1: %\documentclass[12pt,preprint]{aastex}

2: %\documentclass{aastex}

3: %\usepackage{emulateapj}

4: \documentclass{emulateapj}

5: \usepackage{apjfonts}

6: \usepackage{natbib}

7: \usepackage{lscape}

8: 

9: \def\chandra{{\it Chandra\/}}

10: \def\rosat{{\it ROSAT\/}}

11: \def\asca{{\it ASCA\/}}

12: \def\lum{ergs~s$^{-1}$}

13: \def\flux{ergs~cm$^{-2}$~s$^{-1}$}

14: \def\etal{{et\,al.\,}}

15: \begin{document}

16: \slugcomment{Data and images available at http://www.astro.psu.edu/users/niel/cdfs/cdfs-chandra.html}

17: 

18: \title{The {\bf {\em Chandra}} Deep Field-South Survey: 2 Ms Source Catalogs}

19: 

20: \author{

21: B.~Luo,\altaffilmark{1} 

22: F.~E.~Bauer,\altaffilmark{2} 

23: W.~N.~Brandt,\altaffilmark{1}

24: D.~M.~Alexander,\altaffilmark{3}

25: B.~D.~Lehmer,\altaffilmark{3}

26: D.~P.~Schneider,\altaffilmark{1}

27: M.~Brusa,\altaffilmark{4,5} 

28: A.~Comastri,\altaffilmark{6}

29: A.~C.~Fabian,\altaffilmark{7} 

30: A.~Finoguenov,\altaffilmark{4,5} 

31: R.~Gilli,\altaffilmark{6} 

32: G.~Hasinger,\altaffilmark{4}

33: A.~E.~Hornschemeier,\altaffilmark{8} 

34: A.~Koekemoer,\altaffilmark{9} 

35: V.~Mainieri,\altaffilmark{10} 

36: M.~Paolillo,\altaffilmark{11}

37: P.~Rosati,\altaffilmark{10} 

38: O.~Shemmer,\altaffilmark{1} 

39: J.~D.~Silverman,\altaffilmark{12} 

40: I.~Smail,\altaffilmark{13} 

41: A.~T.~Steffen,\altaffilmark{14} 

42: \& C.~Vignali\altaffilmark{15}

43: }

44: \altaffiltext{1}{Department of Astronomy \& Astrophysics, 525 Davey Lab, 

45: The Pennsylvania State University, University Park, PA 16802, USA}

46: \altaffiltext{2}{Columbia Astrophysics Laboratory, Columbia University, 

47: Pupin Laboratories, 550 W. 120th St., New York, NY 10027, USA}

48: \altaffiltext{3}{Department of Physics, Durham University, 

49: Durham, DH1 3LE, UK}

50: \altaffiltext{4}{Max-Planck-Institut f\"ur Extraterrestrische Physik, 

51: Giessenbachstrasse, D-85748 Garching b. M\"unchen, Germany}

52: \altaffiltext{5}{University of Maryland, Baltimore County, 1000 

53: Hilltop Circle, Baltimore, MD 21250, USA}

54: \altaffiltext{6}{INAF---Osservatorio Astronomico di Bologna, Via Ranzani 1, 

55: Bologna, Italy}

56: \altaffiltext{7}{Institute of Astronomy, Madingley Road, Cambridge, 

57: CB3 0HA, UK}

58: \altaffiltext{8}{Laboratory for X-ray Astrophysics, NASA Goddard Space 

59: Flight Center, Code 662, Greenbelt, MD 20771, USA}

60: \altaffiltext{9}{Space Telescope Science Institute, 3700 San Martin Drive, 

61: Baltimore, MD 21218, USA}

62: \altaffiltext{10}{European Southern Observatory, Karl-Schwarzschild-Strasse 2, 

63: Garching, D-85748, Germany}

64: \altaffiltext{11}{Dipartimento de Scienze Fisiche, Universit\`a di Napoli, 

65: Via Cinthia, 80126 Napoli, Italy}

66: \altaffiltext{12}{ETH Zurich, Institute of Astronomy, Department of Physics, 

67: Wolfgang-Pauli-Strasse 16, 8093 Zurich, Switzerland}

68: \altaffiltext{13}{Institute of Computational Cosmology, Durham University, 

69: Durham, DH1 3LE, UK}

70: \altaffiltext{14}{Spitzer Science Center, California Institute of Technology, 

71: Mail Code 220-6, 1200 East California Boulevard, Pasadena, CA 91125, USA}

72: \altaffiltext{15}{Universit\'a di Bologna, Via Ranzani 1, Bologna, Italy}

73: 

74: \begin{abstract}

75: We present point-source catalogs for the $\approx2$~Ms exposure of 

76: the \chandra\ Deep Field-South (\hbox{CDF-S}); this is one of the 

77: two most-sensitive \hbox{X-ray} surveys ever performed. The survey covers an

78: area of $\approx436$ arcmin$^{2}$ and reaches on-axis sensitivity limits of 

79: $\approx1.9\times10^{-17}$ and $\approx1.3\times10^{-16}$ \flux\ 

80: for the \hbox{0.5--2.0} and 2--8~keV bands, respectively. Four hundred and sixty-two

81: \hbox{X-ray} point sources are detected in at least one of 

82: three \hbox{X-ray} bands that were searched; 

83: 135 of these sources are new compared to the previous $\approx1$~Ms

84: \hbox{CDF-S} detections. Source positions are determined using centroid

85: and matched-filter techniques; the median positional uncertainty is 

86: $\approx0\farcs36$. The X-ray--to--optical flux ratios of the newly detected

87: sources indicate a variety of source types; $\approx$55\% of them appear to

88: be active galactic nuclei while $\approx$45\% appear to be starburst 

89: and normal galaxies. In addition to the main

90: \chandra\ catalog, we provide a supplementary catalog of 86 \hbox{X-ray}

91: sources in the $\approx2$~Ms \hbox{CDF-S} footprint that was created 

92: by merging the $\approx250$~ks 

93: Extended \chandra\ Deep Field-South with the 

94: \hbox{CDF-S}; this approach provides 

95: additional sensitivity in the outer portions

96: of the \hbox{CDF-S}. A second supplementary catalog that contains 

97: 30 \hbox{X-ray} 

98: sources was constructed by matching lower significance \hbox{X-ray} sources

99: to bright optical counterparts ($R<23.8$); the majority of these sources

100: appear to be starburst and normal galaxies. 

101: The total number of sources in the main and supplementary catalogs is 578.

102: $R$-band optical counterparts and basic optical and infrared photometry 

103: are provided for the \hbox{X-ray} sources in the main and 

104: supplementary catalogs. 

105: We also include existing spectroscopic redshifts for 224 of the X-ray sources.

106: The average backgrounds in the

107: 0.5--2.0 and 2--8~keV bands are 0.066 and 0.167~counts~Ms$^{-1}$~pixel$^{-1}$,

108: respectively, and the background counts follow Poisson distributions.

109: The effective exposure times and sensitivity limits of 

110: the \hbox{CDF-S} are now comparable 

111: to those of the $\approx2$~Ms \chandra\ Deep Field-North (\hbox{CDF-N}). 

112: We also present

113: cumulative number counts for the main catalog and compare the results 

114: to those for

115: the \hbox{CDF-N}.

116: The soft-band number counts for these two fields 

117: agree well with each other at fluxes higher than 

118: $\approx2\times10^{-16}$~\flux, 

119: while the \hbox{CDF-S}

120: number counts are up to $\approx25\%$ smaller

121: than those for the \hbox{CDF-N}

122: at fluxes below $\approx2\times10^{-16}$~\flux\ in the soft

123: band and $\approx2\times10^{-15}$~\flux\ in the hard band,

124: suggesting small field-to-field variations. 

125: \end{abstract}

126: \keywords{cosmology: observations --- diffuse radiation --- galaxies:active ---

127: surveys --- \hbox{X-rays}: galaxies}

128: 

129: 

130: \section{INTRODUCTION}

131: 

132: One of the greatest successes of the {\it Chandra X-Ray Observatory} 

133: (\chandra) has been the characterization of the

134: sources creating the 0.5--8 keV cosmic \hbox{X-ray} background (CXRB), and the deepest

135: \chandra\ surveys form a central part of this effort. The two 

136: deepest \chandra\ surveys, the \chandra\ Deep Field-North and \chandra\ 

137: Deep Field-South (\hbox{CDF-N} and \hbox{CDF-S}, jointly CDFs; see 

138: \citealt{Brandt2005} for a review), have each detected hundreds of 

139: \hbox{X-ray} sources over $\approx 450$~arcmin$^2$ areas with enormous 

140: multiwavelength observational investments. They have measured the highest 

141: sky density of accreting supermassive black holes (SMBHs) to date and have 

142: also enabled novel X-ray studies of starburst and normal galaxies, groups 

143: and clusters of galaxies, large-scale structures in the distant universe, 

144: and Galactic stars. 

145: 

146: As part of an effort to create still deeper \hbox{X-ray} surveys, we proposed

147: for substantial additional exposure on the \hbox{CDF-S} during \chandra\ 

148: Cycle~9. The \hbox{CDF-S} has superb and improving coverage at optical, 

149: infrared, and radio wavelengths; it will continue to be a premiere 

150: multiwavelength deep-survey field for the coming decades as additional 

151: large facilities are deployed in the southern hemisphere. Furthermore, 

152: owing to the 1~Ms of \chandra\ exposure already available 

153: \citep[][hereafter G02]{Giacconi2002}, the 

154: \hbox{CDF-S} is a natural field to observe more sensitively. 

155: Although our 

156: proposal was not approved in the peer review, subsequently 

157: 1~Ms of Director's Discretionary Time was allocated 

158: for deeper \hbox{CDF-S} 

159: observations. The allocated observations were successfully executed in 

160: 2007 September, October and November, raising the \hbox{CDF-S} 

161: exposure to $\approx 2$~Ms and 

162: improving its sensitivity to be comparable to that of the \hbox{CDF-N} 

163: \citep[e.g.,][hereafter A03]{Alexander2003}.

164: Additional sky coverage at such flux levels 

165: is critically important as it substantially improves the statistical sample 

166: sizes of the faintest X-ray sources and also allows a basic assessment of 

167: the effects of cosmic variance. Furthermore, approximately doubling the 

168: exposure on previously detected sources substantially improves the constraints

169: on their positions, spectral properties, and variability properties.

170: 

171: In this paper, we present up-to-date \chandra\ source catalogs and data 

172: products derived from the full $\approx 2$~Ms \hbox{CDF-S} data set along 

173: with details of the observations, data processing, and technical analysis. 

174: Detailed subsequent investigations and scientific

175: interpretation of the new \hbox{CDF-S} sources will be presented in future

176: papers, e.g., studies of heavily obscured and Compton-thick 

177: active galactic nuclei (AGNs), high-redshift AGNs, AGN spectra and variability, 

178: starburst and normal galaxies, and clusters and groups of galaxies.

179: In \S2 we describe the observations and data reduction, and in \S3 we

180: present the main and supplementary point source catalogs and 

181: describe the methods used to create these catalogs. In \S4 we estimate

182: the background and sensitivity across the survey region. We also present 

183: basic number-count results for point sources in \S5. We summarize in \S6.

184: 

185: The Galactic column density along the line of sight to the 

186: \hbox{CDF-S} is remarkably low: 

187: \hbox{$N_{\rm H}=8.8\times 10^{19}$~cm$^{-2}$} \citep[e.g.,][]{Stark1992}.

188: The coordinates throughout this paper are J2000. 

189: A $H_0=70$~km~s$^{-1}$~Mpc$^{-1}$, $\Omega_{\rm M}=0.3$, and 

190: $\Omega_{\Lambda}=0.7$ cosmology is adopted.

191: 

192: 

193: 

194: \section{OBSERVATIONS AND DATA REDUCTION}

195: 

196: 

197: \subsection{Observations and Observing Conditions}

198: The \hbox{CDF-S} consists of 23 separate observations described

199: in Table~\ref{tbl-obs}. The $\approx 1$~Ms catalogs for the first 11 

200: observations taken between 1999 October 14 and 2000 December 23 

201: were presented in G02 and A03. 

202: Note that observation 581 (1999 October 14)

203: was excluded from the data reduction and is not listed in Table~\ref{tbl-obs}

204: due to telemetry saturation and other problems. The second $\approx 1$~Ms 

205: exposure consisted of 12 observations taken between 2007 September 20 and

206: 2007 November 4.

207: 

208: The Advanced CCD Imaging Spectrometer imaging array (ACIS-I; 

209: \citealt{Garmire2003}) was used

210: for all of the \chandra\ observations.

211: The ACIS-I is composed of 

212: four $1024\times1024$ pixel CCDs (CCDs \hbox{I0--I3}), covering a 

213: field of view of $16\farcm 9\times 16\farcm 9$ ($\approx$285

214: arcmin$^2$), and the pixel size of the CCDs is $\approx$$0\farcs 492$. 

215: The focal-plane 

216: temperature was $-110\degr$C for observations 1431-0 and 1431-1, and 

217: $-120\degr$C for the others. The 12 new observations were taken in Very

218: Faint mode to improve the screening of background events and thus increase

219: the sensitivity of ACIS in detecting faint \hbox{X-ray} sources 

220: \citep{Vikhlinin2001}.

221: 

222: The background light curves for all 23 observations were inspected using

223: EVENT BROWSER in the Tools for ACIS Real-time Analysis ({\sc TARA}; 

224: \citealt{Broos2000}) software package. Aside from a mild flare 

225: during observation 1431-0

226: (factor of $\approx3$ increase for $\approx5$~ks), all data sets are free 

227: from significant

228: flaring, and the background is stable within $\approx$20\% of 

229: typical quiescent {\it Chandra}

230: values. After filtering on good-time intervals and removing the one mild

231: flare, we are left with 1.911~Ms of total exposure time for the 

232: 23 observations.

233: 

234: Because of the differences in pointings and roll angles for the individual

235: exposures, the total region

236: covered by the entire \hbox{CDF-S} is 435.6 arcmin${^2}$, considerably

237: larger than the ACIS-I field of view. Combining the 23 observations, the 

238: average aim point (weighted by exposure time) is $\alpha_{\rm J2000.0}=

239: 03^{\rm h}32^{\rm m}28\fs80$, $\delta_{\rm J2000.0}=-27\degr48\arcmin23\farcs0$. 

240: 

241: 

242: \subsection{Data Reduction}

243: 

244: The basic archive data products were processed with the

245: \chandra\ X-ray Center (CXC) pipeline software versions listed in Table 1.

246: The reduction and analysis of the data used \chandra\ Interactive Analysis

247: of Observations (CIAO) tools whenever possible\footnote{See

248: http://cxc.harvard.edu/ciao/ for details on CIAO.}; however, custom

249: software, including the {\sc TARA} package, was also used. Each

250: observation was

251: reprocessed using the CIAO tool {\sc acis\_process\_events}, to correct for the

252: radiation damage sustained by the CCDs during the first few months of

253: \chandra\ operations using a Charge Transfer Inefficiency (CTI)

254: correction procedure \citep{Townsley2000,Townsley2002}

255: \footnote{Note that the CXC CTI correction procedure

256: is only available for $-120\degr$C data; thus we did not CTI-correct

257: observations 1431-0 and 1431-1.}, to remove the standard pixel

258: randomization which blurs the {\it Chandra} point spread function (PSF),

259: and to apply a modified bad-pixel file as detailed below.

260: 

261: One important deviation from the standard {\it Chandra} reduction

262: procedure outlined by the CXC is implementation of a stripped-down 

263: bad-pixel file. 

264: We note that the standard bad-pixel file supplied with all

265: {\it Chandra} data currently excludes $\approx$ 6--7\% of the total

266: effective area on front-illuminated devices (e.g., ACIS-I). A large

267: fraction of the bad-pixel locations identified in this file, however,

268: appear to be flagged solely because they show a few extra events (per Ms)

269: almost exclusively below 0.5--0.7 keV.\footnote{See

270: http://cxc.harvard.edu/cal/Acis/Cal\_prods/badpix/index.html} Good events

271: with energies above 0.7~keV that fall on these bad pixels are likely to be

272: perfectly acceptable for source searching, as well as for photometry and

273: spectral analysis albeit with a few mild caveats regarding

274: misinterpretation. Rather than reject all events falling on such columns,

275: we instead adopted a procedure to only exclude events below a

276: row-dependent energy of 0.5--0.7 keV.

277: \footnote{The energy range of 0.5--0.7~keV

278: and frequency of occurrence were verified by visual inspection of such

279: columns in our $\approx2$~Ms data set. We found that such ``hot'' soft 

280: columns were

281: not clearly seen in any individual observations. The upper energy bound

282: appears to vary as a function of distance from the readout edge of the

283: front-illuminated CCDs, such that rows closest to the readout edge only

284: have extra events below $\approx$0.5~keV, while those furthest away have

285: extra events extending up to $\approx$0.7~keV.} To this end, we generated

286: a stripped-down bad-pixel file, only selecting obvious bad columns and pixels

287: above 1~keV; this excluded $\approx1.5\%$ of the total effective area on

288: front-illuminated devices. Once the entire $\approx2$~Ms data set 

289: was combined, we

290: isolated ``hot'' soft columns as those where the total number of events with

291: energies below 0.7~keV was 5$\sigma$ or more above the mean. 

292: We then rejected

293: any events in those columns that fell below a row-dependent

294: 0.5--0.7~keV; this removed 1\% of all events.

295: 

296: Through inspection of the data in CCD coordinates, we

297: additionally discovered that the CXC-preferred CIAO tool {\sc 

298: acis\_run\_hotpix}

299: failed to flag a substantial number of obvious cosmic-ray afterglows

300: ($\sim$100--200 per observation, depending on exposure length),

301: elevating the overall background and, in egregious cases, leaving

302: afterglows to be mistaken as real sources. 

303: This problem appeared to be worse for

304: Faint mode data, presumably because the additional $5\times5$ screening

305: applied in Very Faint mode rejects the strongest afterglows 

306: \citep{Vikhlinin2001}.

307: To remedy this situation, we reverted to using the more stringent

308: {\sc acis\_detect\_afterglow} algorithm on all of our data. 

309: Notably, none of our

310: sources has a count rate high enough that {\sc acis\_detect\_afterglow} would

311: reject true source counts, which we verified by inspection of events

312: flagged by this routine. Even {\sc acis\_detect\_afterglow} failed

313: to reject all afterglows, and thus we created custom

314: software to remove many remaining faint afterglows from

315: the data. Working in CCD coordinates, we removed additional faint

316: afterglows with three or more total counts occuring within 20~s (or

317: equivalently 6 consecutive frames). In total, we removed 229 

318: total events associated with afterglows.

319: In all cases, we inspected the data set and found that such flagged events were

320: isolated and not associated with apparent legitimate X-ray sources.

321: 

322: 

323: \section{PRODUCTION OF THE POINT-SOURCE CATALOGS}

324: The production of the point-source catalogs largely followed the procedure

325: described in \S3 of A03. The main differences in the catalog-production 

326: procedure used here are the following:

327: 

328: \begin{enumerate}

329: 

330: \item

331: Our main \chandra\ catalog includes sources detected by running {\sc

332: wavdetect} \citep{Freeman2002} at a false-positive probability 

333: threshold of 10$^{-6}$, less

334: conservative than the 10$^{-7}$ value adopted by A03. Even with this

335: revised threshold, we expect the fraction of false sources to be small; 

336: see \S3.2 for details.

337: 

338: \item

339: Additional sensitivity can be obtained by merging the $\approx 250$~ks

340: Extended \chandra\ Deep Field-South (\hbox{E-CDF-S}; \citealt{Lehmer2005}, 

341: hereafter L05) with the $\approx2$~Ms \hbox{CDF-S}. An additional 86

342: \hbox{X-ray} sources were detected with this approach. 

343: These sources are presented

344: in a supplementary catalog described in \S3.3.2.

345: 

346: \end{enumerate}

347: 

348: 

349: 

350: \subsection{Image and Exposure Map Creation}

351: We registered the observations in the following manner. {\sc wavdetect}

352: was run on each individual cleaned image to generate an initial source list.

353: Centroid positions for each detected source were determined using the

354: reduction tool

355: {\sc acis extract} (AE; \citealt{Broos2000}).\footnote{The {\sc acis extract} 

356: software can be accessed from

357: http://www.astro.psu.edu/xray/docs/TARA/ae\_users\_guide.html}

358: The observations were registered to a common astrometric

359: frame by matching X-ray centroid positions to optical sources detected in

360: deep $R$-band images taken with the Wide Field Imager (WFI) of the MPG/

361: ESO telescope at La Silla (see \S2 of \citealt{Giavalisco2004}). The

362: matching was performed using the CIAO tools {\sc reproject\_aspect} and

363: {\sc wcs\_update} adopting a 3$\arcsec$ matching radius and a 

364: residual rejection limit \footnote{This is a parameter used in {\sc wcs\_update}

365: to remove source pairs based on pair positional offsets.}

366: of 0\farcs6; 50--100 sources were typically used in each observation 

367: for the final

368: astrometric solution. The tool {\sc wcs\_update} applied linear translations

369: ranging from 0\farcs05 to 0\farcs34, rotations ranging from $-0\fdg239$ to

370: 0\fdg009, and scale stretches ranging from 0.999563 to 1.000714; individual

371: registrations are accurate to $\approx$0\farcs3. All of the observations were

372: then reprojected to the frame of observation 2406, since this data set required

373: the smallest translation to align it with the optical astrometric frame.

374: 

375: We constructed images using the standard

376: \asca\ grade set (\asca\ grades 0, 2, 3, 4, 6) for three standard bands: 

377: \hbox{0.5--8.0~keV} (full band; FB), \hbox{0.5--2.0~keV}

378: (soft band; SB), and \hbox{2--8~keV} (hard band; HB). 

379: Figure~\ref{fbimg} shows the full-band raw image.

380: Exposure maps in the

381: three standard bands were created following the basic procedure

382: outlined in \S3.2 of \citet{Hornschemeier2001} and were normalized to the

383: effective exposures of a source located at the average aim point. 

384: Briefly, this

385: procedure takes into account the effects of vignetting, gaps between the CCDs,

386: bad-column filtering, bad-pixel filtering, and the spatially 

387: dependent degradation in quantum efficiency

388: due to contamination on the ACIS optical-blocking filters.

389: A photon index of

390: $\Gamma=1.4$ was assumed in creating the exposure maps, which is

391: approximately

392: the slope of the \hbox{X-ray} background in the

393: \hbox{0.5--8.0~keV} band 

394: \citep[e.g.,][]{Marshall1980,Gendreau1995,Hasinger1998}.

395: We show the full-band exposure map in Figure~\ref{fbemap}.

396: Using the full-band exposure map, we calculated the 

397: survey solid angle as a 

398: function of the minimum full-band effective exposure; the result is

399: plotted in Figure~\ref{emapcum}. Approximately 56\% and 42\% of the 

400: \hbox{CDF-S} field has a full-band effective exposure 

401: greater than 1~Ms and 1.5~Ms, 

402: respectively, with a maximum effective exposure of $\approx1.884$~Ms

403: (note this is slightly smaller than the 1.911~Ms total exposure since the

404: aim points of all the \chandra\ observations were not exactly the same).

405: The survey solid angles are comparable to those of the 

406: $\approx2$~Ms \hbox{CDF-N} (A03; dashed curve in Fig.~\ref{emapcum}).

407: 

408: Adaptively smoothed images were created using the CIAO

409: tool {\sc csmooth} on the raw images. 

410: Exposure-corrected smoothed images were then constructed following

411: \S3.3 of \citet{Baganoff2003}.

412: We show in 

413: Figure~\ref{clrimg} a color composite of the exposure-corrected 

414: smoothed images

415: in the 0.5--2.0 keV ({\it red}), 2--4 keV ({\it green}), and 4--8 

416: keV ({\it blue}) bands.

417: Source searching was performed using only the raw images, while

418: many of the detected \hbox{X-ray} sources are shown

419: more clearly in the adaptively smoothed images.

420: 

421: 

422: %

423: \begin{figure}

424: \centerline{

425: \includegraphics[scale=0.4]{f01.ps}

426: }

427: \figcaption{

428: Full-band (0.5--8.0~keV) raw image of the $\approx$2~Ms \hbox{CDF-S}.

429: The gray scales are linear.

430: The apparent scarcity of sources near the field center is largely due

431: to the small PSF at that location 

432: (see Figs.~\ref{clrimg} and \ref{pos} for clarification).

433: The black outline surrounding the image indicates

434: the extent of all the \hbox{CDF-S} observations.

435: The large rectangle indicates the GOODS-S

436: \citep{Giavalisco2004} region, and

437: the central square indicates the {\it Hubble} Ultra Deep Field

438: (UDF; \citealt{Beckwith2006}) region. The cross near the center of the images

439: indicates the average aim point, weighted by exposure time (see

440: Table~\ref{tbl-obs}).\label{fbimg}}

441: \end{figure}

442: %

443: 

444: %

445: \begin{figure}

446: \centerline{

447: \includegraphics[scale=0.4]{f02.ps}

448: }

449: \figcaption{

450: Full-band (0.5--8.0~keV) exposure map of 

451: the $\approx2$~Ms \hbox{CDF-S}.

452: The darkest areas represent the highest effective 

453: exposure times (the maximum value is 1.884~Ms). The gray scales

454: are logarithmic.

455: The regions and the cross symbol have the same meaning as those in

456: Fig. \ref{fbimg}.

457: \label{fbemap}}

458: \end{figure}

459: %

460: 

461: \begin{figure}

462: \centerline{

463: \includegraphics[scale=0.4]{f03.ps}

464: }

465: \figcaption{

466: Amount of survey solid angle having at least a given amount of

467: full-band effective exposure for the $\approx2$~Ms \hbox{CDF-S}

468: ({\it solid curve}).

469: The maximum exposure is $\approx1.884$~Ms.

470: The vertical dotted

471: line shows an effective exposure of 1~Ms. About 245 arcmin$^2$ 

472: ($\approx 56\%$) of the \hbox{CDF-S} survey area has $>1$~Ms 

473: effective exposure. Corresponding data from 

474: the $\approx2$~Ms \hbox{CDF-N} (A03) are

475: plotted as a dashed curve for comparison.

476: \label{emapcum}}

477: \end{figure}

478: %

479: 

480: %

481: \begin{figure}

482: \centerline{

483: \includegraphics[scale=0.4]{f04.ps}

484: }

485: \figcaption{

486: \chandra\ ``false-color'' image of the $\approx$2~Ms \hbox{CDF-S}. This image

487: is a color composite of the  exposure-corrected adaptively smoothed images

488: in the 0.5--2.0 keV ({\it red}), 2--4 keV ({\it green}), and 4--8

489: keV ({\it blue}) bands. 

490: The apparent smaller size and lower brightness of sources near the 

491: field center is due to the small PSF at that location. 

492: The regions and the cross symbol have the same meaning as those in

493: Fig. \ref{fbimg}.

494: \label{clrimg}}

495: \end{figure}

496: %

497: 

498: 

499: \subsection{Point-Source Detection}

500: 

501: Point-source detection was performed in each of the three standard bands

502: with {\sc wavdetect} using a

503: ``$\sqrt{2}$~sequence'' of wavelet scales (i.e.,\ 1, $\sqrt{2}$, 2,

504: $2\sqrt{2}$, 4, $4\sqrt{2}$, 8, $8\sqrt{2}$, and 16 pixels). 

505: The criterion for source

506: detection is that a source must

507: be found with a given false-positive probability threshold in at least one of

508: the three standard bands. For the main \chandra\ source catalog discussed in

509: \S3.3.1, the false-positive probability threshold in

510: each band was set to $1\times 10^{-6}$.

511: 

512: If we conservatively consider the three images searched 

513: to be independent, $\approx$18 false detections are expected in the main

514: \chandra\ source catalog for the case of a uniform background.

515: However, this false-source estimate is conservative, since 

516: a single pixel usually should not

517: be considered a source-detection cell, particularly at large off-axis angles

518: ({\sc wavdetect} suppresses fluctuations on scales smaller than the PSF).

519: As quantified in

520: \S3.4.1 of A03, the number of false-sources is likely

521: \hbox{$\approx$2--3} times less than our conservative estimate.

522: We also provide additional source-significance information 

523: by running {\sc wavdetect} using false-positive probability

524: thresholds of $1\times 10^{-7}$ and $1\times 10^{-8}$. These results

525: are presented in $\S$3.3.1, which can be utilized

526: to perform more conservative source screening if desired.

527: 

528: \subsection{Point-Source Catalogs}

529: 

530: \subsubsection{Main Chandra Source Catalog}

531: 

532: 

533: The source 

534: lists resulting from the {\sc wavdetect} runs discussed in \S3.2 

535: with false-positive probability 

536: threshold of $1\times 10^{-6}$ were 

537: merged to create the main point-source catalog presented in 

538: Table~\ref{tbl-mcat}, which consists of 462 point sources. 

539: Whenever possible, we have quoted 

540: the position determined in 

541: the full band; when a source is not detected in the full band, we used, in 

542: order of priority, the soft-band position or hard-band position. 

543: For cross-band 

544: matching, we used a matching radius of $2\farcs 5$ for sources 

545: within $6\arcmin$ of the average aim point and $4\farcs 0$ for 

546: larger off-axis angles. These matching radii were chosen by

547: inspecting histograms showing the number of matches obtained as 

548: a function of angular separation \citep[e.g., see \S2 of][]{Boller1998}; 

549: the mismatch probability is $\la 1\%$ over the entire field. 

550: A few mismatches near the edge of the field were removed through 

551: visual inspections. 

552: 

553: We improved the {\sc wavdetect} source positions using the centroid and

554: matched-filter positions generated with AE. 

555: The centroid is simply the mean position of all

556: events within the AE extraction region, while the matched-filter position

557: is the

558: position found by correlating the full-band image in the vicinity of each

559: source with a combined PSF. The combined PSF is produced by combining the

560: ``library'' PSF of a

561: source for each observation, weighted by the number of detected

562: counts.\footnote{The PSFs are

563: taken from the CXC PSF library; see

564: http://cxc.harvard.edu/ciao/dictionary/psflib.html.} This technique takes

565: into account the fact that, due to the complex PSF at large off-axis

566: angles, the X-ray source position is not always located at the peak of the

567: X-ray emission. The {\sc wavdetect}, centroid, and matched-filter techniques

568: provide comparable accuracy on-axis, while the matched-filter technique 

569: performs better off-axis. We chose the matched-filter positions as

570: our default, and then visually inspected each source. 

571: When the adopted position appeared to deviate from the

572: apparent center of the source by more than 0\farcs1, 

573: we modified the position manually such that it was

574: visually consistent with the apparent center.

575: 

576: We refined the absolute \hbox{X-ray} source positions

577: by matching the \hbox{X-ray} sources in the main \chandra\ catalog

578: to the WFI $R$-band optical sources (see \S3.1). 

579: There are $\approx30\,000$ optical sources across the \hbox{CDF-S} field,

580: which have accurate positions with positional error 

581: $\Delta_{\rm o}\approx0\farcs1$.\footnote{See 

582: http://archive.stsci.edu/pub/hlsp/goods/v1/h\_goods\_v1.0\_rdm.html.}

583: We selected relatively bright

584: optical sources with AB magnitudes $R\le 24$ ($\approx5\,500$ sources), 

585: and matched them to the 

586: \hbox{X-ray} sources using a 2$\farcs$5

587: matching radius. 

588: There are eight cases where one \hbox{X-ray} source has

589: two optical counterparts. The $R$-band 

590: magnitudes of the two counterparts differ by less than one in all cases, and

591: thus we selected the closer one 

592: as the most-probable counterpart. 

593: We also visually inspected the optical counterparts and, 

594: for purposes of positional checking, only keep those

595: sources that are point-like or slightly extended; ten extended sources 

596: were removed. Under these criteria, 

597: 229 \hbox{X-ray} sources have bright optical counterparts.

598: We estimated the expected number of false matches by

599: manually shifting the \hbox{X-ray} source coordinates in right 

600: ascension and declination by $5\farcs0$ (both positive and negative

601: shifts) and recorrelating with the optical sources. On average, the 

602: number of false matches is $\approx35$ ($\approx15\%$), and the median

603: offset of these false matches is $\approx$1$\farcs$71. By comparing

604: the \hbox{X-ray} and optical source positions,

605: we found small shift and plate-scale corrections. These corrections have

606: been applied to the positions of all the \hbox{X-ray} sources

607: in the main and supplementary catalogs, resulting in

608: small ($<0\farcs2$) astrometric shifts.

609: 

610: We investigated the accuracy of the \hbox{X-ray} source positions

611: using these 229 \hbox{X-ray} detected bright optical sources.

612: Figure~\ref{dpos} shows the positional offset between 

613: the \hbox{X-ray} sources and their optical

614: counterparts as a function of the off-axis angle. The median offset

615: is $\approx$0$\farcs$36. However,

616: there are clear off-axis angle and source-count

617: dependencies. The off-axis angle dependence is due to the degradation of

618: the \chandra\ PSF

619: at large off-axis angles, while the count dependence is due to the

620: difficulty of finding the centroid of a faint \hbox{X-ray} source. 

621: Simulations have shown that the offsets of {\sc wavdetect} positions appear

622: to increase exponentially with off-axis angle and decrease with the 

623: number of source

624: counts in a power-law form \citep[e.g.,][]{Kim2007}.

625: Based on Figure~\ref{dpos} and taking into account the 

626: probability of false matches, we derived an empirical relation 

627: for the positional uncertainties

628: of the \hbox{X-ray} sources in our sample, which is 

629: \begin{equation}

630: \log \Delta_{\rm X}=0.0326 \theta-0.2595\log C+0.1625~,

631: \end{equation}

632: 

633: \noindent

634: where $\Delta_{\rm X}$ is the positional uncertainty in arcseconds, $\theta$ 

635: the off-axis angle in arcminutes, and $C$ the source counts in the energy

636: band where the source position was determined. 

637: We set an upper limit of 2000 on $C$ as the positional accuracy

638: does not improve significantly beyond that level.

639: Positional uncertainties

640: for $C=20$, 200, and 2000 are shown in Figure~\ref{dpos}.

641: The stated positional uncertainties are for the 

642: $\approx85\%$ confidence level, and are smaller than the {\sc wavdetect}

643: positional errors, especially at large off-axis angles, 

644: because of our positional refinement described above.

645: A few sources in Figure~\ref{dpos} have unexpectedly

646: large positional offsets; they could be false matches.\footnote{For example, 

647: the source with $>200$ counts and a positional offset of $\approx1\farcs9$ in 

648: Figure~\ref{dpos} is source

649: ``289'' in the main \chandra\ catalog (see Table~2). This source does not have 

650: any optical counterpart after adopting a more appropriate matching radius,

651: as shown in the catalog.} There is also the possibility that a few of

652: them are off-nuclear \hbox{X-ray} sources 

653: \citep[e.g.,][]{Hornschemeier2004,Lehmer2006}.

654: Figure~\ref{poshist} shows the distributions of the positional 

655: offsets in four bins of different X-ray positional uncertainties, as well

656: as the expected numbers of false matches assuming a uniform spatial 

657: distribution of the $R\le 24$ optical sources. These histograms illustrate

658: clearly the reliability of our positional error estimates calculated using 

659: equation~(1).

660: 

661: 

662: %

663: \begin{figure}

664: \centerline{

665: \includegraphics[scale=0.5]{f05.ps}

666: }

667: \figcaption{

668: Positional offset vs. off-axis angle for

669: sources in the main {\it Chandra} catalog that were matched to WFI $R$-band

670: optical sources with AB magnitude $R\le24$ to within 2$\farcs$5. 

671: Black, dark gray, light gray, and open circles represent {\it

672: Chandra} sources with $\ge2000$, $\ge200$, $\ge20$, and $<20$ 

673: counts in the energy

674: band where the source position was determined, respectively.

675: The dotted curve shows the running median of all sources in

676: bins of 2\arcmin. The median offset of the expected false matches 

677: ($\approx$1$\farcs$71) is indicated by the 

678: dashed line. These data were

679: used to derive the $\approx85\%$ confidence-level positional uncertainties 

680: of the \hbox{X-ray} sources in the main catalog; see eq. (1).

681: Three solid curves indicate the $\approx85\%$ confidence-level positional 

682: uncertainties for sources with counts of 20, 200 and 2000.

683: The number of black, dark-gray, and light-gray circles lying below/above

684: their corresponding solid curves are 11/1, 48/4 and 116/20, respectively.

685: Note that sources with more than 20 or 200 counts will have expected 

686: positional 

687: uncertainties smaller than those indicated by the corresponding solid curves.

688: \label{dpos}}

689: \end{figure}

690: %

691: 

692: 

693: %

694: \begin{figure}

695: \centerline{

696: \includegraphics[scale=0.5]{f06.ps}

697: }

698: \figcaption{

699: Histograms showing the distributions of positional offset for

700: sources in the main {\it Chandra} catalog that were matched to WFI $R$-band 

701: optical sources with $R\le24$ to within 2$\farcs$5. X-ray sources were

702: divided into four bins based on their positional uncertainties estimated

703: using eq. (1): $0\arcsec$--$0\farcs5$, $0\farcs5$--$1\farcs0$,

704: $1\farcs0$--$1\farcs5$, and $1\farcs5$--$2\farcs0$. 

705: The vertical dashed line indicates the median positional uncertainty for

706: X-ray sources in each bin.

707: Dotted lines show how many random $R\le24$ optical sources

708: are expected as a function of the positional offset.

709: Less than 20\% of

710: the optical counterparts lie beyond the median X-ray positional uncertainties

711: in all cases. 

712: \label{poshist}}

713: \end{figure}

714: %

715: 

716: 

717: The main {\it Chandra} \hbox{X-ray} source catalog 

718: is presented in Table~2, 

719: with the details of the columns given below.

720: 

721: \begin{enumerate}

722: 

723: \item

724: Column~1: the source number. Sources are listed in order of

725: increasing right ascension.

726: 

727: \item Columns~2 and 3: the right ascension and declination of 

728: the \hbox{X-ray} source,

729: respectively. These positions have been determined following the procedure 

730: described above. To avoid truncation error, we

731: quote the positions to higher precision than in the International Astronomical

732: Union (IAU) registered names beginning with the acronym ``CXO CDFS''. 

733: 

734: \item Column~4: the $\approx85\%$ confidence-level 

735: positional uncertainty in arcseconds. 

736: As shown above, the positional

737: uncertainty depends on off-axis angle and the number of detected counts,

738: and is estimated following equation (1). The minimum positional uncertainty

739: is $\approx0\farcs23$ for sources in the main catalog, and the 

740: maximum value is $\approx1\farcs90$.

741: 

742: 

743: \item Column~5: the off-axis angle of

744: the \hbox{X-ray} source in arcminutes. This is

745: calculated using the source position given in columns~2 and 3 and the 

746: average aim point of the \hbox{CDF-S} (see Table~1).

747: 

748: \item Columns~6--14: the source counts and the corresponding $1\sigma$

749: statistical errors \citep{Gehrels1986} or the upper limits on source counts

750: for the three standard bands, respectively.

751: The entries have not been corrected for vignetting. Source counts and 

752: statistical

753: errors have been calculated using circular-aperture photometry; extensive

754: testing has shown that this method is more reliable than the {\sc wavdetect}

755: photometry (e.g., \citealt{Brandt2001}; A03). 

756: The circular aperture was centered at the position given in

757: columns 2 and 3 for all bands. 

758: We have also computed photometry using AE, and the results are

759: in good agreement with this circular-aperture photometry.

760: 

761: 

762: The local background is determined in an annulus outside of the

763: source-extraction region. The mean number of background counts per pixel is

764: calculated from a Poisson model using ${n_1}/{n_0}$, where $n_0$ is the

765: number of pixels with 0 counts and $n_1$ is the number of pixels with 1 count

766: (e.g., A03).

767: By ignoring all pixels with more than 1~count, this technique

768: is robust against background contamination from sources. 

769: The principal

770: requirement for using this Poisson-model technique is that the 

771: background counts are

772: low and follow a Poisson distribution; 

773: we show in \S4 that the background of the $\approx2$~Ms exposure 

774: meets this criterion.

775: We note that the background estimation is problematic for several sources

776: which are located close to bright sources or near the edge of the 

777: survey field

778: where there is a strong gradient in exposure time. For each of these sources,

779: we have measured its background counts in the background maps described in 

780: \S4, using an annulus outside of the source-extraction region. Note 

781: that when constructing the background maps, we filled in the masked 

782: regions with a local background assuming a probability distribution; 

783: thus small additional uncertainties could be introduced during this process and 

784: will be carried on to the background estimation here. 

785: There are 17 such sources and they are marked with ``B''

786: in column 49 of Table~2. The net number of source counts is calculated

787: by subtracting the background counts from the source counts.

788: 

789: For sources with fewer than 1000 full-band counts, we have chosen the aperture

790: radii based on the encircled-energy function of the \chandra\ PSF as determined

791: using the CXC's {\sc mkpsf} software \citep{Feigelson2000,Jerius2000}.

792: In the soft band, where the background is lowest, the aperture radius

793: was set to the 95\% encircled-energy radius of the PSF. In the full and 

794: hard bands, the

795: 90\% encircled-energy radius of the PSF was used. Appropriate aperture

796: corrections were applied to the source counts by dividing the extracted source

797: counts by the encircled-energy fraction for which the counts were extracted.

798: 

799: For sources with more than 1000 full-band counts, systematic errors in the

800: aperture corrections often exceed the expected errors from photon statistics

801: when the apertures described in the previous paragraph are used. Therefore, for

802: such sources we used larger apertures to minimize the importance of the

803: aperture corrections; this is appropriate since these bright sources dominate

804: over the background. We set the aperture radii to be twice the 90\% 

805: encircled-energy full-band

806: radii and inspected these sources to verify that the

807: measurements were not contaminated by neighboring objects. No aperture

808: corrections were applied to these sources.

809: 

810: 

811: Manual correction of the source photometry was performed 

812: for sources having overlapping PSFs.

813: We manually separated 18 close doubles and 4 close triples, and these

814: sources are flagged with ``S'' in column 49 of Table~2.

815: 

816: We have performed several consistency tests to verify the quality of the

817: photometry. For example, we have checked that the sum of the counts measured in

818: the soft and hard bands does not differ from the counts measured in the full

819: band by an amount larger than that expected from measurement error. Systematic

820: errors that arise from differing full-band counts and soft-band plus hard-band

821: counts are estimated to be $\la4\%$.

822: 

823: When a source is not detected in a given band, an upper limit is calculated;

824: upper limits are indicated as a ``$-1.00$'' in the error columns. 

825: All upper limits are

826: determined using the circular apertures described above. When the number of

827: counts in the aperture is $\leq 10$, the upper limit is calculated using the

828: Bayesian method of \citet{Kraft1991} for 99\% confidence. The

829: uniform prior used by these authors results in fairly conservative upper limits

830: \citep[see][]{Bickel1992}, and other reasonable choices of priors 

831: do not materially

832: change our scientific results.  For larger numbers of counts in the aperture,

833: upper limits are calculated at the $3\sigma$ level for Gaussian statistics.

834: 

835: \item Columns~15 and 16: the right ascension and declination 

836: of the optical counterpart,

837: which was obtained by matching the \hbox{X-ray} source positions (columns~2 and

838: 3) to WFI $R$-band source positions using a matching radius that is 1.5 times

839: the quadratic sum of 

840: the positional errors of the X-ray and optical sources (i.e.,

841: $r_{\rm m}=1.5\sqrt{\Delta_{\rm_X}^2+\Delta_{\rm o}^2}$).

842: This matching radius was chosen to provide a large number of 

843: optical counterparts without introducing too many false matches.

844: The WFI $R$-band observations have a 5$\sigma$ limiting AB magnitude of

845: $27.3$ over the entire \hbox{CDF-S} field.

846: For 4 sources (our sources ``74'', ``283'', ``328'', and ``431'') 

847: that have more

848: than one optical match, the magnitude difference 

849: between the counterparts is less than three in all cases, 

850: and therefore the source with the smallest offset was

851: selected as the most-probable counterpart.

852: Using these criteria, 344

853: ($\approx74\%$) of the sources have optical counterparts. 

854: Sources with no

855: optical counterparts have these right ascension and declination values set to 

856: \hbox{``00 00 00.00''} and \hbox{``$-$00 00 00.0''}. 

857: We tested the reliability of the matching

858: by shifting the \hbox{X-ray} source coordinates

859: and recorrelating with the optical sources.

860: The matching is reliable (false-match probability $\la 8\%$) to        

861: $R\approx24$. The false-match probability rises to $\approx18\%$,

862: $\approx27\%$, and $\approx35\%$ at $R\approx25$, 26, and 27, respectively.

863: %A more detailed analysis of \hbox{X-ray} counterpart matching using optical,

864: %ground-based near-infrared, and {\it Spitzer} data will be 

865: %presented in M.~Brusa et~al. (2008, in preparation).

866: 

867: 

868: \item Column~17: the measured offset between the optical and \hbox{X-ray}

869: sources in arcseconds. Sources with no optical counterparts have

870: a value set to ``$-1.00$''. The offsets for all matches are below $2\farcs0$.

871: 

872: \item Column~18: the $R$-band AB magnitude of the optical counterpart.

873: Sources with no optical counterparts have a value set to ``$-1.00$''.

874: 

875: \item Columns~19 and 20: the corresponding source number and 

876: $i$-band AB magnitude from the GOODS-S

877: v2.0 $i$-band source catalog.\footnote{See \citet{Giavalisco2004} and

878: http://archive.stsci.edu/pub/hlsp/goods/catalog\_r2/.}

879: We matched the positions of the optical counterparts (see columns~15 

880: and 16) to the GOODS-S source positions using

881: a matching radius of $0\farcs5$. 

882: In 6 cases (our sources ``88'', ``120'', 

883: ``135'', ``155'', ``313'', and ``322'') 

884: where there is more than one GOODS-S source matching to an

885: optical counterpart,

886: we selected the GOODS-S source with the smallest offset 

887: as the most-probable match. 218 matches were found for the 344 

888: optical counterparts; note that the GOOD-S field does not cover the

889: whole \hbox{CDF-S}.

890: By shifting the coordinates of the optical counterparts

891: and recorrelating with the GOODS-S sources, we estimated the 

892: false-match probability to be

893: $\la 5\%$.

894: The GOODS-S $i$-band observations have a 5$\sigma$ limiting AB magnitude of

895: $28.5$.

896: The $i$-band magnitude is the SExtractor \citep{Bertin1996}

897: corrected isophotal magnitude.

898: Sources with no GOODS-S match have 

899: these two columns set to ``$-1$'' and ``$-1.00$'', respectively.

900: 

901: \item Columns~21 and 22: the corresponding coordinate-based source 

902: name and $z$-band AB magnitude from the

903: Galaxy Evolution from Morphologies and SEDs (GEMS)

904: source catalog \citep{Caldwell2008}.

905: We matched the positions of the optical counterparts (see columns~15

906: and 16) to the GEMS source positions using

907: a matching radius of $0\farcs5$.

908: In 1 case (our source ``74'') where there is more than one 

909: GEMS source matching to an

910: optical counterpart,

911: we selected the GEMS source with the smallest offset

912: as the most-probable match. 297 matches were found for the 344

913: optical counterparts.

914: By shifting the coordinates of the optical counterparts

915: and recorrelating with the GEMS sources, we estimated the

916: false-match probability to be

917: $\la 2\%$. 

918: The GEMS $z$-band observations have a 5$\sigma$ limiting AB magnitude of

919: $27.3$ over the entire \hbox{CDF-S} field.

920: The $z$-band magnitude is the SExtractor MAG\_BEST magnitude.

921: Sources with no GEMS match have 

922: these two columns set to ``$-1$'' and ``$-1.00$'', respectively.

923: 

924: \item Columns~23 and 24: the corresponding source number and

925: $K_s$-band AB magnitude from the source catalog for the ESO/NTT

926: SOFI survey of the \hbox{CDF-S} region.\footnote{See 

927: http://www.eso.org/sci/activities/projects/eis/surveys/summary\_DPS.html.}

928: We matched the positions of the optical counterparts (see columns~15

929: and 16) to the SOFI source positions using

930: a matching radius of $0\farcs75$.

931: 266 matches were found for the 344

932: optical counterparts.

933: By shifting the coordinates of the optical counterparts

934: and recorrelating with the SOFI sources, we estimated the

935: false-match probability to be

936: \hbox{$\la1\%$}.

937: The SOFI $K_s$-band observations have a 5$\sigma$ limiting AB magnitude of

938: $23.0$ over the entire \hbox{CDF-S} field. 

939: The $K_s$-band magnitude is the SExtractor corrected isophotal magnitude.

940: Sources with no SOFI match have  

941: these two columns set to ``$-1$'' and ``$-1.00$'', respectively.

942: 

943: \item Columns~25 and 26: the corresponding source number and IRAC

944: $5.8$ $\mu$m flux density ($f_{58}$) from the {\it Spitzer} 

945: IRAC/MUSYC Public Legacy Survey in the \hbox{E-CDF-S} (SIMPLE)

946: source catalog.\footnote{See http://ssc.spitzer.caltech.edu/legacy/simplehistory.html.}

947: We matched the positions of the optical counterparts (see columns~15

948: and 16) to the SIMPLE source positions using

949: a matching radius of $0\farcs75$.

950: 306 matches were found for the 344

951: optical counterparts.

952: By shifting the coordinates of the optical counterparts

953: and recorrelating with the SIMPLE sources, we estimated the

954: false-match probability to be

955: $\la 2\%$. 

956: The SIMPLE $5.8$ $\mu$m observations have a 5$\sigma$ limiting AB magnitude of

957: 21.9--22.5 over the entire \hbox{CDF-S} field; the limiting magnitude

958: is spatially dependent for SIMPLE.

959: The $5.8$ $\mu$m flux density is the aperture flux density

960: in a $2\farcs0$ circular aperture,

961: normalized to an AB magnitude zero point of 25. Note that 

962: an aperture correction of $\approx1.5$ was not applied to these fluxes;

963: i.e., the aperture-corrected AB magnitude is 

964: $m({\rm AB})=25-2.5\log10(1.5\times f_{58})$.

965: Sources with no SIMPLE match have

966: these two columns set to ``$-1$'' and ``$-1.00$'', respectively.

967: 

968: \item Columns~27 and 28: the corresponding spectroscopic redshift

969: and the reference for the redshift.

970: Secure spectroscopic redshifts were collected from

971: \citet{LeFevre2004}, \citet{Szokoly2004}, \citet{Mignoli2005},

972: \citet{Ravikumar2007}, \citet{Popesso2008}, and \citet{Vanzella2008}, 

973: with the reference numbers of 1--6 in column~28,

974: respectively. A matching radius of $0\farcs5$ was used when matching the

975: optical counterparts (see columns~15

976: and 16) to the redshift catalogs.

977: 190 of the 344 optical counterparts have redshift measurements.

978: By shifting the coordinates of the optical counterparts

979: and recorrelating with the redshift catalogs, we estimated the

980: false-match probability to be

981: $\la 1\%$.

982: Sources with no secure spectroscopic redshift have

983: these two columns set to ``$-1.000$'' and ``$-1$'', respectively.

984: Note that there are also photometric redshifts available in the literature

985: \citep[e.g.,][]{Mobasher2004,Wolf2004}, but these are

986: not included in our catalogs.

987: 

988: 

989: \item Column~29: the corresponding $\approx$1~Ms \hbox{CDF-S} source 

990: number from the main

991: \chandra\ catalog presented in A03 (see column~1 of Table~A2a in A03). We 

992: matched our \hbox{X-ray} source positions to A03 source positions using 

993: a matching radius that is the

994: quadratic sum of

995: the $\approx3\sigma$ positional errors of the \hbox{CDF-S} 

996: and A03 \hbox{X-ray} sources. 

997: The $3\sigma$ positional error of a \hbox{CDF-S} 

998: source is approximately

999: twice the positional error quoted in column~4 (i.e., $2\Delta_{\rm_X}$), 

1000: and that of an A03  

1001: source is approximately

1002: twice the positional error quoted in Table~A2a of A03.

1003: The false match probability is less than 1\% with this matching radius.

1004: Only one A03 match was found for each matched source.

1005: In one case where two close-double sources matched to one A03 source, we 

1006: chose the source with the smallest offset (source ``433'') 

1007: as the most-probable match. 

1008: We manually set the counterpart of the source with 

1009: source number ``437''

1010: to be source ``312'' in A03, 

1011: because A03 apparently underestimated the positional error of this source.

1012: Sources with no A03 match have a value of ``$-1$''.

1013: 

1014: \item Columns~30 and 31: the right ascension and declination of 

1015: the corresponding 

1016: A03 source indicated in column~29. Sources with no A03 match have right 

1017: ascension and

1018: declination values set to \hbox{``00 00 00.00''} and \hbox{``$-$00 00 00.0''}.

1019: 

1020: \item Columns~32 and 33: the corresponding $\approx$1~Ms \hbox{CDF-S} 

1021: source ``ID'' number and ``XID'' number from the main

1022: \chandra\ catalog presented in G02.

1023: When matching our

1024: \hbox{CDF-S} source positions with G02 counterparts, we removed offsets

1025: to the G02 positions of $-1\farcs2$ in right ascension and $+0\farcs8$ in 

1026: declination (see $\S$A3

1027: of A03); these positions are corrected in the quoted source positions in

1028: columns~34 and 35. We used a matching radius that is the

1029: quadratic sum of 

1030: the $\approx3\sigma$ positional errors of the \hbox{CDF-S} 

1031: and G02 \hbox{X-ray} sources. 

1032: The $3\sigma$ positional error of a \hbox{CDF-S}

1033: source is approximately

1034: twice the positional error quoted in column~4, and that of a G02

1035: source is quoted in Table~2 of G02.

1036: Only one G02 match was found for 

1037: each matched source.

1038: In three cases where two close-double sources matched to one G02 source, we

1039: chose the source with the smallest offset 

1040: (sources ``142'', ``195'' and ``275'') as the most-probable match.

1041: Sources with no G02 match have a value of ``$-1$''.

1042: 

1043: \item Columns~34 and 35: the right ascension and declination of 

1044: the corresponding G02 

1045: source indicated in columns~32 and 33. Note that the quoted positions have

1046: been corrected by the offsets described in columns~32 and 33 (see $\S$A3 of

1047: A03). Sources with no G02 match have right ascension and declination 

1048: values set to \hbox{``00 00

1049: 00.00''} and \hbox{``$-$00 00 00.0''}.

1050: 

1051: \item Columns~36--38: the effective exposure times determined from the

1052: standard-band exposure maps (see \S3.1 for details on the exposure maps).

1053: Dividing the counts listed in columns~6--14 by the corresponding effective

1054: exposures will provide vignetting-corrected and quantum-efficiency

1055: degradation corrected count rates.

1056: 

1057: \item Columns~39--41: the band ratio, defined as the ratio of counts

1058: between the hard and soft bands, and the corresponding upper and lower errors,

1059: respectively. Quoted band ratios have been corrected for differential

1060: vignetting between the hard band and soft band using the appropriate exposure

1061: maps. Errors for this quantity are calculated following the ``numerical

1062: method'' described in \S1.7.3 of \citet{Lyons1991}; this avoids 

1063: the failure of the

1064: standard approximate variance formula when the number of counts is small (see

1065: \S2.4.5 of \citealt{Eadie1971}). Note that the error distribution is not

1066: Gaussian when the number of counts is small. Upper limits are calculated for

1067: sources detected in the soft band but not the hard band, and lower limits are

1068: calculated for sources detected in the hard band but not the soft band. For

1069: these sources, the upper and lower errors are set to the computed band ratio.

1070: Sources detected only

1071: in the full band have band ratios and corresponding errors set to ``$-1.00$''.

1072: 

1073: \item Columns~42--44: the effective photon index ($\Gamma$) with upper and

1074: lower errors, respectively, for a power-law model with the Galactic column

1075: density given in \S1. When the number of source counts is not low, 

1076: the effective photon index has been calculated based on the band ratio

1077: in column~39 using the CXC's Portable, Interactive, Multi-Mission Simulator 

1078: (PIMMS). Upper limits are calculated for

1079: sources detected in the hard band but not the soft band, and lower limits are

1080: calculated for sources detected in the soft band but not the hard band. For

1081: sources with only limits on the effective photon index, 

1082: the upper and lower errors are set to the computed effective

1083: photon index.

1084: 

1085: A source with a low number of counts is defined as being (1) detected in the

1086: soft band with $<30$ counts and not detected in the hard band, (2) detected in

1087: the hard band with $<15$ counts and not detected in the soft band, (3) detected

1088: in both the soft and hard bands, but with $<15$ counts in each, or (4) detected

1089: only in the full band.  When the number of counts is low, the photon index is

1090: poorly constrained and is set to $\Gamma=1.4$, a representative value for faint

1091: sources that should yield reasonable fluxes. In this case, the upper and lower 

1092: errors are set to ``$0.00$''.

1093: 

1094: \item Columns~45--47: observed-frame fluxes in the three standard bands;

1095: quoted fluxes are in units of \flux.

1096: Fluxes have been computed using the counts in columns~6--14, the

1097: appropriate exposure maps (columns~36--38), and the effective photon indices 

1098: given in column~42. The

1099: fluxes have not been corrected for absorption by the Galaxy or material

1100: intrinsic to the source. For a power-law model with $\Gamma=1.4$, the soft-band

1101: and hard-band Galactic absorption corrections are $\approx$2.1\% and $\approx

1102: 0.1$\%, respectively. More accurate fluxes for these sources would require

1103: direct fitting of the \hbox{X-ray} spectra for each observation, which is

1104: model dependent and beyond the scope of this paper.

1105: %

1106: 

1107: \item Column~48: the logarithm of the minimum false-positive probability

1108: run with {\sc wavdetect} in which each source was detected (see $\S$3.2).  A

1109: lower false-positive probability indicates a more significant source detection.

1110: 398 ($\approx86\%$) and 357 ($\approx77\%$) of our

1111: sources are detected with false-positive probability thresholds of

1112: 1~$\times$~10$^{-7}$ and 1~$\times$~10$^{-8}$, respectively.

1113: 

1114: \item Column~49: notes on the sources. ``E'' refers to sources at the edge

1115: that lie partially outside of the survey area.

1116:  ``S'' refers to 

1117: close doubles or triples where manual separation was required. 

1118: ``B'' refers to sources with background counts estimated using the 

1119: background maps (see columns~6--14 of Table~2).

1120: 

1121: \end{enumerate}

1122: 

1123: In Table~3 we summarize

1124: the source detections in the three standard bands.  

1125: In total 462 point sources

1126: are detected, 327 of which

1127: were present in the main {\it Chandra} catalogs for the

1128: $\approx$1~Ms \hbox{CDF-S} (G02 and A03), and thus 

1129: 135 sources are new.

1130: For the 308 sources that were detected in the main catalog of A03,

1131: we find general agreement between the derived \hbox{X-ray}

1132: properties presented here and in A03. For example, we have compared the

1133: full-band count rates of these 308 sources between the two catalogs.

1134: The median ratio of the count rates is $\approx0.98$

1135: with an interquartile range of $\approx$0.85--0.12.

1136: Furthermore, the approximately doubled exposure improves the 

1137: source positions and spectral constraints significantly, and thus the 

1138: $\approx$2~Ms \hbox{CDF-S} catalogs presented here supersede those in A03.

1139: 

1140: Eighteen of the 326 sources detected in the 

1141: main catalog of A03

1142: are undetected here. Nine of these were detected in

1143: {\sc wavdetect} runs with a false-positive probability threshold of

1144: $1\times 10^{-5}$ in the present analysis. 

1145: The other nine sources were weakly detected in A03

1146: with less than 17 full-band counts. We examined the

1147: regions of these nine sources in the three $\approx2$~Ms images and found no

1148: emission clearly distinct from the background. 

1149: Ten of the eighteen sources have optical counterparts in the

1150: WFI $R$-band source catalog within $1\farcs3$, and three of them

1151: are present in the

1152: supplementary optically bright \chandra\ catalog (see \S3.3.3), suggesting

1153: that they are likely true \hbox{X-ray} sources.

1154: As the second $\approx1$~Ms

1155: exposure was taken $\approx7$ years later, these eighteen sources could be

1156: below our detection limit due to source variability or background 

1157: fluctuations. A 30\% median flux variability has been observed for sources 

1158: in the first $\approx1$~Ms data set \citep{Paolillo2004}, which is expected

1159: to increase here owning to the long observation interval. 

1160: There is also the possibility that some of the missing sources were false

1161: detections in A03, since $\approx$3--9 false detections were expected (A03).

1162: 

1163: Four of the 304 sources in the main catalog of G02 are not detected here, 

1164: two of which were detected in

1165: {\sc wavdetect} runs with a false-positive probability threshold of

1166: $1\times 10^{-5}$. All four sources lie at large off-axis angles, and none of 

1167: them is in the A03 main catalog. These sources could be

1168: below our detection limit due to source variability or background

1169: fluctuations. Note that 19 G02 sources that were not detected in A03 

1170: are detected here, suggesting that these are likely true sources. 

1171: These sources were probably not reported in the A03 main catalog

1172: due to the conservative {\sc wavdetect} false-positive 

1173: probability threshold ($1\times 10^{-7}$) adopted in that work.

1174: 

1175: 

1176: In Table~4 we summarize 

1177: the number of sources detected in one band but not

1178: another. There are three sources detected only in the hard band. 

1179: For comparison, there is one source in the $\approx1$~Ms \hbox{CDF-S}

1180: that was detected only in the hard band (A03).

1181: In Figure~\ref{cnthist} we show the distributions of detected 

1182: counts in the three standard

1183: bands. The median numbers of counts for the full band, soft band and hard 

1184: band are $\approx101$, $\approx53$ and $\approx89$, respectively.

1185: There are 202 sources with $>100$ full-band counts, for which basic

1186: spectral analyses are possible, and 33 sources with $>1000$ full-band

1187: counts. 

1188: In Figure~\ref{fluxhist} we show the distributions of \hbox{X-ray} flux 

1189: in the three standard

1190: bands. The \hbox{X-ray} fluxes in this survey span roughly four orders of

1191: magnitude, with $\approx$50\% of the sources having soft-band and hard-band

1192: fluxes of less than $2.5\times 10^{-16}$ \flux\ and 

1193: $1.7\times 10^{-15}$ \flux, respectively.

1194: %

1195: \begin{figure}

1196: \centerline{\includegraphics[scale=0.5]{f07.ps}

1197: } \figcaption{Histograms showing the distributions of detected source 

1198: counts for sources in the main

1199: {\it Chandra} catalog in the full ({\it top}), soft ({\it middle}), and 

1200: hard ({\it

1201: bottom}) bands. Sources with upper limits have not been included in these

1202: diagrams.

1203: The vertical dotted lines indicate median numbers of counts in each band

1204: (see Table~3).

1205: \label{cnthist}}

1206: \end{figure}

1207: %

1208: \begin{figure}

1209: \centerline{\includegraphics[scale=0.5]{f08.ps}} 

1210: \figcaption{Histograms showing the distributions of \hbox{X-ray} fluxes for 

1211: sources in the main

1212: {\it Chandra} catalog in the full ({\it top}), soft ({\it middle}), 

1213: and hard ({\it

1214: bottom}) bands. Sources with upper limits have not been included in this figure.

1215: The vertical dotted lines indicate the median fluxes of 

1216: $1.3\times$~10$^{-15}$, $2.5\times$~10$^{-16}$ and $1.7\times$~10$^{-15}$~\flux\

1217: for the full, soft, and hard bands, respectively.

1218: \label{fluxhist}}

1219: \end{figure}

1220: %

1221: 

1222: In Figure~\ref{ps} we show ``postage-stamp'' images from the WFI 

1223: $R$-band image with

1224: adaptively smoothed full-band contours overlaid for sources in the

1225: main {\it Chandra} catalog. The wide range of \hbox{X-ray} source sizes

1226: observed in these images is largely due to PSF broadening with off-axis angle.

1227: Figure~\ref{pos}{\it a} shows the positions of sources detected in the main {\it

1228: Chandra} catalog. The source density is highest 

1229: close to the average aim point where the sensitivity is highest. 

1230: Different symbol sizes represent different significances of

1231: source detection with {\sc wavdetect} (see column~48 of Table~2).

1232: New \hbox{X-ray} sources that are not present in the 

1233: G02 or A03 main catalogs are indicated as filled circles; 135 new 

1234: sources are detected, of which 15 lie outside the solid-angle 

1235: coverage of the first $\approx1$~Ms exposure.

1236: 

1237: 

1238: %

1239: \begin{figure}

1240: \centerline{\includegraphics[scale=0.45]{f09a.ps}}

1241: \figcaption{WFI $R$-band postage-stamp images for the sources in the main {\it Chandra}

1242: catalog with full-band adaptively smoothed \hbox{X-ray} contours overlaid. The

1243: contours are logarithmic in scale and range from \hbox{$\approx$0.003\%--30\%} 

1244: of the maximum pixel value. The label at the top of each image gives the 

1245: source name, which is composed of

1246: the source coordinates, while numbers at the bottom left and right-hand corners

1247: correspond to the source number (see column~1 of Table~2) and the full-band 

1248: counts or upper limits (with a ``$<$'' sign) on the full-band counts, 

1249: respectively. 

1250: In several cases no \hbox{X-ray} contours are 

1251: present, either because these sources were not detected in the full band or 

1252: the full-band counts are low and {\sc

1253: csmooth} has suppressed the observable emission in the adaptively smoothed

1254: images. Each image is $25\arcsec$ on a side, and the

1255: source of interest is always located at the center of the image. Only one of

1256: the 8 pages of cutouts is included here; all 8 pages are

1257: available in the electronic edition.

1258: \label{ps}}

1259: \end{figure}

1260: %

1261: 

1262: %

1263: \begin{figure*}

1264: \centerline{\includegraphics[scale=0.4]{f10a.ps}

1265: \includegraphics[scale=0.4]{f10b.ps}}

1266: \figcaption{

1267: Positions of the sources in ({\it a}) the main {\it Chandra} catalog

1268: and ({\it b}) the supplementary {\it Chandra} catalogs.

1269: Circles represent \hbox{X-ray} sources in ({\it a}) the main \chandra\

1270: catalog and ({\it b}) the supplementary CDF-S plus E-CDF-S \chandra\ catalog.

1271: Open circles represent \hbox{X-ray} sources that were

1272: previously detected in ({\it a}) the main catalogs of G02 or A03

1273: and ({\it b}) the main catalogs of G02, A03, or L05. Filled circles

1274: represent new sources.

1275: Sizes indicate the maximum detection significance corresponding to

1276: {\sc wavdetect} false-positive probability detection thresholds of 

1277: $1\times10^{-8}$ ({\it large circles}),

1278: $1\times10^{-7}$ ({\it medium circles}), and $1\times10^{-6}$ 

1279: ({\it small circles}).

1280: Sources in the optically bright catalog

1281: are shown as open triangles (previously detected in the main catalog of

1282: A03) and filled triangles (new sources) in ({\it b}).

1283: For sources in the CDF-S plus E-CDF-S catalog, 

1284: their detection significances are preferentially higher near the edge of 

1285: the field due to the contribution of the

1286: \hbox{E-CDF-S} exposure.

1287: The regions and the cross symbol have the same meaning as those in

1288: Fig. \ref{fbimg}.

1289: \label{pos}}

1290: \end{figure*}

1291: %

1292: 

1293: Figure~\ref{bratio} shows the band ratio as a function of full-band count

1294: rate for sources in the main \chandra\ catalog. 

1295: We also derived average band ratios by stacking the individual 

1296: sources together using a procedure similar to that of 

1297: \citet{Lehmer2008}.

1298: The average

1299: band ratio rises at lower count rates. The corresponding 

1300: average photon index flattens from $\Gamma\approx1.8$ to 

1301: $\Gamma\approx0.8$ for full-band count rates of $\approx10^{-2}$

1302: to $\approx2\times10^{-4}$ counts s$^{-1}$.

1303: This trend has been

1304: reported in other studies (e.g., \citealt{Tozzi2001}; A03; L05) and 

1305: is due to an increase in the number of absorbed AGNs 

1306: detected at fainter fluxes. 

1307: The average photon index does not continue getting flatter 

1308: below full-band count rates of

1309: $\approx2\times10^{-4}$ counts s$^{-1}$, probably due to the increased 

1310: contribution from normal and starburst galaxies at these lowest count rates

1311: \citep{Bauer2004}.

1312: In Figure~\ref{fox}{\it a} we show the WFI $R$-band magnitude versus 

1313: soft-band flux for \hbox{X-ray} sources

1314: in the main catalog, as well as  

1315: the approximate flux ratios for AGNs and galaxies 

1316: \citep[e.g.,][]{Maccacaro1988,Stocke1991,Hornschemeier2001,Bauer2004}.

1317: More than half (304) of the \hbox{X-ray}

1318: sources lie in the region expected for AGNs, 74 of which are new sources.

1319: A significant minority (158) of the sources lie in the region for normal and 

1320: starburst galaxies, 61 of which are new sources. The new sources 

1321: have an increased fraction of normal and starburst galaxies.

1322: This source characterization, based only on the X-ray--to--optical flux 

1323: ratio, is only approximate and will be refined in future studies.

1324: 

1325: %

1326: \begin{figure}

1327: \centerline{\includegraphics[scale=0.5]{f11.ps}

1328: }

1329: \figcaption{

1330: Band ratio vs. full-band count rate for sources in the main \chandra\ 

1331: catalog. Open circles represent \hbox{X-ray} sources that were

1332: detected in the main catalogs of G02 or A03.

1333: Filled circles represent new sources.

1334: Plain arrows indicate upper or lower limits. Sources detected only

1335: in the full band cannot be plotted. 

1336: The open stars show average band ratios as a function of full-band

1337: count rate derived from stacking analyses.

1338: Horizontal dotted lines show the band ratios 

1339: corresponding to given effective photon indices; these were 

1340: calculated using PIMMS.

1341: \label{bratio}}

1342: \end{figure}

1343: %

1344: 

1345: 

1346: %

1347: \begin{figure*}

1348: \centerline{\includegraphics[scale=0.5]{f12a.ps}

1349: \includegraphics[scale=0.5]{f12b.ps}

1350: }

1351: \figcaption{

1352: WFI $R$-band magnitude vs. soft-band flux for \hbox{X-ray} sources

1353: in ({\it a}) the main catalog 

1354: and ({\it b}) the supplementary optically bright catalog. 

1355: Open circles represent \hbox{X-ray} sources that were

1356: detected in the main catalogs of G02 or A03.

1357: Filled circles represent new sources.

1358: Sources without an optical counterpart are plotted as upward arrows.

1359: Diagonal lines indicate constant flux ratios. The shaded areas show the 

1360: approximate flux ratios for AGNs (dark gray) and galaxies (light gray).

1361: \label{fox}}

1362: \end{figure*}

1363: %

1364: 

1365: 

1366: 

1367: \subsubsection{Supplementary CDF-S plus E-CDF-S Chandra Source Catalog}

1368: 

1369: We can gain additional sensitivity in the outer

1370: portions of the $\approx2$~Ms CDF-S footprint by including 

1371: the $\approx250$~ks \hbox{E-CDF-S} 

1372: (L05) observations. To this end, we processed and registered 

1373: the \hbox{E-CDF-S} exposures in the same manner as 

1374: our \hbox{CDF-S} observations. Notably, because of

1375: the different coverage of the \hbox{CDF-S} and \hbox{E-CDF-S} 

1376: (see Figure~2 of L05), the PSF sizes for the \hbox{E-CDF-S} near

1377: the average aim point for the CDF-S are substantially

1378: larger than those for the \hbox{CDF-S}.

1379: The \hbox{E-CDF-S} will likely only contribute additional background for

1380: all but the strongest sources around the center of the field. 

1381: Thus, we excluded the \hbox{E-CDF-S} event

1382: lists within 4$\arcmin$ of the \hbox{CDF-S} average aim point. 

1383: We also masked out portions of the \hbox{E-CDF-S} where the \hbox{CDF-S}

1384: exposure time was zero. Images and exposure maps were cropped in 

1385: a similar manner.  

1386: 

1387: We ran {\sc wavdetect} with a false-positive probability

1388: threshold of $1\times 10^{-6}$ on the three standard-band images 

1389: for the combined \hbox{CDF-S} plus \hbox{E-CDF-S}, detecting 86

1390: sources not present in the main \chandra\ source

1391: catalog. 

1392: The positions of these sources have been improved following the procedure

1393: described in \S3.3.1.

1394: Due to the drastically different overlapping PSFs, the derived 

1395: properties of these

1396: \hbox{X-ray} sources are not as reliable as those in the main catalog. 

1397: Therefore

1398: we present these sources in Table~\ref{tbl-sp1} as 

1399: a supplementary CDF-S plus E-CDF-S 

1400: \chandra\ source catalog. For sources already detected in 

1401: the \hbox{E-CDF-S} (L05), we took the photometry data from L05 directly.

1402: For new sources, photon counts and effective exposure times

1403: were extracted separately from the \hbox{CDF-S} and \hbox{E-CDF-S} 

1404: data sets and then summed to give a total

1405: number of counts and a total effective exposure time.

1406: The format of Table~\ref{tbl-sp1} is very

1407: similar to that of Table~2, with a few details given below.

1408: 

1409: \begin{enumerate}

1410: 

1411: \item Columns~1--28: the format of these columns is exactly the same as 

1412: that of columns~1--28 in Table~2, so the column descriptions in \S3.3.1 

1413: are applicable.

1414: Note that for sources detected in the \hbox{E-CDF-S} (see column~29 or 52), 

1415: the source

1416: counts and their uncertainties were taken from L05 directly.

1417: 

1418: \item Column~29: the corresponding $\approx$250~ks \hbox{E-CDF-S} source

1419: number from the main

1420: \chandra\ catalog presented in L05 (see column~1 of Table~2 in L05). We

1421: matched our \hbox{X-ray} source positions to L05 source positions using

1422: a matching radius that is the

1423: quadratic sum of

1424: the $\approx3\sigma$ positional errors of the \hbox{CDF-S}

1425: and L05 \hbox{X-ray} sources.

1426: The $3\sigma$ positional error of a \hbox{CDF-S}

1427: source is approximately

1428: twice the positional error quoted in column~4, and that of an L05

1429: source is approximately

1430: twice the positional error quoted in Table~2 of L05.

1431: Only one L05 match was found for each matched source.

1432: Sources with no L05 match have a value of ``$-1$''.

1433: 

1434: \item Columns~30 and 31: the right ascension and declination of

1435: the corresponding

1436: L05 source indicated in column~29. Sources with no L05 match have right

1437: ascension and

1438: declination values set to \hbox{``00 00 00.00''} and \hbox{``$-$00 00 00.0''}.

1439: 

1440: \item Columns~32--51: the format of these columns is exactly the same as

1441: that of \hbox{columns~29--48} in Table~2, so the column descriptions in \S3.3.1 

1442: are applicable. Note that for sources detected in the \hbox{E-CDF-S} 

1443: (see column~29 or 52), the source exposure times, band ratios, photon indices,

1444: and fluxes were taken from L05 directly.

1445: 

1446: \item Column~52: notes on the sources. ``L'' refers to sources that 

1447: were detected in the $\approx$250~ks \hbox{E-CDF-S} (L05). 

1448: 

1449: \end{enumerate}

1450: 

1451: The 86 CDF-S plus E-CDF-S sources have 

1452: effective exposures up to $\approx1.9$~Ms. 

1453: Their

1454: positional uncertainties were estimated following equation (1),

1455: though the positional accuracy of the off-axis sources will often have been

1456: improved due to the small PSF sizes of the \hbox{E-CDF-S}. 60 ($\approx70\%$)

1457: of the sources have optical counterparts. Two of the 86 sources have 

1458: counterparts in 

1459: the A03 main catalog and another two have counterparts 

1460: in the G02 main catalog. In addition, 53 of the sources were detected in 

1461: the main catalog of L05. There are thus 30 new sources in this 

1462: supplementary catalog.

1463: 50 ($\approx57\%$) and 41 ($\approx47\%$) of these

1464: sources are detected with false-positive probability thresholds of

1465: 1~$\times$~10$^{-7}$ and 1~$\times$~10$^{-8}$, respectively.

1466: 

1467: Figure~\ref{pos}{\it b} shows the positions of sources detected in the 

1468: supplementary \hbox{CDF-S} plus \hbox{E-CDF-S} catalog. 

1469: Different symbol sizes represent different significances of the

1470: source detection with {\sc wavdetect} (see column~51 of Table~\ref{tbl-sp1}).

1471: 

1472: \subsubsection{Supplementary Optically Bright Chandra Source Catalog}

1473: 

1474: Since the density of optically bright sources on the sky is comparatively

1475: low, we constructed a supplementary \chandra\ source catalog including 

1476: \hbox{X-ray} sources detected at a lower \hbox{X-ray} significance 

1477: threshold than that used in the main catalog and having bright optical

1478: counterparts. We ran {\sc wavdetect} with a false-positive probability

1479: threshold of $1\times 10^{-5}$ on the three \hbox{CDF-S} images, 

1480: and we found 132 lower

1481: significance \hbox{X-ray} sources not present in the main \chandra\ source

1482: catalog or the supplementary \hbox{CDF-S} plus \hbox{E-CDF-S}

1483: catalog.

1484:  

1485: Bright optical sources were selected from the WFI $R$-band source 

1486: catalog described in \S3.1, with an $R$-band magnitude brighter than 

1487: 23.8. This $R$-band cutoff was empirically determined to provide a good 

1488: balance between the number of detected sources and the

1489: expected number of false sources. We

1490: searched for bright optical counterparts to the low-significance 

1491: \hbox{X-ray} sources using a matching radius of $1\farcs 3$.

1492: A matching radius of $1\farcs3$ was chosen as a compromise between having too

1493: few matches and too many false matches.

1494: In total 30 optically bright \hbox{X-ray} sources were found. 

1495: We estimated the expected number of false matches by

1496: manually shifting the \hbox{X-ray} source coordinates in right

1497: ascension and declination by $5\farcs0$ and $10\arcsec$ 

1498: (both positive and negative

1499: shifts) and recorrelating with the optical sources. On average, the

1500: number of false matches is $\approx3$ ($\approx10\%$), demonstrating 

1501: that the majority

1502: of the 30 \hbox{X-ray} matches are real \hbox{X-ray} sources. 

1503: 

1504: The supplementary optically bright \chandra\ source catalog is presented in

1505: Table~\ref{tbl-sp2}. These sources typically have 4--35 counts in the 

1506: band in which they were detected.

1507: The format of Table~\ref{tbl-sp2} is 

1508: similar to that of Table~2, 

1509: with the details of the columns given below.

1510: 

1511: \begin{enumerate}

1512: 

1513: \item

1514: Column~1: the source number. Sources are listed in order of increasing

1515: right ascension.

1516: 

1517: \item Columns~2 and 3: the right ascension and declination of 

1518: the \hbox{X-ray} source,

1519: respectively. The {\sc wavdetect} positions are used here for these faint

1520: \hbox{X-ray} sources. Whenever possible, we have quoted

1521: the position determined in

1522: the full band; when a source is not detected in the full band, we used, in

1523: order of priority, the soft-band position or hard-band position.

1524: 

1525: \item Column~4: the positional uncertainty. For these faint

1526: \hbox{X-ray} sources, the positional uncertainty is set to $1\farcs2$, 

1527: the approximate 90th percentile of the optical--X-ray positional offsets 

1528: given in column~17.

1529: 

1530: \item Column~5: the off-axis angle of the \hbox{X-ray} source in arcminutes 

1531: (see column~5 of Table~2 for details).

1532: 

1533: \item Columns~6--14: the source counts and the corresponding $1\sigma$

1534: statistical errors \citep{Gehrels1986} or the upper limits on source counts

1535: for the three standard bands, respectively.

1536: When a source is detected in a given band, the photometry is taken 

1537: directly from {\sc wavdetect}. When a source is not detected, an upper limit

1538: is calculated (see columns~6--14 of Table~2 for details).

1539: 

1540: \item Columns~15 and 16: the right ascension and declination of the 

1541: optical counterpart.

1542: 

1543: \item Column~17: the measured offset between the optical and \hbox{X-ray}

1544: sources in arcseconds.

1545: 

1546: \item

1547: Column~18: the $R$-band AB magnitude of the optical counterpart.

1548: 

1549: \item

1550: Columns~19--26: the $i$, $z$, and $K_s$ band AB magnitudes and the IRAC

1551: 5.8 $\mu$m flux density of the optical counterpart, and the correspoding

1552: source ID in the optical and infrared catalogs (see columns~19--26

1553: of Table~2 for details).

1554: 

1555: \item

1556: Columns~27 and 28: the corresponding spectroscopic redshift

1557: and the reference for the redshift (see columns~27 and 28 of Table~2 

1558: for details).

1559: 

1560: 

1561: \item Column~29: the corresponding $\approx$1~Ms CDF-S source number 

1562: from the main

1563: \chandra\ catalog presented in A03 (see column~1 of Table~3a in A03).  

1564: We used a

1565: matching radius that is the quadratic sum of the $\approx3\sigma$ 

1566: positional errors of the

1567: \hbox{CDF-S} and A03 \hbox{X-ray} sources.  

1568: The $3\sigma$ positional error of a \hbox{CDF-S}

1569: source is $\approx1\farcs3$, and that of an A03

1570: source is approximately

1571: twice the positional error quoted in Table~A2a of A03.

1572: Only one A03 match was found for each matched source. Supplementary sources 

1573: with no A03 match have a value of ``$-1$''. There are no matches to the 

1574: main source catalog in G02, so we do not list the match results in this

1575: table.

1576: 

1577: \item Columns~30 and 31: the right ascension and declination of the 

1578: corresponding A03 source indicated in column~29. Sources with no A03 match have right ascension and declination values set to \hbox{``00 00 00.00''} 

1579: and \hbox{``$-$00 00 00.0''}.

1580: 

1581: 

1582: \item Columns~32--34: the effective exposure times derived from the

1583: standard-band exposure maps.

1584: 

1585: \item Column~35: the photon index used to calculate source fluxes

1586: (columns~36--38). We used a constant photon index of $\Gamma=2.0$ since our

1587: source-selection technique preferentially selects objects with flux-ratios

1588: $f_{\rm 0.5-2.0~keV}/f_R < 0.1$, which are observed to have effective photon

1589: indices of $\Gamma \approx 2$ (e.g., $\S$~4.1.1 of Bauer \etal~2004).

1590: 

1591: \item Column~36--38: observed-frame fluxes in the three standard bands;

1592: quoted fluxes are in units of \flux\ and have been calculated assuming

1593: $\Gamma=2.0$. The fluxes have not been corrected for absorption by the Galaxy

1594: or material intrinsic to the sources (see \hbox{columns~45--47} of Table~2 for

1595: details).

1596: 

1597: \end{enumerate}

1598: 

1599: 

1600: The WFI $R$-band magnitudes of these supplementary sources 

1601: span $R=$18.7--23.8.

1602: In Figure~\ref{fox}{\it b} we show the $R$-band magnitude versus

1603: soft-band flux for the 30 optically bright \hbox{X-ray} sources.

1604: The approximate flux ratios for AGNs and galaxies are also plotted.

1605: The majority of the sources have the X-ray--to--optical flux ratios expected

1606: for normal and starburst galaxies.

1607: Some of these sources may be low-luminosity AGNs; only one source is 

1608: detected in the hard band, suggesting that they are

1609: unlikely to be luminous absorbed AGNs.

1610: Note that the supplementary optically bright sources are not 

1611: representative of the faintest \hbox{X-ray} sources as a whole, because our

1612: selection criteria preferentially select optically bright and \hbox{X-ray}

1613: faint non-AGNs (e.g., A03; \citealt{Hornschemeier2003}).

1614: The positions of the sources in the

1615: supplementary optically bright catalog are shown in Figure~\ref{pos}{\it b}.

1616: 

1617: 

1618: 

1619: \section{BACKGROUND AND SENSITIVITY ANALYSIS}

1620: Background maps were created for the three standard bands. We 

1621: first masked out the point sources from the main \chandra\ catalog 

1622: using apertures with radii twice

1623: that of the $\approx$90\% PSF encircled-energy radii; approximately

1624: 12\% of the pixels were masked out. The resultant images

1625: should include minimum contributions from detected point sources.

1626: However, they will include contributions from a few extended sources 

1627: \citep[e.g.,][]{Bauer2002}, 

1628: which will cause a slight overestimation of the measured

1629: background. 

1630: Even with a $\approx2$~Ms exposure, about 79\%

1631: of the pixels have no background counts in the full band.

1632: For such a small number of detected counts per pixel, the 

1633: expected counts distribution is Poissonian. 

1634: We compared the background-count 

1635: distributions to Poisson distributions with the mean number

1636: of background counts per pixel using the Kolmogorov-Smirnov test, 

1637: and we found them to be very similar in 

1638: all three standard bands for various regions across the survey field

1639: (see $\S$4.2 of A03 for more details on the tests).

1640: We filled in the masked regions for each source with a local background

1641: estimate by constructing a probability distribution of counts 

1642: using an annulus with

1643: inner and outer radii of 2 and 4 times the $\approx$90\% PSF encircled-energy

1644: radius, respectively. 

1645: The background properties are summarized in

1646: Table~\ref{tbl-bkg}. The total background includes contributions

1647: from the unresolved cosmic background, particle background, and instrumental

1648: background \citep[e.g.,][]{Markevitch2001,Markevitch2003}. For our analyses we

1649: are only interested in the total background and do not distinguish between

1650: these different components.

1651: The mean background count rates are $\approx20\%$--$30\%$ higher

1652: compared to the $\approx2$~Ms \hbox{CDF-N} (A03) 

1653: or the $\approx250$~ks \hbox{E-CDF-S} (L05), which are reasonable 

1654: variations given the variability of the particle and instrumental

1655: background components over the past several years. 

1656: 

1657: The faintest sources in the main \chandra\ catalog have $\approx5$

1658: counts in the soft band and $\approx8$ counts in the hard band 

1659: (see Table~\ref{tbldet}).

1660: For a $\Gamma=$~1.4 power law with Galactic absorption, the corresponding

1661: soft-band and hard-band fluxes at the average 

1662: aim point are $\approx1.6\times10^{-17}$

1663: \flux\ and $\approx9.0\times10^{-17}$ \flux,

1664: respectively. This provides an estimate of the ultimate sensitivity of this 

1665: survey.

1666: However, these numbers are only relevant for a small area close to the 

1667: average aim

1668: point. To determine the sensitivity across the field it is necessary

1669: to take into account the broadening of the PSF with off-axis angle, 

1670: as well as

1671: changes in the effective exposure and background rate across the field. 

1672: Following L05, we

1673: estimated the sensitivity across the field by employing a Poisson model,

1674: The resulting relation can be approximately represented by

1675: 

1676: \begin{equation}

1677: \log N~=\alpha + \beta \log b + \gamma (\log b)^2+ \delta (\log b)^3

1678: \label{eqsen}

1679: \end{equation}

1680: 

1681: \noindent where $N$ is the required number of counts for detection, and

1682: $b$ is the number of background counts in a source cell; $\alpha=0.917$,

1683: $\beta=0.414$, $\gamma=0.0822$, and $\delta=0.0051$ are fitting constants.

1684: For the sensitivity calculations here, we measured the number of background 

1685: counts $b$ in the background

1686: maps using an aperture size of 70\% of the PSF encircled-energy radius. The 

1687: 70\% encircled-energy radius was chosen as a compromise between having too

1688: few source counts and too many background counts. 

1689: 

1690: Following equation (\ref{eqsen}), 

1691: we constructed sensitivity maps using the background and

1692: exposure maps, assuming a $\Gamma=$~1.4 power-law model with Galactic

1693: absorption. Since we do not filter

1694: out detected sources with our sensitivity maps, a small fraction 

1695: of sources have

1696: fluxes slightly below

1697: these sensitivity limits (4 sources in the full band, 

1698: 14 sources in the soft band, and 7 sources in the

1699: hard band). 

1700: The full-band sensitivity map is shown in Figure~\ref{senmap},

1701: and in

1702: Figure~\ref{senhist} we show plots of solid angle versus flux limit 

1703: for the full, soft,

1704: and hard bands. The $\approx$1~arcmin$^2$ region at the average 

1705: aim point has

1706: soft-band and hard-band sensitivity limits of

1707: $\approx1.9\times10^{-17}$ \flux\ and

1708: $\approx1.3\times10^{-16}$ \flux, respectively. Solid angles for

1709: the $\approx2$~Ms \hbox{CDF-N} have been plotted for comparison in

1710: Figure~\ref{senhist} (dotted curves),

1711: which appear to be similar to those for the \hbox{CDF-S}.\footnote{

1712: The \hbox{CDF-N} sensitivity limits were calculated following the same method

1713: described above.}

1714: 

1715: %

1716: \begin{figure}

1717: \centerline{\includegraphics[scale=0.4]{f13.ps}}

1718: \figcaption{

1719: Full-band sensitivity map of the 2~Ms \hbox{CDF-S}. This

1720: sensitivity map has been created following $\S$4. 

1721: The gray-scale levels (from

1722: black to light gray) represent areas with flux limits (in units of \flux) 

1723: of $<10^{-16}$, \hbox{$10^{-16}$--$3.3\times10^{-16}$}, \hbox{$3.3\times

1724: 10^{-16}$--$10^{-15}$}, and $>10^{-15}$, respectively. 

1725: The regions and the cross symbol have the same meaning as those in

1726: Fig.~\ref{fbimg}.

1727: \label{senmap}}

1728: \end{figure}

1729: %

1730: 

1731: %

1732: \begin{figure}

1733: \centerline{\includegraphics[scale=0.5]{f14.ps}}

1734: \figcaption{

1735: Survey solid angle as a function of the flux limit for the 

1736: full ({\it top}), soft

1737: ({\it middle}), and hard ({\it bottom}) bands, determined following $\S$4.

1738: Data are plotted as solid curves for the $\approx2$~Ms \hbox{CDF-S}, 

1739: and as dotted curves for the $\approx2$~Ms \hbox{CDF-N}.

1740: The flux limits at the average aim point of the \hbox{CDF-S}

1741: are $\approx7.1\times10^{-17}$

1742: \flux\ (full band), $\approx1.9\times10^{-17}$ \flux\ (soft band), 

1743: and $\approx1.3\times10^{-16}$ \flux\ (hard band).

1744: \label{senhist}}

1745: \end{figure}

1746: %

1747: 

1748: 

1749: 

1750: \section{NUMBER COUNTS FOR THE MAIN {\bf {\em CHANDRA}} CATALOG}

1751: Cumulative number counts, $N(>S)$, for the soft and hard

1752: bands were calculated for the $\approx2$~Ms \hbox{CDF-S}.

1753: To quantify the effects of incompleteness and bias, 

1754: we took a similar approach to the one in \citet{Bauer2004} and created

1755: 200 Monte Carlo simulated observations in both the soft and hard bands.

1756: We added simulated sources at random positions to the background maps

1757: described in \S4. The fluxes of these simulated sources were drawn randomly

1758: from the total number-count models of \citet{Moretti2003} between

1759: $1.6\times10^{-17}$ and $10^{-11}$ \flux\ in the soft band 

1760: and $9\times10^{-17}$ and $10^{-11}$ \flux\ in the hard band. These

1761: fluxes were converted to \hbox{X-ray} photon counts using the exposure maps

1762: and a photon index of $\Gamma=1.4$. Statistical errors were added to 

1763: the counts to account for the effect of Eddington bias. Finally, counts

1764: for each simulated source were added to the background map following

1765: a PSF probability distribution function derived from the combined model PSF

1766: of the nearest real X-ray source in the main catalog. These model PSFs

1767: were produced using AE.

1768: 

1769: Source searching and photometry were performed on the simulated images 

1770: using the same method as that used to produce the main catalog. A 

1771: completeness correction factor ($F$) was estimated by comparing the number 

1772: of simulated input sources with the number of simulated detected sources as a 

1773: function of detected counts. A flux recovery correction factor ($R$)

1774: was calculated by comparing the simulated input counts with simulated measured

1775: counts. The correction factors are position- and count-dependent. 

1776: For each of the 462 X-ray sources in the main catalog, we determined

1777: the two correction factors based on a sample of simulated 

1778: sources within $2\arcmin$ of the source position and having similar exposure

1779: times. 

1780: Sources close to the edge of the survey field are not well sampled, and 

1781: thus we calculated cumulative number

1782: counts using only the 428 X-ray sources that are located

1783: within $10\arcmin$ of the average aim point.

1784: The completeness and flux 

1785: recovery corrections remain close to unity above $\sim$50--100 counts.

1786: Below this point, \chandra's varying PSF size and spatially dependent

1787: vignetting begin to affect source detection and photometry.

1788: 

1789: We set our minimum flux levels to $3\times10^{-17}$ \flux\ in

1790: the soft band and $2.5\times10^{-16}$ \flux\ in the hard band.

1791: These limits were chosen since at lower fluxes there are less than 10--15

1792: additional sources contributing to the number counts, and thus the 

1793: number counts at fainter levels have large uncertainties.

1794: The cumulative number of

1795: sources, $N(>S)$, brighter than a given flux, $S$, weighted by the appropriate

1796: aerial coverage, is

1797: \begin{equation}

1798: N(>S) = \sum_{S_i > S} (F_i\Omega_i)^{-1}~,

1799: \end{equation}

1800: \noindent where $\Omega_i$ is the maximum solid angle for which a source with

1801: flux, $S_i$, could be detected. Each flux $S$ has been corrected for flux bias

1802: assuming

1803: \begin{equation}

1804: S_i=R_iS_i^0~,

1805: \end{equation}

1806: \noindent where $S_i^0$ is the original flux quoted in the main catalog.

1807: The maximum solid angles were computed

1808: using the inner $10\arcmin$ radius regions of the sensitivity maps. 

1809: We have also calculated $1\sigma$ errors for 

1810: the cumulative distributions following Gehrels (1986).

1811: 

1812: Figure~\ref{srccounts} displays the cumulative

1813: number counts and the corresponding $1\sigma$ errors for 

1814: the main \chandra\ catalog.

1815: Cumulative number counts for several other surveys have also been shown for

1816: comparison. The derived $\approx$2~Ms \hbox{CDF-S} cumulative

1817: number counts are in general agreement with previous survey results

1818: for the $\approx$1~Ms \hbox{CDF-S}  

1819: \citep{Rosati2002} and the $\approx$250~ks \hbox{E-CDF-S} (L05), at

1820: around the $1\sigma$ confidence level over the entire flux range in

1821: the soft and hard bands. The apparent deviation between the $\approx$2~Ms

1822: and $\approx$1~Ms \hbox{CDF-S} soft-band number counts mainly comes from

1823: the difference in the count-rate--to--flux conversion factor used in these

1824: two surveys.\footnote{An average photon index of $\Gamma=1.4$ was used to 

1825: calculate fluxes in \citet{Rosati2002}, while in this survey, the photon

1826: index was estimated for each source separately and so was the 

1827: count-rate--to--flux conversion factor (see \S3.3.1). We did a test by

1828: calculating the soft-band fluxes using the conversion factor given by 

1829: \citet{Rosati2002}. The derived fluxes are $\sim90\%$ of those 

1830: presented in the main catalog,

1831: and the resulting soft-band number counts are consistent with those for

1832: the $\approx$1~Ms \hbox{CDF-S} to within $1\sigma$.}

1833: The {\it XMM-Newton} observations in the

1834: COSMOS field \citep{Cappelluti2007} provide similar  

1835: number counts, though 

1836: not as deep as the \hbox{CDF-S} observations.

1837: 

1838: To make a consistent comparison with the $\approx$2~Ms \hbox{CDF-N} 

1839: number counts, we analyzed the \hbox{CDF-N} observations in the same way

1840: as in this paper. A main catalog of 575 \hbox{X-ray} sources 

1841: was constructed. Number counts were calculated using the 496 

1842: X-ray sources located within $10\arcmin$ of the average aim point, and these

1843: have been corrected for incompleteness and flux bias based on simulations.

1844: The \hbox{CDF-N} cumulative number counts  

1845: are presented in Figure~\ref{srccounts} (dotted curves), along with the 

1846: ratios of the \hbox{CDF-S} to \hbox{CDF-N} number counts.

1847: In the soft band, the $\approx$2~Ms \hbox{CDF-S} number counts 

1848: appear to be 

1849: consistent with those for the $\approx$2~Ms \hbox{CDF-N} 

1850: to within $\approx1\sigma$ at fluxes above $\approx2\times10^{-16}$~\flux.

1851: Small differences (up to $\approx3\sigma$) exist at fainter fluxes.

1852: In the hard band, the \hbox{CDF-N} number counts 

1853: deviate above the $1\sigma$ errors of the \hbox{CDF-S} number counts 

1854: at fluxes below $\approx2\times10^{-15}$~\flux; the difference at 

1855: the faintest fluxes is $\approx$25\% ($\approx3\sigma$).

1856: Similar findings of differences between the \hbox{CDF-N} and \hbox{CDF-S}

1857: number counts have been

1858: reported in previous studies \citep[e.g.,][]{Cowie2002,Moretti2003,Bauer2004},

1859: and it appears that this results from small field-to-field variations.

1860: Such field-to-field variations

1861: are generally believed to arise from the large-scale structure underlying the

1862: cosmic \hbox{X-ray} source distribution \citep[e.g.,][]{Gilli2003,Yang2003}.

1863: 

1864: %

1865: \begin{figure*}

1866: \centerline{

1867: \includegraphics[scale=0.59]{f15a.ps}

1868: \includegraphics[scale=0.59]{f15b.ps}

1869: }

1870: \figcaption{

1871: {\it Top}: Number of sources, $N(>S)$, brighter than a given flux, $S$, for 

1872: the ({\it a}) soft band and ({\it b}) hard band. 

1873: The $\approx$2~Ms \hbox{CDF-S} data 

1874: are plotted as black solid curves with the 1$\sigma$ errors

1875: plotted as gray shaded areas. The cumulative

1876: number counts were computed using 428 \hbox{X-ray} sources in the main 

1877: \chandra\ catalog that are located within $10\arcmin$ of the average 

1878: aim point, and have been corrected for incompleteness and flux bias.

1879: Also shown are the cumulative number-count results for the 

1880: $\approx$2~Ms \hbox{CDF-N} (dotted curves),

1881: the $\approx$1~Ms \hbox{CDF-S} (red dashed curves; \citealt{Rosati2002}),

1882: the $\approx$250~ks \hbox{E-CDF-S} (green dash-dotted curves; L05),

1883: and {\it XMM-Newton} observations in the COSMOS field (blue long-dashed curves; 

1884: \citealt{Cappelluti2007}). The 2--10 keV fluxes in the 

1885: $\approx$1~Ms \hbox{CDF-S} and the COSMOS field were 

1886: converted to 2--8 keV fluxes assuming a photon index $\Gamma=1.4$.

1887: {\it Bottom}: Ratio of the \hbox{CDF-S} to \hbox{CDF-N} cumulative 

1888: number counts for

1889: the ({\it a}) soft band and ({\it b}) hard band. Ratios were calculated 

1890: only for number counts that were derived from a sample of $\ga50$ sources,

1891: since for smaller numbers of sources there are large statistical errors.

1892: This corresponds to soft-band fluxes $\la1.5\times10^{-15}$ \flux\ and

1893: hard-band fluxes $\la5\times10^{-15}$ \flux. 

1894: \label{srccounts}}

1895: \end{figure*}

1896: 

1897: 

1898: 

1899: \section{SUMMARY}

1900: We have presented catalogs and basic analyses of \hbox{X-ray} point sources 

1901: detected in the $\approx2$~Ms \hbox{CDF-S}, which is one of the 

1902: two deepest \chandra\ surveys. 

1903: The key points from this work are the following:

1904: 

1905: \begin{enumerate}

1906: \item

1907: The entire \hbox{CDF-S} consists of 23 separate observations with

1908: 1.911~Ms of combined exposure. The survey covers an area of 435.6 

1909: arcmin$^{2}$.

1910: 

1911: \item

1912: The main \chandra\ source catalog consists of 462 sources that were

1913: detected using {\sc wavdetect} with a false-positive probability threshold

1914: of $1\times10^{-6}$. These sources were detected in up to three X-ray 

1915: bands: 0.5--8.0~keV, 0.5--2.0~keV,

1916: and 2--8~keV; 135 of these sources are new.

1917: 

1918: \item

1919: The first supplementary \chandra\ source catalog contains 86 

1920: sources that were generated by

1921: merging the $\approx250$~ks \hbox{E-CDF-S} with the \hbox{CDF-S}, which

1922: provides additional sensitivity in the outer regions of the \hbox{CDF-S}.

1923: 

1924: \item

1925: The second supplementary \chandra\ source catalog contains 30 sources that were

1926: detected at a lower X-ray significance threshold of $1\times10^{-5}$ 

1927: and that have bright optical counterparts ($R<23.8$).

1928: 

1929: \item

1930: Source positions for the main 

1931: and supplementary \hbox{CDF-S} plus \hbox{E-CDF-S} \chandra\ catalogs

1932: have been determined using centroid and matched-filter techniques;

1933: the median positional uncertainty is $\approx0\farcs36$.

1934: 

1935: \item

1936: The basic X-ray and optical properties of the point sources indicate a

1937: variety of source types. More than half of the sources in the

1938: main \chandra\ catalogs

1939: appear to be AGNs. Of the 135 newly detected sources,  

1940: $\approx$55\% appear to be AGNs while $\approx$45\% appear to be starburst

1941: and normal galaxies. 

1942: The majority of the sources in the 

1943: supplementary optically bright catalog are expected to be normal

1944: and starburst galaxies.

1945: 

1946: \item

1947: The average backgrounds in the

1948: 0.5--2.0 and 2--8~keV bands are 0.066 and 0.167~counts Ms$^{-1}$~pixel$^{-1}$,

1949: respectively. Thus these observations are nearly photon limited near the aim

1950: point and could be extended to substantially greater depths with 

1951: further exposure. The background count distributions are very close to 

1952: Poisson distributions. The on-axis flux limits in the 0.5--2.0~keV and

1953: 2--8~keV bands are $\approx1.9\times10^{-17}$ \flux\ and

1954: $\approx1.3\times10^{-16}$ \flux, respectively. 

1955: 

1956: \item

1957: Compared to the other deepest \chandra\ survey, the $\approx2$~Ms \hbox{CDF-N},

1958: the \hbox{CDF-S} has similar effective exposure coverage and sensitivity 

1959: limits. The cumulative number counts of these two fields are consistent

1960: with each other to within $\approx$1~$\sigma$ at fluxes above 

1961: $\approx2\times10^{-16}$~\flux\ in the soft band.

1962: The \hbox{CDF-N} number counts

1963: are up to $\approx25\%$ higher than the  

1964: \hbox{CDF-S} 

1965: number counts at the faintest fluxes in the soft and hard bands, 

1966: indicating small

1967: field-to-field variations.

1968: 

1969: \end{enumerate}

1970: 

1971: 

1972: \section{ACKNOWLEDGEMENTS}

1973: Support for this work was provided by NASA through \chandra\ Award SP8-9003A 

1974: (BL, FEB, WNB) issued by the \chandra\ X-ray Observatory Center, which 

1975: is operated by the Smithsonian Astrophysical Observatory. We also 

1976: acknowledge the financial support of

1977: the Royal Society (DMA and IRS),

1978: the Science and Technology Facilities Council fellowship program (BDL),

1979: NSF grant 06-7634 (DPS), contract ASI--INAF I/023/05/0 and grant 

1980: PRIN-MIUR 2006-02-5203 (AC, RG, and CV). 

1981: %

1982: We thank H.D. Tananbaum for allocating the time for these observations

1983: and T.~L. Aldcroft, P. Broos and L.~K. Townsley for helpful discussions. 

1984: 

1985: 

1986: %\clearpage

1987: 

1988: 

1989: %\bibliographystyle{apj} 

1990: %\bibliography{refs} 

1991: 

1992: \begin{thebibliography}{56}

1993: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi

1994: 

1995: \bibitem[{{Alexander} {et~al.}(2003)}]{Alexander2003}

1996: {Alexander}, D.~M., {et~al.} 2003, \aj, 126, 539 (A03)

1997: 

1998: \bibitem[{{Baganoff} {et~al.}(2003)}]{Baganoff2003}

1999: {Baganoff}, F.~K., {et~al.} 2003, \apj, 591, 891

2000: 

2001: \bibitem[{{Bauer} {et~al.}(2004){Bauer}, {Alexander}, {Brandt}, {Schneider},

2002:   {Treister}, {Hornschemeier}, \& {Garmire}}]{Bauer2004}

2003: {Bauer}, F.~E., {Alexander}, D.~M., {Brandt}, W.~N., {Schneider}, D.~P.,

2004:   {Treister}, E., {Hornschemeier}, A.~E., \& {Garmire}, G.~P. 2004, \aj, 128,

2005:   2048

2006: 

2007: \bibitem[{{Bauer} {et~al.}(2002)}]{Bauer2002}

2008: {Bauer}, F.~E., {et~al.} 2002, \aj, 123, 1163

2009: 

2010: \bibitem[{{Beckwith} {et~al.}(2006)}]{Beckwith2006}

2011: {Beckwith}, S.~V.~W., {et~al.} 2006, \aj, 132, 1729

2012: 

2013: \bibitem[{{Bertin} \& {Arnouts}(1996)}]{Bertin1996}

2014: {Bertin}, E., \& {Arnouts}, S. 1996, \aaps, 117, 393

2015: 

2016: \bibitem[{{Bickel}(1992)}]{Bickel1992}

2017: {Bickel}, P.~J. 1992, in Statistical Chanllenges in Modem Astronomy, ed, E. D.

2018:   Feigelson \& G. J. Babu (New York: Springer), 320

2019: 

2020: \bibitem[{{Boller} {et~al.}(1998){Boller}, {Bertoldi}, {Dennefeld}, \&

2021:   {Voges}}]{Boller1998}

2022: {Boller}, T., {Bertoldi}, F., {Dennefeld}, M., \& {Voges}, W. 1998, \aaps, 129,

2023:   87

2024: 

2025: \bibitem[{{Brandt} \& {Hasinger}(2005)}]{Brandt2005}

2026: {Brandt}, W.~N., \& {Hasinger}, G. 2005, \araa, 43, 827

2027: 

2028: \bibitem[{{Brandt} {et~al.}(2001)}]{Brandt2001}

2029: {Brandt}, W.~N., {et~al.} 2001, \aj, 122, 1

2030: 

2031: \bibitem[{{Broos} {et~al.}(2000)}]{Broos2000}

2032: {Broos}, P., {et~al.} 2000, User's Guide for the TARA Package (University Park:

2033:   Pennsylvania Sate Univ.)

2034: 

2035: \bibitem[{{Caldwell} {et~al.}(2008)}]{Caldwell2008}

2036: {Caldwell}, J.~A.~R., {et~al.} 2008, \apjs, 174, 136

2037: 

2038: \bibitem[{{Cappelluti} {et~al.}(2007)}]{Cappelluti2007}

2039: {Cappelluti}, N., {et~al.} 2007, \apjs, 172, 341

2040: 

2041: \bibitem[{{Cowie} {et~al.}(2002){Cowie}, {Garmire}, {Bautz}, {Barger},

2042:   {Brandt}, \& {Hornschemeier}}]{Cowie2002}

2043: {Cowie}, L.~L., {Garmire}, G.~P., {Bautz}, M.~W., {Barger}, A.~J., {Brandt},

2044:   W.~N., \& {Hornschemeier}, A.~E. 2002, \apjl, 566, L5

2045: 

2046: \bibitem[{{Eadie} {et~al.}(1971){Eadie}, {Dryard}, {James}, {Roos}, \&

2047:   {Sadoulet}}]{Eadie1971}

2048: {Eadie}, W.~T., {Dryard}, D., {James}, F.~E., {Roos}, M., \& {Sadoulet}, B.

2049:   1971, Statistical Methods in Expermental Physics (Amsterdam: North-Holland)

2050: 

2051: \bibitem[{{Feigelson} {et~al.}(2000){Feigelson}, {Broos}, \&

2052:   {Gaffney}}]{Feigelson2000}

2053: {Feigelson}, E.~D., {Broos}, P.~S., \& {Gaffney}, J. 2000, Memo on the Optimal

2054:   Extraction Radius for ACIS Point Souces (University Park: Pennsylvania State

2055:   Univ.)

2056: 

2057: \bibitem[{{Freeman} {et~al.}(2002){Freeman}, {Kashyap}, {Rosner}, \&

2058:   {Lamb}}]{Freeman2002}

2059: {Freeman}, P.~E., {Kashyap}, V., {Rosner}, R., \& {Lamb}, D.~Q. 2002, \apjs,

2060:   138, 185

2061: 

2062: \bibitem[{{Garmire} {et~al.}(2003){Garmire}, {Bautz}, {Ford}, {Nousek}, \&

2063:   {Ricker}}]{Garmire2003}

2064: {Garmire}, G.~P., {Bautz}, M.~W., {Ford}, P.~G., {Nousek}, J.~A., \& {Ricker},

2065:   Jr., G.~R. 2003, Proc. SPIE, 4851, 28

2066: 

2067: \bibitem[{{Gehrels}(1986)}]{Gehrels1986}

2068: {Gehrels}, N. 1986, \apj, 303, 336

2069: 

2070: \bibitem[{{Gendreau} {et~al.}(1995)}]{Gendreau1995}

2071: {Gendreau}, K.~C., {et~al.} 1995, \pasj, 47, L5

2072: 

2073: \bibitem[{{Giacconi} {et~al.}(2002)}]{Giacconi2002}

2074: {Giacconi}, R., {et~al.} 2002, \apjs, 139, 369 (G02)

2075: 

2076: \bibitem[{{Giavalisco} {et~al.}(2004)}]{Giavalisco2004}

2077: {Giavalisco}, M., {et~al.} 2004, \apjl, 600, L93

2078: 

2079: \bibitem[{{Gilli} {et~al.}(2003)}]{Gilli2003}

2080: {Gilli}, R., {et~al.} 2003, \apj, 592, 721

2081: 

2082: \bibitem[{{Hasinger} {et~al.}(1998){Hasinger}, {Burg}, {Giacconi}, {Schmidt},

2083:   {Trumper}, \& {Zamorani}}]{Hasinger1998}

2084: {Hasinger}, G., {Burg}, R., {Giacconi}, R., {Schmidt}, M., {Trumper}, J., \&

2085:   {Zamorani}, G. 1998, \aap, 329, 482

2086: 

2087: \bibitem[{{Hornschemeier} {et~al.}(2001)}]{Hornschemeier2001}

2088: {Hornschemeier}, A.~E., {et~al.} 2001, \apj, 554, 742

2089: 

2090: \bibitem[{{Hornschemeier} {et~al.}(2003)}]{Hornschemeier2003}

2091: ---. 2003, \aj, 126, 575

2092: 

2093: \bibitem[{{Hornschemeier} {et~al.}(2004)}]{Hornschemeier2004}

2094: ---. 2004, \apjl, 600, L147

2095: 

2096: \bibitem[{{Jerius} {et~al.}(2000){Jerius}, {Donnelly}, {Tibbetts}, {Edgar},

2097:   {Gaetz}, {Schwartz}, {Van Speybroeck}, \& {Zhao}}]{Jerius2000}

2098: {Jerius}, D., {Donnelly}, R.~H., {Tibbetts}, M.~S., {Edgar}, R.~J., {Gaetz},

2099:   T.~J., {Schwartz}, D.~A., {Van Speybroeck}, L.~P., \& {Zhao}, P. 2000, Proc.

2100:   SPIE, 4012, 17

2101: 

2102: \bibitem[{{Kim} {et~al.}(2007)}]{Kim2007}

2103: {Kim}, M., {et~al.} 2007, \apjs, 169, 401

2104: 

2105: \bibitem[{{Kraft} {et~al.}(1991){Kraft}, {Burrows}, \& {Nousek}}]{Kraft1991}

2106: {Kraft}, R.~P., {Burrows}, D.~N., \& {Nousek}, J.~A. 1991, \apj, 374, 344

2107: 

2108: \bibitem[{{Le F{\`e}vre} {et~al.}(2004)}]{LeFevre2004}

2109: {Le F{\`e}vre}, O., {et~al.} 2004, \aap, 428, 1043

2110: 

2111: \bibitem[{{Lehmer} {et~al.}(2006){Lehmer}, {Brandt}, {Hornschemeier},

2112:   {Alexander}, {Bauer}, {Koekemoer}, {Schneider}, \& {Steffen}}]{Lehmer2006}

2113: {Lehmer}, B.~D., {Brandt}, W.~N., {Hornschemeier}, A.~E., {Alexander}, D.~M.,

2114:   {Bauer}, F.~E., {Koekemoer}, A.~M., {Schneider}, D.~P., \& {Steffen}, A.~T.

2115:   2006, \aj, 131, 2394

2116: 

2117: \bibitem[{{Lehmer} {et~al.}(2005)}]{Lehmer2005}

2118: {Lehmer}, B.~D., {et~al.} 2005, \apjs, 161, 21 (L05)

2119: 

2120: \bibitem[{{Lehmer} {et~al.}(2008)}]{Lehmer2008}

2121: ---. 2008, \apj, 681, 1163

2122: 

2123: \bibitem[{{Lyons}(1991)}]{Lyons1991}

2124: {Lyons}, L. 1991, Data Analysis for Physical Science Students (Cambridge:

2125:   Cambridge Univ. Press)

2126: 

2127: \bibitem[{{Maccacaro} {et~al.}(1988){Maccacaro}, {Gioia}, {Wolter}, {Zamorani},

2128:   \& {Stocke}}]{Maccacaro1988}

2129: {Maccacaro}, T., {Gioia}, I.~M., {Wolter}, A., {Zamorani}, G., \& {Stocke},

2130:   J.~T. 1988, \apj, 326, 680

2131: 

2132: \bibitem[{{Markevitch}(2001)}]{Markevitch2001}

2133: {Markevitch}, M. 2001, CXC Memo, (Cambridge: CXC)

2134:   http://cxc.harvard.edu/cal/calreview/mm{\_}calrev{\_}bg.ps

2135: 

2136: \bibitem[{{Markevitch} {et~al.}(2003)}]{Markevitch2003}

2137: {Markevitch}, M., {et~al.} 2003, \apj, 583, 70

2138: 

2139: \bibitem[{{Marshall} {et~al.}(1980){Marshall}, {Boldt}, {Holt}, {Miller},

2140:   {Mushotzky}, {Rose}, {Rothschild}, \& {Serlemitsos}}]{Marshall1980}

2141: {Marshall}, F.~E., {Boldt}, E.~A., {Holt}, S.~S., {Miller}, R.~B., {Mushotzky},

2142:   R.~F., {Rose}, L.~A., {Rothschild}, R.~E., \& {Serlemitsos}, P.~J. 1980,

2143:   \apj, 235, 4

2144: 

2145: \bibitem[{{Mignoli} {et~al.}(2005)}]{Mignoli2005}

2146: {Mignoli}, M., {et~al.} 2005, \aap, 437, 883

2147: 

2148: \bibitem[{{Mobasher} {et~al.}(2004)}]{Mobasher2004}

2149: {Mobasher}, B., {et~al.} 2004, \apjl, 600, L167

2150: 

2151: \bibitem[{{Moretti} {et~al.}(2003){Moretti}, {Campana}, {Lazzati}, \&

2152:   {Tagliaferri}}]{Moretti2003}

2153: {Moretti}, A., {Campana}, S., {Lazzati}, D., \& {Tagliaferri}, G. 2003, \apj,

2154:   588, 696

2155: 

2156: \bibitem[{{Paolillo} {et~al.}(2004){Paolillo}, {Schreier}, {Giacconi},

2157:   {Koekemoer}, \& {Grogin}}]{Paolillo2004}

2158: {Paolillo}, M., {Schreier}, E.~J., {Giacconi}, R., {Koekemoer}, A.~M., \&

2159:   {Grogin}, N.~A. 2004, \apj, 611, 93

2160: 

2161: \bibitem[{{Popesso} {et~al.}(2008)}]{Popesso2008}

2162: {Popesso}, P., {et~al.} 2008, \aap, submitted

2163: 

2164: \bibitem[{{Ravikumar} {et~al.}(2007)}]{Ravikumar2007}

2165: {Ravikumar}, C.~D., {et~al.} 2007, \aap, 465, 1099

2166: 

2167: \bibitem[{{Rosati} {et~al.}(2002)}]{Rosati2002}

2168: {Rosati}, P., {et~al.} 2002, \apj, 566, 667

2169: 

2170: \bibitem[{{Stark} {et~al.}(1992){Stark}, {Gammie}, {Wilson}, {Bally}, {Linke},

2171:   {Heiles}, \& {Hurwitz}}]{Stark1992}

2172: {Stark}, A.~A., {Gammie}, C.~F., {Wilson}, R.~W., {Bally}, J., {Linke}, R.~A.,

2173:   {Heiles}, C., \& {Hurwitz}, M. 1992, \apjs, 79, 77

2174: 

2175: \bibitem[{{Stocke} {et~al.}(1991){Stocke}, {Morris}, {Gioia}, {Maccacaro},

2176:   {Schild}, {Wolter}, {Fleming}, \& {Henry}}]{Stocke1991}

2177: {Stocke}, J.~T., {Morris}, S.~L., {Gioia}, I.~M., {Maccacaro}, T., {Schild},

2178:   R., {Wolter}, A., {Fleming}, T.~A., \& {Henry}, J.~P. 1991, \apjs, 76, 813

2179: 

2180: \bibitem[{{Szokoly} {et~al.}(2004)}]{Szokoly2004}

2181: {Szokoly}, G.~P., {et~al.} 2004, \apjs, 155, 271

2182: 

2183: \bibitem[{{Townsley} {et~al.}(2000){Townsley}, {Broos}, {Garmire}, \&

2184:   {Nousek}}]{Townsley2000}

2185: {Townsley}, L.~K., {Broos}, P.~S., {Garmire}, G.~P., \& {Nousek}, J.~A. 2000,

2186:   \apjl, 534, L139

2187: 

2188: \bibitem[{{Townsley} {et~al.}(2002){Townsley}, {Broos}, {Nousek}, \&

2189:   {Garmire}}]{Townsley2002}

2190: {Townsley}, L.~K., {Broos}, P.~S., {Nousek}, J.~A., \& {Garmire}, G.~P. 2002,

2191:   Nuclear Instruments and Methods in Physics Research A, 486, 751

2192: 

2193: \bibitem[{{Tozzi} {et~al.}(2001)}]{Tozzi2001}

2194: {Tozzi}, P., {et~al.} 2001, \apj, 562, 42

2195: 

2196: \bibitem[{{Vanzella} {et~al.}(2008)}]{Vanzella2008}

2197: {Vanzella}, E., {et~al.} 2008, \aap, 478, 83

2198: 

2199: \bibitem[{{Vikhlinin}(2001)}]{Vikhlinin2001}

2200: {Vikhlinin}, A. 2001, CXC Memo, (Cambridge: CXC)

2201:   http://cxc.harvard.edu/cal/Acis/Cal\_prods/vfbkgrnd/

2202: 

2203: \bibitem[{{Wolf} {et~al.}(2004)}]{Wolf2004}

2204: {Wolf}, C., {et~al.} 2004, \aap, 421, 913

2205: 

2206: \bibitem[{{Yang} {et~al.}(2003){Yang}, {Mushotzky}, {Barger}, {Cowie},

2207:   {Sanders}, \& {Steffen}}]{Yang2003}

2208: {Yang}, Y., {Mushotzky}, R.~F., {Barger}, A.~J., {Cowie}, L.~L., {Sanders},

2209:   D.~B., \& {Steffen}, A.~T. 2003, \apjl, 585, L85

2210: 

2211: \end{thebibliography}

2212: 

2213: %\clearpage

2214: %\LongTables

2215: %\begin{landscape}

2216: \begin{deluxetable}{lccccccl}

2217: \tabletypesize{\small}

2218: %\tabletypesize{\footnotesize}

2219: \tablecaption{Journal of {\it Chandra} Deep Field-South Observations}

2220: %

2221: \tablehead{

2222: \colhead{}                                 &

2223: \colhead{Obs. Start}                                 &

2224: \colhead{Exposure}                             &

2225: \multicolumn{2}{c}{Aim Point$^{\rm b}$}                 &

2226: \colhead{Roll Angle$^{\rm c}$}                                 &

2227: \colhead{Obs.}                                &

2228: \colhead{Pipeline}                             \\

2229: \cline{4-5}

2230: %

2231: \colhead{Obs. ID}                                 &

2232: \colhead{(UT)}                         &

2233: \colhead{Time$^{\rm a}$ (ks)}               &

2234: \colhead{$\alpha$ (J2000.0)}                &

2235: \colhead{$\delta$ (J2000.0)}                &

2236: \colhead{(deg)}         &

2237: \colhead{Mode$^{\rm d}$}                             &

2238: \colhead{Version$^{\rm e}$}                             

2239: }

2240: %

2241: \tablewidth{0pt}

2242: \startdata

2243: %

2244: 1431-0\dotfill \ldots \ldots & 1999 Oct 15, 17:38 & \phantom{0}24.6 & 03 32 29.44 & $-$27 48 21.8 & \phantom{0}47.3 & VF & R4CU5UPD11 \\ 

2245: 1431-1\dotfill \ldots \ldots & 1999 Nov 23, 02:30 & \phantom{0}93.6 & 03 32 29.44 & $-$27 48 21.8 & 353.9 & \phantom{V}F & R4CU5UPD11 \\       

2246: 441\dotfill \ldots \ldots & 2000 May 27, 01:18 & \phantom{0}56.0 & 03 32 26.91 & $-$27 48 19.4 & 166.7 & \phantom{V}F & 7.6.10 \\       

2247: 582\dotfill \ldots \ldots & 2000 June 03, 02:38 & 130.6 & 03 32 26.97 & $-$27 48 18.5 & 162.9 & \phantom{V}F & 7.6.10 \\    

2248: 2406\dotfill \ldots \ldots & 2000 Dec 10, 23:35 & \phantom{0}29.7 & 03 32 28.33 & $-$27 48 36.5 & 332.2 & \phantom{V}F & 7.6.10 \\    

2249: 2405\dotfill \ldots \ldots & 2000 Dec 11, 08:14 & \phantom{0}59.6 & 03 32 28.82 & $-$27 48 43.5 & 331.8 & \phantom{V}F & 7.6.10 \\    

2250: 2312\dotfill \ldots \ldots & 2000 Dec 13, 03:28 & 123.7 & 03 32 28.28 & $-$27 48 36.9 & 329.9 & \phantom{V}F & 7.6.10 \\    

2251: 1672\dotfill \ldots \ldots & 2000 Dec 16, 05:07 & \phantom{0}95.1 & 03 32 28.73 & $-$27 48 44.5 & 326.9 & \phantom{V}F & 7.6.10 \\    

2252: 2409\dotfill \ldots \ldots & 2000 Dec 19, 03:55 & \phantom{0}69.0 & 03 32 28.08 & $-$27 48 38.6 & 319.2 & \phantom{V}F & 7.6.10 \\    

2253: 2313\dotfill \ldots \ldots & 2000 Dec 21, 02:08 & 130.4 & 03 32 28.08 & $-$27 48 38.6 & 319.2 & \phantom{V}F & 7.6.10 \\    

2254: 2239\dotfill \ldots \ldots & 2000 Dec 23, 17:28 & 130.8 & 03 32 28.08 & $-$27 48 38.6 & 319.2 & \phantom{V}F & 7.6.10 \\    

2255: 8591\dotfill \ldots \ldots & 2007 Sep 20, 05:26 & \phantom{0}45.4 & 03 32 28.20 & $-$27 48 06.9 & \phantom{0}72.7 & VF & 7.6.11.1 \\    

2256: 9593\dotfill \ldots \ldots & 2007 Sep 22, 20:34 & \phantom{0}46.4 & 03 32 28.20 & $-$27 48 06.9 & \phantom{0}72.7 & VF & 7.6.11.1 \\    

2257: 9718\dotfill \ldots \ldots & 2007 Oct 03, 13:56 & \phantom{0}49.4 & 03 32 28.61 & $-$27 48 07.4 & \phantom{0}62.0 & VF & 7.6.11.1 \\    

2258: 8593\dotfill \ldots \ldots & 2007 Oct 06, 02:04 & \phantom{0}49.5 & 03 32 28.61 & $-$27 48 07.4 & \phantom{0}62.0 & VF & 7.6.11.1 \\    

2259: 8597\dotfill \ldots \ldots & 2007 Oct 17, 07:07 & \phantom{0}59.3 & 03 32 29.25 & $-$27 48 10.4 & \phantom{0}44.2 & VF & 7.6.11.2 \\    

2260: 8595\dotfill \ldots \ldots & 2007 Oct 19, 14:16 & 115.4 & 03 32 29.35 & $-$27 48 11.2 & \phantom{0}41.2 & VF & 7.6.11.2 \\    

2261: 8592\dotfill \ldots \ldots & 2007 Oct 22, 12:14 & \phantom{0}86.6 & 03 32 29.62 & $-$27 48 13.8 & \phantom{0}32.4 & VF & 7.6.11.2 \\    

2262: 8596\dotfill \ldots \ldots & 2007 Oct 24, 13:20 & 115.1 & 03 32 29.62 & $-$27 48 13.8 & \phantom{0}32.4 & VF & 7.6.11.2 \\    

2263: 9575\dotfill \ldots \ldots & 2007 Oct 27, 05:43 & 108.7 & 03 32 29.62 & $-$27 48 13.8 & \phantom{0}32.4 & VF & 7.6.11.2 \\    

2264: 9578\dotfill \ldots \ldots & 2007 Oct 30, 22:35 & \phantom{0}38.6 & 03 32 29.84 & $-$27 48 16.7 & \phantom{0}24.2 & VF & 7.6.11.2 \\    

2265: 8594\dotfill \ldots \ldots & 2007 Nov 01, 11:51 & 141.4 & 03 32 29.84 & $-$27 48 16.7 & \phantom{0}24.2 & VF & 7.6.11.2 \\    

2266: 9596\dotfill \ldots \ldots & 2007 Nov 04, 04:11 & 111.9 & 03 32 29.95 & $-$27 48 18.5 & \phantom{0}19.8 & VF & 7.6.11.2 \\    

2267: 

2268: %

2269: \enddata

2270: 

2271: \tablecomments{

2272: The focal-plane temperature was $-110\degr$C during the first two 

2273: observations and $-120\degr$C during the others. Units of right

2274: ascension are hours, minutes, and seconds, and units of declination are 

2275: degrees, arcminutes, and arcseconds.

2276: }

2277: \par \tablenotetext{a}{All observations were continuous. The data were filtered

2278: on good-time intervals, and one mild flare was removed in observation 1431-0.

2279: The short time

2280: intervals with bad satellite aspect are negligible and have not been removed.

2281: The total exposure time for the 23 observations is 1.911~Ms.}

2282: \tablenotetext{b}{The average aim point, weighted by exposure time, is

2283: $\alpha_{\rm J2000.0}=03^{\rm h}32^{\rm m}28\fs80$, 

2284: $\delta_{\rm J2000.0}=-27\degr48\arcmin23\farcs0$.}

2285: 

2286: \tablenotetext{c}{Roll angle describes the orientation of the \chandra\ 

2287: instruments

2288: on the sky. The angle is between 0--360$^{\circ}$, and it increases to the west

2289: of north (opposite to the sense of traditional position angle).}

2290: \tablenotetext{d}{The observing mode: F=Faint mode and VF=Very Faint mode.}

2291: \tablenotetext{e}{The version of the CXC pipeline software used for basic 

2292: processing of the data.}

2293: \label{tbl-obs}

2294: \end{deluxetable}

2295: %\clearpage

2296: %\end{landscape}

2297: 

2298: \begin{deluxetable}{lllcccccccc}

2299: \tabletypesize{\scriptsize}

2300: \tablewidth{0pt}

2301: \tablecaption{Main {\it Chandra} Catalog}

2302: 

2303: \tablehead{

2304: \colhead{} &

2305: \multicolumn{2}{c}{X-ray Coordinates} &

2306: \colhead{}                   &

2307: \colhead{}                   &

2308: \multicolumn{6}{c}{Counts}      \\

2309: \\ \cline{2-3} \cline{6-11} \\

2310: \colhead{No.}                    &

2311: \colhead{$\alpha_{2000}$}       &

2312: \colhead{$\delta_{2000}$}       &

2313: \colhead{Pos Err}       &

2314: \colhead{Off-Axis}       &

2315: \colhead{FB}          &

2316: \colhead{FB Upp Err}          &

2317: \colhead{FB Low Err}          &

2318: \colhead{SB}          &

2319: \colhead{SB Upp Err}          &

2320: \colhead{SB Low Err}          \\

2321: \colhead{(1)}         &

2322: \colhead{(2)}         &

2323: \colhead{(3)}         &

2324: \colhead{(4)}         &

2325: \colhead{(5)}         &

2326: \colhead{(6)}         &

2327: \colhead{(7)}         &

2328: \colhead{(8)}         &

2329: \colhead{(9)}        &

2330: \colhead{(10)}        &

2331: \colhead{(11)}

2332: }

2333: 

2334: \startdata

2335:        1 \dotfill \ldots   &03 31 34.19 &$-$27 50 04.2& 1.6&12.19&     26.1&    11.1&    11.8&     13.5&    $-$1.0&    $-$1.0\\

2336:        2 \dotfill \ldots   &03 31 35.79 &$-$27 51 34.7& 1.9& 12.14&    14.9&    $-$1.0&    $-$1.0&    12.1&    \phantom{0}7.3&    \phantom{0}7.2\\

2337:        3 \dotfill \ldots   &03 31 40.15 &$-$27 47 46.3& 1.3&10.77&     33.6&    11.8&    11.8&     25.8&     \phantom{0}8.4&     \phantom{0}8.1\\

2338:        4 \dotfill \ldots   &03 31 40.93 &$-$27 46 21.8& 1.1& 10.77&     61.2&    14.0&    14.0&     16.0&    $-$1.0&     $-$1.0\\

2339:        5 \dotfill \ldots   &03 31 44.23 &$-$27 49 25.5& 1.0& \phantom{0}9.91&     79.5&    19.4&    19.4&     37.8&     12.5&     12.5\\

2340: \enddata

2341: 

2342: \tablecomments{

2343: Units of right

2344: ascension are hours, minutes, and seconds, and units of declination are

2345: degrees, arcminutes, and arcseconds.

2346: Table~2 is presented in its entirety in the electronic edition. An 

2347: abbreviated version of the table is shown here for guidance as 

2348: to its form and content. The full table contains 49 columns of 

2349: information on the 462 \hbox{X-ray} sources.}

2350: \label{tbl-mcat}

2351: 

2352: \end{deluxetable}

2353: 

2354: 

2355: 

2356: \begin{deluxetable}{lccccc}

2357: 

2358: \tabletypesize{\small}

2359: \tablewidth{0pt}

2360: \tablecaption{Summary of {\it Chandra} Source Detections \label{tbldet}}

2361: 

2362: \tablehead{

2363: \colhead{} &

2364: \colhead{Number of} &

2365: \multicolumn{4}{c}{Detected Counts Per Source} \\

2366: \cline{3-6}   

2367: \colhead{Band (keV)} &

2368: \colhead{Sources} &

2369: \colhead{Maximum} &

2370: \colhead{Minimum} &

2371: \colhead{Median} &

2372: \colhead{Mean}

2373: }

2374: 

2375: \startdata

2376: Full (0.5--8.0)   & 403 & 21579.7 & 11.4 & 101.0 & 410.6 \\

2377: Soft (0.5--2.0)  & 392 & 15929.7 & \phantom{0}4.7 & \phantom{0}53.0 & 269.9  \\

2378: Hard (2--8)  & 265 & \phantom{0}5664.3 & \phantom{0}7.7 & \phantom{0}88.6 & 216.9  \\

2379: \enddata

2380: \end{deluxetable}

2381: 

2382: 

2383: \begin{deluxetable}{lccc}

2384: 

2385: \tabletypesize{\small}

2386: \tablewidth{0pt}

2387: 

2388: \tablecaption{Sources Detected in One Band but not Another \label{tblundet}}

2389: 

2390: \tablehead{

2391: \colhead{Detection Band} &

2392: \multicolumn{3}{c}{Nondetection Energy Band} \\

2393: \cline{2-4}

2394: \colhead{(keV)} &

2395: \colhead{Full} &

2396: \colhead{Soft} &

2397: \colhead{Hard} 

2398: }

2399: 

2400: \startdata

2401: Full (0.5--8.0)  & \ldots & 67 & 141 \\

2402: Soft (0.5--2.0)  & ~~~~56~~~~ & \ldots & 166 \\

2403: Hard (2--8)   & ~~~~\phantom{0}3~~~~ & 39  & \ldots \\

2404: \enddata

2405: \tablecomments{For example, there were 67 sources detected in the full band

2406: that were not detected in the soft band.}

2407: \end{deluxetable}

2408: %

2409: 

2410: \begin{deluxetable}{lllcccccccc}

2411: \tabletypesize{\scriptsize}

2412: \tablewidth{0pt}

2413: \tablecaption{Supplementary \hbox{CDF-S} plus \hbox{E-CDF-S} \chandra\ Catalog}

2414: 

2415: \tablehead{

2416: \colhead{} &

2417: \multicolumn{2}{c}{X-ray Coordinates} &

2418: \colhead{}                   &

2419: \colhead{}                   &

2420: \multicolumn{6}{c}{Counts}      \\

2421: \\ \cline{2-3} \cline{6-11} \\

2422: \colhead{No.}                    &

2423: \colhead{$\alpha_{2000}$}       &

2424: \colhead{$\delta_{2000}$}       &

2425: \colhead{Pos Err}       &

2426: \colhead{Off-Axis}       &

2427: \colhead{FB}          &

2428: \colhead{FB Upp Err}          &

2429: \colhead{FB Low Err}          &

2430: \colhead{SB}          &

2431: \colhead{SB Upp Err}          &

2432: \colhead{SB Low Err}          \\

2433: \colhead{(1)}         &

2434: \colhead{(2)}         &

2435: \colhead{(3)}         &

2436: \colhead{(4)}         &

2437: \colhead{(5)}         &

2438: \colhead{(6)}         &

2439: \colhead{(7)}         &

2440: \colhead{(8)}         &

2441: \colhead{(9)}        &

2442: \colhead{(10)}        &

2443: \colhead{(11)}

2444: }

2445: 

2446: \startdata

2447:        1 \dotfill \ldots   &03 31 40.98 &$-$27 44 34.8& 1.0&11.24&     118.4&    12.8&    11.6&     56.1&    \phantom{0}8.8&    \phantom{0}7.7\\

2448:        2 \dotfill \ldots   &03 31 42.76 &$-$27 53 40.7& 1.6& 11.47&    \phantom{0}17.4&    \phantom{0}5.9&    \phantom{0}4.7&    \phantom{0}7.5&    \phantom{0}4.2&    \phantom{0}3.0\\

2449:        3 \dotfill \ldots   &03 31 43.21 &$-$27 54 05.1& 0.9& 11.58&     152.3&    14.2&    13.1&     49.7&     \phantom{0}8.4&     \phantom{0}7.3\\

2450:        4 \dotfill \ldots   &03 31 44.64 &$-$27 45 19.1& 1.2& 10.23&     \phantom{0}39.9&   \phantom{0}8.1&    \phantom{0}6.9&     \phantom{0}7.5&    $-$1.0&     $-$1.0\\

2451:        5 \dotfill \ldots   &03 31 48.14 &$-$27 52 32.1& 1.6& \phantom{0}9.90&     \phantom{0}10.8&   $-$1.0&    $-$1.0&     \phantom{0}8.1&    \phantom{0}4.4&     \phantom{0}3.2\\

2452: \enddata

2453: \tablecomments{

2454: Units of right

2455: ascension are hours, minutes, and seconds, and units of declination are

2456: degrees, arcminutes, and arcseconds.

2457: Table~5 is presented in its entirety in the electronic edition. An

2458: abbreviated version of the table is shown here for guidance as

2459: to its form and content. The full table contains 52 columns of

2460: information on the 86 \hbox{X-ray} sources.}

2461: \label{tbl-sp1}

2462: 

2463: \end{deluxetable}

2464: 

2465: 

2466: 

2467: \begin{deluxetable}{lllcccccccc}

2468: \tabletypesize{\scriptsize}

2469: \tablewidth{0pt}

2470: \tablecaption{Supplementary Optically Bright \chandra\ Catalog}

2471: 

2472: \tablehead{

2473: \colhead{} &

2474: \multicolumn{2}{c}{X-ray Coordinates} &

2475: \colhead{}                   &

2476: \colhead{}                   &

2477: \multicolumn{6}{c}{Counts}      \\

2478: \\ \cline{2-3} \cline{6-11} \\

2479: \colhead{No.}                    &

2480: \colhead{$\alpha_{2000}$}       &

2481: \colhead{$\delta_{2000}$}       &

2482: \colhead{Pos Err}       &

2483: \colhead{Off-Axis}       &

2484: \colhead{FB}          &

2485: \colhead{FB Upp Err}          &

2486: \colhead{FB Low Err}          &

2487: \colhead{SB}          &

2488: \colhead{SB Upp Err}          &

2489: \colhead{SB Low Err}          \\

2490: \colhead{(1)}         &

2491: \colhead{(2)}         &

2492: \colhead{(3)}         &

2493: \colhead{(4)}         &

2494: \colhead{(5)}         &

2495: \colhead{(6)}         &

2496: \colhead{(7)}         &

2497: \colhead{(8)}         &

2498: \colhead{(9)}        &

2499: \colhead{(10)}        &

2500: \colhead{(11)}

2501: }

2502: 

2503: \startdata

2504:        1 \dotfill \ldots   &03 31 50.82 &$-$27 47 03.8& 1.2&8.50&     47.2&    $-$1.0&    $-$1.0&     22.1&    5.8&    4.7\\

2505:        2 \dotfill \ldots   &03 31 52.03 &$-$27 50 37.6& 1.2& 8.43&    40.9&    $-$1.0&    $-$1.0&    20.8&     5.6&    4.5\\

2506:        3 \dotfill \ldots   &03 31 57.23 &$-$27 45 36.9& 1.2&7.51&     41.6&    $-$1.0&    $-$1.0&     22.7&     5.8&     4.7\\

2507:        4 \dotfill \ldots   &03 32 00.32 &$-$27 46 11.4& 1.2& 6.67&     35.9&   $-$1.0&    $-$1.0&     18.7&    5.4&     4.3\\

2508:        5 \dotfill \ldots   &03 32 06.59 &$-$27 50 37.3& 1.2& 5.39&     24.3&   $-$1.0&    $-$1.0&     12.0&    4.6&     3.4\\

2509: \enddata

2510: \tablecomments{

2511: Units of right

2512: ascension are hours, minutes, and seconds, and units of declination are

2513: degrees, arcminutes, and arcseconds.

2514: Table~6 is presented in its entirety in the electronic edition. An

2515: abbreviated version of the table is shown here for guidance as

2516: to its form and content. The full table contains 38 columns of

2517: information on the 30 \hbox{X-ray} sources.}

2518: \label{tbl-sp2}

2519: 

2520: \end{deluxetable}

2521: 

2522: 

2523: 

2524: 

2525: 

2526: 

2527: 

2528: 

2529: \begin{deluxetable}{lcccc}

2530: %\tabletypesize{\small}

2531: 

2532: \tabletypesize{\footnotesize}

2533: \tablecaption{Background Parameters}

2534: %

2535: \tablehead{

2536: \colhead{}                                 &

2537: \multicolumn{2}{c}{Mean Background}                 &

2538: \colhead{Total Background$^{\rm c}$}                                 &

2539: \colhead{Count Ratio$^{\rm d}$}                                \\

2540: \cline{2-3}

2541: %

2542: \colhead{Band (keV)}                                 &

2543: \colhead{(counts pixel$^{-1}$)$^{\rm a}$}                         &

2544: \colhead{(counts Ms$^{-1}$ pixel$^{-1}$)$^{\rm b}$}               &

2545: \colhead{(10$^5$ counts)}                &

2546: \colhead{(background/source)}                

2547: }

2548: %

2549: \tablewidth{0pt}

2550: \startdata

2551: Full (0.5--8.0)  & 0.248 & 0.242  & 16.1 & \phantom{0}9.7  \\

2552: Soft (0.5--2.0)   & 0.067 & 0.066  & \phantom{0}4.3 & \phantom{0}4.1  \\

2553: Hard (2--8)   & 0.179 & 0.167  & 11.6 & 20.2  \\

2554: \enddata

2555: \label{tbl-bkg}

2556: \par \tablenotetext{a}{The mean numbers of background counts per pixel. These are 

2557: measured from the background images described in $\S$4.}

2558: \tablenotetext{b}{The mean numbers of counts per pixel 

2559: divided by the mean effective exposure. 

2560: These are measured from the exposure maps and 

2561: background images described in $\S$4.}

2562: \tablenotetext{c}{Total number of background counts.}

2563: \tablenotetext{d}{Ratio of the total number of background 

2564: counts to the total number of source counts.}

2565: 

2566: \end{deluxetable}

2567: 

2568: 

2569: \end{document}

2570: