1: \documentstyle[emulateapj,psfig,apjfonts]{article}
2: \submitted{Received 2003 -- --; Accepted: 2003 -- --}
3:
4: \def\gs{\mathrel{\raise0.35ex\hbox{$\scriptstyle >$}\kern-0.6em
5: \lower0.40ex\hbox{{$\scriptstyle \sim$}}}}
6: \def\ls{\mathrel{\raise0.35ex\hbox{$\scriptstyle <$}\kern-0.6em
7: \lower0.40ex\hbox{{$\scriptstyle \sim$}}}}
8:
9: \newenvironment{inlinefigure}{%
10: \def\@captype{figure}%
11: \noindent\begin{minipage}{0.999\linewidth}\begin{center}\small}
12: {\end{center}\end{minipage}\smallskip}
13: \makeatother
14:
15: \lefthead{Smail et al.}
16: \righthead{{\it Chandra}/SCUBA sources around HzRGs}
17:
18: \begin{document}
19:
20: \title{{\it Chandra} Detections of SCUBA Galaxies around High-$z$ Radio Sources}
21:
22: \author{
23: Ian Smail,\altaffilmark{1} C.\,A.\ Scharf,\altaffilmark{2}
24: R.\,J.\ Ivison,\altaffilmark{3} J.\,A.\ Stevens,\altaffilmark{3}
25: R.\,G.\ Bower\altaffilmark{1} \& J.\,S.\ Dunlop\altaffilmark{4}}
26:
27: \altaffiltext{1}{Institute for Computational Cosmology, University of Durham, South Road,
28: Durham DH1 3LE UK}
29: \altaffiltext{2}{Columbia Astrophysics Laboratory, Columbia University, MC5247, 550
30: West 120th St., New York, NY 10027, USA}
31: \altaffiltext{3}{Astronomy Technology Centre, Royal Observatory,
32: Blackford Hill, Edinburgh EH9 3HJ UK}
33: \altaffiltext{4}{Institute for Astronomy, University of Edinburgh,
34: Blackford Hill, Edinburgh EH9 3HJ UK}
35:
36: \setcounter{footnote}{4}
37:
38: \begin{abstract}
39: The most massive galaxies in the present day universe are the giant
40: ellipticals found in the centers of rich clusters. These have old,
41: coeval stellar populations, suggesting they formed at high redshift,
42: and are expected to host supermassive black holes (SMBHs). The recent
43: detection of several high-redshift radio galaxies (HzRGs) at
44: submillimeter (submm) wavelengths confirms that indeed some massive
45: galaxies may have formed the bulk of their stellar populations in
46: spectacular dust-enshrouded starbursts at high redshift. In this letter
47: we compare sensitive {\it Chandra} X-ray images -- which identify
48: actively-fueled SMBHs -- and submm observations -- capable of detecting
49: obscured activity in luminous galaxies at high redshift -- of the
50: environments of three HzRGs. These observations exhibit overdensities of
51: X-ray sources in all three fields and a close correspondence between
52: the {\it Chandra} and SCUBA populations. This suggests that both
53: substantial star formation and nuclear activity may be occuring in
54: these regions. We identify possible pairs of {\it Chandra} sources
55: with each of two SCUBA sources, suggesting that their ultraluminous
56: activity may be triggered by the interaction of two {\it massive}
57: galaxies, each of which hosts an accreting SMBH. The presence of two
58: SMBHs in the resulting remanent is predicted to produce a flattened
59: stellar core in the galaxy, a morphological signature frequently seen
60: in luminous cluster ellipticals. Hence the confirmation of pairs of
61: {\it Chandra} sources within individual, luminous SCUBA galaxies would
62: provide additional evidence that these galaxies at $z\sim 2$--4 are the
63: progenitors of the giant elliptical galaxies found in clusters at the
64: present-day.
65: \end{abstract}
66:
67: \keywords{cosmology: observations --- galaxies: individual (SMM\,J06509+4130,
68: SMM\,J06509+4131, SMM\,J11409$-$2629, SMM\,J17142+5016) ---
69: galaxies: evolution --- galaxies: formation}
70:
71: %
72: %
73: %
74: \section{Introduction}
75:
76: %
77: % FIGURE 1
78: %
79: \begin{figure*}[tbh]
80: \centerline{\psfig{file=f1.ps,width=6.0in,angle=270}}
81: \caption{\small The 0.5--2\,keV {\it Chandra} X-ray images of the
82: regions around 4C\,41.17 (Scharf et al.\ 2003), MRC1138$-$262 (P02)
83: and 53W002 (White et al.\ 2003), each overlayed as a contour map on a
84: grayscale representation of the relevant 850$\mu$m SCUBA map from
85: Stevens et al.\ (2003). We mark with \fbox{} the positions of the
86: hard-band sources detected in these fields. Notice the X-ray
87: counterparts to four of the SCUBA sources in these field (numbered as in
88: Table~1), as well as the X-ray emission from the radio galaxies at the
89: center of each field.}
90: \end{figure*}
91:
92: Studies of luminous ellipticals in high-density regions out to $z\sim
93: 1$ suggest that their stellar populations are formed via a short burst
94: of star formation at high-$z$ (probably $z>2$), with little additional
95: star formation at $z<1$ (Bower, Lucey \& Ellis 1992; Ellis et al.\
96: 1997; Stanford et al.\ 1998; Poggianti et al.\ 2001; Kelson et al.\
97: 2001). Further support for this conclusion comes from the apparent
98: lack of evolution in the metallicity of the intracluster medium out to
99: $z\sim 1$ (Tozzi et al.\ 2003). The short dynamical times expected in
100: the densest regions at high-$z$ coupled with the intense star
101: formation necessary to form the stellar population of a luminous
102: elliptical suggest that this formation phase may resemble a classical
103: ``monolithic'' collapse -- even within a hierarchical cosmogony
104: (Kauffmann 1996; Baugh et al.\ 1998). A clear test of this paradigm
105: is to {\it directly} observe the epoch of elliptical formation in
106: high-$z$ protoclusters. There are expected to be two distinct
107: observational signatures of this formation phase.
108:
109: Firstly, the intensity of the star formation event and the high
110: metallicity of the stars in ellipticals today (as well as the
111: intracluster medium) all argue that this activity will
112: produce a large amount of dust. This will obscure the later
113: phases of the starburst and the most active systems, those with
114: star formation rates of $>10^2$--$10^3M_\odot$\,yr$^{-1}$, would
115: be classed as ultraluminous infrared galaxies (ULIRGs) with
116: far-infrared luminosities of $>10^{12}L_\odot$. The sensitivity of
117: submillimeter (submm) cameras such as SCUBA on the JCMT is sufficient
118: to detect these systems out to $z\gg 1$ (Chapman et al.\ 2003).
119:
120: The second observational signature derives from the apparently close
121: correlation seen locally between the mass of central black holes and
122: the mass of the spheroid which hosts them (Kormendy \& Richstone 1995;
123: Magorrian et al.\ 1998; Gebhardt et al.\ 2000; Ferrarese \& Merritt
124: 2000), which in turn suggests a tight linkage between the
125: formation of these components. As the most massive spheroids, luminous
126: cluster ellipticals are expected to host supermassive black holes
127: (SMBH), $>10^9$M$_\odot$. Thus many young ellipticals may also exhibit
128: signatures of nuclear activity as their SMBH will be actively fueled
129: and growing during the dusty starburst phase, and should be detectable in
130: deep X-ray imaging.
131:
132: To investigate the formation of the elliptical galaxy population of
133: rich clusters using these observational tools we need to
134: identify the likely site of their formation: proto-clusters at $z>2$.
135: In standard cosmological models, massive galaxies at $z>2$ are
136: expected to strongly cluster in the deepest potential wells, onto which other
137: galaxies later accrete, assembling the rich clusters we see at $z\sim
138: 0$. The strongly clustered population of luminous high-$z$ radio
139: galaxies (HzRG) are believed to host massive black holes and thus may
140: represent some of the most massive galaxies at these epochs (e.g.\ de
141: Breuck et al.\ 2000) which in turn could trace some of the highest
142: density environments. The association of luminous radio galaxies with
143: overdense structures at high-$z$ is supported by the identification of
144: modest excesses of companion galaxies around HzRGs in the
145: optical/near-infrared (e.g.\ R\"ottgering et al.\ 1996), X-ray
146: (Pentericci et al.\ 2002, P02; Fig.~1) and submm wavebands (Ivison et
147: al.\ 2000, I00; Stevens et al.\ 2003; Fig.~1).
148:
149: We have brought together these two observational tools: submm and X-ray
150: imaging, to search for the formation phase of luminous elliptical
151: galaxies in putative high density environments around HzRGs. We have
152: identified X-ray counterparts to the SCUBA sources in the fields of
153: three HzRGs. We discuss the properties of these sources and the
154: information they provide on the formation of SMBHs and the early stages
155: of the growth of massive cluster elliptical galaxies. We assume a
156: cosmology with $\Omega_m=0.27$, $\Omega_\Lambda=0.73$ and
157: $H_o=71$\,km\,s$^{-1}$\,Mpc$^{-1}$.
158:
159: %
160: %
161: %
162: \section{Analysis and Results}
163:
164: A search of the {\it Chandra} archive shows that three fields from the
165: SCUBA survey of HzRG by Stevens et al.\ (2003) have moderate or deep
166: X-ray imaging observations. These are 4C\,41.17 ($z=3.79$),
167: MRC\,1138$-$262 ($z=2.16$) and 53W002 ($z=2.39$), with useable
168: integration times of 135.1\,ks, 32.5\,ks and 32.8\,ks respectively.
169: The {\it Chandra} observations of 4C\,41.17 are discussed in Scharf et
170: al.\ (2003), while those for MRC\,1138$-$262 are detailed in P02 and
171: Carilli et al.\ (2002), and 53W002 is presented in White et al.\
172: (2003). We refer the reader to these papers for more details of the
173: observations and their reduction. Sources have been detected and
174: measured in these images using the {\sc ciao} software {\sc wavdetect}
175: algorithm with default settings. Hardness ratios (HR) are determined
176: as $(H-S)/(H+S)$ for a 0.5--2 keV soft band $(S)$, and 2--10 keV hard
177: band $(H)$. To convert count rates to fluxes we assume a Galactic
178: $\log N(\rm H{\sc i})=21.0$ for all sources and a power law spectrum
179: with photon index $\Gamma=1.5$ and intrinsic $\log N(\rm H{\sc
180: i})=23.0$ for HR$>-0.5$ and $\Gamma=2.2$ with no intrinsic absorption
181: for HR$<-0.5$ (c.f.\ Tozzi et al.\ 2001). Luminosities are calculated
182: in the 0.3--10 keV rest frame using {\sc xspec}, errors assume
183: Poissonian count rate uncertainties.
184:
185: In all three fields excesses of soft-band sources are seen: P02 show
186: that the MRC\,1138$-$262 field has a modest excess of soft-band
187: sources amounting to a factor of 50\% (in an $8'\times 8'$ field),
188: while White et al.\ (2003) estimate an overdensity of 2--3$\times$ in
189: soft-band sources around 53W002. In the 4C\,41.17 field we find 11
190: soft-band sources brighter than $S_{0.5-2}=0.2\times
191: 10^{-15}$\,erg\,s$^{-1}$\,cm$^{-2}$ within the SCUBA field (Fig.~1,
192: 2.1$'$, 1\,Mpc diameter), compared to a blank field expectation of
193: $\sim 2$ (Rosati et al.\ 2002). Thus we have a factor of 5--6$\times$
194: overdensity of soft-band sources in the vicinity of 4C\,41.17 (there
195: is also a $2\times$ overdensity of hard-band sources in this field).
196:
197: As described by P02, the X-ray sources in MRC\,1138$-$262 also display
198: a strongly filamentary pattern around the radio galaxy suggestive of
199: associated large scale structure. A similar, but weaker filamentary
200: structure may extend to the south of 4C\,41.17 (Fig.~1). The
201: existance of overdense structures around the HzRGs is supported by
202: follow-up of the X-ray sources in the MRC\,1138$-$262 field where at
203: least 2, and possibly up to 5, sources lie at the same redshift as the
204: HzRG (P02). Similarly, for 53W002, White et al.\ (2003) identify at
205: least 3 X-ray bright AGN at the same redshift as the HzRG. On-going
206: spectroscopy of the 4C\,41.17 field has so far confirmed one {\it
207: Chandra} companion at $z=3.8$.
208:
209: %
210: % Table 1
211: %
212: \begin{table*}
213: {\scriptsize
214: \centerline{\sc Catalog of SCUBA/{\it Chandra} Sources}
215: \begin{center}
216: \begin{tabular}{lcccccccccl}
217: \hline\hline
218: \noalign{\smallskip}
219: Source & R.A.\ & Dec.\ & S$_{850}$ & S$_{0.5-2}$ & S$_{2-10}$ & HR & $L_X(0.3-10)$ & $\alpha_{SX}^a$ & P$^b$ & Comment \cr
220: & (J2000) & (J2000) & (mJy) & \multispan2{ ~~(10$^{-15}$\,erg\,s$^{-1}$\,cm$^{-2}$) } & & (10$^{43}$\,erg\,s$^{-1}$) & & & \cr
221: \hline
222: \noalign{\smallskip}
223: 1)~ {\bf SMM\,J06509+4130} & 06 50 51.2 & +41 30 05 & 15.6$\pm$1.8 & ... & ... & ... & ... & 1.18 & ... & 4C\,41.17 HzRG850.1, I00 \cr
224: ~~CXJ\,065051.3+412958 & 06 50 51.30 & +41 29 58.1 & ... & 0.26$\pm$0.09 & 0.93$\pm$0.40 & $-0.3$ & 3.2$\pm$1.0 & 1.28 & 0.004 & HzRG850.1/K3, I00 \cr %R>26
225: ~~CXJ\,065052.1+412951 & 06 50 52.09 & +41 29 51.2 & ... & 0.38$\pm$1.0 & 1.9$\pm$0.6 & $-0.1$ & 5.6$\pm$1.2 & 1.24 & 0.047 & \cr %R>26
226: ~~CXJ\,065052.2+413004 & 06 50 52.19 & +41 30 04.5 & ... & 0.85$\pm$1.6 & 0.44$\pm$0.25 & $-0.8$ &13.0$\pm$2.0 & 1.25 & 0.221 & LR1, Lacy \& Rawlings (1996) \cr %R=18.3
227: \noalign{\smallskip}
228: 2)~ {\bf SMM\,J06509+4131} & 06 50 49.3 & +41 31 27 & 8.7$\pm$1.2 & ... & ... & ... & ... &1.18 & ... & 4C\,41.17 HzRG850.2, I00 \cr
229: ~~CXJ\,065049.1+413127 & 06 50 49.13 & +41 31 26.5 & ... & 0.43$\pm$0.11 & 3.3$\pm$0.8 & $+0.1$ & 8.0$\pm$1.2 & 1.18 & 0.016 & HzRG850.2/K1, I00 \cr %R=24.0
230: \noalign{\smallskip}
231: 3)~ {\bf SMM\,J11409$-$2629} & 11 40 53.3 & $-$26 29 11 & 7.4$\pm$1.5 & ... & ... & ... & ... & 1.07 & ... & MRC\,1138$-$262, Stevens et al.\ (2003)\cr
232: ~~CXJ\,1114052.8$-$262911 & 11 40 52.84 & $-$26 29 11.2 & ... & 5.1$\pm$0.7 & 5.2$\pm$1.7 & $-0.7$ & 4.2$\pm$0.4 & 1.07 & 0.009 & \#9 ($z\sim 2.16$), P02 \cr %I=21.7
233: \noalign{\smallskip}
234: 4)~ {\bf SMM\,J17142+5016} & 17 14 12.1 & +50 16 02 & 5.6$\pm$0.9 & ... & ... & ... & ... & 1.14 & ... & 53W002 \#18 ($z=2.39$), S03a\cr
235: ~~CXJ\,171412.0+501602 & 17 14 11.97 & +50 16 02.3 & ... & 0.48$\pm$0.24 & 3.1$\pm$1.6 & $~0.0$ & 4.6$\pm$1.5 & 1.14 & 0.002 & \cr %I=22.5
236: \hline
237: \end{tabular}
238: \end{center}
239: \addtolength{\baselineskip}{-1.0mm}
240: $a$) The submm--X-ray spectral index (Fabian et al.\ 2000). The values for the SCUBA galaxies assume all the X-ray sources contribute to the submm emission, while those quoted for
241: the individual {\it Chandra} sources assume all the submm emission
242: arises just from that source.\\
243: $b$) Likelihood of detecting the observed submm flux at the X-ray position by random chance.
244:
245: }
246: \end{table*}
247:
248: Submm maps with noise levels of $\sim 1.0$--1.5\,mJy are available of
249: these three fields from the 850$\mu$m SCUBA survey of Stevens et al.\
250: (2003), with the first results on 4C\,41.17 reported in I00. These
251: cleaned maps also show modest overdensities of submm sources compared
252: to blank fields, with at least one example, SMM\,J17142+5016 in the
253: 53W002 field, spectroscopically confirmed at the same redshift as the
254: radio galaxy (Smail et al.\ 2003a, S03a). There are 3 $>4$-$\sigma$
255: submm sources around 4C\,41.17 (I00), 3 around MRC\,1137$-$262 (Stevens
256: et al.\ 2003) and 4 around 53W002 (S03a). Using the flux limits for
257: the three fields and the compilation of SCUBA counts in Smail et al.\
258: (2002), we estimate that the HzRG fields show $2\times$
259: overdensities of SCUBA galaxies compared to the field, with
260: roughly 4--5 of the sources likely to be unrelated field galaxies.
261:
262: To compare the X-ray and submm emission in these fields we align the
263: two maps. The astrometric precision of the 850$\mu$m maps is $\pm
264: 3''$ rms and small shifts of less than this magnitude have been
265: applied to align the submm emission with the radio centroids for the
266: luminous HzRGs in these fields. The X-ray images are similarly tied
267: to the radio frame with an absolute precision of 0.5$''$ rms using the
268: positions of several X-ray and radio bright AGN in each field.
269:
270: Due to the extended nature of the submm emission in many of the SCUBA
271: sources in our fields (I00; Stevens et al.\ 2003) we choose to test
272: the association of {\it Chandra} sources with the SCUBA galaxies by
273: searching for statistical excesses of submm emission around the
274: precisely-located X-ray source. To calculate these fluxes we use a
275: 3$''$-radius aperture which represents the relative uncertainty of the
276: astrometry between the {\it Chandra} and SCUBA maps. We measure the
277: flux in $\sim 10^3$ randomly-placed apertures across the SCUBA map and
278: calculate the fraction of these which exceed that measured at the
279: positions of the {\it Chandra} sources. This procedure provides a
280: likelihood for each association and allows us to identify
281: statistically significant ($>$98\% c.l.) associations between X-ray
282: sources and 4 out of the 10 SCUBA galaxies in these fields (we
283: ignore the central HzRG in each field each of which are detected in
284: both the X-ray and submm wavebands, Fig.~1). It should be
285: noted that these associations may result either from a direct correspondence
286: between the sources in the two wavebands, or because both classes
287: populate the same structures around the HzRGs. We list in Table~1 the
288: details of these associations and briefly discuss the individual cases
289: below.
290:
291: \noindent{\it 1) SMM\,J06509+4130} -- this is the very bright
292: ($L_{bol}\sim 10^{13}L_\odot$), southern SCUBA source in Fig.~1a
293: (HzRG850.1 from I00). This resolved submm source (70\,kpc diameter)
294: coincides with the soft-band {\it Chandra} source
295: CXJ\,065051.3+412958, which is identified as the $K=19.4$ very red
296: object HzRG850.1/K3 from I00. A second soft-band {\it Chandra} source
297: to the South-East, CXJ\,065052.1+412951, is also associated with the
298: submm emission ($<$5\% chance of this being random coincidence). This
299: source has no obvious optical or near-IR counterparts in the imaging
300: discussed by I00 indicating $K>20.0$ and $R>26.0$. A third, bright
301: soft-band {\it Chandra} source lies just outside the region of submm
302: emission to the East (Fig.~1a). This source is coincident with a
303: red stellar object (LR1 from Lacy \& Rawlings 1996) with $UVR$ colors
304: consistent with either a foreground star or a $z\sim 3.8$ AGN (there
305: is a 6\% chance of an unrelated X-ray source falling this close to the
306: SCUBA position). Finally, a hard-band source with a faint $K$-band
307: counterpart is detected 18$''$ to the South-East, although this is
308: unlikely to be directly related to the SCUBA galaxy.
309:
310: \noindent{\it 2) SMM\,J06509+4131} -- this is the second, bright and
311: spatially-extended, SCUBA galaxy (HzRG850.2 in I00) in
312: the 4C\,41.17 field (Fig.~1a) and is associated with the {\it Chandra} source
313: CXJ\,065049.1+413127. The X-ray source is detected in both the hard
314: and soft-bands and is coincident with the K1 counterpart of
315: HzRG850.2 from I00. K1 is one of a pair of faint, very red
316: galaxies separated by $\sim 2''$.
317:
318: \noindent{\it 3) SMM\,J11409$-$2629} -- this is the brightest submm
319: source in the vicinity of the $z=2.16$ radio galaxy MRC1138$-$262
320: mapped by Stevens et al.\ (2003; Fig.~1b). The {\it Chandra} source
321: CXJ\,1114052.8$-$262911 (\#9 from P02) is statistically associated
322: with the SCUBA source (Table~1). P02 report that \#9 exhibits an Ly$\alpha$
323: emission-line excess, indicating that it probably lies
324: at $z\sim 2.16$, which would place this SCUBA galaxy in the striking
325: filament seen across this field.
326:
327: \noindent{\it 4) SMM\,J17142+5016} -- this relatively faint SCUBA
328: galaxy has been identified by S03a with a weak AGN, \#18 from the catalog of
329: Keel et al.\ (2002). The {\it Chandra}
330: source coincides exactly with the optical/near-IR counterpart and
331: hence we can unequivocally associate both the X-ray and luminous submm
332: emission with this galaxy which is spectroscopically-confirmed as a
333: companion to the radio galaxy at $z=2.39$. A much brighter X-ray
334: source is detected 9$''$ (75\,kpc) to the North-West, coincident with
335: the broad-line $z=2.39$ AGN \#19 (Keel et al.\ 2002), although this
336: source is undetected in the SCUBA map (Fig.~1c). There is a 2\% chance
337: that this source would fall within 9$''$ of the SCUBA galaxy by
338: random chance.
339:
340: In summary, we detect X-ray counterparts to four luminous submm sources
341: in the fields of three $z=2.2$--3.8 HzRGs. Moreover, one of these
342: X-ray sources has the same redshift as the radio galaxy in its vicinity
343: and a second is also likely to be at the same redshift as its
344: neighboring HzRG. These identifications confirm the presence of
345: luminous dust-obscured galaxies containing actively-fueled SMBHs in the
346: environments of radio galaxies at $z>2$. We give their hardness ratios
347: and submm/X-ray spectral indices ($\alpha_{SX}$) in Table~1 and also
348: estimate their X-ray luminosities assuming they lie at the same
349: redshifts as the relevant HzRG. For all four sources these X-ray
350: properties are consistent with obscured, Seyfert-like systems, $L_X\sim
351: 10^{43}$--$10^{44}$\,ergs\,s$^{-1}$ with absorbing columns of $N(\rm
352: H{\sc i})\gs 10^{23}$\,cm$^{-2}$ (Alexander et al.\ 2003a; Fabian et
353: al.\ 2000). We obtain a lower limit on the black hole masses in these
354: systems of $\gs 10^{7}$M$_\odot$ by assuming that they are accreting at
355: their Eddington limit, or $\sim 10^{8}$M$_\odot$ if their SMBHs are
356: accreting at a similar rate to local Seyferts.
357:
358:
359: %
360: %
361: %
362: \section{Discussion and Conclusions}
363:
364: Our identifications of {\it Chandra} counterparts to the SCUBA galaxies
365: in these fields are statistical in nature and hence the possibility
366: exists that the two populations are simply tracing the same large-scale
367: structures, rather than being {\it directly} related. However, the
368: relatively small spatial scale of the associations,$\ls 100$\,kpc,
369: suggests that there is a direct relationship and so the following
370: discussion proceeds on that assumption.
371:
372: The rate of detection of SCUBA sources in our {\it Chandra} maps is
373: $\sim 40$\%, this is somewhat higher than the $\sim 5$--10\% detection
374: rate reported from comparable-depth {\it Chandra} and {\it XMM-Newton}
375: observations (Barger et al.\ 2001; Ivison et al.\ 2002; Almaini et al.\
376: 2003; Waskett et al.\ 2003) and is similar to that achieved in the
377: far-deeper, 2-Ms {\it Chandra} integration discussed by Alexander et
378: al.\ (2003), who detect four out of ten $\geq 5$\,mJy SCUBA sources
379: above a flux limit of $S_{0.5-8}\geq 1\times
380: 10^{-15}$\,erg\,s$^{-1}$\,cm$^{-2}$. Our high detection rate could
381: either reflect enhanced X-ray emission from SCUBA galaxies in these
382: environments, or may instead simply point to an increase in the
383: detection rate of $\gs 5$\,mJy SCUBA sources just below the flux limits
384: of the shallower surveys, e.g.\ $S_{0.5-8}\sim 3\times
385: 10^{-15}$\,erg\,s$^{-1}$\,cm$^{-2}$ for the Almaini et al.\ (2003)
386: study (Manners et al.\ 2003).
387:
388: We also identify a significant overdensity of {\it Chandra} sources
389: within a 1-Mpc region around the radio source in the 4C\,41.17 field,
390: in both the soft and hard-bands. Similar overdensities have been
391: previously reported around MRC\,1138$-$262 (P02) and 53W002 (White et
392: al.\ 2003), suggesting that the environment of HzRGs at $z>2$ are also
393: fertile ground for triggering AGN activity. This activity appears to
394: be linked to the excess of luminous submm sources found in these regions,
395: as can be seen if we measure the mean submm flux of {\it all} {\it
396: Chandra} sources within our SCUBA maps. We estimate a mean flux of
397: $2.6\pm 0.7$\,mJy from the 15 soft-band sources (excluding the 3 radio
398: galaxies). This flux is roughly twice that of blank-field {\it Chandra}
399: sources at similar X-ray flux limits (Almaini et al.\ 2003; Barger et
400: al.\ 2001; Waskett et al.\ 2003), although the result is only
401: significant at the $\sim 2$-$\sigma$ level due to the small sample
402: size.
403:
404: Two of the SCUBA sources detected by {\it Chandra} are particularly
405: noteworthy. The first is SMM\,J06509+4130 (HzRG850.1 from I00), a very
406: luminous and spatially-extended submm source which is apparently
407: associated with two X-ray sources (separated by 11$''$ or 80\,kpc if at
408: $z=3.8$), with two other X-ray sources in close proximity. A second
409: X-ray detected SCUBA source is SMM\,J17142+5016, which also has an
410: X-ray bright companion, 53W002\#19, at a projected distance of 80\,kpc
411: and a velocity offset of only 400\,km\,s$^{-1}$ (Keel et al.\ 2002).
412: Two similar pairs of {\it Chandra} sources associated with SCUBA
413: galaxies, with separations of 2--3$''$, have been reported by Alexander
414: et al.\ (2003a) from a sample of 10 SCUBA galaxies in the CDF-N. Thus
415: from a combined sample of 21 SCUBA sources we have 3 (14\%) with pairs
416: of {\it Chandra} counterparts brighter than $S_{0.5-8}\sim
417: 10^{-15}$\,erg\,s$^{-1}$\,cm$^{-2}$, this is $7\times$ higher than the
418: occurence of similar separation pairs in the general {\it Chandra}
419: population at this flux limit (9/460 or 2\%, Alexander et al.\ 2003b).
420:
421: These SCUBA galaxies with pairs of {\it Chandra} counterparts are
422: remarkable systems -- suggesting that the ultraluminous SCUBA events
423: are triggered by the interaction between multiple(?) massive galaxies.
424: The activity in these paired {\it Chandra} sources implies that they
425: are currently interacting and their separations then suggests that the
426: pericentric passage probably occured within the last $\sim 100$\,Myrs,
427: consistent with the expected life time of the starbursts in SCUBA
428: galaxies (Smail et al.\ 2003b). Moreover, the presence of SMBHs in the
429: progenitors means they have either been formed on the timescale of the
430: current interaction, or these galaxies had sufficient prior activity to
431: build a SMBH (and by implication a massive spheroid), yet still retain
432: significant quantities of gas needed to power the current, obscured
433: starburst.
434:
435: The final merger of these pairs of galaxies, probably within the next
436: 100--200\,Myrs or by $z\sim 3.4$ for the 4C\,41.17 pair, will produce
437: a remanent which hosts two SMBHs. The interaction of the two SMBHs is
438: predicted to produce a flattening of the stellar distribution in the
439: core of the remanent (Ravindranath, Ho \& Filippenko 2002). This is a
440: structural feature common to bright elliptical galaxies, and also one
441: which is preferentially found in ellipticals in clusters (Laine et
442: al.\ 2003; Quillen, Bower \& Stritzinger 2000). Hence the
443: identification of multiple {\it Chandra} sources within a modest
444: fraction of the SCUBA population is consistent with the expectation
445: that these galaxies will evolve into luminous ellipticals at the
446: present-day and that the environments we find them in at high-$z$ are
447: regions which will go on to become the cores of rich clusters.
448: However, the relatively modest mass estimates we derive for these
449: SMBH, $\gs 10^7$--$10^8$M$_\odot$, compared to the $\sim
450: 10^9$M$_\odot$ SMBHs found in bright ellipticals today gives scope for
451: subsequent growth of the SMBHs in these systems during the final
452: stages of the merger.
453:
454: The success rate for identifying {\it Chandra} counterparts to SCUBA galaxies
455: in our X-ray images suggests that X-ray observations may
456: provide a useful route to localise the submm-emitting galaxies in
457: these regions. This would normally be achieved through sensitive
458: radio observations (e.g.\ Ivison et al.\ 2002), but the presence of
459: the very bright radio source in these fields restricts the depth that
460: can be achieved due to dynamic-range limitations.
461:
462: We summarise the main results of this work. We study moderate depth
463: X-ray observations of luminous submm galaxies in the fields of three
464: luminous radio galaxies at $z=2.2$--3.8. We identify a significant
465: overdensity of X-ray sources around the most distant target, 4C\,41.17
466: at $z=3.8$. We also identify possible X-ray counterparts to 4 submm
467: galaxies in these fields (as well as the 3 submm-luminous central radio
468: galaxies). These associations either indicate a direct relationship
469: between the {\it Chandra} and SCUBA sources, or simply that both
470: populations are tracing the same large-scale structures. Spectroscopic
471: and narrow-band imaging suggests that two of these {\it Chandra}
472: sources lie at the same redshifts as their neighbouring HzRGs, hence
473: our identification of these objects as submm sources extends the number
474: of dusty, ultraluminous galaxies likely to reside in structures around
475: radio galaxies at $z>2$. We also highlight the properties of two
476: of these submm galaxies, one appears to be associated with at least
477: {\it two} {\it Chandra} sources, while the second has a {\it Chandra}
478: counterpart and also a bright X-ray companion in close proximity.
479: These results suggest that the ultraluminous activity in some fraction
480: of the SCUBA population may result from the interaction of pairs of
481: massive galaxies, each of which can be identified by the X-ray emission
482: from their central supermassive black holes. The presence of two SMBHs
483: in the resulting merger remanent is predicted to produce a clear
484: structural signature, a flattened core, in the stellar profile. Such
485: features are commonly observed in luminous elliptical galaxies in
486: clusters, suggesting a link between SCUBA galaxies at $z>2$--4 and the
487: population of luminous elliptical galaxies seen in clusters today.
488:
489:
490: %
491: %
492: %
493: \acknowledgments
494: We thank Dave Alexander, Omar Almaini, Wil van Breugel, Arjun Dey and
495: Bill Keel for help and useful conversations. We also
496: thank an anonymous referee for suggestions which improved the
497: presentation of this work. IRS acknowledges support
498: from the Royal Society and the Leverhulme Trust, JAS, RGB and JSD
499: acknowledge support from PPARC. This research was supported by
500: NASA/{\it Chandra} grant SAO G02-3267X. This work was made possible
501: by the exceptional capabilities of the NASA/{\it Chandra} observatory,
502: and the dedicated support of the CXC.
503:
504: %
505: %
506: %
507: \begin{thebibliography}{}
508:
509: \bibitem[]{} Alexander, D.M., Bauer, F.E., Brandt, W.N.,
510: Hornschemeier, A.E., Vignali, C., Garmire, G.P., Schneider, D.P.,
511: Chartas, G., Gallagher, S.C., 2003a, AJ, 125, 383
512:
513: \bibitem[]{} Alexander, D.M., Bauer, F.E., Brandt, W.N., Schneider, D.P.,
514: Hornschemeier, A.E., Vignali, C., Barger, A.J., Broos, P.S., et al., 2003b, AJ, in press
515:
516: \bibitem[]{} Almaini, O., Scott, S.E., Dunlop, J.S., Manners, J.C.,
517: Willott, C.J., Lawrence, A., Ivison, R.J., Johnson, O., et al., 2003,
518: MNRAS, 338, 303
519:
520: \bibitem[]{} Barger, A.J., Cowie, L.L., Steffen, A.T.,
521: Hornschemeier, A.E., Brandt, W.N., Garmire, G.P., 2001,
522: ApJ, 560, L23
523:
524: \bibitem[]{} Barger, A.J., Cowie, L.L., Mushotzky,
525: R.F., Richards, E.A., 2001, AJ, 121, 662
526:
527: \bibitem[]{} Baugh, C.M., Cole, S.M., Frenk, C.S., Lacey, C.G., 1998,
528: ApJ, 498, 504
529:
530: \bibitem[]{} Bower, R.G., Lucey, J.R., Ellis, R.S., 1992, MNRAS, 254,
531: 589
532:
533: \bibitem[]{} de Breuck, C., R\"ottgering, H., Miley, G., van Breugel,
534: W. Best, P., 2000, A\&A, 362, 519
535:
536: \bibitem[]{} Carilli, C., Harris, D.E., Pentericci, L., R\"ottgering,
537: H.J.A., Miley, G.K., Kurk, J.D., van Breugel, W., 2002, ApJ, 567, 781
538:
539: \bibitem[]{} Chapman, S.C., Blain, A.W., Ivison, R.J., Smail, I., 2003,
540: Nature, 422, 695
541:
542: \bibitem[]{} Ellis, R.S., Smail, I., Dressler, A., Couch, W.J.,
543: Oemler, A., Butcher, H., Sharples, R.M., 1997, ApJ, 483, 582
544:
545: \bibitem[]{} Fabian, A.C., Smail, I., Iwasawa, K., Allen, S.W., Blain,
546: A.W., Crawford, C.S., Ettori, S., Ivison, R.J., et al., 2000, MNRAS,
547: 315, L8
548:
549: \bibitem[]{} Ferrarese, L., Merritt, D., 2000, ApJ, 539, L9
550:
551: \bibitem[]{} Gebhardt, K., Bender, R., Bower, G., Dressler, A., Faber,
552: S.M., Filippenko, A.V., Green, R., Grillmair, C., et al., 2000, ApJ,
553: 539, L13
554:
555: \bibitem[]{} Ivison, R.J., Dunlop, J.S., Smail, I., Dey, A., Lui,
556: M.C., Graham, J.R., 2000, ApJ, 542, 27 [I00]
557:
558: \bibitem[]{} Ivison, R.J., Greve, T.R., Smail, I., Dunlop, J.S.,
559: Roche, N.D., Scott, S.E., et al., 2002, MNRAS, 337, 1
560:
561: \bibitem[]{} Kauffmann, G., 1996, MNRAS, 281, 487
562:
563: \bibitem[]{} Kauffmann, G., Haehnelt, M.G., 2000, MNRAS, 311, 578
564:
565: \bibitem[]{} Keel, W.C., Wu, W., Waddington, I., Windhorst, R.A.,
566: Pascarelle, S.M., 2002, AJ, 123, 3041
567:
568: \bibitem[]{} Kelson, D.,D., Illingworth, G.D., Franx, M., van Dokkum,
569: P.G., 2001, 552, L17
570:
571: \bibitem[]{} Kormendy, J., Richstone, D., 1995, AR\&AA, 33, 581
572:
573: \bibitem[]{} Lacy, M., Rawlings, S., 1996, MNRAS, 280, 888
574:
575: \bibitem[]{} Laine, S., van der Marel, R., Lauer, T.R., Postman, M.,
576: O'Dea, C.P., Owen, F.N., 2003, AJ,125, 478
577:
578: \bibitem[]{} Magorrian, J., Tremaine, S., Richstone, D., Bender, R.,
579: Bower, G., Dressler, A., Faber, S.M., Gebhardt, K., et al., 1998, AJ,
580: 115, 2285
581:
582: \bibitem[]{} Manners, J.C., Johnson, O., Almaini, O., Willott, C.J.,
583: Gonzalez-Solares, E., Lawrence, A., Mann, R.G., Perez-Fournon, I.,
584: et al., 2003, MNRAS, in press.
585:
586: \bibitem[]{} Pentericci, L., Kurk, J.D., Carilli, C.L., Harris, D.E.,
587: Miley, G.K., R\''ottgering, H.J.A., 2002, A\&A, 396, 109 [P02]
588:
589: \bibitem[]{} Pentericci, L., Kurk, J.D., R\''ottgering, H.J.A., Miley,
590: G.K., van Breugel, W., Carilli, C.L., Ford, H., Heckmna, T., et al.,
591: 2000, A\&A, 361, L25
592:
593: \bibitem[]{} Poggianti, B.M., Bridges, T.J., Mobasher, B., Carter, D.,
594: Doi, M., Iye, M., Kashikawa, N., Komiyama, Y., et al., 2001, ApJ, 563,
595: 118
596:
597: \bibitem[]{} Quillen, A.C., Bower, G.A., Stritzinger, M., 2000, ApJS, 128, 85
598:
599: \bibitem[]{} Ravindranath, S., Ho, L.C., Filippenko, A.V., 2002, ApJ, 566, 801
600:
601: \bibitem[]{} Rosati, P., Tozzi, P., Giacconi, R., Gilli, R.,
602: Hasinger, G., Kewley, L., Mainieri, V., Nonino, M., et al., 2002, ApJ, 566, 667
603:
604: \bibitem[]{} R\"ottgering, H.J.A., West, M.J., Miley, G.K., Chambers,
605: K.C., 1996, A\&A, 307, 376
606:
607: \bibitem[]{} Scharf, C., Smail, I., Ivison, R.J., Bower, R.G., van
608: Breugel, W., Reuland, M., 2003, ApJ, in press
609:
610: \bibitem[]{} Smail, I., Ivison, R.J., Blain, A.W.,
611: Kneib, J.-P., 2002, MNRAS, 331, 495
612:
613: \bibitem[]{} Smail, I., Ivison, R.J., Gilbank, D.G., Dunlop, J.S.,
614: Keel, W.C., Motohara, K., Stevens, J.A., 2003a, ApJ, 583, 551 [S03a]
615:
616: \bibitem[]{} Smail, I., Chapman, S.C., Ivison, R.J., Blain, A.W., Takata, T., Heckman,
617: T.M., Dunlop, J.S., Sekiguchi, K., 2003b, MNRAS, 342, 1185.
618:
619: \bibitem[]{} Stevens, J.A., Ivison, R.J., Dunlop, J.S., Smail, I.,
620: Percival, W.J., Hughes, D., Rottgering, H., van Breugel, W., et al.,
621: 2003, Nature, in press
622:
623: \bibitem[]{} Stanford, S.A., Eisenhardt, P.R., Dickinson, M.E., 1998,
624: ApJ, 492, 461
625:
626: \bibitem[]{} Tozzi, P., Rosati, P., Nonino. M., Bergeron, J.,
627: Borgani, S., Gilli, R., Gilmozzi, R., Hasinger, G., et al., 2001,
628: ApJ, 562, 42.
629:
630: \bibitem[]{} Tozzi, P., Rosati, P., Ettori, S.,
631: Borgani, S., Mainieri, V., Norman, C., 2003, ApJ, in press.
632:
633: \bibitem[]{} Veilleux, S., Kim, D.-C., Sanders, D.B., 2002, ApJS, 143,
634: 315
635:
636: \bibitem[]{} Waskett, T.J., Eales, S.A.,
637: Gear, W.K., Puchnarewicz, E.M., Lilly, S., Flores,
638: H., Webb, T., Clements, D., et al., 2003, MNRAS, 341, 1217
639:
640: \bibitem[]{} White, R., et al., 2003, in prep
641:
642: \end{thebibliography}
643:
644: \end{document}
645:
646:
647: