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

2: %\usepackage{color}

3: \citestyle{aa}

4: 

5: %\slugcomment{Version of \today}

6: \shorttitle{\ion{O}{6} Absorption in AGN Host Galaxy}

7: \shortauthors{Bonamente \& Dixon}

8: 

9: %% INSERT PERSONAL DEFINITIONS HERE

10: 

11: \newcommand{\cf}{cf.}

12: \newcommand{\eg}{e.g.}

13: \newcommand{\etal}{et~al.}

14: \newcommand{\etc}{etc.}

15: \newcommand{\fig}[1]{Fig.~\ref{#1}}

16: \newcommand{\h}{$^{\rm h}$}

17: \newcommand{\ie}{i.e.}

18: \newcommand{\m}{$^{\rm m}$}

19: 

20: \newcommand{\cfour}{\ion{C}{4}}

21: \newcommand{\cthree}{\ion{C}{3}}

22: \newcommand{\ctwo}{\ion{C}{2}}

23: \newcommand{\hetwo}{\ion{He}{2}}

24: \newcommand{\hone}{\ion{H}{1}}

25: \newcommand{\htwo}{H$_2$}

26: \newcommand{\nfive}{\ion{N}{5}}

27: \newcommand{\nho}{$N$(\hone)}

28: \newcommand{\nht}{$N$(H$_2$)}

29: \newcommand{\nfour}{\ion{N}{4}}

30: \newcommand{\ntwo}{\ion{N}{2}}

31: \newcommand{\nosix}{$N$(\osix)}

32: \newcommand{\oone}{\ion{O}{1}}

33: \newcommand{\osix}{\ion{O}{6}}

34: 

35: \newcommand{\ebv}{$E($\bv)}

36: \newcommand{\f}{{\it f}\/}

37: \newcommand{\flux}{erg cm$^{-2}$ s$^{-1}$ \AA$^{-1}$}

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

39: \newcommand{\logg}{$\log g$}

40: \newcommand{\lu}{photons cm$^{-2}$ s$^{-1}$ sr$^{-1}$}

41: \newcommand{\mdot}{$\dot{M}$}

42: \newcommand{\mdotq}{$\dot{M}q$}

43: \newcommand{\rv}{$R_V$}

44: \newcommand{\specfit}{{\small SPECFIT}}

45: \newcommand{\teff}{$T_{\rm eff}$}

46: \newcommand{\vinf}{$v_{\infty}$}

47: \newcommand{\vlsr}{$V_{\rm LSR}$}

48: \newcommand{\vrot}{$v_{\rm rot}$}

49: 

50: \newcommand{\chips}{{\it CHIPS}}

51: \newcommand{\cop}{{\it Copernicus}}

52: \newcommand{\euve}{{\it EUVE}}

53: \newcommand{\fuse}{{\it FUSE}}

54: \newcommand{\hst}{{\it HST}}

55: \newcommand{\hut}{HUT}

56: \newcommand{\iue}{{\it IUE}}

57: \newcommand{\orf}{{ORFEUS}}

58: \newcommand{\rosat}{{\it ROSAT}}

59: \newcommand{\heao}{{\it HEAO-1}}

60: \newcommand{\na}{New~Astronomy}

61: \newcommand{\source}{2MASX~J21362313-6224008}

62: 

63: %\received{2003 August 21}

64: \begin{document}

65:  

66: \title{FUSE Observations of Galactic and Intrinsic Absorption

67: in the Spectrum of the Seyfert 1 Galaxy 2MASX~J21362313-6224008\/\footnotemark}

68: 

69: \footnotetext[1]{Based on observations made with the NASA-CNES-CSA

70: {\it Far Ultraviolet Spectroscopic Explorer.  FUSE}\/ is operated for NASA by

71: the Johns Hopkins University under NASA contract NAS5-32985.}

72: 

73: \author{Massimiliano Bonamente}

74: \affil{Physics Department, The University of Alabama in Huntsville,

75: John Wright Dr., OB201, Huntsville, AL 35899}

76: \email{bonamem@email.uah.edu}

77: 

78: \and

79: 

80: \author{W.\ Van Dyke Dixon}

81: \affil{Department of Physics and Astronomy, The Johns Hopkins University,

82: 3400 North Charles Street, Baltimore, MD 21218} 

83: \email{wvd@pha.jhu.edu}

84: 

85: \begin{abstract}

86: 

87: We present the far-ultraviolet spectrum of the Seyfert 1 galaxy \source\/ obtained

88: with the {\it Far Ultraviolet Spectroscopic Explorer (FUSE)}.

89: The spectrum features absorption from Galactic \ion{O}{6} at two velocities and 

90: redshifted \hone\ Lyman $\beta$ and $\gamma$, \ion{C}{2}, \ion{C}{3}, and

91: \ion{O}{6}.

92: The redshifted absorption features represent a single kinematic component 

93: blueshifted by $\sim 310$ km s$^{-1}$ relative to the AGN. We 

94: use photoionization models to derive constraints on the physical parameters of the absorbing gas.

95: An alternative interpretation for the absorption lines is also proposed, whereby the

96: absorbing gas is associated with an intervening galaxy cluster.

97: 

98: \end{abstract}

99: \keywords{galaxies: individual (2MASX J21362313-6224008)---galaxies: ISM --- ultraviolet: galaxies --- galaxies: clusters: general}

100: 

101:  

102: %\keywords{ galaxies: individual (\objectname{2MASX~J21362313-6224008}) ---

103: %galaxies: ISM --- ultraviolet: galaxies ---  galaxies: clusters: general}

104: 

105: 

106: \section{INTRODUCTION}

107: A large fraction of Seyfert 1 galaxies exhibits intrinsic UV absorption associated

108: with the active galactic nucleus (AGN).  The absorption lines are commonly

109: blueshifted with respect 

110: to the source, indicating that the absorbing material undergoes a net radial 

111: outflow (\eg, \citealt{Crenshaw:99}).  The absorbing material is interpreted

112: as photoionized gas with ionization parameter $U \sim 0.01-0.1$ and 

113: total hydrogen column density $N_H \sim 10^{18}-10^{21}$ cm$^{-2}$ 

114: \citep{Kriss:00,Kriss:03,Romano:02}.

115: 

116: In this paper we present the results of a 37 ks observation of \source, a

117: Seyfert 1 galaxy with a measured redshift $z=0.0588$ \citep{Hewitt:91}

118: and visual magnitude V=15.2 mag \citep{Remillard:86}, with the

119: {\it Far Ultraviolet Spectroscopic Explorer (FUSE)}.

120: The available X-ray HEAO-1 data  

121: of the source (also identified as 1H2129-624 

122: and H2132-626) were analyzed by \citet{Remillard:86}, who derive a luminosity

123: $L_{2-10\; {\rm keV}}=3 \times 10^{44}$ erg s$^{-1}$,

124: which is on the bright end of the quasar luminosity function \citep{George:00}.

125: The object was observed in soft X-rays by \rosat; the \rosat\/ All-Sky Bright Source Catalogue \citep{Voges:99} and

126: the \rosat\/ Bright Survey \citep{Schwope:00} detect a strong X-ray source at the

127: location of our target, which is identified as 1RXS J213623.1-622400.

128: (Its quoted X-ray flux may be contaminated by up to 10\% by the faint source

129: 1RXS J213530.1-623005 approximately 12 arcmin to the southwest.)

130: \euve\/ observed the source during its  All-Sky Survey, but the short exposure time

131: resulted only in an upper limit \citep{Marshall:95}.

132: The spectral energy distribution (SED) of \source\ is shown in Fig. \ref{sed}.

133: 

134: \section{OBSERVATIONS AND DATA REDUCTION}

135: 

136: \fuse\/ consists of four separate optical systems.  Two employ LiF

137: coatings and are sensitive to wavelengths from 990 to 1187 \AA, while

138: the other two use SiC coatings, which provide reflectivity to

139: wavelengths as short as 905 \AA.  The four channels overlap between 990

140: and 1070 \AA.  For a complete description of the instrument, see \citet{Moos:00}

141: and \citet{Sahnow:00}.

142: 

143: \footnotetext[2]{{\it The CalFUSE Pipeline Reference Guide}\/ is available at

144: \url{http://fuse.pha.jhu.edu/analysis/pipeline\_reference.html}.}

145: 

146: The \fuse\/ spectrum of 2MASX~J21362313-6224008 (data set D9030401) was

147: obtained in 10 separate exposures on 2003 May 25-26.  The total

148: integration time was 37 ks, of which 12 ks were obtained during orbital

149: night.  All observations were made through the 30\arcsec\ $\times$

150: 30\arcsec\ (LWRS) aperture.  The data were reduced using version 2.4 of

151: the CalFUSE calibration software pipeline, described in {\it The

152: CalFUSE Pipeline Reference Guide}\/\footnotemark\

153: \citep{calfuse_pipeline_reference}, but with the following

154: modification:  The first half of the pipeline, which corrects for

155: time-dependent effects (such as spacecraft jitter) was run separately

156: on each exposure.  The resulting position-corrected photon-event lists

157: were combined, using the program {\small TTAG\_COMBINE}, into a single

158: data file.  The second half of the pipeline, which performs background

159: subtraction and spectral extraction (among other tasks), was run on the

160: combined data file.  By thus using the entire data set to scale the

161: background model, we optimize its fidelity, an important consideration

162: for faint continuum sources.

163: 

164: The \fuse\/ flux calibration, based on theoretical models of

165: white-dwarf stellar atmospheres, is believed accurate to about 10\%

166: \citep{Sahnow:00}.  Error bars are assigned to the data assuming

167: Gaussian statistics.  To increase their signal-to-noise ratio, we bin

168: the spectra by four detector pixels, about half of a resolution

169: element.  The FUV spectrum of \source\ (\fig{full_spectrum}) 

170: shows a power-law continuum with a broad \osix\ emission feature 

171: at the redshift of the AGN.

172: Selected spectral regions showing redshifted absorption

173: features are presented in \fig{absorption}.

174: 

175: \section{SPECTRAL ANALYSIS \label{spectral}}

176: 

177: Because the four channels are essentially independent spectrographs,

178: they have different line-spread functions, and their data cannot be

179: safely combined into a single spectrum.  Instead, we use only the

180: spectrum from the channel with the highest sensitivity at the

181: wavelength of interest.  We identify statistically-significant

182: absorption features with a simple routine that bins each spectrum to

183: the instrument resolution and flags regions whose flux lies more than

184: three standard deviations below the local median.  Most of these

185: features are due to the interstellar medium (ISM) of our Galaxy, but a

186: handful share the redshift of the AGN:

187: Ly$\gamma$ $\lambda 973$, \ion{C}{3} $\lambda 977$, Ly$\beta$ $\lambda

188: 1026$, \ion{O}{6} $\lambda \lambda 1032, 1038$, and \ion{C}{2} $\lambda

189: 1036$ (\fig{absorption}).

190: 

191: Absorption-line profiles are modeled with the interstellar

192: line-fitting package written by M. Hurwitz and V. Saba.  Wavelengths,

193: oscillator strengths, and other atomic data are taken from

194: \citet{Morton:91}.  Given a column density and Doppler broadening

195: parameter, the program computes a Voigt profile for each absorption

196: feature and produces a high-resolution (0.001 \AA) spectrum of $\tau$

197: versus wavelength.  The model spectra are convolved with a Gaussian of

198: FHWM = 0.08 \AA, roughly the spectral resolution of our data,

199: and rebinned to 0.01 \AA.

200: 

201: \footnotetext[3]{The Image Reduction and Analysis Facility (IRAF) is

202: distributed by the National Optical Astronomy Observatories, which is

203: operated by the Association of Universities for Research in Astronomy,

204: Inc., (AURA) under cooperative agreement with the National Science

205: Foundation.}

206: 

207: Model spectra are fit to the data using the nonlinear curve-fitting

208: program \specfit\ \citep{Kriss:94}, which runs in the

209: IRAF\footnotemark\ environment.  \specfit\ performs a $\chi^2$

210: minimization of the model parameters.  Error bars for a particular

211: parameter are derived by fixing that parameter at the best-fit value,

212: then raising it, allowing the other model parameters to vary freely,

213: until $\chi^2$ is increased by 1, which corresponds to a 1 $\sigma$

214: deviation for a single interesting parameter \citep{Avni:76}.

215: 

216: We begin with the segment of the LiF~1A spectrum shown in the top panel

217: of \fig{absorption}, which includes the redshifted Ly$\gamma$ $\lambda

218: 973$ and \ion{C}{3} $\lambda 977$ lines and Galactic \osix\ $\lambda

219: \lambda 1032, 1038$ absorption at two velocities.  Comparison with data

220: taken during orbital night (not shown) indicates that Ly$\gamma$ is

221: well separated from the geocoronal \ion{O}{1} $\lambda 1028$ emission

222: line.  The redder \osix\ $\lambda 1038$ line is contaminated by $J$=1

223: \htwo\ absorption.  Assuming a linear continuum over the wavelength

224: region 1028.4--1038.6 \AA, we model all nine species/velocity

225: components labeled in \fig{absorption}.  We constrain the depth of the

226: \htwo\ line by fitting two other $J$=1 \htwo\ lines in the 1049--1052

227: \AA\ range simultaneously.  Derived parameters for the Galactic

228: \osix\ lines are presented in Table \ref{osix_lines} and those for the

229: redshifted features in Table \ref{red_lines}.

230: 

231: To model the redshifted Ly$\beta$ $\lambda 1026$ feature, we use data

232: from the SiC~1A channel (middle panel of \fig{absorption}), as the more

233: sensitive LiF channels do not include this wavelength range.  The

234: Ly$\beta$ line falls between a pair of geocoronal emission features due

235: to \ion{N}{2}* $\lambda \lambda 1084.6, 1087.5$.  The pair of

236: \ion{N}{2} $\lambda 1084$ absorption features is interstellar. A

237: comparison with the night-only spectrum (not shown) indicates that

238: \ion{N}{2}* $\lambda 1084.6$ does not contribute significantly to

239: this spectrum and that the narrow peak on the blue shoulder of the

240: \ion{N}{2}* $\lambda 1087.5$ line is probably not geocoronal, but

241: intrinsic to the target spectrum.  Diffuse emission filling the LWRS

242: aperture yields a line profile that is well approximated by a top-hat

243: function with a width of $\sim 106$ \kms.  The \ion{N}{2}* $\lambda 1087.5$ 

244: line is even broader, which also suggests that a second emission component is

245: present.  We model the region between 1081 and 1088 \AA\ with a linear

246: continuum, a pair of interstellar \ion{N}{2} features, and a redshifted

247: Ly$\beta$ line.  We model the emission feature with a 106

248: \kms\ top hat and a narrow Gaussian.  Derived parameters for the

249: redshifted Ly$\beta$ line are presented in Table \ref{red_lines}.

250: 

251: Finally, we turn to the redshifted \ion{O}{6} $\lambda \lambda 1032,

252: 1038$ and \ion{C}{2} $\lambda 1036$ features in the LiF~2A spectrum

253: (bottom panel of \fig{absorption}).  These features fall on the broad

254: peak of the redshifted \osix\ emission feature, which is well fit by a linear

255: continuum.  The apparent absorption features at 1095.2 and 1098.5 \AA\ 

256: are detector artifacts.  The pair of \ion{Fe}{2} $\lambda

257: 1097$ lines represents absorption from two velocity components in the

258: ISM of our Galaxy.  This velocity structure is repeated in \ion{Fe}{2}

259: lines at longer wavelengths.  As with the Galactic \osix\ lines, we fit

260: both components of the \ion{O}{6} doublet simultaneously and quote a

261: single set of derived parameters in Table \ref{red_lines}.

262: 

263: If the absorbing cloud(s) cover only part of the emitting region, then our

264: derived column densities are lower limits.  Alternatively, if the absorption features

265: that we have fit with a single velocity component are due to multiple unresolved

266: clouds, then our derived column densities are upper limits.  To investigate

267: the importance of these effects, we calculate the covering fraction of the \osix 

268: -absorbing gas as a function of velocity $C_f (v)$ by comparing the depths

269: of the two components with their expected 2:1 ratio.  Following the recipe given by

270: \citet{Hamann:97b}, we find that $C_f$ for the \osix -absorbing gas ranges from

271: 0.5 to 1.0 across the line.  Though the error bars are large, this range of values

272: suggests that multiple velocity components are partially resolved in our data.

273: A second velocity component fit to the \osix\ doublet is, however, significant

274: at only the 2 $\sigma$ level.  We therefore assume a single velocity component

275: with a covering fraction $0.5 \leq C_f \leq 1.0$ in the analysis that follows.

276: 

277: Misalignments among the four \fuse\ channels can lead to offsets in the

278: wavelength scales of their spectra.  To correct for such offsets, we

279: measure the positions of ISM features in each spectrum and adjust their

280: velocities to a common scale.  Our standard is the LiF~1A spectrum, as

281: data from this channel are used to guide the spacecraft.  Specifically,

282: we measure the \ion{Ar}{1} $\lambda \lambda 1048, 1067$ lines in the

283: LiF~1A spectrum, \ion{Ar}{1} $\lambda 1048$ in SiC~1A, and \ion{Fe}{2}

284: $\lambda 1097$ in LiF~2A.  Absolute velocities remain uncertain,

285: however, because an offset in the position of the target relative to

286: the center of the LWRS aperture can lead to a zero-point uncertainty in

287: the \fuse\ wavelength scale of up to $\pm 0.15$ \AA.  This uncertainty

288: is not included in our quoted error bars.

289:                                                                                                     

290: Though a detailed analysis of intrinsic emission in the spectrum of

291: \source\ is beyond the scope of this paper, we can provide a brief

292: description of the salient features.  

293: At long wavelengths, the continuum follows a power law

294: of the form $f \propto \lambda^{-\alpha}$, where $\alpha = 5.3 \pm 0.2$.

295: There is a spectral break at the wavelength of Galactic

296: Lyman $\beta$, and the spectrum continues flat to the Lyman limit.

297: The \osix\ feature that dominates \fig{full_spectrum} is well fit by a

298: pair of broad \osix\ emission lines (FWHM = $6280 \pm 160$ \kms) and a

299: pair of narrow \osix\ emission lines (FWHM = $1060 \pm 150$ \kms),

300: where all lines are fit with Gaussian profiles.  Our target is similar

301: to NGC~3783 in that broad emission from Ly $\beta$ and higher-order

302: Lyman lines is negligible \citep{Gabel:03}.

303: 

304: \section{DISCUSSION}

305: 

306: Derived parameters for the Galactic \osix\ lines are presented in Table

307: \ref{osix_lines}, and those for the redshifted features in Table

308: \ref{red_lines}.  

309: %We would expect the redshifted Lyman $\beta$ and

310: %$\gamma$ lines to have the same Doppler parameter, but they do not.  The

311: %discrepancy may reflect differences in the line-spread functions of the

312: %two channels that are not included in our models.  Nevertheless, the

313: The Lyman $\beta$ and

314: $\gamma$ lines  yield consistent values for the redshifted \hone\ column

315: density and Doppler parameter $b$.  We adopt a value of

316: $N($\hone$) = 10^{15}$ cm$^{-2}$ in our analysis.

317: 

318: \subsection{Photoionization modeling \label{phot_model}}

319: 

320: The redshifted \ion{H}{1}, \ion{C}{2}, \ion{C}{3}, and \ion{O}{6} lines

321: indicate the presence of ionized gas in the neighborhood

322: of the AGN (\eg, \citealt{Crenshaw:99}) moving away from the central

323: source with a speed of $\sim$ 310 km s$^{-1}$ (Table \ref{red_lines}).

324: Photoionization models can be used to constrain the parameters of the

325: absorbing gas. We use Cloudy v.94 \citep{Ferland:96} to calculate

326: the fractional abundance $f_{ion}$ of an element in a given ionization

327: state, assuming the illuminating continuum described by the Cloudy ``table agn''

328: model, an absorbing gas with solar abundances and total hydrogen density

329: %$n($H$) \equiv n($\ion{H}{1} + \ion{H}{2}$)=10^5$ cm$^{-3}$, and a grid of values for the

330: %total hydrogen column density $N_H \equiv N($\ion{H}{1} + \ion{H}{2}$)$ 

331: $n($H$)=10^5$ cm$^{-3}$  and a grid of values for the

332: total hydrogen column density $N_H$\footnote{The hydrogen density $n($H)

333: and column density $N_H$

334: include atomic and molecular hydrogen in all ionization stages, as defined

335: in the Cloudy v.94 manual \citep{Ferland:96}. In the present model, hydrogen is 

336: mostly in atomic form.}

337: and ionization parameter $U$.

338: The measured SED of \source\/ (\fig{sed}) is consistent with

339: the ``table agn'' model employed in the calculation; other studies (\eg, \citealt{Romano:02})

340: indicate that the use of other SEDs for the illuminating spectrum results in only

341: minor changes to the derived values of $N_H$ and $U$.

342: We compare the column densities $N_{ion}$ presented in

343: Table \ref{red_lines} with the predictions of the

344: photoionization model to constrain $U$ and $N_H$ in the absorbing gas.

345: 

346: In Section 3, the covering fraction of the absorbing gas in \source\ was found

347: to be $0.5 \leq C_f \leq 1.0$, consistent with

348: most other intrinsic UV absorbers in AGN \citep{Crenshaw:99}.

349: Following \citet{Arav:01} and \citet{Romano:02}, we plot

350: curves of constant $N_{ion}$ in the log($U$)--log($N_H$) plane

351: in Fig. \ref{photo}.  If the absorbing gas covers only part of the emitting region, then

352: $N_{ion}$ is a lower limit and the allowed region of parameter space lies {\it above} 

353: all the curves in Fig. \ref{photo} (see \citealt{Arav:01} for details). The total

354: hydrogen column density of warm absorbers in Seyfert galaxies is usually

355: lower than $ 10^{21}$ cm$^{-2}$ \citep{Kriss:00,Kriss:03,Romano:02,Blustin:03}.

356: Given the constraints of Fig. \ref{photo}, we tentatively place the 

357: UV absorber at $\log (U) = -2$, $\log (N_H) = 20$, keeping in mind

358: that substantially higher column densities are not excluded by our data.

359: Observations of other UV, optical, or X-ray lines are required

360: to further constrain the state of the absorber in this AGN.

361: 

362: If the

363: absorption is due to multiple velocity components, the assumption of a single,

364: uniform cloud breaks down, and this model is not applicable.

365: 

366: %If $N_{ion}$ is an upper limi.t,

367: %(as may be the case for \ion{C}{2}), then the allowed region lies {\it below}

368: %the curve.  

369: 

370: %If we assume that the covering fraction for all species is unity and 

371: %that the \ctwo\ detection is secure, then the curves in Fig. \ref{photo}

372: %most nearly intersect at $\log (U) = -2$, $\log (N_H) = 19$.  If we

373: %assume that the \hone - and \cthree -absorbing gas covers only a fraction of the

374: %emitting region and that $N$(\ctwo) is an upper limit, then the allowed region

375: %lies between the \ctwo\ and \cthree\ curves.  It is crossed by the \osix\ line

376: %(for which we have argued a covering fraction near unity)

377: %at roughly the same point, $\log (U) = -2$, $\log (N_H) = 19$.  Finally, if the

378: %absorption is due to multiple velocity components, the assumption of a single,

379: %uniform cloud breaks down, and this model is not applicable.

380: %We therefore adopt $\log (U) = -2$, $\log (N_H) = 19$ as the parameters

381: %of the absorbing gas and parametrize

382: %the resulting physical quantities as functions of $N_{{\rm H},19}$, $U_{-2}$

383: %and $n(H)_{5}$.

384: 

385: \subsection{Physical condition of the warm absorber}

386: The ionization parameter is defined as $U=Q(H)/4 \pi r_0^2 n(H) c$, where

387: $Q(H)$ is the number of ionizing photons ($E \geq$ 1 Ry),

388: $n(H)$ is the total hydrogen number density, $c$ is the speed

389: of light, and $r_0$ is the distance of the illuminated face of the cloud from

390: the central source.

391: We use the measured 2-10 keV luminosity (3$\times 10^{44}$ erg s$^{-1}$)

392: in conjunction with the assumed ``table agn'' SED to derive a flux of

393: $Q(H)=10^{56}$ s$^{-1}$ ionizing photons for \source. 

394: 

395: %Precise constraints on $U$, $N_H$  and the density $n(H)$ are also required to derive the distance,

396: %size and mass of the absorbing gas. 

397: %Our data provides no constraints on the density of the absorber; variability

398: %studies or other information such as imaging data are necessary to measure  $n(H)$

399: %(e.g., \citealt{Hamann:97a,Crenshaw:02}).

400: %Our constraints on $U$ and $N_H$ are described in  section \ref{phot_model};

401: %we employ the  "best-guess" values of $\log (U) = -2$, $\log (N_H) = 20$

402: %to provide a parametrization of the physical parameters of the absorbing cloud, also

403: %assumed to have C=1 covering fraction.

404: %The absorber is located at a distance  

405: %$r_0=5 \times 10^{20} \; U_{-2}^{-1/2} \; n(H)_{5}^{-1/2}$ cm from the central source,

406: %where $ n(H)_{5}$ is the density in units of $10^5$ cm$^{-3}$.

407: %The size of the absorbing cloud is 

408: %$L=N_H/n(H)=3.1 \times 10^{19} \; N_{{\rm H},20} \; n(H)_{5}^{-1}$ cm.

409: %Since $L << r_0$ for a wide range of hydrogen densities $n(H)$,

410: %the volume of the warm gas can be approximated as $V \sim 4 \pi r_0^2 L$ and the 

411: % mass of the warm absorber is $M \sim 2.1 \times 10^7  \; U_{-2}^{-1} \;

412: %n(H)_{5}^{-1} \; N_{{\rm H},20} \; M_{\odot}$.

413: 

414: Knowing $U$, $N_H$, and the density $n(H)$, one can derive the distance,

415: size, and mass of the absorbing cloud. 

416: Our data provide no constraints on the density of the absorber --- variability

417: studies or other information such as imaging data are necessary to measure  $n(H)$

418: (e.g., \citealt{Hamann:97a,Crenshaw:02}) --- so we will continue to use the

419: value $n($H$)=10^5$ cm$^{-3}$ assumed by the ``table agn'' model.

420: Adopting the ``best-guess'' values of $\log (U) = -2$, $\log (N_H) = 20$

421: derived above,

422: % and assuming (for the moment) a covering fraction of unity, 

423: we find that the illuminated face of the absorber lies  

424: $r_0 = 160 \; U_{-2}^{-1/2} \; n(H)_{5}^{-1/2}$ pc from the central source.

425: The size of the absorbing cloud is  simply given by

426: $L=N_H/n(H)=10 \; N_{{\rm H},20} \; n(H)_{5}^{-1}$ pc.

427: Since $L << r_0$ for a wide range of hydrogen densities $n(H)$,

428: the volume of the warm gas can be approximated as $V \sim 4 \pi r_0^2 L \times C_f$,

429: and the mass of the warm absorber is $M \sim 2.1 \times 10^7  \; U_{-2}^{-1} \;

430: n(H)_{5}^{-1} \; N_{{\rm H},20} C_f\; M_{\sun}$.

431: %If the covering fraction $C_f < 1.0$, these values become upper/lower limits.

432: 

433: \subsection{An intergalactic origin for the absorption lines?}

434: 

435: The line of sight to \source\ intersects several clusters of galaxies, two of which

436: have measured redshifts: A3782 at $z=0.0557$ and APMCC684 at $z=0.056$.

437: The difference in redshift between the measured absorption lines

438: ($z \sim 0.0578$) and the two clusters is so small that an alternative 

439: interpretation of the absorption lines can be entertained, viz., the source shines

440: through the foreground cluster(s), which cause the redshifted absorption lines.

441: 

442: The center of A3782 lies $\sim 26$ arcmin from \source, which corresponds to

443: approximately 1.8 Mpc (for a Hubble constant of $H_0=70$ km s$^{-1}$ Mpc$^{-1}$).

444: The center of APMCC684 is $\sim$ 17.5 arcmin distant, which corresponds to 1.2 Mpc.

445: Clusters of galaxies are known to host hot intergalactic gas ($T \geq 10^7$ K) -- too hot to

446: produce the observed absorption lines -- and oftentimes a lower temperature phase 

447: ($T \sim 10^5-10^6$ K;

448: \citealt{Lieu:96a,Lieu:96b,Bonamente:02,Bonamente:03})

449: that could contain substantial amounts of \ion{O}{6}, \ion{C}{2}, \ion{C}{3},

450: and neutral atomic hydrogen. 

451: This warm intergalactic gas has been observed in emission in several

452: clusters (i.e., the ``soft excess'' phenomenon; \citealt{Bonamente:02}; \citealt{Kaastra:03}; \citealt{Nevalainen:03}), 

453: notably in the Coma cluster,

454: where it extends some 2.6 Mpc from the cluster center \citep{Bonamente:03}.

455: Detection of \ion{O}{6} absorption associated with the Local Group of galaxies has been recently

456: reported by \citet{Nicastro:03}, and detection of an \ion{O}{6} absorption system associated

457: with another galaxy group was reported by \citet{Tripp:00}. Here we investigate 

458: the association of the detected absorption systems with the two galaxy clusters A3782 and APMCC684.

459: 

460: We use the soft excess measurements of \citet{Bonamente:03} for the Coma cluster to estimate the

461: amount of low-ionization gas in clusters of galaxies. Several other

462: clusters contain warm gas in amounts comparable (within a factor of a few) to those of the Coma cluster.

463: The warm gas is generally more diffuse than the hot gas \citep{Bonamente:02}, and it

464: could reside either inside the cluster (i.e., {\it mixed} with the hot gas)

465: or in filamentary structures outside the cluster, as is often seen in

466: hydrodynamical simulations (\eg, \citealt{Cen:99}). 

467: If the warm gas is bound to the cluster, \citet{Bonamente:03} show that it must have a

468: density of $1.5 \times 10^{-3} - 6 \times 10^{-5}$ cm$^{-3}$ throughout the cluster.

469: Assuming a cluster radius

470: of $\sim$3 Mpc, in agreement with current X-ray measurements, this density

471: implies \ion{H}{1} column densities of $\sim 10^{20}-10^{22}$ cm$^{-2}$.

472: The metal abundance of the warm gas is

473: $\sim 0.1$ solar (\citealt{Bonamente:03}); this implies

474: total carbon column densities of $N_C \sim 5 \times 10^{15}-5 \times 10^{17}$ cm$^{-2}$

475: and oxygen column densities of $N_O \sim 10^{16}- 10^{18}$ cm$^{-2}$.

476: Figure \ref{mazzotta} shows the

477: ionization fractions of the ions of concern as a function of temperature, assuming ionization

478: equilibrium \citep{Mazzotta:98}. Column densities of 

479: \ion{H}{1}, \ion{C}{2}, \ion{C}{3}, and \ion{O}{6} predicted by this scenario are

480: therefore consistent, for a wide range of temperatures,  with those detected toward 

481: \source~(Table \ref{red_lines}). 

482: The warm gas may alternatively reside in filamentary structures with densities

483: of $\sim 10^{-4} - 10^{-5}$ cm$^{-3}$ \citep{Dave:01}.

484: In this case, \citet{Bonamente:03} show that

485: the filaments will extend for $\sim$ 10 Mpc outside the cluster, yielding similar

486: ion column densities.

487: In addition, the Doppler parameters $b$ of a warm gas at $T \sim 10^4-10^6$ K are 

488: fully consistent with the values derived in Table 2 \citep{Spitzer:78}, 

489: and similar to those of \citet{Tripp:00}. 

490: 

491: We conclude that the current data for \source~are consistent with the redshifted

492: absorption lines of Table \ref{red_lines} originating from warm gas associated with an intervening

493: galaxy cluster.

494: 

495: \section{CONCLUSIONS}

496: We detect Galactic \ion{O}{6} absorption in the direction of \source\

497: and \hone\ Lyman $\beta$ and $\gamma$, \ion{C}{2},

498: \ion{C}{3}, and \ion{O}{6} absorption at a redshift of $z \sim 0.0578$.

499: The redshifted absorption lines are consistent with a circumnuclear

500: absorber outflowing with a relative velocity of $-310$ km s$^{-1}$.

501: We derive constraints on its physical parameters through photoionization modeling.

502: %The total hydrogen column density is $N_H \sim 10^{19}$ cm$^{-2}$ and the ionization parameter is $U \sim 10^{-2}$.  

503: Alternatively, we suggest that the redshifted absorption may originate in or around a 

504: cluster of galaxies  located along the line of sight.

505: 

506: \acknowledgments

507: 

508: This research has made use of NASA's Astrophysics Data System

509: Bibliographic Services and the NASA/IPAC Extragalactic Database (NED),

510: which is operated by the Jet Propulsion Laboratory, California

511: Institute of Technology, under contract with the National Aeronautics

512: and Space Administration.

513: The authors thank Pat Romano for insightful discussions and for

514: sharing the Cloudy commands for the photoionization calculations. 

515: This work is supported by a \fuse\ Cycle 4 Guest Investigator grant from NASA.

516: 

517: \nocite{Mihalas:Binney:81}

518: 

519: %%%%%%%%%%%%% REFERENCES %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

520: 

521: \clearpage 

522: 

523: \bibliographystyle{apj}

524: \bibliography{apjmnemonic,ms}	

525: 

526: %%%%%%%%%%%%%%%% FIGURES %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

527: 

528: \clearpage

529: 

530: \begin{figure}

531: %\epsscale{0.75}

532: \plotone{f1.eps}

533: \caption{Spectral energy distribution of  \source.

534: To the right of the \fuse\/ data point are the \euve\/, \rosat\/ and \heao\/ measurements, respectively.  The lower-frequency measurements were derived from the available

535: literature through the NASA/IPAC Extragalactic Database

536: (http://nedwww.ipac.caltech.edu).

537: \label{sed}}

538: \end{figure}

539: 

540: \begin{figure}

541: %\epsscale{0.75}

542: \plotone{f2.eps}

543: \caption{\fuse\ spectrum of the Seyfert 1 galaxy \source.

544: The data are smoothed by 32 pixels, or about four resolution elements, and

545: geocoronal emission lines (defined as resolved features extending more than three

546: standard deviations above the local median) are excluded.

547: Wavelengths around 1085 \AA\ are not covered by the LiF channels.

548: %The spectrum exhibits a power-law continuum, broad redshifted \osix\ emission,

549: %and a spectral break at the wavelength of Galactic Lyman $\beta$.  The spectrum

550: %remains flat to the Lyman limit.  Most absorption lines are due to the ISM of our Galaxy. 

551: The spectrum exhibits a power-law continuum with a spectral break at the

552: wavelength of Galactic Lyman $\beta$.  At shorter wavelengths, the

553: spectrum is flat to the Lyman limit.  The broad emission feature is

554: redshifted \osix\ emission.  Most absorption lines are due to the ISM of

555: our Galaxy.

556: \label{full_spectrum}}

557: \end{figure}

558: 

559: \begin{figure}

560: \epsscale{0.75}

561: \plotone{f3.eps}

562: \caption{

563: Redshifted absorption features in the \fuse\ spectrum of 2MASX~J21362313-6224008.

564: Interstellar absorption and geocoronal emission features are marked with dotted lines,

565: redshifted absorption features with solid lines.

566: {\it Top:}\/ The Ly$\gamma$ $\lambda 973$ and \ion{C}{3} $\lambda 977$

567: features are blueshifted by 310 \kms\ relative to the AGN.  The Galactic \ion{O}{6}

568: $\lambda \lambda 1032, 1038$, \ion{C}{2} $\lambda 1036$, and \ion{C}{2}*

569: $\lambda 1037$ features show two velocity components.

570: {\it Middle:}\/ Redshifted Ly$\beta$ absorption falls

571: between the geocoronal \ion{N}{2} $\lambda \lambda 1094.6, 1085.7$ emission

572: lines.  {\it Bottom:}\/ Redshifted \ion{O}{6} $\lambda \lambda 1032,

573: 1038$ and \ion{C}{2} $\lambda 1036$ with Galactic \ion{Fe}{2} $\lambda

574: 1097$ (two velocity components).  The apparent absorption features at

575: 1095.2 and 1098.5 \AA\ are detector artifacts.

576: \label{absorption}}

577: \end{figure}

578: 

579: \begin{figure}

580: %\epsscale{0.75}

581: \rotatebox{90}{

582: \plotone{f4.eps}}

583: \caption{Photoionization curves at constant ionic column density. $U$ is the ionization parameter and $N_H$ the total hydrogen column density. The column densities are from Table \ref{red_lines}; the \ion{H}{1} curve represents a neutral hydrogen 

584: column density of $10^{15}$ cm$^{-2}$. The cross-hatched area is the region where all constraints are met.

585: \label{photo}}

586: \end{figure}

587: 

588: \begin{figure}

589: %\epsscale{0.75}

590: \plotone{f5.eps}

591: \caption{Ionization fractions of  \ion{H}{1}, \ion{C}{2}, \ion{C}{3}, and \ion{O}{6} for a gas

592: in collisional equilibrium (from \citealt{Mazzotta:98}).

593: \label{mazzotta}}

594: \end{figure}

595: %%%%%%%%%%%%%%%%%%%%%%%%%%%%% TABLES %%%%%%%%%%%%%%%%%%%%%%%%%

596: 

597: \clearpage 

598: 

599: \begin{deluxetable}{ccc}

600: \tablewidth{0pt}

601: \tablecaption{Galactic \ion{O}{6} Absorption Features\label{osix_lines}}

602: \tablehead{

603: \colhead{Velocity} & \colhead{Doppler Parameter} & \colhead{Column Density}\\

604: \colhead{(\kms)} & \colhead{(\kms)} & \colhead{$\log N$ (cm$^{-2}$)}

605: }

606: \startdata

607: $\phn27 \pm 3$ & $53 \pm \phn6$ & $14.44 \pm 0.03$ \\

608: $175 \pm 6$ & $26 \pm 15$ & $13.77 \pm 0.09$

609: \enddata

610: 

611: \tablecomments{Velocities are quoted relative to the local standard of

612: rest (LSR; Mihalas \& Binney 1981).}

613: 

614: \end{deluxetable}

615: 

616: %\begin{deluxetable}{lcccc}

617: %\tablewidth{0pt}

618: %\tablecaption{Redshifted Absorption Features\label{red_lines}}

619: %\tablehead{

620: %& & & \colhead{Doppler}\\

621: %& & \colhead{Velocity}& \colhead{Parameter} & \colhead{Column Density}\\

622: %\colhead{Feature} & \colhead{Redshift} & \colhead{(\kms)} &

623: %\colhead{(\kms)} &

624: %\colhead{$\log N$ (cm$^{-2}$)}

625: %}

626: %\startdata

627: %Lyman $\gamma$ $\lambda 972.5$ & 0.057760 $\pm$ 7E-6 & $-312 \pm 2$ & $32 \pm 2$ & $14.97 \pm 0.06$ \\

628: %\ion{C}{3} $\lambda 977.0$     & 0.057783 $\pm$ 5E-6 & $-305 \pm 1$ & $24 \pm 3$ & $13.95 \pm 0.06$ \\

629: %Lyman $\beta$ $\lambda 1025.7$ & 0.05777 $\phn\pm$ 1E-5 & $-309 \pm 3$ & $19 \pm 3$ & $15.07 \pm 0.28$ \\

630: %\ion{O}{6} $\lambda \lambda 1031.9,1037.6$ & 0.057783 $\pm$ 9E-6 & $-305 \pm 3$ &  $20 \pm 2$ & $13.95 \pm 0.05$ \\

631: %\ion{C}{2} $\lambda 1036.4$ & 0.05776 $\phn\pm$ 9E-5 & $-312 \pm 3$  & $6 \pm 10$ & $13.45 \pm 0.16$ \\

632: %\enddata

633: 

634: %\tablecomments{Velocities are quoted relative to the systemic velocity of \source\ ($z = 0.0588, cz = 17,630$ \kms).}

635: 

636: %\end{deluxetable}

637: 

638: \begin{deluxetable}{llccc}

639: \tablewidth{0pt}

640: \tablecaption{Redshifted Absorption Features\label{red_lines}}

641: \tablehead{

642: & & & \colhead{Doppler}\\

643: & & \colhead{Velocity}& \colhead{Parameter} & \colhead{Column Density}\\

644: \colhead{Feature} & \colhead{Redshift} & \colhead{(\kms)} &

645: \colhead{(\kms)} &

646: \colhead{$\log N$ (cm$^{-2}$)}

647: }

648: \startdata

649: Lyman $\gamma$ $\lambda 972.5$ & 0.057760(7) & $-312 \pm 2$ & $27 \pm 5$ & $14.97 \pm 0.05$ \\

650: \ion{C}{3} $\lambda 977.0$     & 0.057783(5) & $-305 \pm 1$ & $24 \pm 3$ & $13.95 \pm 0.06$ \\

651: Lyman $\beta$ $\lambda 1025.7$ & 0.057770(15) & $-309 \pm 3$ & $26 \pm 9$ & $14.89 \pm 0.14$ \\

652: \ion{O}{6} $\lambda \lambda 1031.9,1037.6$ & 0.057783(9) & $-305 \pm 3$ &  $20 \pm 2$ & $13.95 \pm

653: 0.05$ \\

654: \ion{C}{2} $\lambda 1036.4$ & 0.057760(87) & $-312 \pm 3$  & $6 \pm 10$ & $13.45 \pm 0.16$ \\

655: \enddata

656:                                                                                                            

657: \tablecomments{In column 2, the digits in parentheses represent the one-standard-deviation uncertainty in the final digits of the redshift.  In column 3, velocities are quoted relative to the systemic velocity of \source\ ($z = 0.0588, cz = 17,630$ \kms).}

658:                                                                                                            

659: \end{deluxetable}

660: 

661: \end{document}

662: