApJ,
501: 490, L5{}{}
502: \bibitem[]{} Smail, I., Ivison, R.J., Kneib, J.-P., Cowie, L.L.,
503: Blain, A.W., Barger, A.J., Owen, F.N.\ \& Morrison, G.\
504: 1999. MNRAS, 308, 1061{}{}
505: \bibitem[]{} Smail, I., Ivison, R.\ J., Blain, A.\ W., \& Kneib, J.-P.\
506: 2001, MNRAS, submitted {}{}
507: \bibitem[]{} Smith, G.\ P., et al., 2001, MNRAS, submitted {}{}
508: \bibitem[]{} Steidel, C.\ C., Pettini, M., \& Hamilton, D.\
509: 1995, AJ, 110, 2519{}{}
510: \bibitem[]{} Townsend, R.\ H.\ D., Ivison, R.\ J., Smail, I., Frayer, D.\
511: T., \& Blain, A.\ W., 2001, MNRAS, submitted
512: (astro-ph/0106112){}{}
513: \bibitem[]{} van der Werf, P., Knudsen, K.\ K., Labb\'e, I., \& Franx, M.\
514: 2001, pre-print (astro-ph/0010459){}{}
515:
516: \end{thebibliography}
517:
518: \end{document}
519: