1: % instructions: http://www.journals.uchicago.edu/ApJ/instruct.html

2: %\documentclass{emulateapj}

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

4: 

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

6: 

7: %%\documentclass[12pt,manuscript]{aastex}

8: 

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

10: 

11: % \documentclass[preprint2]{aastex}

12: 

13: %% If you want to create your own macros, you can do so

14: %% using \newcommand. Your macros should appear before

15: %% the \begin{document} command.

16: %%

17: %% If you are submitting to a journal that translates manuscripts

18: %% into SGML, you need to follow certain guidelines when preparing

19: %% your macros. See the AASTeX v5.x Author Guide

20: %% for information.

21: 

22: %\newcommand{\vdag}{(v)^\dagger}

23: %%\newcommand{\kms}{km~s$^{-1}$}

24: %\newcommand{\HI}{\ion{H}{1}}

25: %\newcommand{\HII}{\ion{H}{2}}

26: %\newcommand{\myemail}{bianchi@pha.jhu.edu}

27: %\newcommand{\secpoint}{\mbox{$^{\prime\prime}$}}

28: %\newcommand{\as}{\mbox{$^{\prime\prime}$}}

29: %%%\newcommand{\dlam}{\mbox{$\lambda\lambda$}}

30: %\newcommand{\ebv}{$E(B-V)$}

31: %\newcommand{\ala}{$\lambda$}

32: %\newcommand{\Ha}{\mbox{H${\alpha}$}}

33: %\newcommand{\ha}{\mbox{H${\alpha}$}}

34: %%\newcommand{\Te}{\mbox{$T_{\rm e}$}}

35: %%\newcommand{\Ne}{\mbox{$N_{\rm e}$}}

36: %\newcommand{\Teff}{\mbox{$T_{\rm eff}$}}

37: %\newcommand{\Lbol}{\mbox{$L_{\rm bol}$}}

38: %\newcommand{\lbol}{\mbox{$L_{\rm bol}$}}

39: %\newcommand{\Lsun}{\mbox{$L_{\odot}$}}

40: %\newcommand{\Lx}{\mbox{{$L_x$}/$L_{\odot}$}}

41: %\newcommand{\Rstar}{\mbox{$R_{\ast}$}}

42: %\newcommand{\Rsun}{\mbox{$R_{\odot}$}}

43: %\newcommand{\Msun}{\mbox{$M_{\odot}$}}

44: %\newcommand{\msun}{\mbox{$M_{\odot}$}}

45: %%\newcommand{\wzero}{\mbox{$w_{\rm 0}$}}

46: %%\newcommand{\wone}{\mbox{$w_{\rm 1}$}}

47: %%\newcommand{\aone}{\mbox{$\alpha_{\rm 1}$}}

48: %%\newcommand{\atwo}{\mbox{$\alpha_{\rm 2}$}}

49: %\newcommand{\Vinf}{\mbox{$V_{\infty}$}}

50: %\newcommand{\vinf}{\mbox{$V_{\infty}$}}

51: %\newcommand{\MV}{\mbox{$M_V$}}

52: %%\newcommand{\Rt}{\mbox{$R_t$}}

53: %%

54: %\newcommand{\kms}{km s$^{-1}$}

55: %\newcommand{\Myr}{M$_{\odot}$ yr$^{-1}$}

56: %\newcommand{\myr}{M$_{\odot}$ yr$^{-1}$}

57: %%

58: %\newcommand{\mdot}{$\dot M$}

59: %\newcommand{\Mdot}{$\dot M$}

60: 

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

62: 

63: %\slugcomment{To appear in the GALEX special issue of ApJ Letters}

64: 

65: %% If you wish, you may supply running head information, although

66: %% this information may be modified by the editorial offices.

67: %% The left head contains a list of authors,

68: %% usually a maximum of three (otherwise use et al.).  The right

69: %% head is a modified title of up to roughly 44 characters.  Running heads

70: %% will not print in the manuscript style.

71: 

72: %\shorttitle{Star Formation in the Extreme Outer Disk} 

73: %\shortauthors{Thilker \& Bianchi}

74: 

75: %% This is the end of the preamble.  Indicate the beginning of the

76: %% paper itself with \begin{document}.

77: 

78: %\received{2003 June 27}

79: 

80: %\lefthead{Thilker et al.}

81: %\righthead{Star Formation in the Extreme Outer Disk}

82: %\slugcomment{Version of  \today}

83: %\received{2004 May 4}

84: \begin{document}

85: %% LaTeX will automatically break titles if they run longer than

86: %% one line. However, you may use \\ to force a line break if

87: %% you desire.

88: \title{Recent Star Formation in the Extreme Outer Disk of M83}

89: %% Use \author, \affil, and the \and command to format

90: %% author and affiliation information.

91: %% Note that \email has replaced the old \authoremail command

92: %% from AASTeX v4.0. You can use \email to mark an email address

93: %% anywhere in the paper, not just in the front matter.

94: %% As in the title, you can use \\ to force line breaks.

95: \author{

96: David A. Thilker\altaffilmark{1},

97: Luciana Bianchi\altaffilmark{1}, 

98: Samuel Boissier\altaffilmark{2},

99: Armando Gil de Paz\altaffilmark{2},

100: Barry F. Madore\altaffilmark{2},

101: D. Christopher Martin\altaffilmark{3},

102: Gerhardt R. Meurer\altaffilmark{1},

103: Susan G. Neff\altaffilmark{4},

104: R. Michael Rich\altaffilmark{5},

105: David Schiminovich\altaffilmark{3},

106: Mark Seibert\altaffilmark{3},

107: Ted K. Wyder\altaffilmark{3},

108: Tom A. Barlow\altaffilmark{3},

109: Yong-Ik Byun\altaffilmark{6},

110: Jose Donas\altaffilmark{7},

111: Karl Forster\altaffilmark{3},

112: Peter G. Friedman\altaffilmark{3},

113: Timothy M. Heckman\altaffilmark{8},

114: Patrick N. Jelinsky\altaffilmark{9},

115: Young-Wook Lee\altaffilmark{6},

116: Roger F. Malina\altaffilmark{7},

117: Bruno Milliard\altaffilmark{7},

118: Patrick Morrissey\altaffilmark{3},

119: Oswald H. W. Siegmund\altaffilmark{9},

120: Todd Small\altaffilmark{3},

121: Alex S. Szalay\altaffilmark{8}, and

122: Barry Y. Welsh\altaffilmark{9}

123: }

124:  

125: 

126: \altaffiltext{1}{Center for Astrophysical Sciences, The Johns Hopkins

127: University, 3400 N. Charles St., Baltimore, MD 21218, dthilker@pha.jhu.edu}

128: 

129: \altaffiltext{2}{Observatories of the Carnegie Institution of Washington,

130: 813 Santa Barbara St., Pasadena, CA 91101}

131: 

132: \altaffiltext{3}{California Institute of Technology, MC 405-47, 1200 East

133: California Boulevard, Pasadena, CA 91125}

134: 

135: \altaffiltext{4}{Laboratory for Astronomy and Solar Physics, NASA Goddard

136: Space Flight Center, Greenbelt, MD 20771}

137: 

138: \altaffiltext{5}{Department of Physics and Astronomy, University of

139: California, Los Angeles, CA 90095}

140: 

141: \altaffiltext{6}{Center for Space Astrophysics, Yonsei University, Seoul

142: 120-749, Korea}

143: 

144: \altaffiltext{7}{Laboratoire d'Astrophysique de Marseille, BP 8, Traverse

145: du Siphon, 13376 Marseille Cedex 12, France}

146: 

147: \altaffiltext{8}{Department of Physics and Astronomy, The Johns Hopkins

148: University, Homewood Campus, Baltimore, MD 21218}

149: 

150: \altaffiltext{9}{Space Sciences Laboratory, University of California at

151: Berkeley, 601 Campbell Hall, Berkeley, CA 94720}

152: 

153: 

154: 

155: 

156: 

157: %% Notice that each of these authors has alternate affiliations, which

158: %% are identified by the \altaffilmark after each name.  Specify alternate

159: %% affiliation information with \altaffiltext, with one command per each

160: %% affiliation.

161: 

162: %% Mark off your abstract in the ``abstract'' environment. In the manuscript

163: %% style, abstract will output a Received/Accepted line after the

164: %% title and affiliation information. No date will appear since the author

165: %% does not have this information. The dates will be filled in by the

166: %% editorial office after submission.

167: 

168: \begin{abstract}

169: 

170: Ultraviolet imaging with the {\it Galaxy Evolution Explorer} (GALEX)

171: has revealed an extensive sample of UV-bright stellar complexes in the

172: extreme outer disk of M83, extending to about four times the

173: radius where the majority of \ion{H}{2} regions are detected ($R_{HII} =

174: 5.1\arcmin$ or 6.6 kpc).  These sources are typically associated with

175: large-scale filamentary \ion{H}{1} structures in the warped outer disk of M83,

176: and are distributed beyond the galactocentric radii at which molecular

177: ISM has yet been detected.  We present measured properties of these

178: stellar complexes, including FUV and NUV magnitudes and local gas

179: surface density.  Only a subset of the outer disk UV sources have

180: corresponding \ion{H}{2} regions detected in H$\alpha$ imaging,

181: consistent with a sample of mixed age in which some sources are a few

182: Myr old and others are much more evolved ($\sim 10^8$ yr). 

183: % We compare

184: %the outer disk regions with the ensemble of well-studied inner disk

185: %star-forming complexes and the fainter, apparently diffuse UV emission

186: %at large galactocentric radii. These sites of recent (occasionally

187: %ongoing) star formation are driving galaxy evolution at unexpectedly

188: %large galactocentric radii via chemical enrichment and feedback

189: %processes, and call for re-examination of the star formation threshold

190: %issue.

191: 

192: 

193: 

194: 

195: \end{abstract}

196: 

197: %% example has been keyed in ApJ style. See the instructions to authors

198: %% for the journal to which you are submitting your paper to determine

199: %% what keyword punctuation is appropriate.

200: 

201: \keywords{galaxies: individual (M83) --- galaxies: evolution --- ultraviolet: galaxies}

202: 

203: %% From the front matter, we move on to the body of the paper.

204: %% In the first two sections, notice the use of the natbib \citep

205: %% and \citet commands to identify citations.  The citations are

206: %% tied to the reference list via symbolic KEYs. The KEY corresponds

207: %% to the KEY in the \bibitem in the reference list below. We have

208: %% chosen the first three characters of the first author's name plus

209: %% the last two numeral of the year of publication as our KEY for

210: %% each reference.

211: 

212: \section{Introduction}

213: \label{sintro}

214: 

215: The {\it Galaxy Evolution Explorer} (GALEX), with its $1.25\arcdeg$

216: field-of-view and sensitivity to stellar populations younger than a

217: few hundred Myr, is remarkably well-suited for addressing the topic of

218: star formation at large galactocentric radii. Recent star formation

219: within such environments has been detected in deep H$\alpha$ and

220: broadband observations of a few galaxies (NGC 628, NGC1058, NGC6946:

221: Ferguson et al. 1998a, also Lelievre \& Roy 2000 for NGC628; M31:

222: Cuillandre et al. 2001; NGC6822: de Blok \& Walter 2003).  One of the

223: goals of the GALEX Nearby Galaxy Survey (NGS: Bianchi et al. 2004) is

224: to conduct a census of outer disk star formation in a statistically

225: significant sample of galaxies.

226: 

227: The confirmed occurrence of star formation in the outer disk of

228: ordinary spiral galaxies has far-reaching implications.  First, if

229: massive stars are forming at large radii, then the outer disk

230: continues to undergo chemical enrichment.  This may clarify the origin of dust detected in the extended HI disks of spiral

231: galaxies (Popescu \& Tuffs 2003, A\&A 410L, 21P). Secondly, radiative

232: and mechanical feedback processes associated with massive stars are

233: expected to be more efficient in the

234: comparatively rarified interstellar medium (ISM) and uncrowded

235: environment at large radii.  In particular, the unresolved issue of

236: \ion{H}{1} production via H$_2$ dissociation (Smith et al. 2000, Allen

237: 1986) can be studied in detail.

238: %Within the less-crowded,

239: %low-density outer disk, a fiducial stellar cluster can sweep up a much

240: %faster expanding bubble having smaller probability of disruption,

241: %relative to inner disk counterparts (Thilker et al. 1998).  This, in

242: %turn, enhances porosity in the ISM.  The degree of porosity has a

243: %direct influence on the radiative transfer of Lyman continuum emission

244: %from massive stars, and presumably also leakage of ionizing photons to

245: %the intergalactic medium (IGM).  Even without a change in porosity, a

246: %low density environment allows photoionization to proceed more

247: %effectively (volume ionized per photon source).  

248: %Insight gained from our uncomplicated view of outer disk star

249: %forming regions can be applied to the inner disk, to improve

250: %modeling accuracy at all radii.

251: 

252: The presence of recently formed stellar complexes at large galactocentric

253: radii also provides a simplified laboratory for investigating the star

254: formation threshold, the (minimum) gas column density required for

255: star formation to occur spontaneously.  The precise workings of this

256: mechanism remain a matter of debate.  Kennicutt (1989) demonstrated

257: that star formation is rare in disk environments having total gas

258: surface density $\Sigma_{gas} < 5-10$ M$_\sun$ pc$^{-2}$.  Martin \&

259: Kennicutt (2001) showed the observed threshold gas density varies by

260: about an order of magnitude at the ``edge'' of the star-forming disk

261: ($R = R_{HII}$) in different galaxies, but the ratio of $\Sigma_{gas}$

262: to a critical density (which depends on the physical and dynamical

263: conditions of each galaxy in particular) is approximately constant at

264: this position. 

265: % Boissier et al. (2003) concluded that the instability

266: %criterion depends sensitively on the method used to convert from CO to

267: %H$_2$ and whether the pre-existing stellar disk is considered.

268: 

269: Deep H$\alpha$ imaging by Ferguson et al. (1998a) and Lelievre \& Roy

270: (2000) showed that star-formation regions only a few Myr old exist

271: beyond two optical radii (R$_{25}$) in NGG~628, NGC~1058, and

272: NGC~6946.  Lelievre \& Roy (2000) demonstrated that the \ion{H}{2}

273: region luminosity function (LF) slope is significantly steeper in the

274: outer disk, relative to a galaxy-wide sample, possibly indicating a

275: genuine change in the initial cluster mass function (CMF) or a

276: population which is preferentially older.  Ferguson et al. (1998a)

277: noted that the outer \ion{H}{2} regions appear small, isolated, and

278: faint compared to inner disk analogs. 

279: 

280: 

281: In this Letter, we present far-UV (FUV) and near-UV (NUV) wide-field

282: GALEX imagery which reveals an extensive population of recent

283: star-forming regions in the extreme outer disk of M83, out to $\sim

284: 3\times$ the Holmberg radius.  Some of these sources were discovered

285: by Wamsteker et al. (1983) in deep red (127-04+RG630) plates, but

286: observed properties were never published.  We analyze the GALEX data

287: in conjuction with visible and $\lambda$21cm imagery, to place the outer disk

288: sources in context relative to the underlying stellar population and

289: extended \ion{H}{1} disk.  {\it The GALEX data is of particular importance

290: because it demonstrates that H$\alpha$ observations will fail to

291: detect a significant population of moderate age clusters in the

292: outermost disk of spiral galaxies, leading to an underestimate of

293: galactic evolution in these locales.}  Processes such as chemical

294: enrichment (Ferguson et al. 1998b), disk heating (Sellwood \& Balbus

295: 1999), and ISM phase balance (Allen 1986) may be significantly

296: influenced, given that the (predominantly) B star population traced by

297: GALEX effectively drives these feedback mechanisms.  GALEX NGS

298: observations show that M83 is not unique in terms of outer disk UV

299: morphology.  We will analyze UV-visible-IR outer disk properties of a

300: representative galaxy sample in a subsequent paper.

301: 

302: 

303: 

304: \section{Observations and Data Analysis}

305: 

306: M83 (NGC5236) was observed with GALEX on 7 June 2003 for a single

307: orbital night as part of the NGS, with an exposure of 1352 seconds in

308: both FUV (1350--1750~\AA) and NUV (1750--2750~\AA).  Figure 1 shows

309: the color-composite GALEX image of M83.  Two bright stars eastward of

310: M83 prevented us from centering the target in the field, but our

311: data includes most of the outer disk.  RMS sensitivity is 26.6(26.8)

312: AB mag for FUV(NUV), or 27.5(27.6) AB mag arcsec$^{-2}$ expressed in

313: units of surface brightness evaluated at the scale of the PSF.

314: % FUV 0.00078 cnts/s

315: % NUV 0.0021 cnts/s

316: 

317: Boissier et al. (2004, this volume) also analyze the GALEX M83

318: dataset, emphasizing the inner disk in regard to extinction

319: and chemical evolution models.

320: 

321: Courtesy of P. Crosthwaite, we gained access to the $\lambda21$cm VLA

322: observations discussed by Tilanus \& Allen (1993).  For the inner

323: $14\arcmin\times14\arcmin$ of M83, we also used the CO map of

324: Crosthwaite et al. (2002) to estimate $\Sigma_{gas}$ as a function of

325: position.  Figures 1, 4, and 5 pertain to the \ion{H}{1} and CO

326: observations.  Narrowband H$\alpha$ and R-band imaging of M83 was

327: obtained using the Swope 1-m telescope at Las Campanas Observatory

328: (see Fig. 5)

329: 

330: We adopt a distance of 4.5 Mpc to M83 (Thim et al. 2003), thus the

331: $\sim 5\arcsec$ GALEX PSF corresponds to a linear scale of 110

332: pc.  Accordingly, GALEX detections in M83 are generally not single

333: clusters, but rather aggregates possibly consisting of a few discrete

334: (unresolved) clusters.

335: 

336: Photometry of the discrete UV sources within M83 was accomplished

337: using Source Extractor (Bertin \& Arnouts 1996) to define an isophotal

338: aperture for every object detected within our NUV image. These

339: irregular region boundaries were then applied to both the NUV and FUV

340: data for the measurement of background-subtracted source flux, after

341: applying a distortion correction to the FUV image to ensure

342: registration with our NUV data.  The background for every source was

343: estimated locally (within $\sim 250$ pc). We measured median surface

344: brightness as a function of galactocentric radius as described by

345: Bianchi et al. (2004, this volume) with $i = 25\arcdeg$ and PA =

346: $226\arcdeg$ (Crosthwaite et al. 2002).

347: 

348: 

349: 

350: 

351: \section{Recent star formation in the\\ outer disk of M83}

352: 

353: 

354: M83 is considered prototypical of those galaxies showing a

355: well-defined edge of the star-forming disk (eg. Martin \& Kennicutt

356: 2001).  The surface density of \ion{H}{2} regions declines abruptly

357: near R$_{HII} = 5.1\arcmin$ (6.6 kpc) radius (cf. R$_{Holm} =

358: 7.3\arcmin$, 9.6 kpc), suggesting that massive star formation is now

359: confined to M83's inner disk, despite the prominence of a rather

360: extended \ion{H}{1} disk (Rogstad et al. 1974, Huchtmeier \& Bohnenstengel

361: 1981, see also Fig. 1).  Our GALEX observations modify this view.

362: While the distribution of UV-bright stellar complexes is highly

363: concentrated in the area occupied by \ion{H}{2} regions (tracing O

364: stars), many UV sources (tracing O and B stars) are easily

365: recognizable far beyond the ``edge'' of this distribution.

366: Furthermore, diffuse UV emission is also detected beyond the

367: optically-bright star-forming disk.

368: 

369: 

370: \subsection{Outer disk, UV-bright stellar complexes}

371: 

372: Our GALEX observations of M83 revealed $>100$ UV-bright

373: star-forming regions and stellar clusters beyond R$_{HII}$.  Figure 1

374: shows that the outer disk star-forming regions appear preferentially

375: located on local maxima or filaments in the structure of the warped

376: \ion{H}{1} disk.  The distribution of GALEX sources extends to the

377: limit of the field surveyed at the VLA by Tilanus \& Allen (1993).  We

378: suspect that additional M83 star-forming regions may be located in the

379: remainder of the \ion{H}{1} envelope (Huchtmeier \& Bohnenstengel

380: 1981), but this is challenging to confirm without corroborating

381: \ion{H}{1} morphology.

382: 

383: Upon inspection of POSS2 images, we found that the majority of

384: the outer disk GALEX detections could be discerned in IIIaJ+GG385

385: (blue) plates.  However, the apparent contrast of the

386: sources is lower in the visible than at UV wavelengths.  An

387: undetermined fraction of the outer disk star-forming complexes were

388: discovered by Wamsteker et al. (1983) in their analysis of deep red

389: (127-04+RG630) Schmidt plates. To our knowledge, this is the only

390: mention of the outer disk sources in this galaxy prior to the GALEX data.

391: 

392: Figure 2 presents the FUV luminosity function for M83 star-forming

393: complexes in various disk locations.  The outer disk complexes (dashed

394: line) appear systematically fainter than their inner disk counterparts

395: (dotted line). The median FUV luminosity of the inner disk population

396: is a factor of 3 greater than the corresponding figure beyond

397: R$_{HII}$ (only correcting for foreground extinction).  Internal

398: extinction is likely more severe in the inner disk.  Thus, the

399: intrinsic luminosity difference may be even higher than that shown in

400: Fig 2, although we caution that blending and photometric

401: incompleteness (more significant at $R < R_{HII}$) would have an

402: opposite effect.  The shape of the apparent FUV LF differs between the

403: inner and outer disk populations. Our M83 FUV luminosity function of

404: outer disk stellar complexes is steeper than for the inner disk

405: population.  A similar trend was noted in NGC~628 for the H$\alpha$ LF

406: of \ion{H}{2} regions (Levievre \& Roy 2000).  Because our GALEX measurements

407: represent the aggregate flux of more than one independent cluster (or

408: simply an extended distribution of massive stars), it is premature to

409: interpret the precise slope of these FUV LFs in terms of the CMF. We

410: note that blending tends to flatten observed LFs, but the magnitude of

411: this effect may vary with environment.  High-resolution follow-up

412: observations would resolve ambiguities related to the FUV LF and

413: median luminosity estimates. Nevertheless, the maximum UV luminosity

414: exhibited by the population of outer disk sources implies an upper

415: limit of about $10^5 M_\sun$ for the most recently formed complexes

416: (see also Fig. 3).

417: 

418: We compare the observed (NUV, FUV--NUV) color magnitude diagram (CMD)

419: for UV-bright complexes with population synthesis models of Bruzual \&

420: Charlot (2003) in Figure 3.  M83's outer disk sources have FUV--NUV

421: colors generally consistent with ages from a few Myr up to 400 Myr and

422: are generally less massive than sources with $R < R_{HII}$.

423: 

424: Only a subset of the outer disk UV-bright sources have detected

425: counterparts in H$\alpha$ imagery.  This effect could be partially due

426: to a combination of limited sensitivity and the lower expected

427: emission measure of \ion{H}{2} nebulae in a low-density environment, but

428: supports the view that our GALEX observations trace a diverse

429: population, comprised of both young and moderate-age complexes.  A distinct advantage of UV observations is to directly trace stellar, rather than nebular, emission.

430: %  The production

431: %rate of Lyman continuum photons in a coeval cluster drops

432: %precipitously after 6 Myr, when all stars more massive than $\sim 20

433: %M_\sun$ are gone.  The decline of FUV, NUV luminosity is notably less

434: %severe, as B stars dominate this part of the spectrum.

435: 

436: Figure 4 presents the distribution of total gas surface density at the

437: locations of all inner(outer) disk sources using a dotted(dashed)

438: line.  The local surface densities at which outer disk star-forming

439: complexes appear to be forming are typically subcritical, with respect

440: to the disk instability criterion (Martin \& Kennicutt 2001).

441: However, we caution that the outer disk measurements of $\Sigma_{gas}$

442: are lower limits in two important ways.  Most obviously, the estimate

443: of surface density at positions beyond $\sim R_{HII}$ is based only on

444: \ion{H}{1} observations, neglecting the contribution from molecular

445: gas traced by CO (unmapped outside of $14\arcmin\times14\arcmin$).

446: Also, the resolution of the \ion{H}{1} aperture synthesis map is

447: coarse in comparison to the scale of our detections (50$\arcsec$

448: vs. 5$\arcsec$).  Higher-resolution 21cm observations could plausibly

449: boost the apparent gas surface density in the local environment of the

450: UV-bright complexes.

451: 

452: 

453: 

454: 

455: 

456: 

457: \subsection{Underlying low surface brightness disk}

458: 

459: The discrete outer disk stellar complexes lie projected against a

460: fainter, slowly-varying field population.  To place the UV-bright

461: complexes in the context of M83's overall star formation history, we

462: have plotted the median FUV, NUV and H$\alpha$ surface brightness

463: (Fig. 5), plus FUV--NUV (Fig. 6), as a function of galactocentric

464: radius.  The underlying stellar population becomes substantially bluer

465: (in FUV--NUV) with increasing radius.  Dust is known to exist in the

466: outer disks (and even in the extended HI disks) of spiral galaxies

467: (Popescu \& Tuffs 2003) and thus is likely to be present in the outer

468: disk of M83, too, particularly given the star formation traced there

469: by GALEX.  Accordingly, scattered light may contribute to the emission

470: at each position (see also Popescu et al., this volume), tending to

471: smooth the UV profile.  Nevertheless, the FUV--NUV profile (Fig. 6)

472: shows a sequence of breaks within the bright disk of M83.  Outside of

473: R$_{HII}$ the FUV--NUV color becomes bluer almost monotonically,

474: indicating recent star formation dominates the UV emission.  The

475: median H$\alpha$ surface brightness profile (Fig 5.) is observed to

476: drop off much more rapidly than either UV band.  Given that the

477: dynamical timescale beyond $R_{HII}$ is $\ga 60$ Myr, the dropout of

478: H$\alpha$ emission compared to UV cannot likely be attributed to a

479: (recent) coherent decrease in the SFR at large radii.  FUV--NUV

480: measurements for discrete sources are overplotted on Fig. 6.  Within

481: $10\arcmin$ galactocentric radius, the UV-bright star-forming regions

482: generally appear bluer (younger) than or consistent with the FUV--NUV

483: color of the underlying diffuse emission.

484: 

485: 

486: \acknowledgments

487: 

488:  

489: GALEX (Galaxy Evolution Explorer) is a NASA Small Explorer,

490: launched in April 2003.  We gratefully acknowledge NASA's support for

491: construction, operation, and science analysis for the GALEX mission,

492: developed in cooperation with the Centre National d'Etudes Spatiales

493: of France and the Korean Ministry of Science and Technology.

494: 

495:  

496: 

497:  

498: 

499: \begin{references}

500: \reference{}Allen, R. J. 1986, Nature 319, 296

501: \reference{}Bertin, E. \& Arnouts, S. 1996, \aaps ~117, 393

502: \reference{NGS} Bianchi, L. et al. 2004, in "The Local Group as an

503: Astrophysical Laboratory", in press (STScI, M. Livio editor) 

504: \reference{}Boissier, S. et al. 2003, \mnras ~321, 733

505: \reference{BC03}Bruzual, G. \& Charlot, G. 2003, \mnras ~344, 1000

506: \reference{}Crosthwaite, L. P. et al. 2002, \aj ~123, 1892

507: \reference{}Cuillandre, J-C. et al. 2001, \apj ~554, 190

508: \reference{}de Blok \& Walter 2003, \mnras ~341, 39

509: \reference{}Ferguson, A. M. N. et al. 1998a, \apj ~506, 19

510: \reference{}Ferguson, A. M. N. et al. 1998b, \aj ~116, 673

511: al. 1998), 

512: \reference{}Harris et al. 2001, \aj ~122, 3046

513: \reference{}Huchtmeier \& Bohnenstengel 1981, \aap ~100, 72

514: \reference{}Kennicutt, R. C. 1989, \apj ~344, 685

515: \reference{}Lelievre \& Roy 2000, \aj ~120, 1306

516: \reference{}Martin, C. L. \& Kennicutt, R. C. 2001, \apj ~555, 301

517: \reference{}Popescu, C. \& Tuffs, R. 2003, \aap ~410, 21

518: \reference{}Rogstad, D. H., Lockart, I. A. \& Wright, M. C. H. 1974, \apj ~193, 309

519: \reference{}Sellwood, J. A. \& Balbus, S. A. 1999, \apj ~511, 660 

520: \reference{}Smith, D. A. et al. 2000, \apj ~538, 608

521: \reference{}Tilanus, R. P. J. \& Allen, R. J. 1993, \aap ~274, 707

522: \reference{}Thim, F. et al. 2003, \apj ~590, 256

523: \reference{}Wamsteker, W., Lorre, J. J., and Schuster, H. E. 1983, in ``Internal Kinematics and Dynamics of Galaxies'' (Dordrecht, D. Reidel)

524: 

525: \end{references}

526: 

527: \clearpage

528: 

529: 

530: \begin{figure}

531: \epsscale{0.9}

532: \plotone{f1.lowres.ps}

533: \caption{GALEX FUV and NUV color-composite image of M83, highlighting

534: the newly re-discovered sites of recent star-formation in the extreme

535: outer disk. We show FUV in blue and NUV in red (along with their

536: average in green).  A red contour represents the extent of the

537: \ion{H}{1} distribution detected by Tilanus \& Allen (1993). The blue

538: \ion{H}{1} contour is drawn at $1.8\times10^{20}$ cm$^{-2}$,

539: corresponding to $\sim 0.4 M_\sun$ pc$^{-2}$.  We also include a

540: yellow contour from Crosthwaite et al. (2002), indicating total

541: neutral gas surface density, $\Sigma_{gas}$ of $10 M_\sun$ pc$^{-2}$.

542: The yellow $\Sigma_{gas}$ contour lies immediately within R$_{HII}$.

543: The projected spatial extent of the field ($44\arcmin$) is 57.6 kpc

544: (R$_{HII}$ = 6.6 kpc) at a distance of 4.5 Mpc.  }

545: \end{figure}

546: 

547: 

548: 

549: \begin{figure}

550: \epsscale{.85}

551: \plotone{f2.ps}

552: \caption{GALEX FUV luminosity functions of the UV-bright stellar

553: complexes in M83, having photometric error $< 0.25$ mag.  We show the

554: FUV LF for all sources with a solid line, inner disk ($R < R_{HII}$)

555: sources with a dotted red line, and outer disk ($R \ge R_{HII}$)

556: sources with a dashed blue line.  Outer disk complexes appear

557: typically fainter and have a steeper slope in the FUV LF, relative to

558: inner disk objects, although detailed assessment of completeness and

559: blending is needed to bolster these statements.}

560: \end{figure}

561: 

562: \begin{figure}

563: \epsscale{.85}

564: \plotone{f3.ps}

565: \caption{GALEX (FUV--NUV, NUV) color-magnitude diagram for M83's

566: UV-bright stellar complexes (same sample and radial classification as

567: Fig. 2).  Inner disk sources ($R < R_{HII}$) are plotted with red

568: circles, whereas outer disk sources ($R \ge R_{HII}$) are indicated

569: with blue diamonds.  Our data points have been corrected only for

570: Galactic foreground extinction, E(B-V)=0.066. The black curves show

571:  model predictions for starburst populations of 10$^3$, 10$^4$, 10$^5$,

572: and 10$^6$ $M_\sun$ as a function of age (log age = 5.7, 6.0, 6.3,

573: 6.5, 6.6, 6.75, 7.0, 7.3, 7.6, 7.75, 7.9, 8.0, 8.7, 9.0 indicated for

574: 10$^6$ $M_\sun$).  Dashed lines indicate the same models reddened

575: with E(B-V)=0.5, a value slightly higher than the median E(B-V)

576: determined for a large sample of clusters in M83 (Harris et

577: al. 2001).}

578: \end{figure}

579: 

580: \begin{figure}

581: \epsscale{.85}

582: \plotone{f4.ps}

583: \caption{Local $\Sigma_{gas}$ at sites of recent star formation in

584: M83, evaluated at 1.1 kpc (50$\arcsec$) resolution.  The overall

585: distribution (solid) appears largely bimodal, with inner disk UV

586: sources (within R$_{HII}$, dotted red) forming in locales with total

587: surface density greater than $\sim 10 M_\sun$~pc$^{-2}$.  Recently

588: formed, outer disk stellar complexes (dashed blue) exist within

589: environments characterized by local surface density about an order of

590: magnitude smaller, although the CO map used to compute the H$_2$

591: contribution does not extend to large galactocentric radii.

592: High-resolution \ion{H}{1} and CO observations toward the outer

593: UV-bright complexes are critically needed.  Such data would enable

594: investigation of \ion{H}{1} production via H$_2$ photodissociation

595: (Smith et al. 2000), as well as a fundamental check on the star

596: formation threshold (Kennicutt 1989).  We reiterate that this plot

597: shows local measurements of $\Sigma_{gas}$, whereas Fig. 5 presents

598: azimuthally averaged surface density as a function of galactocentric

599: radius.}

600: \end{figure}

601: 

602: \begin{figure}

603: \epsscale{.8}

604: \plotone{f5.ps}

605: \caption{Radial profiles of the {\it median} FUV (blue), NUV (red),

606: and H$\alpha$ (black) surface brightness in M83, plus {\it average}

607: surface densities derived from \ion{H}{1} (green) and CO (orange)

608: data.  Azimuthally averaged $\Sigma_{gas}$ is shown with a dashed

609: line.  The UV profiles are presented in absolute units, while the

610: H$\alpha$ profile is arbitrarily normalized.  We indicate R$_{HII}$

611: with a vertical line, and the UV sky background with horizontal lines.

612: Slight deviations from flatness across the GALEX field, coupled with

613: the fact that M83 is positioned on the edge of the field, make it

614: impractical to follow the profiles beyond $\sim 10\arcmin$. Note the

615: remarkable difference between the H$\alpha$ and UV profiles at

616: (subcritical) galactocentric radii $>5$ kpc.}

617: \end{figure}

618: 

619: \begin{figure}

620: \epsscale{.8}

621: \plotone{f6.ps}

622: \caption{Median FUV--NUV as a function of galactocentric radius, with

623: individual inner (dots) and outer (diamonds) stellar complexes plotted

624: for comparison.  We also show the FUV--NUV color for model starburst

625: populations of varied age and three different periods of continuous

626: star-formation (CSP: 100 Myr, 1 Gyr, and 10 Gyr).  Intrinsic and

627: reddened, E(B-V) = 0.2, colors are indicated with thick and thin line

628: segments, respectively, although the difference is minimal compared to

629: the observed run of FUV--NUV.  The $<\Sigma_{gas}>$ profile (dotted)

630: is duplicated from Fig. 5.}

631: \end{figure}

632: 

633: 

634: \end{document}