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

2: 

3: %% manuscript produces a one-column, double-spaced document:

4: %%\documentclass[manuscript]{aastex}

5: 

6: %% preprint2 produces a double-column, single-spaced document:

7: %% \documentclass[preprint2]{aastex}

8: 

9: \newcommand{\degree}{$^{\circ}$}

10: \newcommand{\msol}{\hbox{$M_\odot$}}

11: 

12: %% You can insert a short comment on the title page using the command below.

13: \slugcomment{ApJS, in press}

14: 

15: \shorttitle{Point Sources GC Spitzer Survey}

16: \shortauthors{Ram\'{\i}rez et al.}

17: 

18: \begin{document}

19: 

20: \title{Point Sources from a {\it Spitzer} IRAC Survey of the Galactic Center}

21: 

22: \author{Solange V. Ram\'{\i}rez}

23: \affil{IPAC/Caltech, Pasadena, CA 91125}

24: \email{solange@ipac.caltech.edu}

25: 

26: \author{Richard G. Arendt}

27: \affil{CRESST/UMBC/GSFC, Code 665, Greenbelt, MD 20771}

28: 

29: \author{Kris Sellgren}

30: \affil{The Ohio State University, Columbus, OH 43210}

31: 

32: \author{Susan R. Stolovy}

33: \affil{{\it Spitzer} Science Center, Caltech, Pasadena, CA 91125}

34: 

35: \author{Angela Cotera}

36: \affil{SETI Institute, Mountain View, CA 94043}

37: 

38: \author{Howard A. Smith}

39: \affil{Harvard-Smithsonian Center for Astrophysics, Cambridge, MA 02138}

40: 

41: \author{Farhad Yusef-Zadeh}

42: \affil{Northwestern University, Evanston, IL 60208}

43: 

44: 

45: %%ABSTRACT

46: \begin{abstract}

47: We have obtained $Spitzer$/IRAC observations of the central

48: 2.0\degree $\times$ 1.4\degree ($\sim$ 280 $\times$ 200 pc) 

49: of the Galaxy at 3.6\micron--8.0\micron.

50: A point source catalog of 1,065,565 objects is presented.

51: The catalog includes magnitudes for the point sources

52: at 3.6, 4.5, 5.8, and 8.0\micron , as well as $JHK_s$ 

53: photometry from 2MASS. 

54: The point source catalog is confusion limited with

55: average limits of 12.4, 12.1, 11.7, and 11.2 magnitudes

56: for [3.6], [4.5], [5.8], and [8.0], respectively.

57: We find that the confusion limits are spatially variable

58: because of stellar surface density, background surface brightness 

59: level, and extinction variations across the survey region.

60: The overall distribution of point source density with Galactic

61: latitude and longitude is essentially constant, but structure

62: does appear when sources of different magnitude ranges 

63: are selected.

64: Bright stars show a steep decreasing gradient with Galactic

65: latitude, and a slow decreasing gradient with Galactic

66: longitude, with a peak at the position of the Galactic center.

67: From IRAC color-magnitude and color-color diagrams,

68: we conclude that most of the point sources in our

69: catalog have IRAC magnitudes and colors characteristic

70: of red giant and AGB stars. 

71: \end{abstract}

72: 

73: \keywords{Galaxy: center --- stars: late-type}

74: 

75: %INTRODUCTION

76: \section{Introduction}

77: 

78: Our Galactic center (GC), at a distance of $\sim$ 8.0 kpc \citep{rei93}, 

79: is the closest galactic nucleus, observable at spatial 

80: resolutions unapproachable in other galaxies (1 pc$\approx$26\arcsec). 

81: The region has been intensely studied at wavelengths outside

82: the optical and UV regime, because it is unobservable with

83: optical telescopes due to obscuring dust in the Galactic plane.

84: The typical extinction toward the inner 200 pc 

85: is 25-30 visual magnitudes \citep{sch99,dut03}, and it is 

86: considerably higher towards molecular clouds located close to the GC.

87: 

88: The extent of the GC region is defined by a region of relatively 

89: high density molecular gas ($n_{H_2}$  $\sim$ $10^{4}$ cm$^{-3}$;

90: \citealt{bal87}), covering the inner 200 pc 

91: (170\arcmin $\times$ 40\arcmin, centered on the GC), 

92: called the Central Molecular Zone (CMZ). 

93: The CMZ produces 5\%-10\% of the Galaxy's infrared and Lyman 

94: continuum luminosity and contains 10\% of its molecular gas

95: \citep{bal87,mor96}.

96: The CMZ contains extremely dense giant molecular clouds 

97: \citep{mar04,oka05,bol06}, which are also very turbulent. 

98: Strong tidal shearing forces arise within the CMZ from a

99: gravitational potential well that increases as the

100: galactocentric radius decreases \citep{gus80},

101: culminating in the central black hole, Sgr A$^*$

102: \citep[e.g. ][]{sch03,ghe05}.

103: 

104: In the past, the study of the GC stellar population has been

105: concentrated primarily on the spatial regions surrounding 

106: three clusters of stars.

107: The Central Cluster contains the dense core of stars within

108: a few parsecs of the GC.  The cluster is composed of a mixture of

109: red supergiant and giant stars \citep[e.g. ][]{leb82,sel87} 

110: and young massive stars which exhibit

111: energetic winds as observed in their emission line spectra 

112: \citep[e.g. ][]{all90,kra91,lib95,blu95,tam96}. 

113: These bright, hot emission line stars trace an epoch of 

114: star formation that occurred about $10^7$ yr ago, while

115: the bright cool stars may be associated with either

116: the most recent epoch of star formation or older ones

117: \citep{hal92,kra95}. 

118: The separation of bright cool stars into

119: M supergiants (tracers of recent star formation) and 

120: less massive giants (tracers of older star formation)

121: has been used to study the star formation history

122: within the central cluster \citep{leb82,sel87,blu96,blu03}.

123: The Quintuplet and Arches Clusters are located at about

124: 30 pc in projection from the GC.

125: Both clusters contain hundreds of massive O-B stars, 

126: and have ages of 2--4 Myr \citep{fig99}

127: These clusters are thought to be

128: the low-mass analog of the young ``super star clusters'' 

129: found in external galaxies \citep{all90,nag96,cot96,fig99}.

130: 

131: One stellar population that has been studied across a broader 

132: area ($\sim$200 pc) centered on the GC is the OH/IR stars

133: \citep[among others]{hab83,lin92,sjo98}.

134: OH/IR stars are oxygen rich Asymptotic Giant Branch (AGB) stars that 

135: are characterized by long period pulsations and high mass loss. 

136: Studies suggest that there are two distinct populations

137: of OH/IR stars observed towards the GC, which are

138: separated both spatially and kinematically \citep{lin92}.

139: The OH/IR stars that are more closely confined to the

140: Galactic plane and that have a net prograde rotational

141: velocity in the GC are also found to have higher OH maser expansion 

142: velocities than other OH/IR stars in the GC \citep{lin92}.

143: A higher expansion velocity requires

144: either that the star is more luminous than the average (thus a more massive

145: and younger star), or that it has a higher dust-to-gas ratio

146: (and thus a higher metallicity).

147: 

148: Four other infrared studies have surveyed areas within

149: 200 pc including the CMZ. 

150: The 2MASS all sky survey \citep{skr06} and the DENIS survey

151: \citep{epc97} were limited by their wavelength range 

152: between 1.2 and 2.2\micron\, which was inadequate to characterize

153: the more obscured regions.

154: The {\it Midcourse Space Experiment} (MSX) 

155: observed between 6 and 25\micron\ and 

156: included the CMZ in its survey of the Galactic plane \citep{pri01}.

157: The angular resolution of MSX (18\arcsec\ at 8.3\micron), however,

158: was only sufficient to identify the brightest isolated individual stars.

159: Finally, portions of the CMZ were observed with ISOCAM as part 

160: of the ISOGAL survey, with $\sim$6\arcsec\ angular 

161: resolution at 7\micron\ and $\sim$13\arcsec\ angular

162: resolution at 15\micron\ \citep{omo03}.

163: The ISOGAL survey has been used to select young stellar object

164: (YSO) candidates in the GC \citep{fel02,sch06} within the 

165: restricted area coverage of the survey.

166: 

167: We have obtained $Spitzer$/IRAC observations of the central 

168: 2.0\degree $\times$ 1.4\degree

169: ($\sim$ 280 $\times$ 200 pc, including the CMZ) of the Galaxy at

170: 3.6\micron--8\micron\ in Cycle 1 (GO 3677, PI: Stolovy).

171: These data represent the highest spatial resolution ($\sim$2\arcsec) 

172: and sensitivity uniform large-scale map made to date of the GC 

173: at mid-infrared wavelengths.

174: The IRAC data display complex filamentary structures in the interstellar

175: medium (S. Stolovy et al. 2007, in preparation) 

176: and allow us to detect optically obscured stellar sources.

177: In this paper, we present details on the data reduction and

178: point source extraction (Section \ref{sec_obs}). 

179: A catalog of the IRAC point sources band merged with 2MASS photometry

180: is presented in Section \ref{sec_cat}.

181: This catalog contains 1,065,565 point sources uniformly covering the CMZ.

182: The point source magnitude distributions are discussed in Section 

183: \ref{sec_num_den} and the point source distributions with Galactic

184: coordinates are examined in Section \ref{sec_gcoord}.

185: A discussion of the nature of the point sources in the catalog

186: is presented in Section \ref{sec_cmd_col}.

187: This is the first paper in an upcoming series on the $Spitzer$/IRAC 

188: observations of the Galactic center.

189: 

190: %OBSERVATIONS

191: \section{Observations and Data Processing \label{sec_obs}}

192: 

193: The InfraRed Array Camera \citep[IRAC,][]{faz04} on board the 

194: {\it Spitzer Space Telescope} \citep{wer04} was used to map the 

195: central regions of the Galaxy, with a spatial coverage of about 

196: 2.0\degree \ in Galactic longitude by 1.4\degree \ in Galactic 

197: latitude. 

198: Details of the observations and data processing are given in 

199: S. Stolovy et al. (2007, in preparation) but we provide a brief summary here.

200: 

201: Each IRAC detector has a 5.2\arcmin $\times$ 5.2\arcmin \ field 

202: of view comprised of 256$\times$256 pixels and a mean pixel scale of 

203: 1.22\arcsec\ per pixel. 

204: The four cameras have wavelengths of 3.6\micron, 4.5\micron, 

205: 5.8\micron, and 8.0 \micron \  for Channels 1, 2, 3, and 4, 

206: respectively. 

207: Because all four cameras do not see the exact same region of sky

208: simultaneously and because of orientation constraints, a larger 

209: region was mapped to cover fully the desired central 

210: 2.0\degree \ $\times$ 1.4\degree \  region. 

211: We used the shortest frame time (2 sec.) available for the full-array 

212: mode, corresponding to an on-source effective integration time of 1.2 

213: seconds per pixel. 

214: We took five dithered exposures (or frames) on the sky for each 

215: pointing, giving a total average 

216: on-sky integration time of 6 seconds. 

217: This dithering strategy allows us to correct for bad pixels, 

218: scattered light, and latent images and provided improved sampling 

219: of the point spread function. 

220: Additional processing as described in S. Stolovy et al. 

221: (2007, in preparation) was performed 

222: on the {\it Spitzer} Science Center (SSC) pipeline version S13.2 Basic 

223: Calibrated Data (BCD) products to

224: correct various artifacts (scattered light, latent images, column 

225: pulldown, and banding), producing much improved BCD frames and mosaics. 

226: One electronic artifact that was not corrected due to its non-linear nature

227: was the `bandwidth effect'.

228: This artifact causes extra `sources' to appear 4 

229: pixels away (and in some cases 8 pixels away) from very bright sources 

230: in Channels 3 and 4 along the readout direction (IRAC Data

231: Handbook, Version 3.0, Section 4.3.3). 

232: Thus, a few of these artifacts remain in our final mosaics.

233: 

234: Additional observations were taken in IRAC's sub-array mode for

235: areas in the survey that were affected by saturation.

236: These regions include the Central Cluster and the Quintuplet

237: cluster, plus 12 individual pointings.

238: In sub-array mode, a small section of the array is read out

239: (32 $\times$ 32 pixels = 40\arcsec\ $\times$ 40\arcsec),

240: and we used the shortest exposure time available of 0.02 sec.

241: 

242: \subsection{Mosaicking}

243: 

244: We used the SSC Mosaicking and Point source Extraction (MOPEX) package, 

245: version 030106 (available from the SSC web page 

246: \footnote{ssc.spitzer.caltech.edu/postbcd/download-mopex.html}) 

247: to create mosaics, extract point sources, and create source 

248: subtracted mosaics of the full-array data. 

249: MOPEX is composed of a series of PERL scripts. 

250: We used mosaic.pl to create mosaics, apex.pl to detect 

251: and measure fluxes of point sources, and apex\_qa.pl to create 

252: source subtracted images.

253: The script mosaic.pl performs interpolation and co-addition of FITS

254: images, with the additional functionality of detection of outliers.

255: The outliers are due to radiation hits, hot pixels, and bad pixels.

256: There are three outlier detection algorithms implemented within mosaic.pl.

257: The single frame outlier detection algorithm performs spatial filtering

258: within an individual BCD frame, flagging outliers above a user defined flux 

259: threshold, and below a user defined size. 

260: The multiframe outlier and the dual outlier detection algorithms determine

261: outlier pixels by stacking pixels taken in different exposures but at the same 

262: spatial location. 

263: The BCD frames are spatially matched by coincident point sources.

264: The multiframe outlier computes the statistics of the stacked pixels and 

265: finds the outliers above a user defined $\sigma$ threshold.

266: The dual outlier detection algorithm first detects all sources above a

267: user defined $\sigma$ threshold and then compares the number of 

268: detections for each spatial location to discriminate the outliers 

269: from the real sources.

270: 

271: The optimization of the parameters of the outlier detection algorithms

272: for our confusion-limited data was the most challenging part of the 

273: creation of the mosaics. 

274: The single frame outlier and the dual outlier algorithms flagged many 

275: real point sources as outliers, even using the most conservative set

276: of input parameters, mainly due to crowding of point sources in our images. 

277: Therefore, we adopted only the multiframe outlier detection algorithm 

278: with a conservative threshold of 50$\sigma$, 50$\sigma$, 60$\sigma$, and

279: 25$\sigma$, for Channels 1, 2, 3, and 4 respectively, which gave satisfactory 

280: results in terms of the number of pixels flagged as outliers per BCD frame per 

281: unit of exposure time and the lack of real point sources flagged as outliers.

282: The expected number of radiation hits in one single IRAC frame is 

283: approximately 3 to 6 pixels per second in Channels 1 and 2, and 

284: approximately 4 to 8 pixels per second in Channels 3 and 4. 

285: Our observations comprise a total of 2,895 individual BCD frames per channel,

286: with an integration time of 1.2 s per frame. 

287: This predicts 

288: a total number of radiation hits of about 10,000-21,000 pixels 

289: in Channels 1 and 2, and about 14,000--28,000 pixels in Channels 3 and 4. 

290: The total numbers of flagged outliers determined by MOPEX using the

291: parameters described above were about 26,000, 26,000, 32,000, 

292: and 14,000 pixels in Channels 1, 2, 3, and 4, respectively, 

293: which are similar to expected values.

294: 

295: It is crucial to create a clean mosaic at each channel

296: before attempting to extract the point sources.

297: Our final mosaics represent a significant improvement over available 

298: SSC pipeline data products.

299: The top panels of Figures \ref{fig_ch1} and \ref{fig_ch4}

300: show the final full-array mosaics for Channels 1 and 4, covering a field of

301: view of 2.0\degree $\times$ 1.4\degree, centered on $l$=0,0 and $b$=0.0

302: (see S. Stolovy et al. 2007, in preparation, for final mosaics including 

303: the sub-array observations).

304: 

305: \subsection{Source extraction from full-array data \label{full_extr}}

306: 

307: The source extraction was performed using the MOPEX script apex.pl, set up

308: so it utilizes data products created with the MOPEX script mosaic.pl, in 

309: particular those concerning outlier detection. 

310: The script apex.pl performs the source detection in a background subtracted

311: mosaic.

312: It provides two measurements of the flux: one comes from a point response

313: function (PRF) fitting and the other comes from an aperture measurement.

314: 

315: The PRF is the telescope point spread function convolved with the instrument

316: response function. 

317: The algorithm that fits the PRF to the BCD sources allows a determination of

318: the local background, which is advisable to use in crowded fields

319: with variable background level such as those observed in the GC.

320: Note that the flux uncertainty and the signal to noise ratio (SNR) 

321: as provided by apex.pl are the uncertainty and SNR

322: of the PRF fitting algorithm.

323: The PRFs used in the flux measurement were provided by the SSC.

324: 

325: The aperture flux measurement is performed on the mosaic image.

326: The usual background estimates include performing the aperture 

327: measurement on a median filtered mosaic or using an annulus around

328: the detected sources. 

329: Neither of these two methods for estimating the background are 

330: appropriate for our survey due to point source crowding and 

331: variable and high background levels. 

332: Instead, we used the local background determined by the PRF fitting to

333: subtract the background contribution to the corresponding aperture flux.

334: We measured the flux within a small aperture of 2 pixels radius 

335: (about 2.44\arcsec) to avoid confusion, then subtracted the background

336: contribution, and finally applied the corresponding aperture correction.

337: The value of the aperture corrections are

338: 1.213, 1.234, 1.379, and 1.584 for Channels 1, 2, 3, and 4, respectively, 

339: as provided in the IRAC Data Handbook, Version 3.0. 

340: The script apex.pl does not provide a measure of the uncertainty of 

341: the aperture flux. 

342: We estimate the aperture flux uncertainty by performing an aperture 

343: measurement of the same size as the photometric aperture directly in the 

344: mosaic of the uncertainty images (data product of mosaic.pl). 

345: 

346: The source extraction was performed in each of the 12 AORs separately

347: because of the lack of enough computing memory to process the entire

348: data set simultaneously. 

349: The resulting source lists for each AOR were combined to obtain a 

350: total source list for the whole survey.

351: During this process we also rejected sources that were within a certain 

352: radius from a bright source ($\sigma$ threshold $>$ 30) and sources that 

353: were detected on top of extended emission in Channels 3 and 4. 

354: The radius of avoidance was 4.5\arcsec, 5.8\arcsec, 6.4\arcsec, and 

355: 8.3\arcsec \ for Channels 1, 2, 3, and 4, respectively, which 

356: corresponds roughly to the radius of the second minimum of the diffraction 

357: pattern of the PRF.

358: Note that rejecting faint sources within the radius of avoidance 

359: will also discard possible artifacts due to the 4-pixel bandwidth

360: effect, present in Channels 3 and 4.

361: The `bandwidth effect' causes extra `sources' to appear 4 pixels away 

362: (and in some cases 8 pixels away) from very bright sources

363: in Channels 3 and 4 along the readout direction (IRAC Data

364: Handbook, Version 3.0, Section 4.3.3).

365: It is possible that some saturated sources were misidentified as many 

366: individual sources, each of them too faint to trigger the avoidance radius.

367: To discriminate between a point source and an extended source, we use the fact 

368: that a point source should have the same aperture corrected flux independent 

369: of the size of the aperture used.

370: We flagged a source as extended if the aperture corrected flux from a 3 pixel 

371: radius aperture differs by more than 15\% from a 2 pixel radius aperture 

372: corrected flux. 

373: Sources flagged as extended are not included in the catalog.

374: 

375: We found that the PRF fluxes and aperture fluxes

376: agreed to within 12\% overall.  We did, however, find that

377: the PRF fluxes were systematically lower than the aperture 

378: fluxes by 13-12\% for Channels 1 and 2, and higher by

379: 7\% for Channel 4.  No significant difference was found

380: for Channel 3. This difference is likely to arise from errors 

381: in PRF normalization. We measured the difference between the

382: aperture fluxes and the PRF fluxes by first determining

383: the IRAC colors of foreground sources with low amounts

384: of reddening.

385: Most of the foreground stars are expected to be red giant stars,

386: whose IRAC colors should be near zero (M. Cohen, private communication;

387: IRAC Handbook).

388: \citet{sch99} and \citet{dut03} have determined extinction maps at

389: the Galactic center distance.

390: The minimum extinction observed for a source located at the distance of 

391: the GC is $A_K$ = 1 magnitudes \citep{sch99,dut03}.

392: We therefore selected as foreground stars those sources with 

393: $A_K <$ 1 or $(J-K_s) <$ 1.5.

394: There are 6816 sources in our source list that have $(J-K) < $1.5,

395: and their mean IRAC colors are listed in Table \ref{tab1}, for

396: both PRF and aperture magnitudes.

397: Aperture IRAC colors are closer to zero than PRF IRAC colors.

398: The PRF method, however, is generally superior than aperture photometry

399: in crowded fields, and also shows less scatter at fainter magnitudes.

400: We adjusted the PRF fluxes for Channels 1, 2, and 4 such that the 

401: median ratio of the two extraction methods was 1. 

402: The multiplicative factors applied to the PRF fluxes were 1.13, 

403: 1.12, 1.00, and 0.93 for Channels 1, 2, 3, and 4 respectively.

404: The IRAC colors of the final photometry are also listed in 

405: Table \ref{tab1}.

406: 

407: We produced point source-subtracted images, using the MOPEX script apex\_qa.pl, 

408: to assess the effectiveness of the extraction and to compare the flux results 

409: obtained with the PRF fitting and aperture photometry. 

410: We found that the PRF fitting occasionally failed, producing a flux that 

411: was much too high. 

412: For the cases where the adjusted-PRF/aperture flux ratio exceeded 1.5, 

413: we adopted the aperture value of the flux and its corresponding 

414: uncertainty was derived as described above.

415: The source-subtracted residual images show even fainter sources but 

416: we did not attempt to extract them. 

417: Additionally, the brighter sources close to saturation are in the 

418: nonlinear regime and therefore do not match the shape of the PRF.

419: We did not attempt to subtract highly saturated sources. 

420: 

421: The source subtracted mosaics for Channels 1 and 4 are shown in 

422: the bottom panels of Figures \ref{fig_ch1} and \ref{fig_ch4}.

423: The four circular areas shown in the bottom panel of Figure \ref{fig_ch1}

424: will be used in Section \ref{sec_num_den} to study the distribution of

425: point sources in different locations within our field of view.

426: The circular areas have a radius of 5\arcmin\ and they are centered on 

427: $l$=359.946, $b=-0.0378$; $l$=0.166, $b$=0.1162; 

428: $l$=0.386, $b$=0.2702; and $l$=0.606, $b$=0.4242.

429:  

430: Figure \ref{fig_det} shows a 10\arcmin $\times$ 10\arcmin\ field of view

431: centered on ($l$=0.3523, $b$=$-0.17427$), marked as a box

432: in the bottom panel of Figure \ref{fig_ch4}.

433: Figure \ref{fig_det} shows the differences in source densities and 

434: extended emission among the different IRAC channels.

435: Residuals in Channel 4 are the smallest because the PRF is better sampled than 

436: in the other channels.

437: 

438: The total number of sources detected at a level of 3$\sigma$ or above

439: in each channel was: 735,020, 700,923, 493,207, and 323,512 for Channels

440: 1, 2, 3, and 4, respectively.

441: All the sources detected and measured by MOPEX at the 3$\sigma$ level 

442: are listed in Table \ref{tab2}, \ref{tab3}, \ref{tab4}, and \ref{tab5} 

443: (shown partially, available entirely in the electronic version).

444: The columns of Table \ref{tab2}--\ref{tab5} are as follows: 

445: Source Identification, IRAC Channel, Position (Equatorial and Galactic), 

446: Flux in mJy, Flux uncertainty in mJy, Number of Observations (BCD frames 

447: used in the Flux measurement), Signal-to-noise ratio, and Flux Method 

448: as explained in the previous section.

449: 

450: The cumulative distribution of positional uncertainties is shown 

451: in Figure \ref{fig_pos_unc}, 

452: for both right ascension (open symbols)

453: and declination (filled symbols).

454: The overall distributions of positional uncertainties are

455: similar between right ascension and declination,

456: for all four IRAC channels.

457: We found that 90\% of the sources in our survey have positional 

458: uncertainties less than 0.13\arcsec, 0.16\arcsec, 0.48\arcsec, 

459: and 1.18\arcsec \ for Channels 1, 2, 3, and 4, respectively,

460: corresponding to the typical positional uncertainties in our survey.

461: Also, 99\%, 96\%, 59\%, and 36\% of the Channel 1, 2, 3, and 4

462: sources, respectively, have a positional uncertainty less than 

463: 0.2\arcsec.

464: In the process of merging the IRAC point source lists with 2MASS

465: (see details below), we found systematic offsets between IRAC positions

466: in different channels and between the IRAC and 2MASS positions. 

467: These offsets are $\sim -$0.25\arcsec \ in 

468: right ascension, and $\sim$0.15\arcsec \ in declination,

469: and have been applied to Ch. 1-4 such that the IRAC astrometry 

470: in all channels should 

471: now match the 2MASS astrometry in our final catalog.

472: 

473: Figure \ref{fig_flux_unc} shows the cumulative distribution of percentage 

474: flux uncertainty, derived as described above.

475: The cumulative distribution of the percentage flux uncertainty for the

476: four IRAC channels is shown in separate panels.

477: The open symbols denote the cumulative distribution of percentage flux

478: uncertainties for all the sources in each of the IRAC channels.

479: We found that 90\% of the sources in our survey have a percentage

480: flux uncertainty less than 4.0\%, 5.0\%, 31\% and 28\%  for

481: Channels 1, 2, 3, and 4, respectively, corresponding to the typical 

482: percentage flux uncertainties in our survey.

483: Also, 99\%, 99\%, 64\%, 56\% of the Channel 1, 2, 3, and 4

484: sources, respectively, have a percentage flux uncertainty less than 10\%.

485: The solid, dashed, and dotted lines in Figure \ref{fig_flux_unc} 

486: show the cumulative distribution of percentage flux uncertainties for 

487: sources of 

488: three different source brightness ranges, {\it bright}, {\it medium}, and 

489: {\it faint}, respectively, as defined in Section \ref{sec_num_den}. 

490: The distribution of percentage flux uncertainties for Channels 3 and 4 is

491: dominated by the distribution of percentage flux uncertainties for faint

492: sources. 

493: Channels 3 and 4 mosaics show a wide range in background levels

494: on top of which faint sources are measured. 

495: Variations in the local background may be the cause of the larger

496: uncertainties in the flux measured in those channels.

497: 

498: \subsection{Source extraction from sub-array data}

499: 

500: In order to recover useful photometry from the small saturated

501: regions in the full-array observations, we performed photometry

502: on sub-array data, which consists of mosaics of the Central

503: Cluster (SgrA) and the Quintuplet Cluster, plus 12 individual 

504: pointings.

505: We used the interactive IDL program xstarfinder \citep{dio00},

506: because the parameters used for the full-array data using MOPEX were

507: not appropriate for the small sub-array observations.

508: We also tested photometry using the IRAF source extraction program

509: ``daophot'' but found that daophot gave less reliable results than 

510: xstarfinder.

511:  

512: The PRF was constructed from a composite of sub-array observations of well

513: exposed, isolated single sources.

514: These sources were chosen from the twelve individual pointing observations, 

515: excluding observations with higher than typical noise or

516: with other stars within a $\sim$10\arcsec\ radius of the main source.

517: PRF's were made using both xstarfinder and daophot and it was determined

518: that the point source subtracted residual images were superior for the

519: daophot PRFs; thus, the PRFs from daophot were used in the source extraction.

520: 

521: For each input mosaic, the surface brightness error per pixel was

522: computed from each input mosaic directly.

523: This error is photon-noise-dominated and well fit with a gaussian distribution.

524: The flux errors for the extracted point sources are statistical only and

525: do not reflect differences in the flux estimate that may arise from

526: methodology, e.g., using a different set of extraction parameters such

527: as the background smoothing box.

528: We expect that the systematic errors may

529: exceed the quoted random errors.

530: The source extraction computes a correlation factor, which is a measure

531: of the goodness of fit to the PRF, with 1.0 being a perfect fit.

532: The correlation factor for all extracted sources was 0.75 at minimum,

533: but most sources had a factor exceeding 0.9.

534: A median smoothing size of 7 times the FWHM was used for the background

535: determination.

536: 

537: Table \ref{tab6} lists the sub-array photometric results including 13

538: sources in the dozen individual pointings or ``sat''  fields, 104 sources

539: located in the Central Cluster or ``sgra'' field, and 90 sources 

540: located in the Quintuplet Cluster or ``quint'' field.

541: The sub-array source table lists the brightness and its uncertainty

542: in magnitudes for each IRAC Channel.

543: 

544: %RESULTS

545: 

546: \section{Catalog of Point Sources in the Galactic Center \label{sec_cat}}

547: 

548: We bandmerged our IRAC full-array source list for 

549: each channel with the sources 

550: in the 2MASS catalog \citep{skr06} located in the same field of view 

551: as our IRAC observations. 

552: The merging procedure was done as follows: We first matched and merged,

553: via a positional association, Channel 1 and 2, then Channel 3, then 

554: Channel 4, and finally 2MASS. 

555: A match was defined as the closest counterpart within a 1\arcsec\ radius.

556: The position used for the final merged list was always that corresponding 

557: to the shortest IRAC wavelength at which a source is seen.

558: The sub-array photometry was incorporated into the band merged list.

559: The full-array photometry of each of the 13 sources in the ``sat'' 

560: fields was individually replaced by the sub-array photometry.

561: All the sources with full-array photometry located within 

562: 40\arcsec\ of the Central Cluster 

563: (RA=17 45 40.0, DEC=$-29$ 00 28) and within 

564: 42\arcsec\ of the Quintuplet Cluster 

565: (RA=17 46 16.1, DEC=$-28$ 53 43) were discarded. 

566: The sub-array photometry of the ``sgra'' and ``quint'' fields were added

567: to the band merged list.

568: 

569: The merged source list was further studied and additional flags 

570: regarding the detection reliability, position comments, and

571: photometric quality were added.

572: Magnitudes for each of the IRAC channels were computed using 

573: zero point fluxes of 280.9 Jy, 179.7 Jy, 115.0 Jy, and 64.13 Jy,

574: for IRAC Channels 1, 2, 3, and 4 respectively, as provided by

575: \citet{rea05}.

576: 

577: One of the qualities studied in the band merged source list was

578: flux saturation for the full-array photometry. 

579: We needed to explore whether the saturation values for point sources 

580: provided by the IRAC documentation were appropriate for our survey.

581: Figure \ref{fig_sat} shows color magnitude diagrams (CMDs) using each 

582: IRAC magnitude and the $K_s$ magnitude from 2MASS

583: \citep{skr06}. 

584: Only non-saturated $K_s$ magnitudes with 2MASS photometric quality 

585: flags (ph\_qual) equal to `A' (SNR$>$10) are included in Figure \ref{fig_sat}.

586: The gray scale shows the number density distribution of

587: sources, with white being the highest density.

588: Any anomalies in the bright regime of the CMDs

589: are due to non-linear and saturation effects in the IRAC magnitudes.

590: The horizontal dotted lines show the magnitudes 

591: corresponding to the saturation 

592: fluxes of 190 mJy, 200 mJy, 1400 mJy, and 740 mJy 

593: (7.92, 7.38, 4.79, and 4.84 magnitudes) for IRAC Channels 1, 2, 3, 

594: and 4 respectively, as provided by the $Spitzer$ Observer's Manual, 

595: Version 7.1.

596: The slanted dashed lines correspond to the completeness limit of $K_s$=12.3

597: for the 2MASS point source catalog within a 6\degree \ radius

598: of the Galactic center \citep[see Sec IV.7, ][]{cut03}. 

599: Figure \ref{fig_sat} also demonstrates that the saturation fluxes 

600: provided by the $Spitzer$ Observer's Manual (Version 7.1) are indeed

601: appropriately applied to our survey, as can be seen from the 

602: anvil-shaped tops of the CMDs for Channels 1 and 2.

603: Note that the point source fluxes are superimposed on a high

604: background, and the combination of both is likely to explain 

605: the saturation level appearing to be conservative when it really is 

606: appropriate.

607: Saturated sources are retained in our point source catalog, as long as

608: they are recognized as a point source by the apex.pl script.

609: If the flux of a source is greater than the saturation flux

610: provided by the $Spitzer$ Observer's Manual, 

611: then all its measured quantities are kept in the catalog, but

612: the flux flag

613: of that source in that channel is set to `3' in our final catalog.

614: 

615: The coverage of our survey has some incompleteness due to the fact

616: that the four IRAC cameras do not see exactly the same region of the sky. 

617: The coverage for each source at each channel was determined by

618: measuring the value of the pipeline coverage map at the position

619: of each source. 

620: The coverage value is the same as the number of available BCD frames at 

621: the position of each source, and it is also listed in our

622: final catalog for each channel.

623: The location of each of the sources is flagged in our catalog

624: by the Position flag. 

625: If a source is located in the area of incomplete coverage

626: (coverage value is equal to zero in at least one channel),

627: then the Position flag is set to `0'.

628: Sub-array photometry has been incorporated into the catalog,

629: in particular in the location near the

630: Central Cluster and near the Quintuplet Cluster.

631: If a source is located within 40\arcsec \ of the Central Cluster

632: (RA=17 45 40.0, DEC=$-29$ 00 28), the Position flag is set to `2'.

633: If a source is located within 42\arcsec \ of the Quintuplet Cluster

634: (RA=17 46 16.1, DEC=$-28$ 53 43), the Position flag is set to `3'.

635: The sources with Position flag set to `2' and `3' have photometry 

636: from the sub-array observations.

637: 

638: The reliability of the sources in our band merged list can

639: be estimated by applying the ``2+1" criterion as defined

640: by the Galactic Legacy Infrared Mid-Plane Survey Extraordinaire

641: (GLIMPSE) v1.5 Data Products Description (available from 

642: the GLIMPSE Documents web page 

643: \footnote{http://www.astro.wisc.edu/sirtf/docs.html}).

644: The $M/N$ ratio is defined as the ratio between $M$ number of

645: detections over $N$ number of possible observations

646: (coverage value).

647: The ``2+1" criterion requires $M/N\geq0.6$ in one IRAC

648: band, $M/N\geq0.4$ in an adjacent band.

649: Sources that satisfy the GLIMPSE ``2+1" criterion have the

650: ``2+1" flag set to `1' in our final catalog.

651: 

652: Our final catalog of point sources in the Galactic center is

653: listed in Table \ref{tab7}.

654: The columns of our point source catalog are explained as follows:

655: 

656: \begin{description}

657: 

658: \item[Column 1, Source ID:] Designation of the detected source.

659: 

660: \item[Column 2, R.A.:] Right Ascension in J2000.

661: 

662: \item[Column 3, Dec.:] Declination in J2000.

663: 

664: \item[Column 4, $l$:] Galactic Longitude.

665: 

666: \item[Column 5, $b$:] Galactic Latitude.

667: 

668: \item[Column 6, ``2+1" Flag:] Set to `1' when the source 

669: satisfies the GLIMPSE ``2+1" criterion ($M/N\geq0.6$ in one IRAC

670: band, $M/N\geq0.4$ in an adjacent band), set to `0' otherwise.

671: 

672: \item[Column 7, Pos. Flag.:] Set to `0' when the source

673: is located in areas of incomplete coverage, set to `2'

674: when the source is within 40\arcsec \ of the Central Cluster, 

675: set to `3' when the source is within 42\arcsec \ of the Quintuplet 

676: Cluster, otherwise is set to `1'.

677: 

678: \item[Column 8, 2MASS ID:] 2MASS identification number from the 2MASS Catalog,

679: set to `none' when there is no 2MASS counterpart.

680: 

681: \item[Column 9, $J$:] $J$ magnitude from the 2MASS Catalog, set to `$-9.999$'

682: when not detected.

683: 

684: \item[Column 10, $J$ unc.:] Uncertainty of $J$ magnitude from the 2MASS 

685: Catalog, set to `$-9.999$' when not measured.

686: 

687: \item[Column 11, $H$:] $H$ magnitude from the 2MASS Catalog, set to `$-9.999$'

688: when not detected.

689: 

690: \item[Column 12, $H$ unc.:] Uncertainty of $H$ magnitude from the 2MASS 

691: Catalog, set to `$-9.999$' when not measured.

692: 

693: \item[Column 13, $K_s$:] $K_s$ magnitude from the 2MASS Catalog, set to 

694: `$-9.999$' when not detected.

695: 

696: \item[Column 14, $K_s$ unc.:] Uncertainty of $K_s$ magnitude from the 

697: 2MASS Catalog, set to `$-9.999$' when not measured.

698: 

699: \item[Column 15, Qual. Flag:] Photometric quality flag (ph\_qual) 

700: from the 2MASS Catalog. 

701: It is composed of three letters, one for each 2MASS filter ($JHK_s$). 

702: The letters can be: 

703: A (SNR $>$ 10), B (SNR $>$ 7), C (SNR $>$ 5), D (no SNR requirement),

704: E (poor profile-fit photometry), 

705: F (detection without photometric uncertainty), 

706: U (detection with upper limit on magnitude), X (detection without 

707: brightness estimate). Set to `ZZZ' when there is no 2MASS counterpart.

708: 

709: \item[Column 16, ch1 ID:] Channel 1 identification number 

710: from point source list, also listed in Column 1 of Table \ref{tab2},

711: set to `none' when there is no Channel 1 counterpart.

712: 

713: \item[{Column 17, [3.6]:}] Channel 1 magnitude, computed using the Flux from 

714: Column 7 of Table \ref{tab2}, and the corresponding zero point flux, 

715: set to `$-9.999$' when not detected in this IRAC channel.

716: 

717: \item[{Column 18, [3.6] unc.:}] Uncertainty of Channel 1 magnitude, 

718: computed using the Flux and its uncertainty from Columns 7 and 8 of 

719: Table \ref{tab2}, and the corresponding zero point flux, 

720: set to `$-9.999$' when not detected in this IRAC channel.

721:  

722: \item[{Column 19, [3.6] SNR:}] Signal to Noise ratio of Channel 1 magnitude.

723: set to `$-9.9$' when not detected in this IRAC channel.

724: 

725: \item[Column 20, ch1 Flag:] Channel 1 magnitude flag. 

726: Set to `1' when flux and hence magnitude comes from PRF fitting algorithm, 

727: set to `2' when flux and hence magnitude comes from aperture corrected 

728: measurement, 

729: set to `3' when the full-array flux is greater than corresponding saturation 

730: limit,

731: set to `4' when the photometry comes from the sub-array observations,

732: set to `0' when there is no detection in this IRAC channel.

733: 

734: \item[Column 21, ch1 Cov.:] Number of available BCD frames at the 

735: position of the source for the corresponding IRAC channel ($N$).

736: 

737: \item[Column 22, ch1 $M/N$:] Ratio between $M$ number of detections 

738: over $N$ number of possible observations (ch1 Cov.).

739: 

740: \item[Column 23, ch2 ID:] Same as Column 16, but for Channel 2.

741: 

742: \item[{Column 24, [4.5]:}] Same as Column 17, but for Channel 2.

743: 

744: \item[{Column 25, [4.5] unc.:}] Same as Column 18, but for Channel 2.

745: 

746: \item[{Column 26, [4.5] SNR:}] Same as Column 19, but for Channel 2.

747: 

748: \item[Column 27, ch2 Flag:] Same as Column 20, but for Channel 2.

749: 

750: \item[Column 28, ch2 Cov.:] Same as Column 21, but for Channel 2.

751: 

752: \item[Column 29, ch2 $M/N$:] Same as Column 22, but for Channel 2.

753: 

754: \item[Column 30, ch3 ID:] Same as Column 16, but for Channel 3.

755: 

756: \item[{Column 31, [5.8]:}] Same as Column 17, but for Channel 3.

757: 

758: \item[{Column 32, [5.8] unc.:}] Same as Column 18, but for channel 3.

759: 

760: \item[{Column 33, [5.8] SNR:}] Same as Column 19, but for Channel 3.

761: 

762: \item[Column 34, ch3 Flag:] Same as Column 20, but for Channel 3.

763: 

764: \item[Column 35, ch3 Cov.:] Same as Column 21, but for Channel 3.

765: 

766: \item[Column 36, ch3 $M/N$:] Same as Column 22, but for Channel 3.

767: 

768: \item[Column 37, ch4 ID:] Same as Column 16, but for Channel 4.

769: 

770: \item[{Column 38, [8.0]:}] Same as Column 17, but for Channel 4.

771: 

772: \item[{Column 39, [8.0] unc.:}] Same as Column 18, but for Channel 4.

773: 

774: \item[{Column 40, [8.0] SNR:}] Same as Column 19, but for Channel 4

775: 

776: \item[Column 41, ch4 Flag:] Same as Column 20, but for Channel 4.

777: 

778: \item[Column 42, ch4 Cov.:] Same as Column 21, but for Channel 4.

779: 

780: \item[Column 43, ch4 $M/N$:] Same as Column 22, but for Channel 4.

781: 

782: \end{description}

783: 

784: There are a total of 1,065,565 sources in our final catalog; 

785: 656,673 of those satisfy the GLIMPSE ``2+1" criterion and have a 

786: SNR$>$ 10; they are considered to be highly reliable sources.

787: We summarize other relevant statistics for our catalog in Table \ref{tab8}.

788: 

789: Two asteroids were found in the field of view at the time of

790: our observations. 

791: Asteroid Alikoski appears in the final catalog as sources SSTGC 0629833 and

792: SSTGC 0636843 (twice because it moved between Channel 3 and 4 coverage) 

793: and asteroid 459 Signe is in the final catalog as source SSTGC 0216539.

794: 

795: Areas close to the edges of our survey overlap with the GLIMPSE II 

796: (PI: Churchwell) observations.

797: We have compared the photometry of our sources with SNR$>$10 to the

798: photometry of the GLIMPSE II Highly Reliable Catalog.

799: There are 184,392, 160,387, 132,077, and 72,065 sources positionally matched

800: within 1\arcsec\  between our catalog and GLIMPSE's. 

801: The mean difference in magnitudes between our photometry

802: and GLIMPSE is 0.11, 0.06, 0.00, and 0.03 for Channels 1, 2, 3, and 4.

803: The standard deviations of the same differences are 0.15, 0.15, 0.17, and

804: 0.18 magnitudes for Channels 1, 2, 3, and 4, respectively. 

805: The GLIMPSE photometry is expected to be have uncertainties less than

806: 0.2 mags for most of its sources, according to the GLIMPSE Quality

807: Assurance Document, v1.0. Thus, the observed differences are therefore less

808: than the expected photometric uncertainties.

809: 

810: 

811: %DISCUSSION

812: 

813: \section{Magnitude Distributions \label{sec_num_den}}

814: 

815: Figure \ref{fig_num_den} shows the magnitude distribution of point source

816: detections for each of the channels in our survey. 

817: Only the sources located within the uniform coverage box 

818: ($-1.0 \leq l \leq 1.0, -0.7 \leq b \leq 0.7$) are included in 

819: the determination of the magnitude distributions. 

820: The open symbols show the magnitude distribution for all the sources 

821: and the filled symbols show the distribution for the sources which

822: satisfy the ``2+1" criterion and have a SNR greater than 10.

823: 

824: Figure \ref{fig_num_den} shows that the magnitude distributions 

825: have a similar shape in all the IRAC channels.

826: There is a steep slope of increasing number of stars with 

827: increasing magnitude at the brightest magnitudes.

828: This steep slope flattens around a {\it bright turnover} magnitude of 

829: 9.0, 8.6, 8.2, and 8.2 magnitudes for Channels 1, 2, 3 and 4, respectively.

830: The number of stars increases more slowly with increasing magnitude

831: at magnitudes fainter than the {\it bright turnover}.

832: Finally, there is a {\it faint cutoff} followed by 

833: a steep slope of decreasing number of stars with increasing magnitude.

834: The {\it faint cutoff} for all the sources (open symbols in Figure

835: \ref{fig_num_den}) is 12.4, 12.1, 11.7, and 11.2 for Channels 1, 2, 

836: 3, and 4, respectively.

837: The {\it faint cutoff} for the sources which satisfy the ``2+1'' 

838: criterion and have SNR $>$ 10 (filled symbols in Figure \ref{fig_num_den}) 

839: is 12.0, 11.8, 11.2, and 10.8 for Channels 1, 2, 3, and 4, respectively.

840: Hereafter we define the {\it bright} magnitude range as those 

841: magnitudes brighter than the {\it bright turnover} magnitude, 

842: the {\it medium} magnitude range as those magnitudes between

843: the {\it bright turnover} and the {\it faint cutoff}, and 

844: the {\it faint} magnitude range as those magnitudes fainter

845: than the {\it faint cutoff}.

846: 

847: We plot a subset of magnitude distributions drawn from

848: small regions within our GC mosaic, to understand

849: features of the magnitude distribution.

850: Only sources satisfying the ``2+1" criterion and with SNR$>$10 are 

851: included in the determination of these magnitude distributions.

852: Figure \ref{fig_num_den_area} shows the magnitude distributions 

853: of four circular areas located along a diagonal going away from the 

854: GC, but avoiding dark clouds, as plotted in the bottom panel of 

855: Figure \ref{fig_ch1}.

856: The solid, dotted, dashed and dashed-dotted lines 

857: show the magnitude distributions of these four

858: locations in the order of increasing distance to the GC.

859: There are about 2800, 2700, 2000, and 1200

860: sources in each 5\arcmin\ radius circular area

861: for Channels 1, 2, 3, and 4, respectively.

862: 

863: Figure \ref{fig_num_den_area} shows that, 

864: at this angular scale (10\arcmin), 

865: the magnitude distributions have the same shape as the 

866: magnitude distribution for the entire field of view 

867: (2.0\degree$\times$1.4\degree) as shown in Figure \ref{fig_num_den}.

868: The main difference among the magnitude distributions of

869: the circular areas is that they seem to shift towards fainter 

870: magnitudes with increasing distance to the GC. 

871: As a consequence, the circular area located at the GC 

872: (solid line) has at least a factor of 3 more bright sources

873: than the circular area located farthest from the GC 

874: (dashed-dotted line). For example at [3.6]=9.0, [4.5]=9.0,

875: [5.8]=7.0, and [8.0]=7.5 magnitudes,

876: the ratio of bright sources between the two areas 

877: is 3.0, 3.5, 5.0, and 5.5, respectively.

878: 

879: The {\it faint cutoff} of the magnitude distributions

880: can be interpreted as due to confusion.

881: Figure \ref{fig_num_den_area} shows that confusion is

882: occurring at brighter magnitudes as one gets closer to the GC.

883: The confusion limits suggested in Figure \ref{fig_num_den_area}

884: for Channels 1 and 2 range from about 8.5 mag in the circular area

885: located at the GC (solid line) to 13 mag in the circular area

886: located the farthest away from the GC (dashed-dotted line).

887: The same range of confusion limits for Channels 3 and 4 in

888: Figure \ref{fig_num_den_area} is 8.0 (solid line) to 12.0

889: (dashed-dotted line).

890: 

891: The fact that we observe more bright sources with decreasing 

892: distance to the GC

893: is consistent with previous population studies based on dereddened

894: $K-$band luminosity functions.

895: \citet{blu96} computed a dereddened $K-$band luminosity function

896: within 1\arcmin\ (2.3 pc) of the GC and compare it with a similar 

897: study at Baade's Window in the bulge.

898: \citet{blu96} found that both $K-$band luminosity functions had

899: the same slope, but there was an overabundance

900: of bright stars in the GC relative to Baade's Window.

901: \citet{nar96} constructed a dereddened $K-$band luminosity

902: function for a region of 16\arcmin $\times$16\arcmin\  (37$\times$37

903: pc) centered on the GC, but excluding the inner 2\arcmin\ of

904: the Galaxy.

905: They found a luminosity function intermediate between that

906: of Baade's Window and the inner 2\arcmin\ of the GC, having

907: an excess of luminous stars over the bulge but not as many

908: luminous stars as the inner 2\arcmin.

909: \citet{fig04} obtained dereddened 2 \micron\ luminosity 

910: functions, using high angular resolution observations.

911: They computed synthetic luminosity functions using stellar evolution

912: models and concluded that the observations were best fitted by models 

913: of continuous star formation.

914: 

915: All of the near infrared luminosity functions outside of the Central

916: Cluster show the presence of a {\it bright turnover}.

917: The luminosity function of the central 200 pc is known 

918: to have an excess of luminous stars relative to

919: bulge fields such as Baade's Window, and this excess 

920: of luminous stars increases closer to the GC

921: \citep{cat90,blu96,nar96,phi99,fig04}.

922: The variation in the number of the brightest stars 

923: with distance to the GC over a large area may cause 

924: the presence of the {\it bright turnover}.

925: This speculation is complicated by several effects.

926: First, the {\it bright turnover} for Channels 1 and 2,

927: at the areas close to the GC, occurs very close to the 

928: saturation limit (Ch. 1 and 2 magnitudes of 7.9 and 7.4, respectively).

929: Second, the extinction and local background in the GC are 

930: highly spatially variable, even within a 10\arcmin\ field of view.

931: Finally, stellar crowding may artificially enhance the bright

932: end of measured luminosity functions as pointed out by \citet{dep93}.

933: Higher angular resolution surveys of the GC area may indeed

934: improve our knowledge of the nature of the {\it bright} range sources

935: seen in this survey.

936: 

937: The magnitude distribution shown in Figure \ref{fig_num_den}

938: can be understood as the integral of individual magnitude

939: distributions such as those plotted in Figure \ref{fig_num_den_area}.

940: The integral of the individual magnitude distributions is

941: non-trivial to calculate due to the wide range in stellar density, 

942: extinction, background levels, and confusion within our field of view.

943: Complete modeling of the observed magnitude distribution is beyond

944: the scope of the present work and may be addressed in a

945: future work.

946: 

947: \section{Source distribution with Galactic Longitude and 

948: Latitude \label{sec_gcoord}}

949: 

950: The overall density of detected point sources with 

951: latitude and longitude is essentially constant,

952: a consequence of being confusion limited.  

953: The relatively constant number of sources per 

954: circular area also indicates that our images are confusion limited.

955: However, interesting structure along Galactic latitude and longitude

956: does appear when we select sources within different magnitude ranges

957: (defined above).

958: The point source distributions along Galactic latitude and longitude 

959: in the different magnitude ranges are shown in Figures 

960: \ref{fig_glat_mag} and \ref{fig_glon_mag}.

961: The different panels show the distribution of the point sources

962: for {\it bright}, {\it medium}, and {\it faint}, as indicated.

963: Circles, triangles, squares, and pentagons correspond to the

964: Galactic coordinate distributions of Channels 1, 2, 3, and 4,

965: respectively.

966: 

967: The structure seen at the {\it bright} range is consistent with 

968: the fact that more {\it bright} sources are observed towards

969: the GC (as discussed in Section \ref{sec_num_den}).

970: This agrees well with previous population studies

971: that find an excess of luminous stars in the GC relative to

972: bulge fields \citep{cat90,blu96,nar96,phi99,fig04}.

973: 

974: The structure seen at the {\it faint} range is set by our ability

975: to detect faint sources. The {\it faint} range contains sources

976: below the lowest confusion limit for each channel, and therefore

977: they follow the trend of the variation of the confusion with Galactic

978: latitude and longitude.

979: 

980: The structure along Galactic latitude has a similar shape for all

981: the IRAC channels, but some features are more prominent at longer

982: wavelengths.

983: The sources in the {\it bright} range show a steeply decreasing

984: gradient with Galactic latitude, with a peak in the position of the

985: Galactic center. The {\it medium} range sources show a slowly

986: decreasing gradient with Galactic latitude.

987: Those sources in the {\it faint}

988: range show an increasing gradient with Galactic latitude, with the

989: minimum at the position of the Galactic center.

990: 

991: Figure \ref{fig_glon_mag} illustrates that

992: the structure along Galactic longitude also has the same shape for all

993: the IRAC channels.

994: The sources in the {\it bright} and {\it medium} brightness range 

995: show a slow decrease with Galactic longitude, 

996: with a peak at the position of the Galactic center. 

997: The sources in the {\it faint} range 

998: show an increase with Galactic longitude, with the 

999: minimum at the position of the Galactic center.

1000: 

1001: \section{Color-Magnitude and Color-Color Diagrams \label{sec_cmd_col}}

1002: 

1003: Figures \ref{fig_cmd1} and \ref{fig_cmd2} show the 

1004: [8.0] vs. [3.6]-[8.0] color-magnitude diagram (CMD) of all the 

1005: sources satisfying the ``2+1" criterion and having both [3.6] 

1006: and [8.0] magnitudes with a SNR greater than 10. 

1007: The gray scale shows the number density distribution of

1008: sources, with white being the highest density.

1009: The arrows show the direction of the reddening vector, using the 

1010: extinction law from \citet{ind05}.

1011: The red arrow shows the amount of extinction for $A_K$=1.0,

1012: while the purple arrow shows the amount of extinction for $A_K$=6.5.

1013: According to the extinction maps of \citet{sch99} and \citet{dut03} an

1014: extinction of $A_K$=1.0 is observed at the edges of our survey

1015: which we adopt as the minimum foreground extinction towards 

1016: GC stars in our field of view.

1017: An extinction of $A_K$=6.5 was measured as the maximum observed

1018: extinction within 2\arcmin\  of the Galaxy by the

1019: the near infrared photometric work of \citet{blu96}.

1020: The highest density of points in the CMD shows a well defined sequence 

1021: of constant [3.6]-[8.0] color, at a color of $\sim$0.2 mag. 

1022: The distribution is skewed towards red colors, which is consistent

1023: with varying amounts of extinction.

1024: 

1025: To determine what types of objects are seen in our survey, we have 

1026: overplotted the location of evolved stars in the CMD of Figure 

1027: \ref{fig_cmd1}. 

1028: The location of evolved stars is taken from the 

1029: [8.0] vs. [3.6]-[8.0] CMD of stars in 

1030: the $Spitzer$ SAGE survey of the Large Magellanic

1031: Cloud \citep{blu06}.

1032: They determined the location of the tip of the red giant branch,

1033: the O-rich and C-rich AGB stars, supergiant stars, and extreme

1034: AGB stars 

1035: from the DENIS and 2MASS analysis of LMC stars by \citet{cio06}. 

1036: We assume a distance modulus to the LMC of 18.48 magnitudes

1037: \citep{bor04}, and a distance to the Galactic center of 8.0 kpc

1038: \citep{rei93} to determine their location in our observed CMD.

1039: In Figure \ref{fig_cmd1}, the solid line boxes show the location

1040: of objects assuming an extinction of $A_K$=1.0 magnitudes, and the dashed

1041: line boxes show the same boxes assuming an extinction of $A_K$=6.5

1042: magnitudes.

1043: 

1044: \citet{blu06} also noted the location of background galaxies

1045: in their CMD.

1046: The cyan line in Figure \ref{fig_cmd1} shows the limit

1047: below which background galaxies should be observed,

1048: assuming an extinction of $A_K$=1.0 magnitudes.

1049: This line is at the edge of our observing limit, and it

1050: would be even lower if more extinction is added to it.

1051: We conclude that our survey is very unlikely to include

1052: background galaxies.

1053: 

1054: In Figure \ref{fig_cmd1}, 

1055: different colored boxes illustrate the location of

1056: different types of stars, including

1057: red giants (red),

1058: O-rich stars (blue), 

1059: C-rich stars (purple), 

1060: extreme AGB stars (yellow), 

1061: and supergiants (green).

1062: The bottom of the solid red giant box (at [8.0]=11.67 mag.) 

1063: marks the [8.0] magnitude of a K0 III star located at the GC 

1064: observed through $A_K$=1.0 magnitudes of extinction.

1065: There are 183,857 points sources plotted in figure \ref{fig_cmd1}.

1066: About 78\% of the point sources shown (143,039 in total)

1067: lie within the limits of the red solid box denoting the location 

1068: of the red giant stars with spectral types later than K0 III.

1069: The location of all the point sources with colors bluer than

1070: [3.6]-[8.0]=2.0 and with [8.0] magnitudes brighter than 8.0

1071: can be understood as evolved stars seen through

1072: varying amounts of extinction with the range of values discussed above.

1073: 

1074: There are 917 sources in our catalog with [3.6]-[8.0] $\geq$ 2.0

1075: and [8.0] $\geq$ 8.0.

1076: To explore the possibility of finding young stellar objects 

1077: among these 917 red objects, we have overplotted the location of 

1078: YSOs in the CMD of Figure \ref{fig_cmd2}.

1079: As in Figure \ref{fig_cmd1}, the solid line boxes show the location

1080: of objects assuming an extinction of $A_K$=1.0 magnitudes, and the dashed

1081: line boxes show the same boxes assuming an extinction of $A_K$=6.5

1082: magnitudes.

1083: The cyan line shows the [8.0] magnitude of the brightest 

1084: low-mass

1085: YSO observed in Taurus \citep{har05}, assuming a distance to 

1086: Taurus of 140 pc.

1087: This line demonstrate that low-mass YSOs cannot be detected in our survey.

1088: \citet{whi04} studied the giant HII region RCW 49, as part of

1089: the GLIMPSE legacy program. 

1090: They determined the location of 2.5, 3.8, and 5.9 \msol \ YSOs, 

1091: using the radiative transfer models from \citet{whi03}.

1092: We assume a distance of 4.2 kpc to RCW 49 \citep{chu04} 

1093: and a Galactic center distance of 8 kpc \citep{rei93} to 

1094: determine their location in our observed CMD.

1095: The red boxes denote the location of the 3.8 \msol \ YSOs, and the

1096: blue boxes denote the location of 5.9 \msol \ YSOs.

1097: The dotted yellow boxes show the location of evolved stars,

1098: as reference, assuming an extinction of $A_K$=6.5 magnitudes, 

1099: as shown in Figure \ref{fig_cmd1}.

1100: 

1101: Figure \ref{fig_colcol} shows the [3.6]-[4.5] vs. [5.8]-[8.0]

1102: color-color diagram.

1103: The gray scale shows the number density distribution of

1104: sources, with white being the highest density.

1105: In Figure \ref{fig_colcol}, the solid line boxes show the location

1106: of objects assuming an extinction of $A_K$=1.0 magnitudes, and the dashed

1107: line boxes show the same boxes assuming an extinction of $A_K$=6.5

1108: magnitudes.

1109: \citet{mar06} derived colors of AGB stars by convolving

1110: observed ISO spectra with IRAC bandpasses.

1111: The location of these derived IRAC colors are coincident with the

1112: models of \citet{gro06} computed using stellar atmosphere

1113: models with dust envelopes of different composition.

1114: The red boxes show the location of AGB star colors from \citet{mar06}.

1115: \citet{mar06} also show that an AGB star with a thick envelope

1116: (V354 Lac) may have a unreddened [5.8]-[8.0] color between 2.4 and 2.9,

1117: and a unreddened [3.6]-[4.5] color between 0.05 and 0.15.

1118: The red line shows the location of V354 Lac with the corresponding

1119: amounts of extinction. 

1120: The blue boxes show the location of 3.8 \msol \ and 5.9 \msol \ 

1121: YSOs \citep{whi04}.

1122: If we consider the typical uncertainties discussed in Section \ref{full_extr},

1123: the typical uncertainty in the [3.6]-[4.5] color is 0.06 magnitudes and

1124: the typical uncertainty in the [5.8]-[8.0] color is 0.42 magnitudes.

1125: There are 176,724 points sources plotted in Figure \ref{fig_colcol}.

1126: About 38\% of the point sources shown (66,988 in total)

1127: have zero IRAC colors within the typical uncertainties.

1128: These point sources have been exposed to little reddening, and hence

1129: they may be foreground objects or objects away from the Galactic plane.

1130: 

1131: As the GC is a known region of recent star formation, the possibility 

1132: of observing a YSO population is an exciting prospect. 

1133: We are likely to be sensitive only to the most massive YSOs (if present),

1134: at the distance of the GC, due to confusion. 

1135: Our survey, however, contains sources observed at different distances 

1136: and over varying amounts of extinction.

1137: We must carefully examine any candidate YSO population that we identify

1138: based on IRAC colors, to distinguish foreground YSOs (in the star forming

1139: arms along the line of sight, for instance) with a range

1140: of masses from massive YSOs at the GC.

1141: In addition, thick envelope AGB stars and YSOs have similar IRAC 

1142: colors, and distinguishing one from the other will require additional 

1143: diagnostics. 

1144: Future work will include incorporating photometry at longer infrared 

1145: wavelengths (e.g. ISOGAL, MSX and eventually MIPS 24 $\mu$m) and 

1146: spectroscopy in order to best determine the nature of this population of 

1147: objects with red IRAC colors.

1148: 

1149: %CONCLUSIONS

1150: \section{Conclusions}

1151: 

1152: Our conclusions can be summarized as follows:

1153: 

1154: \begin{itemize}

1155: 

1156: \item A point source catalog of 1,065,565 objects is presented.

1157: The catalog includes positions, $J$, $H$, $K_s$, [3.6], [4.5],

1158: [5.8], and [8.0] magnitudes, and a series of flags that

1159: assess the quality of the measurements.

1160: 

1161: \item The point source catalog is confusion limited.

1162: The confusion limits vary by 2 to 3 magnitudes

1163: within the field of view.

1164: Nevertheless, the average confusion limits  

1165: are 12.4, 12.1, 11.7, and 11.2 magnitudes

1166: for Channels 1, 2, 3, and 4, respectively.

1167: 

1168: \item The overall distribution of point sources with Galactic

1169: latitude and longitude is essentially constant 

1170: (a consequence of being confusion limited), but structure

1171: does appear when sources of different magnitude ranges 

1172: are selected.

1173: Bright stars show a slow decrease in number density with Galactic

1174: longitude, and a steeper decrease with Galactic

1175: latitude, with a peak at the position of the Galactic center.

1176: 

1177: \item Most of the point sources in our

1178: catalog have IRAC magnitudes and colors characteristic

1179: of red giant stars and AGB stars.

1180: There are several hundreds of extremely red objects, however,

1181: some of which may be massive YSOs.

1182: Follow up observations are needed to determine the nature

1183: of the extremely red objects.

1184: 

1185: \end{itemize}

1186: 

1187: 

1188: %ACKNOWLEDGEMENTS

1189: \acknowledgments

1190: This work is based on observations made with the 

1191: {\it Spitzer Space Telescope}, 

1192: which is operated by the Jet Propulsion Laboratory, 

1193: California Institute of Technology under a contract with NASA. 

1194: Support for this work was provided by NASA through an 

1195: award issued by JPL/Caltech.

1196: This publication makes use of data products from the Two Micron 

1197: All Sky Survey, which is a joint project of the University of 

1198: Massachusetts and the Infrared Processing and Analysis 

1199: Center/California Institute of Technology, funded by the National 

1200: Aeronautics and Space Administration and the National Science Foundation.

1201: The research described in this paper was partially carried out at the

1202: Jet Propulsion Laboratory, California Institute of Technology, under

1203: contract with the National Aeronautics and Space Administration.

1204: KS thanks the NASA Faculty Fellowship Program for financial

1205: support and the hospitality of JPL's Long Wavelength Center

1206: and the Spitzer Science Center.

1207: We thank S. Carey, P. Lowrance, R. Blum, C. Koresko, D. Shupe, M. Meade, 

1208: and B. Babler for enlightening discussions.

1209: 

1210: 

1211: %REFERENCES

1212: %BIBLIOGRAPHY

1213: \begin{thebibliography}{}

1214: 

1215: \bibitem[Allen et al.(1990)]{all90}

1216: Allen, D. A., Hyland, A. R., Hillier, D. J. 

1217: 1990, \mnras, 244, 706

1218: 

1219: %\bibitem[Allen et al.(2004)]{all04} 

1220: %Allen, L. E., et al.

1221: %2004, \apjs, 154, 363

1222: 

1223: \bibitem[Bally et al.(1987)]{bal87}

1224: Bally, J., Stark, A. A., Wilson, R. W., Henkel, C. 

1225: 1987, \apjs, 65, 13

1226: 

1227: \bibitem[Blum et al.(2006)]{blu06} 

1228: Blum, R. D., et al.

1229: 2006, \aj, 132, 2034

1230: 

1231: \bibitem[Blum et al.(2003)]{blu03}

1232: Blum, R. D., Ram\'{\i}rez, S. V., Sellgren, K., Olsen, K.

1233: 2003, \apj, 597, 323

1234: 

1235: \bibitem[Blum et al.(1996)]{blu96} 

1236: Blum, R. D., Sellgren, K., DePoy, D. L.

1237: 1996, \apj, 470, 864

1238: 

1239: \bibitem[Blum et al.(1995)]{blu95}

1240: Blum, R. D., DePoy, D. L., Sellgren, K.

1241: 1995, \apj, 441, 603

1242: 

1243: \bibitem[Boldyrev and Yusef-Zadeh(2006)]{bol06}

1244: Boldyrev, S. \& Yused-Zadeh, F.

1245: 2006, \apj, 637, 101L

1246: 

1247: \bibitem[Borissova et al.(2004)]{bor04} 

1248: Borissova, J., Minniti, D., Rejkuba, M., Alves, D., 

1249: Cook, K. H., Freeman, K. C.

1250: 2004, \aap, 97, 109

1251: 

1252: \bibitem[Catchpole et al.(1990)]{cat90}

1253: Catchpole, R. M., Whitelock, P. A., Glass, I. S.

1254: 1990, \mnras, 247, 479

1255: 

1256: \bibitem[Churchwell et al.(2004)]{chu04} 

1257: Churchwell, E., et al.

1258: 2004, \apjs, 154, 322

1259: 

1260: \bibitem[Cioni et al.(2006)]{cio06} 

1261: Cioni, M.-R. L., Girardi, L., Marigo, P., Habing, H. J. 

1262: 2006, \aap, 448, 77

1263: 

1264: \bibitem[Cotera et al.(1996)]{cot96}

1265: Cotera, A. S., Erickson, E. F., Colgan, S. W. J., Simpson, J. P.,

1266: Allen, D. A., Burton, M. G.

1267: 1996, \apj, 461, 750

1268: 

1269: \bibitem[Cutri et al.(2003)]{cut03} 

1270: Cutri, R. M., et al.

1271: "Explanatory Supplement to the 2MASS All Sky Data Release and 

1272: Extended Mission Products", 

1273: http://www.ipac.caltech.edu/2mass/releases/allsky/doc/explsup.html

1274: 

1275: \bibitem[DePoy et al.(1993)]{dep93}

1276: DePoy, D. L., Terndrup, D. M., Frogel, J. A., Atwood, B., Blum, R.

1277: 1993, \aj, 105, 2121

1278: 

1279: \bibitem[Diolaiti et al.(2000)]{dio00} 

1280: Diolaiti, E., Bendinelli, O., Bonaccini, D., Close, L., Currie, 

1281: D., Parmeggiani, G.

1282: 2000, \aaps, 147, 335

1283: 

1284: \bibitem[Dutra et al.(2003)]{dut03}

1285: Dutra, C. M., Santiago, B. X., Bica, E. L. D., Barbuy, B.

1286: 2003, \mnras, 338, 253

1287: 

1288: \bibitem[Epchtein et al.(1997)]{epc97}

1289: Epchtein, N., et al.

1290: 1997, ESO Messenger, 87, 27

1291: 

1292: \bibitem[Fazio et al.(2004)]{faz04} 

1293: Fazio, G. G., et al.

1294: %Hora, J. L., Allen, L. E., 

1295: %Ashby, M. L. N., Barmby, P., Deutsch, L. K., Huang, J.-S., Kleiner, S., 

1296: %Marengo, M., Megeath, S. T., Melnick, G. J., Pahre, M. A., Patten, B. M., 

1297: %Polizotti, J., Smith, H. A., Taylor, R. S., Wang, Z., Willner, S. P., 

1298: %Hoffmann, W. F., Pipher, J. L., Forrest, W. J., McMurty, C. W., 

1299: %McCreight, C. R., McKelvey, M. E., McMurray, R. E., Koch, D. G., 

1300: %Moseley, S. H., Arendt, R. G., Mentzell, J. E., Marx, C. T., Losch, P., 

1301: %Mayman, P., Eichhorn, W., Krebs, D., Jhabvala, M., Gezari, D. Y., 

1302: %Fixsen, D. J., Flores, J., Shakoorzadeh, K., Jungo, R., Hakun, C., 

1303: %Workman, L., Karpati, G., Kichak, R., Whitley, R., Mann, S., 

1304: %Tollestrup, E. V., Eisenhardt, P., Stern, D., Gorjian, V., 

1305: %Bhattacharya, B., Carey, S., Nelson, B. O., Glaccum, W. J., Lacy, M., 

1306: %Lowrance, P. J., Laine, S., Reach, W. T., Stauffer, J. A., 

1307: %Surace, J. A., Wilson, G., Wright, E. L., Hoffman, A., Domingo, G., Cohen, M.,

1308: 2004, \apjs, 154, 10

1309: 

1310: \bibitem[Felli et al.(2002)]{fel02}

1311: Felli, M., Testi, L., Schuller, F., Omont, A.

1312: 2002, \aap, 392, 971

1313: 

1314: \bibitem[Figer et al.(1999)]{fig99}

1315: Figer, D. F., Kim, S. S., Morris, M., Serabyn, E., Rich, R. M., McLean, I. S.

1316: 1999, \apj, 525, 750

1317: 

1318: \bibitem[Figer et al.(2004)]{fig04}

1319: Figer, D. F., Rich, R. M., Kim, S. S., Morris, M., Serabyn, E.

1320: 2004, \apj, 601, 319

1321: 

1322: \bibitem[Ghez et al.(2005)]{ghe05}

1323: Ghez, A., Salim, S., Hornstein, S. D., Tanner, A., Lu, J. R., Morris, M.,

1324: Becklin, E. E., Duchane, G.

1325: 2005, \apj, 620, 744

1326: 

1327: \bibitem[Groenewegen(2006)]{gro06} 

1328: Groenewegen, M. A. T.

1329: 2006, \aap, 448, 181

1330: 

1331: \bibitem[G\"{u}sten and Downes(1980)]{gus80}

1332: G\"{u}sten, R. \& Downes, D.

1333: 1980, \aap, 99, 27

1334: 

1335: \bibitem[Habing et al.(1983)]{hab83}

1336: Habing, H. J., Olnon, F. M., Winnberg, A., Matthews, H. E., Baud, B.

1337: 1983, \aap, 128, 230

1338: 

1339: \bibitem[Haller(1992)]{hal92}

1340: Haller, J. W. 

1341: 1992, Ph. D. thesis, University of Arizona

1342: 

1343: \bibitem[Hartmannn et al.(2005)]{har05} 

1344: Hartmann, L., Megeath, S. T., Allen, L., Luhman, K., Calvet, N., 

1345: D'Alessio, P., Franco-Hernandez, R., Fazio, G.

1346: 2005, \apj, 629, 881

1347: 

1348: \bibitem[Indebetouw et al.(2005)]{ind05} 

1349: Indebetouw, R., et al.

1350: %Mathis, J. S.,

1351: %Babler, B. L., Meade, M. R., Watson, C., Whitney, B. A., Wolff, M. J., 

1352: %Wolfire, M. G., Cohen, M., Bania, T. M., Benjamin, R. A., Clemens, D. P.,

1353: %Dickey, J. M., Jackson, J. M., Kobulnicky, H. A., Marston, A. P., 

1354: %Mercer, E. P., Stauffer, J. R., Stolovy, S. R., Churchwell, E.,

1355: 2005, \apj, 619, 931

1356: 

1357: \bibitem[Krabbe et al.(1991)]{kra91}

1358: Krabbe, A., Genzel, R., Drapatz, S., Rotaciuc, V.

1359: 1991, \apj, 382, L19

1360: 

1361: \bibitem[Krabbe et al.(1995)]{kra95}

1362: Krabbe, A., et al.

1363: 1995, \apj, 447, L95

1364: 

1365: \bibitem[Lebofsky et al.(1982)]{leb82}

1366: Lebofsky, M. J., Rieke, G. H., Tokunaga, A. T.

1367: 1982, \apj, 263, 736

1368: 

1369: \bibitem[Libonate et al.(1995)]{lib95}

1370: Libonate, S., Pipher, J. L., Forrest, W. J., Ashby, M. L. N. 

1371: 1995, \apj, 439, 202

1372: 

1373: \bibitem[Lindqvist et al.(1992)]{lin92}

1374: Lindqvist, M., Habing, H. J., Winnberg, A. 

1375: 1992, \aap, 259, 118

1376: 

1377: \bibitem[Marengo et al.(2006)]{mar06} 

1378: Marengo, M., Hora, J. L., Barmby, P., Willner, S. P., 

1379: Allen, L. E., Schuster, M. T., Fazio, G. G.

1380: 2006, ASP conference series, in press (astro-ph/0611346)

1381: 

1382: \bibitem[Martin et al.(2004)]{mar04}

1383: Martin, C. L., Walsh, W. M., Xiao, K., Lane, A. P., Walker, C. K., Stark, A.

1384: 2004, \apjs, 150, 239

1385: 

1386: \bibitem[Morris and Serabyn(1996)]{mor96}

1387: Morris, M. \& Serabyn, E. 

1388: 1996, \araa, 34, 645

1389: 

1390: \bibitem[Nagata et al.(1996)]{nag96}

1391: Nagata, T., Kawara, K., Onaka, T., Kitamura, Y., Okuda, H.

1392: 1996, \aap, 315L, 205

1393: 

1394: \bibitem[Narayanan et al.(1996)]{nar96}

1395: Narayanan, V. K., Gould, A., DePoy, D. L.

1396: 1996, \apj, 472, 183

1397: 

1398: \bibitem[Oka et al.(2005)]{oka05}

1399: Oka, T., Geballe, T. R., Goto, M., Usuda, T., McCall, B. J.

1400: 2005 \apj, 632, 882

1401: 

1402: \bibitem[Omont et al.(2003)]{omo03}

1403: Omont, A., et al.

1404: 2003, \aap, 403, 975

1405: 

1406: \bibitem[Philipp et al.(1999)]{phi99}

1407: Philipp, S., Zylka, R., Mezger, P. G., Duschl, W. J., Herbst, T., Tuffs, R. J. 

1408: 1999, \aap, 348, 768

1409: 

1410: \bibitem[Price et al.(2001)]{pri01}

1411: Price, S. D., et al.

1412: 2001, \aj, 121, 2819

1413: 

1414: \bibitem[Reach et al.(2005)]{rea05} 

1415: Reach, W. T., et al.

1416: %Megeath, S. T., 

1417: %Cohen, M., Hora, J., Carey, S., Surace, J., Willner, S. P., Barmby, P., 

1418: %Wilson, G., Glaccum, W., Lowrance, P., Marengo, M., Fazio, G. G.,

1419: 2005, \pasp, 117, 978

1420: 

1421: \bibitem[Reid(1993)]{rei93} 

1422: Reid, M. J.

1423: 1993, \araa, 31, 345

1424: 

1425: \bibitem[Sellgren et al.(1987)]{sel87}

1426: Sellgren, K., Hall, D. N. B., Kleinmann, S. G., Scoville, N. Z. 

1427: 1987, \apj, 317, 881

1428: 

1429: \bibitem[Sch\"{o}del et al.(2003)]{sch03}

1430: Sch\"{o}del, R., Ott, T., Genzel, R., Eckart, A., Mouawad, N., Alexander, T.

1431: 2003, \apj, 596, 1015

1432: 

1433: \bibitem[Schuller et al.(2006)]{sch06}

1434: Schuller, F., Omont, A., Glass, I. S., Schultheis, M., Egan, M. P., Price, S. D.

1435: 2006, \aap, 452, 535

1436: 

1437: \bibitem[Schultheis et al.(1999)]{sch99}

1438: Schultheis, M., et al.

1439: 1999, \aap, 349, 69

1440: 

1441: \bibitem[Sjouwerman et al.(1998)]{sjo98}

1442: Sjouwerman, L. O., van Langevelde, H. J., Winnberg, A., Habing, H. J.

1443: 1998, \apjs, 128, 35

1444: 

1445: \bibitem[Skrutskie et al.(2006)]{skr06} 

1446: Skrutskie, M. F., et al.

1447: %Cutri, R. M., 

1448: %Stiening, R., Weinberg, M. D., Schneider, S., Carpenter, J. M., 

1449: %Beichman, C., Capps, R., Chester, T., Elias, J., Huchra, J., Liebert, J., 

1450: %Lonsdale, C., Monet, D. G, Price, S., Seitzer, P., Jarrett, T., 

1451: %Kirkpatrick, J. D., Gizis, J., Howard, E., Evans, T., Fowler, J., 

1452: %Fullmer, L., Hurt, R., Light, R., Kopan, E. L., Marsh, K. A., 

1453: %McCallon, H. L., Tam, R., Van Dyk, S., \& Wheelock, S., 

1454: 2006, \aj, 131, 1163

1455: 

1456: \bibitem[Tamblyn et al.(1996)]{tam96}

1457: Tamblyn, P., Rieke, G. H., Hanson, M. M., Close, L. M., 

1458: McCarthy, D. W., Jr., Rieke, M. J.

1459: 1996, \apj, 456, 206

1460: 

1461: \bibitem[Werner et al.(2004)]{wer04} 

1462: Werner, M. W., et al.

1463: %Roellig, T. L., 

1464: %Low, F. J., Rieke, G. H., Rieke, M., Hoffmann, W. F., Young, E., 

1465: %Houck, J. R., Brandl, B., Fazio, G. G., Hora, J. L., Gehrz, R. D., 

1466: %Helou, G., Soifer, B. T., Stauffer, J., Keene, J., Eisenhardt, P., 

1467: %Gallagher, D., Gautier, T. N., Irace, W., Lawrence, C. R., Simmons, L., 

1468: %Van Cleve, J. E., Jura, M., Wright, E. L., Cruikshank, D. P.,

1469: 2004, \apjs, 154, 1

1470: 

1471: \bibitem[Whitney et al.(2004)]{whi04} 

1472: Whitney, B. A., et al.

1473: 2004, \apjs, 154, 315

1474: 

1475: \bibitem[Whitney et al.(2003)]{whi03} 

1476: Whitney, B. A., Wood, K., Bjorkman, J. E., Cohen, M. 

1477: 2003, \apj, 598, 1079

1478: 

1479: \end{thebibliography}

1480: 

1481: %TABLES

1482: 

1483: \clearpage

1484: \input{tab1}

1485: \input{stub.tab2}

1486: \input{stub.tab3}

1487: \input{stub.tab4}

1488: \input{stub.tab5}

1489: \input{stub.tab6}

1490: \input{stub.tab7}

1491: \input{tab8}

1492: 

1493: 

1494: %FIGURES

1495: 

1496: \clearpage

1497: \begin{figure}

1498: \epsscale{.75}

1499: \plotone{f1a.ps}\\

1500: ~\\

1501: ~\plotone{f1b.ps}

1502: \epsscale{1.0}

1503: \figcaption{

1504: IRAC Channel 1 (3.6 \micron) mosaic ({\it top panel}) and source subtracted 

1505: mosaic ({\it bottom panel}) of the Galactic center, covering an area of 

1506: 2.0\degree\ $\times$ 1.4\degree, centered on $l$=0.0, $b$=0.0

1507: (Galactic north is up, Galactic east is to the left).

1508: The mosaics are shown in reverse grayscale with the same scale.

1509: The circular areas shown in the {\it bottom panel} are centered on 

1510: $l$=359.946, $b=-0.0378$; $l$=0.166, $b$=0.1162; 

1511: $l$=0.386, $b$=0.2702; and $l$=0.606, $b$=0.4242.

1512: These circular areas are used in this paper to study the distribution of point

1513: sources in locations with different source densities.

1514: \label{fig_ch1}}

1515: \end{figure}

1516: 

1517: \clearpage

1518: \begin{figure}

1519: \epsscale{.75}

1520: \plotone{f2a.ps}\\

1521: ~\\

1522: ~\plotone{f2b.ps}

1523: \epsscale{1.0}

1524: \figcaption{

1525: IRAC Channel 4 (8 \micron) mosaic ({\it top panel}) and source subtracted 

1526: mosaic ({\it bottom panel}) of the Galactic center, covering an area of 

1527: 2.0\degree$\times$1.4\degree, centered on $l$=0.0, $b$=0.0

1528: (Galactic north is up, Galactic east is to the left).

1529: The mosaics are shown in reverse grayscale with the same scale.

1530: The box plotted in the {\it bottom panel} shows the location of the blown up

1531: section detailed on Figure \ref{fig_det}.

1532: \label{fig_ch4}}

1533: \end{figure}

1534: 

1535: \clearpage

1536: \begin{figure}

1537: \epsscale{.50}

1538: \plotone{f3a.ps}\\

1539: ~\\

1540: ~\plotone{f3b.ps}

1541: \epsscale{1.0}

1542: \figcaption{Detail of the IRAC mosaics and source subtracted mosaics

1543: showing 10\arcmin\ $\times$ 10\arcmin\ field of view,

1544: centered on ($l$=0.3523, $b$=$-0.17427$).

1545: The original mosaics are shown in the top set of four panels and the

1546: corresponding source subtracted images are shown in the bottom set. 

1547: Each IRAC channel is labeled in the top right corner of the 

1548: individual images.

1549: Note the differences in source densities and extended emission among the 

1550: different IRAC channels.

1551: The residuals from the point sources are larger in Channel 1, and smaller in

1552: Channel 4, because the PRF is better sampled at longer wavelengths.

1553: \label{fig_det}}

1554: \end{figure}

1555: 

1556: \clearpage

1557: \begin{figure}

1558: \plotone{f4.eps}

1559: \figcaption{Cumulative distribution of positional uncertainties.

1560: Open symbols denote the cumulative distribution of positional 

1561: uncertainties in right ascencsion and the filled symbols denote the cumulative 

1562: distribution of positional uncertainties in 

1563: declination, both in units of arcsec.

1564: All sources observed in each IRAC channel are included in these

1565: cumulative distributions.

1566: \label{fig_pos_unc}}

1567: \end{figure}

1568: 

1569: \clearpage

1570: \begin{figure}

1571: \plotone{f5.eps}

1572: \figcaption{Cumulative distribution of flux uncertainties.

1573: All sources observed in each IRAC channel are included in the

1574: cumulative distributions, plotted with filled symbols. 

1575: The $solid$, $dashed$, and $dotted$ $lines$ show the

1576: cumulative distribution of flux uncertainties for sources of

1577: three different source brightness ranges, respectively bright, medium, and

1578: faint.

1579: \label{fig_flux_unc}}

1580: \end{figure}

1581: 

1582: \clearpage

1583: \begin{figure}

1584: \plotone{f6.ps}

1585: \figcaption{Color-magnitude diagrams (CMDs), 

1586: or plots of the difference between $K_s$ and IRAC magnitude vs. 

1587: IRAC magnitude.

1588: The CMD for the four IRAC magnitudes are shown in the corresponding

1589: panels.

1590: Only non-saturated high quality IRAC and $K_s$ magnitudes (IRAC magnitudes

1591: with SNR$>$10 and 2MASS photometric quality flag equal to `A' (SNR$>$10)) 

1592: are included in this figure.

1593: The gray scale shows the number density distribution of

1594: sources, with white being the highest density.

1595: The {\it dotted line} shows the magnitude corresponding to the saturation

1596: fluxes of 190 mJy, 200 mJy, 1400 mJy, and 740 mJy 

1597: (7.92, 7.38, 4.79, and 4.84 magnitudes) for IRAC Channels

1598: 1, 2, 3, and 4 respectively, as provided by the $Spitzer$

1599: Observer's Manual (SOM).

1600: The {\it dashed lines} corresponds to the completeness limit of $K_s$=12.3

1601: magnitudes of the 2MASS point source catalog within a 6\degree \ radius

1602: of the Galactic center.

1603: \label{fig_sat}}

1604: \end{figure}

1605: 

1606: \clearpage

1607: \begin{figure}

1608: \plotone{f7.eps}

1609: \figcaption{Number density distribution of point sources for each

1610: of the IRAC channels.

1611: Only the sources located within $-1.0 \leq l \leq 1.0$ and 

1612: $-0.7 \leq b \leq 0.7$ are included in the determination of the 

1613: distribution.

1614: The open symbols show the magnitude distribution for all the sources

1615: and the filled symbols show the distribution for sources satisfying

1616: the ``2+1" criterion and having SNR$>$10.

1617: The total number of sources used in the determination of the 

1618: distributions is listed on the top left side of each panel.

1619: The bin size is 0.1 magnitudes.

1620: The {\it dashed lines} show the limits of three brightness ranges

1621: defined to study the distribution of point sources with $l$ and $b$.

1622: \label{fig_num_den}}

1623: \end{figure}

1624: 

1625: \clearpage

1626: \begin{figure}

1627: \plotone{f8.eps}

1628: \figcaption{Magnitude distribution within a 5\arcmin \ radius 

1629: in locations of the survey with different source densities 

1630: located along a diagonal going away from the GC, as plotted 

1631: in the {\it bottom panel} of 

1632: Figure \ref{fig_ch1}.

1633: The bin size is 0.25 magnitudes.

1634: The magnitude distributions of the four circular areas

1635: are plotted  

1636: ($solid~line$: area centered on the Galactic center,

1637: $l$=359.946, $b=-0.0378$;

1638: $dotted~line$: area centered on $l$=0.166, $b$=0.1162;

1639: $dashed~line$: area centered on $l$=0.386, $b$=0.2702; 

1640: $dashed-dotted~line$: area centered on $l$=0.606, $b$=0.4242).

1641: \label{fig_num_den_area}}

1642: \end{figure}

1643: 

1644: \clearpage

1645: \begin{figure}

1646: \plotone{f9.eps}

1647: \figcaption{Distribution of point sources with Galactic latitude

1648: for the three brightness ranges defined in Fig. \ref{fig_num_den}.

1649: Only the sources located within $-1.0 \leq l \leq 1.0$ and

1650: $-0.7 \leq b \leq 0.7$ are included in the determination of the

1651: number density distribution.

1652: Circles, triangles, squares, and pentagons correspond to the

1653: Galactic coordinate distributions of Channels 1, 2, 3, and 4,

1654: respectively.

1655: \label{fig_glat_mag}}.

1656: \end{figure}

1657: 

1658: 

1659: \clearpage

1660: \begin{figure}

1661: \plotone{f10.eps}

1662: \caption{Distribution of point sources with Galactic longitude

1663: for the three defined  brightness ranges, as indicated.

1664: Only the sources located within $-1.0 \leq l \leq 1.0$ and

1665: $-0.7 \leq b \leq 0.7$ are included in the determination of the

1666: number density distribution.

1667: Symbols are the same as in Figure \ref{fig_glat_mag}.

1668: \label{fig_glon_mag}}

1669: \end{figure}

1670: 

1671: \clearpage

1672: \begin{figure}

1673: \epsscale{.90}

1674: \plotone{f11.ps}

1675: \caption{[8.0] vs. [3.6]-[8.0] color-magnitude diagram.

1676: The gray scale shows the number density distribution of

1677: sources, with white being the highest density.

1678: Only sources satisfying the ``2+1" criterion and having SNR$>$10

1679: are plotted.

1680: The arrows show the direction of the reddening vector, using the

1681: extinction law from \citet{ind05}, and the minimum ($A_K$=1.0)

1682: and maximum ($A_K$=6.5) amount of extinction measured towards the GC

1683: \citep{blu96,sch99,dut03}.

1684: The locations of evolved stars are taken from the 

1685: CMD of the $Spitzer$ SAGE LMC survey \citep{blu06}

1686: and placed at the Galactic center distance:

1687: red giant stars ($red~boxes$), O-rich AGB stars ($blue~boxes$),

1688: C-rich AGB stars ($purple~boxes$), extreme AGB stars ($yellow~boxes$),

1689: and supergiant stars ($green~boxes$).

1690: The $solid~line~boxes$ show the location of objects assuming an 

1691: extinction of $A_K$=1.0 magnitudes, and the $dashed~line~boxes$ 

1692: show the same boxes assuming an extinction of $A_K$=6.5

1693: magnitudes. 

1694: The $cyan~line$ shows the position below which background galaxies

1695: should be located, assuming an extinction of $A_K$=1.0 magnitudes.

1696: The location of all the point sources with colors bluer than

1697: [3.6]-[8.0]=2.0 can be understood as evolved stars seen through

1698: varying amounts of extinction.

1699: \label{fig_cmd1}}

1700: \end{figure}

1701: 

1702: \clearpage

1703: \begin{figure}

1704: \epsscale{.90}

1705: \plotone{f12.ps}

1706: \figcaption{[8.0] vs. [3.6]-[8.0] color-magnitude diagram.

1707: The gray scale shows the number density distribution of

1708: sources, with white being the highest density.

1709: Only sources satisfying the ``2+1" criterion and having SNR$>$10

1710: are plotted.

1711: The arrows show the direction of the reddening vector, using the

1712: extinction law from \citet{ind05}, and the minimum ($A_K$=1.0)

1713: and maximum ($A_K$=6.5) amount of extinction measured towards the GC

1714: \citep{blu96,sch99,dut03}.

1715: The $cyan~line$ shows the [8.0] magnitude of the brightest low-mass

1716: YSO observed in Taurus \citep{har05} at Galactic center distance.

1717: The $red~boxes$ denote the location of 3.8 \msol \ YSOs, and the

1718: $blue~boxes$ denote the location of 5.9 \msol \ YSOs,

1719: as observed in the giant HII region RCW 49 by \citet{whi04}

1720: and placed at the distance of the Galactic center.

1721: The dotted $yellow~boxes$ show the location of evolved stars,

1722: assuming an extinction of $A_K$=6.5 magnitudes,

1723: as shown in Figure \ref{fig_cmd1}, for reference.

1724: \label{fig_cmd2}}

1725: \end{figure}

1726: 

1727: \clearpage

1728: \begin{figure}

1729: \epsscale{.90}

1730: \plotone{f13.ps}

1731: \caption{[3.6]-[4.5] vs. [5.8]-[8.0] color-color diagram.

1732: The gray scale shows the number density distribution of

1733: sources, with white being the highest.

1734: Only sources satisfying the ``2+1" criterion and having SNR$>$10 

1735: are plotted.  

1736: The $solid~line~boxes$ show the location

1737: of objects assuming an extinction of $A_K$=1.0 magnitudes, 

1738: and the $dashed~line~boxes$ show the same boxes assuming an 

1739: extinction of $A_K$=6.5 magnitudes.

1740: The $red~boxes$ show the average location of AGB star colors from

1741: \citet{mar06}.

1742: \citet{mar06} also determine the location of an AGB star 

1743: with a thick envelope (V354 Lac), which is shown with the

1744: $red~lines$.

1745: The $blue~boxes$ show the location of 3.8 \msol \ and 5.9 \msol \

1746: YSOs \citep{whi04}.

1747: \label{fig_colcol}}

1748: \end{figure}

1749: 

1750: 

1751: \end{document}

1752: