1:
2: \documentclass[apjl]{emulateapj}
3: \usepackage{apjfonts}
4: \submitted{}
5:
6: \def\hi{\ifmmode {\rm H}\,{\sc i}~ \else H\,{\sc i}~\fi}
7: \def\kms{\rm\,km\,s^{-1}}
8: \def\hubunits{\rm\,km\,s^{-1}\,Mpc^{-1}}
9: \def\K{\,{\rm K}}
10: \def\cm{{\rm cm}}
11: \def\chandra {{\it Chandra}}
12: \def\xmm {{\it XMM}}
13: \def\xmmnewton {{\it XMM--Newton}}
14: \def\fuse {{\it FUSE}}
15: \def\cv {\ion{C}{5}}
16: \def\cvi {\ion{C}{6}}
17: \def\nvi {\ion{N}{6}}
18: \def\neix {\ion{Ne}{9}}
19: \def\nex {\ion{Ne}{10}}
20: \def\nvii {\ion{N}{7}}
21: \def\nvi {\ion{N}{6}}
22: \def\ovi {\ion{O}{6}}
23: \def\ovii {\ion{O}{7}}
24: \def\oviii {\ion{O}{8}}
25: \def\ovihv {\ion{O}{6}$_{\rm HV}$}
26: \def\novi {N_{\rm OVI}}
27: \def\novii {N_{\rm OVII}}
28: \def\noviii {N_{\rm OVIII}}
29:
30: \def\kms {~km~s$^{-1}$}
31:
32:
33: %% You can insert a short comment on the title page using the command below.
34:
35: \slugcomment{Submitted to ApJ Letters}
36:
37:
38: \shorttitle{X--ray absorption in AGN host galaxies}
39: \shortauthors{Williams et al.}
40:
41:
42:
43: \begin{document}
44:
45:
46: \title{On the nature of the $z=0$ X-ray absorbers: II. The contrast between
47: local and AGN host galaxy absorption }
48: %\title{Clues to the nature of $z=0$ X--ray absorbers from AGN host galaxies}
49: %% Use \author, \affil, and the \and command to format
50: %% author and affiliation information.
51:
52: \author{Rik J. Williams\altaffilmark{1},
53: Smita Mathur\altaffilmark{2},
54: Fabrizio Nicastro\altaffilmark{3,4}}
55: \altaffiltext{1}{Leiden Observatory, Leiden University, PO Box 9513,
56: 2300 RA Leiden, The Netherlands}
57: \altaffiltext{2}{Department of Astronomy, The Ohio State University,
58: Columbus, OH, 43210, USA}
59: \altaffiltext{3}{Osservatorio Astronomico di Roma, Istituto Nazionale di
60: AstroFisica, Italy}
61: \altaffiltext{4}{Harvard--Smithsonian Center for Astrophysics, Cambridge,
62: MA, 01238, USA}
63: \email{williams@strw.leidenuniv.nl}
64:
65:
66:
67: \begin{abstract}
68: We search for highly--ionized gas near three AGN host galaxies
69: using the \chandra\ low--energy transmission grating spectrograph.
70: Strong absorption lines from such gas are seen at $z=0$, most likely
71: from one or more of the following components: (1) a Galactic corona,
72: (2) the Local Group medium, and (3) an extended warm--hot intergalactic
73: medium (WHIM) filament passing through our local overdensity.
74: Since AGNs reside within host galaxies that
75: are also expected to sit within cosmically overdense regions, similar
76: absorption resulting from these three
77: components should appear at the AGN redshifts as well.
78: However, no such absorption is seen. The lack of strong absorption
79: lines is likely a result of the gas in these host galaxies and surrounding
80: galaxy clusters being much hotter, and hence more highly ionized, than
81: the gas in the Local Group$+$Galaxy system. We conclude that
82: WHIM filaments produce no measurable absorption lines at the AGN
83: redshifts, and therefore contribute at most a small fraction of the
84: observed $z=0$ warm--hot gas.
85: \end{abstract}
86:
87: \keywords{ Galaxy: general --- intergalactic medium --- Local Group ---
88: X--rays: galaxies }
89:
90:
91:
92: \section{Introduction}
93: The study of local warm--hot gas has been revolutionized by the
94: advent of sensitive space--based observatories such as \chandra,
95: \xmmnewton, and the {\it Far Ultraviolet Spectroscopic Observer} (\fuse).
96: By taking spectra of bright background quasars, absorption lines from
97: highly--ionized heavy elements at $z\sim 0$ can be seen in nearly all
98: directions in both the X--ray
99: \citep{nicastro02,williams05,williams06b,williams07} and
100: far--UV \citep{wakker03}. These lines are thought
101: to trace a variety of mass components; while strong, broad \ovi\ absorption
102: is associated with the Galactic thick disk \citep{savage03},
103: high--velocity \ovi\ clouds more likely reside in the extended Galactic
104: halo, with a subpopulation of these showing evidence of extragalactic
105: (Local Group) kinematics \citep{sembachetal03,nicastro03}.
106: More enigmatic still are the X--ray absorbers, which exhibit kinematics and
107: column densities inconsistent with both the low-- and high--velocity
108: \ovi\ absorbers \citep[e.g.,][]{williams05}. Whether these absorbers
109: arise primarily from a Galactic or an extragalactic medium is still a
110: matter of debate.
111:
112:
113: In the ``extragalactic'' picture, the local X--ray absorption arises
114: primarily from the intragroup medium filling the Local Group (analogous
115: to the extended X--ray emitting gas seen in massive galaxy clusters), or
116: from the larger--scale IGM filament(s) connecting the overdensities traced by the Local
117: Group and nearby structures such as the Virgo cluster
118: \citep{kravtsov02}. Such structure, containing most of the baryonic
119: mass at low redshifts, is now a generic prediction of
120: cosmological simulations and has come to be known as the warm--hot
121: intergalactic medium \citep[WHIM;][]{cen99,dave01}. On the other hand, much
122: of the local X--ray absorption may also be within a few tens of kpc, resulting
123: from a hot Galactic corona or fountain driven by supernova winds
124: \citep{sembach03,collins05,wang05}.
125:
126: In a companion paper to this one \citep{mathur07}, we propose that
127: the X--ray absorption lines seen at $z=0$ are likely to contain contributions
128: from both Galactic and intra--Local Group media. Additional contributions
129: could come from the local large--scale intergalactic
130: filament, and perhaps a supernova remnant lying near the occasional line of
131: sight \citep[e.g., 3C~273;][]{savage93}. All of these components would
132: apparently be blended into a single line within
133: \chandra's $\sim 700$\kms\ resolution. While it appears that
134: poor groups of galaxies similar to our own can produce the observed
135: \ovii\ column densities \citep{mathur07}, the relative
136: contributions from Galactic and extragalactic phenomena in our own Local Group
137: are as yet undetermined.
138:
139: The intervening WHIM lines found by \citet{nicastro05} are quite weak
140: relative to the local absorption; indeed, intervening absorption systems
141: of comparable column density have yet to be discovered.
142: Thus, the strength of the $z=0$ absorption is doubtless related to our
143: peering out from a ``special'' vantage point centered on a dense region ---
144: in the disk
145: of a spiral galaxy within a poor galaxy group. However, inherent to
146: every quasar spectrum is one other such special location: the AGN host
147: galaxy. As AGNs typically occur at the centers of massive galaxies,
148: some within groups and clusters, large column densities of warm--hot gas
149: should also be present at the quasar redshifts as at
150: $z=0$. These components could include intra-- and intergalactic gas
151: surrounding the host galaxy,
152: plus gas directly related to the accretion process (such as a
153: warm absorber or outflow). In other words, if the Galaxy is a typical
154: (non--AGN) system, absorption column densities
155: near AGN host galaxies should be {\it at least} as strong as those
156: observed locally, and significant discrepancies (vis--\`a--vis differences
157: between the host galaxies' and the Milky Way's systems) can provide clues
158: to the nature of the local absorption.
159: In this {\it Letter} we search for absorption lines at the redshifts
160: of three AGNs observed with the \chandra\ grating spectrograph, and contrast
161: these with the absorption lines seen at $z=0$ in the same spectra.
162:
163: \begin{deluxetable*}{lccccccc}
164: \tabletypesize{\footnotesize}
165: \tablecolumns{8}
166: \tablewidth{450pt}
167: \tablecaption{Absorption lines at $z=0$ and $z_{\rm host}$ \label{tab_lines}}
168: \tablehead{
169: \colhead{Object} &
170: \colhead{$z_{\rm abs}$} &
171: \colhead{$W_\lambda$\tablenotemark{a}} &
172: \colhead{$W_\lambda$\tablenotemark{a}} &
173: \colhead{$W_\lambda$\tablenotemark{a}} &
174: \colhead{$W_\lambda$\tablenotemark{a}} &
175: \colhead{$W_\lambda$\tablenotemark{a}} &
176: \colhead{$W_\lambda$\tablenotemark{a}}
177: \\
178: \colhead{} &
179: \colhead{} &
180: \colhead{\nex} &
181: \colhead{\neix} &
182: \colhead{\oviii} &
183: \colhead{\ovii} &
184: \colhead{\cvi} &
185: \colhead{\cv}
186: \\
187: \colhead{} &
188: \colhead{} &
189: \colhead{12.13\,\AA} &
190: \colhead{13.45\,\AA} &
191: \colhead{18.97\,\AA} &
192: \colhead{21.60\,\AA} &
193: \colhead{33.74\,\AA} &
194: \colhead{40.27\,\AA}
195: }
196:
197: \startdata
198: Local \\
199: \hline
200: % columns are: NeX, NeIX, OVIII, OVII, CVI, CV
201: Mrk 421 &0 &\nodata &$1.6\pm 0.5$ &$1.7\pm 0.6$ &$11.4\pm 0.8$ &$7.4\pm 1.5$ &$15.7\pm 3.3$\\
202: PKS 2155--304 &0 &\nodata &$3.8\pm 1.2$ &$6.2\pm 1.4$ &$12.2^{+3.7}_{-2.8}$ &$5.5\pm 2.6$ &$9.4^{+7.1}_{-5.2}$\\
203: 3C 273 &0 &\nodata &$6.9\pm 2.0$ &$7.4\pm 2.9$ &$26.5\pm 3.1$ &$19.5\pm 8.2$ &$50\pm 20$\\
204: Mrk 279\tablenotemark{b} &0 &\nodata &\nodata &\nodata &$26.6\pm 6.2$ &\nodata &\nodata\\
205: \hline\hline
206: Host galaxies \\
207: \hline
208: Mrk 421 &0.030 &$<1.4$ &$<3.3$ &$<1.8$ &$<3.8$ &$<5.4$ &$<12.9$\\
209: PKS 2155--304 &0.116 &$<5.4$ &$<5.8$ &$<7.3$ &$<6.8$ &$<12.6$ &$<8.6$ \\
210: 3C 273 &0.158 &$<7.2$ &$<4.1$ &$<11.7$ &$<14.4$ &$<47.9$ &$<24.9$\\
211: \enddata
212: \tablenotetext{a}{All equivalent widths are quoted in m\AA\ with $1\sigma$
213: errors; upper limits are $2\sigma$ confidence.}
214: \tablenotetext{b}{Redshifted absorption lines are not measured for this
215: object, due to absorption from a strong AGN outflow.}
216: \end{deluxetable*}
217:
218: \section{Data Reduction and Measurements}
219: In this analysis we focus on data from the \chandra\ Low--Energy Transmission
220: Grating (LETG) coupled with its two standard readout detectors, the
221: High Resolution Camera and AXAF CCD Imaging Spectrometer spectroscopic
222: arrays (HRC--S and ACIS--S respectively). While the {\it XMM--Newton}
223: Reflection Grating Spectrometers (RGS1/RGS2) together have comparable
224: sensitivity
225: to the \chandra\ LETG, the RGS contains many narrow detector features and
226: a more restricted wavelength range \citep{williams06a}. Furthermore,
227: there is evidence for significant cross--calibration uncertainties between
228: \xmm/RGS and \chandra/LETG \citep{rasmussen07}; thus, for the sake of
229: consistency we will focus solely on the \chandra\ data for this analysis.
230:
231: Local ($z\sim 0$) absorption has been confidently detected by \chandra\ in
232: four local AGN sight lines: Mrk 421 \citep{williams05}, Mrk 279
233: \citep{williams06b}, PKS 2155--304 \citep{nicastro02,williams07},
234: and 3C 273 \citep{fang03}. The far--UV and X--ray spectra of Mrk 279 indicate
235: the presence of strong intrinsic absorption from an AGN outflow
236: \citep{fields07}; thus, only
237: the $z=0$ absorption seen in this spectrum will be considered in this analysis
238: All \chandra/LETG observations of these four sources employing the ACIS--S
239: and HRC--S readout detectors were downloaded from the \chandra\ archive
240: and fully re--reduced with CIAO
241: 4.0b1\footnote{\url{http://cxc.harvard.edu/ciao/}}. Zeroth--order source
242: positions were checked by eye and corrected if necessary, and grating
243: spectra and response files for all observations generated. The spectra
244: from positive and negative grating orders were added for each observation,
245: and all observations for the same object subsequently added. Response
246: files for the positive and negative ACIS--S orders, and from orders
247: $-6$ through $+6$ for HRC--S\footnote{This is necessary to properly account
248: for contamination from unresolved higher spectral orders; see
249: \url{http://cxc.harvard.edu/ciao/threads/hrcsletg\_orders/}}, were similarly
250: combined to produce average ACIS and HRC instrumental response functions
251: for each object.
252:
253: Spectral continua and absorption lines were modeled with the
254: fitting program
255: {\it Sherpa}\footnote{\url{http://cxc.harvard.edu/sherpa/}}.
256: A single powerlaw plus Galactic absorption model was initially fit to each
257: spectrum over the 10--50\,\AA\ range to properly determine
258: the level of higher--order contamination in HRC--S. Individual
259: lines were fitted with negative Gaussians in $\sim 2$\,\AA\ windows around each
260: wavelength of interest, retaining the continuum shape found with the
261: broadband fit but allowing the amplitude to vary. Any residual broad
262: features in the spectra (much larger than the LETG line--spread function
263: width, $\sim 50$\,m\AA) were removed by including broad Gaussian components.
264: The continua were
265: allowed to vary independently for HRC--S and ACIS--S to allow for
266: flux variations between the different observation epochs, while absorption line
267: amplitudes were determined using a joint fit to both instrumental
268: spectra. Upper limits (at the
269: $2\sigma$/95\% confidence level) were determined by allowing the
270: line wavelength to vary within $\pm 0.02$\AA\ of the expected central
271: wavelength to take into account the systematic uncertainty in the
272: LETG dispersion relation.\footnote{See \url{http://cxc.harvard.edu/cal/}}
273:
274: At the AGN redshifts, we measured six absorption lines that were
275: detected in the Mrk 421 spectrum at $z=0$, and are expected
276: to peak in abundance over a range of temperatures spanning roughly a factor
277: of ten under collisional
278: ionization: $T\sim 10^{6.5}$\,K
279: (\oviii, \nex), $10^{6}$\,K (\ovii, \cvi, \neix) and $10^{5.5}$\,K
280: (\cv). Although this latter temperature is expected to occur in galactic
281: interstellar media \citep[e.g., as observed in the thick disk of the Galaxy
282: through strong \ovi\ absorbers with FUSE;][]{savage03}, it may provide
283: a useful indicator of whether the AGN host galaxies contain similar
284: warm--hot gas column densities as the Galaxy but at lower temperatures.
285: To ensure that consistent reduction methods and calibrations are
286: used throughout, we re--measured the strengths of the same absorption
287: lines at $z=0$ using the same broadband continuum fits.
288: Table~\ref{tab_lines} shows the strengths (or upper limits)
289: for all measured local and redshifted absorption lines.
290:
291: As the table shows, at the AGN redshifts none of these absorption lines are
292: formally detected at a significance larger than $2\sigma$; indeed,
293: visual inspection of the spectra confirms the absence of any apparent
294: absorption at or near the AGN. Note that \citet{rasmussen07} find
295: a weak ($2.6$\,m\AA), $z=0.032$ \ovii\ line at $3.7\sigma$ significance
296: with \xmm;
297: however, this is not seen in the \chandra\ spectrum and may be a $z=0$
298: \ion{O}{5} line near the same wavelength. Figure~\ref{fig_m421res} shows the
299: continuum fit residuals for \oviii, \ovii, and \cv\ in the highest--quality
300: spectrum (Mrk 421). Though absorption in all three atomic species is
301: detected at $z=0$, none is seen at $z=0.03$. Furthermore, these three
302: lines can be present at temperatures ranging from $10^5-10^7$\,K; if the Mrk 421
303: host galaxy contained gas at a column density comparable to that observed at
304: $z=0$ but with a somewhat higher or lower temperature, the strength of the
305: \oviii\ or \cv\ line should be boosted respectively. However, this is
306: clearly not the case.
307:
308: A visual representation of the data in Table~\ref{tab_lines} is shown in
309: Figure~\ref{fig_limits}. The discrepancy is most evident for
310: \ovii: most of the $2\sigma$ upper limits at the AGN redshifts are
311: inconsistent with the typical local absorption strength. Even
312: for the other four lines shown in this figure, the upper limits
313: are nonetheless below or comparable to the $z=0$ measurements. It is
314: thus apparent that the gaseous environments of the AGN host galaxies
315: are in fact dramatically different from that around the Galaxy.
316:
317: \section{AGN host galaxy properties}
318: Since strong $z=0$ \ovii\ absorption ($>10$\,m\AA\ in all cases) is seen toward
319: AGNs anytime sufficient S/N is acquired, it is unlikely that the lack of
320: observed host galaxy absorption is a result of a low \ovii\ covering
321: fraction. Instead, this discrepancy must be due to fundamental differences
322: between the Milky Way and host galaxy systems themselves. A summary of
323: the three host galaxy systems follows.
324:
325: \subsection{Mrk 421 and PKS 2155--304}
326: As the two highest signal--to--noise spectra, the absence of absorption at
327: the redshifts of Mrk 421 and PKS 2155--304 is particularly striking. Not
328: only do both of these spectra exhibit strong similarities in their $z=0$
329: absorption properties (consistent column densities, Doppler parameters,
330: and temperature constraints), but the AGN environments are also remarkably
331: similar. Both are BL Lac objects contained within giant elliptical host
332: galaxies \citep{ulrich75,kikuchi85,falomo91}. Such galaxies
333: are known to contain large quantities of hot gas from X--ray emission
334: studies \citep[e.g.,][]{brighenti97}, with the most massive ellipticals
335: exhibiting gas temperatures of $T\sim 10^7$\,K. At this temperature
336: oxygen is expected to be fully ionized. The nuclei of both
337: these galaxies are powerful blazars; thus, intense X--ray flux from the
338: central source may further contribute to the ionization of gas in the central
339: regions of the galaxy, and jets or outflows from the AGN might evacuate
340: galactic gas from the line of sight. Furthermore, these host galaxies
341: reside at the centers of clusters \citep{ulrich78,falomo93}. Since the
342: intracluster media within massive galaxy clusters
343: typically exhibit similarly high temperatures ($T\ga 10^7$\,K), this cluster
344: environment is unlikely to produce strong absorption lines.
345:
346: \subsection{3C 273}
347: The host galaxy of 3C 273 was studied extensively with the HST Advanced
348: Camera for Surveys (ACS) coronagraph by \citet{martel03}. They find
349: a morphology similar to an early--type galaxy (dominated by an extended
350: stellar halo at large radii), but also evidence for spiral structure
351: and a possible recent major merger in the central region. Although the
352: existence of a spiral and/or
353: merger morphology might increase the likelihood of seeing strong absorption
354: (due to higher star formation rates than in giant ellipticals, and hence
355: more supernova remnants and ``galactic fountain'' activity), it may be
356: that the fully ionized hot gas in the outer regions dominates the
357: total column density. Furthermore, 3C 273 is not known to be a member
358: of a group or cluster, in which case we don't expect to see a contribution
359: from the surrounding intragroup medium like at $z=0$.
360:
361: In any case, the comparatively low sensitivity of this spectrum ($2\sigma$
362: upper limit of $<14.4$\,m\AA\ on redshifted \ovii) is the primary limitation
363: to detecting absorption near the host galaxy. Only two of the four $z=0$
364: systems show \ovii\ absorption stronger than the upper limit found for
365: the host galaxy, so it is perhaps not surprising that no redshifted
366: absorption is seen here.
367:
368: \section{Discussion and Conclusions}
369: We searched for absorption near AGN host galaxies in the three
370: highest--quality \chandra/LETG AGN spectra, but none could be seen within
371: the sensitivity of our observations. This is in stark contrast to $z=0$,
372: where absorption of $W_\lambda$(\ovii)$\ga 10$\,m\AA\ is seen in all
373: directions whenever enough signal is accumulated. Given this discrepancy,
374: and by comparing the characteristics of the AGN host galaxy systems
375: with the $z=0$ Galaxy$+$Local Group system, we propose three likely
376: contributing factors:
377: \begin{enumerate}
378: \item{{\it The host galaxies reside in high--mass, and thus very high--temperature systems.}
379: Mrk 421, PKS 2155--304, and possibly 3C 273 are hosted
380: by massive ellipticals; the former two are also at the centers of
381: rich galaxy groups or clusters. Although both massive ellipticals and
382: clusters contain large amounts of hot gas, at $\ga 10^7$\,K even the heavier
383: elements analyzed herein are fully ionized. On the other hand, the
384: Local Group is a relatively low--mass system and the measured gas
385: temperatures at $z=0$ ($\sim 10^6$\,K) are in a range that maximizes
386: the abundances of ions easily measured by \chandra.
387: }
388: \item{{\it The IGM filaments surrounding the Local Group and host galaxies
389: contribute negligibly to the total absorption.} The extended intergalactic
390: medium filaments surrounding the AGN host galaxies should in principle be
391: similar
392: to that around our own Local Group, and at sufficient distance to be essentially
393: unaffected by the galaxy and/or AGN. Thus, given the ubiquity of absorption
394: lines at $z=0$, any ``local filament'' contribution to these lines should
395: also be visible around most AGN as well. Since it is in fact not visible,
396: we conclude that large--scale filamentary IGM structures are not an
397: important contributor to the $z=0$ X--ray absorption. This is consistent
398: with the weakness of the absorption seen in the intervening WHIM filaments
399: reported by \citet{nicastro05}. However, it stands in conflict with the
400: marginal detection of X-ray absorption (also presumably from a
401: surrounding WHIM filament) toward the Coma cluster in a shallower
402: \xmm\ spectrum by \citet{takei07}.
403: }
404: \item{{\it Lack of star--formation activity in the host galaxies.}
405: As supernovae are known to produce large ``bubbles'' of hot gas, it stands to
406: reason that an occasional AGN sight line will pass near enough to one of these
407: supernova remnants to exhibit strong X--ray absorption. This is likely
408: the cause of the strong absorption seen locally toward 3C 273
409: \citep{savage93}. Not only is the sample size
410: here relatively small, thereby lowering the chance probability of
411: a sight line intercepting a SNR in one of these host galaxies (unless the
412: average SNR covering fraction is very large), but the host galaxies of these
413: AGN are early--type galaxies and hence contain very little star formation,
414: further decreasing the likelihood of a sight line intersecting a SNR.
415: }
416: \end{enumerate}
417:
418: It thus appears that X--ray absorption line strengths are primarily
419: sensitive to galactic and cluster/group properties, and that the systems
420: that typically host bright AGNs (massive ellipticals in high--density
421: environments)
422: are not conducive to producing absorption like that seen at $z=0$. In this
423: case, the Milky Way and Local Group may present a nearly--ideal system
424: for the production of strong absorption: a
425: spiral galaxy with sufficient star formation to produce some ionized
426: gas in the corona, and a low--mass galaxy group that contains a large column
427: density of intra--group medium at $T\sim 10^6$\,K (but not so massive that this
428: medium is fully
429: ionized). Indeed, even slightly larger galaxy groups may already be
430: too hot: the poor group of galaxies \citep[containing four \ion{H}{1}
431: galaxies;][]{shull98} at $z=0.056$ toward PKS 2155--304
432: appears to produce only weak \oviii\ absorption \citep{fang02,fang07},
433: indicating that the oxygen in this system is nearly fully--ionized.
434:
435: Of course, at this point the above explanation for this discrepancy
436: is mostly speculative; quantitative
437: constraints can only be placed with many more high--S/N X--ray spectra.
438: The sightlines analyzed herein were chosen solely on the basis of their
439: being the highest--quality extragalactic \chandra\ grating spectra
440: available. Naturally, a search for X--ray absorption in a massive
441: spiral galaxy hosting a Seyfert--like AGN (or in a quasar sight line
442: passing close to an intervening spiral) would provide a far more direct
443: comparison to our local system, but strong AGN outflows
444: from the brightest known Seyferts \citep[e.g., NGC 3783;][]{kaspi02}
445: hinder the detectability of
446: galactic and intergalactic gas. A similar
447: analysis of the far--UV \ovi\ line with \fuse\ is currently in progress,
448: but this line is expected to trace cooler ($T\sim 10^5$\,K) gas.
449: To place stronger constraints on the origin of local warm--hot gas,
450: soft X--ray spectrometers with both large effective areas and high velocity
451: resolution ($\la 100$\kms) will be critical
452: to distinguish between Galactic and extragalactic absorption at $z=0$
453: and in AGN hosts. Forthcoming missions such as {\it Constellation--X} and
454: {\it XEUS} will be able to make great advances in this field.
455:
456: \acknowledgments
457: R.J.W. acknowledges financial support from the Netherlands Foundation for Scientific
458: Research (NWO). This work is supported in part by \chandra\ grant AR5--6017X
459: from the \chandra\ Science Center to S.M.
460: We are grateful to the \chandra\ team for their efforts on this mission and
461: the accompanying library of software and documentation.
462:
463:
464: \begin{thebibliography}{}
465: \bibitem[Brighenti \& Mathews(1997)]{brighenti97} Brighenti, F. \& Mathews,
466: W.~G.~1997, \apj, 486, L83
467: \bibitem[Cen \& Ostriker(1999)]{cen99} Cen, R., \& Ostriker, J.~P.
468: 1999, \apj, 514, 1
469: \bibitem[Collins et al.(2005)]{collins05} Collins, J.~A., Shull, J.~M.,
470: \& Giroux, M.~L.~2005, \apj, 623, 196
471: \bibitem[Dav\'e et al.(2001)]{dave01} Dav\'e, R., et al.~2001, \apj, 552, 473
472: \bibitem[Falomo et al.(1991)]{falomo91} Falomo, R., Giraud, E., Melnick, J.,
473: Maraschi, L., Tanzi, E.~G., \& Treves, A.~1991, \apj, 380, L67
474: \bibitem[Falomo et al.(1993)]{falomo93} Falomo, R., Pesce, J.~E.,
475: \& Treves, A.~1993, \apj, 411, L63
476: \bibitem[Fang et al.(2002)]{fang02} Fang, T.-T., Marshall, H.~L., Lee, J.~C.,
477: Davis, D.~S., \& Canizares, C.~R.~2002, \apj, 572, L127
478: \bibitem[Fang et al.(2003)]{fang03} Fang, T.-T., Sembach, K.~R., \&
479: Canizares, C.~R.~2003, \apj, 586, L49
480: \bibitem[Fang et al.(2007)]{fang07} Fang, T.-T., Canizares, C.~R., \&
481: Yao, Y.~2007, \apj, in press (arXiv:0708.1800)
482: \bibitem[Fields et al.(2007)]{fields07} Fields, D.~L., Mathur, S., Krongold,
483: Y., Williams, R., \& Nicastro, F.~2007, \apj, 666, 828
484: \bibitem[Kaspi et al.(2002)]{kaspi02} Kaspi, S., et al.~2002, \apj, 574, 643
485: \bibitem[Kikuchi \& Mikami(1985)]{kikuchi85} Kikuchi, S. \&
486: Yoshitaka, M.~1985, \pasj, 39, 237
487: \bibitem[Kravtsov et al.(2002)]{kravtsov02} Kravtsov, A.~V., Klypin,
488: A., \& Hoffman, Y.~2002, \apj, 571, 563
489: \bibitem[Martel et al.(2003)]{martel03} Martel, A.~R., et al.~2003,
490: \aj, 125, 2964
491: \bibitem[Mathur et al.(2007)]{mathur07} Mathur, S.,
492: et al.~2007, ApJL, Submitted (arXiv:0709.2870)
493: \bibitem[Nicastro et al.(2002)]{nicastro02} Nicastro, F., et al.~2002,
494: \apj, 573, 157
495: \bibitem[Nicastro et al.(2003)]{nicastro03} Nicastro, F., et al.~2003,
496: Nature, 421, 719
497: \bibitem[Nicastro et al.(2005)]{nicastro05} Nicastro, F., et al.~2005,
498: \apj, 629, 700
499: \bibitem[Rasmussen et al.(2007)]{rasmussen07} Rasmussen, A.~P., Kahn,
500: S.~M., Paerels, F., den Herder, J.~W., Kaastra, J., \& de Vries,
501: C.~2007, \apj, 656, 129
502: \bibitem[Savage et al.(1993)]{savage93} Savage, B.~D., Lu, L., Weymann, R.~J.,
503: Morris, S.~L., \& Gilliland, R.~L.~1993, \apj, 404, 124
504: \bibitem[Savage et al.(2003)]{savage03} Savage, B.~D., et al.~2003,
505: \apjs, 146, 125
506: \bibitem[Sembach et al.(2003)]{sembachetal03} Sembach, K.~R., et al.~2003,
507: \apjs, 146, 165
508: \bibitem[Sembach(2003)]{sembach03} Sembach, K.~R.~2003, preprint
509: (arXiv:astro-ph/0311089)
510: \bibitem[Shull et al.(1998)]{shull98} Shull, J.~M., Penton, S.~V., Stocke,
511: J.~T., Giroux, M.~L., van Gorkom, J.~H., Lee, Y.~H., \& Carilli, C.~
512: 1998, \aj, 116, 2094
513: \bibitem[Takei et al.(2007)]{takei07} Takei, Y., Henry, J.~P., Finuguenov,
514: A., Mitsuda, K., Tamura, T., Fujimoto, R., \& Briel, U.~G.~2007,
515: \apj, 655, 831
516: \bibitem[Ulrich et al.(1975)]{ulrich75} Ulrich, M.-H., Kinman, T.~D.,
517: Lynds, C.~R., Rieke, G.~H., \& Ekers, R.~D.~1975, \apj, 198, 261
518: \bibitem[Ulrich(1978)]{ulrich78} Ulrich, M.-H.~1978, \apj, 222, L3
519: \bibitem[Wakker et al.(2003)]{wakker03} Wakker, B.~P., et al.~2003,
520: \apjs, 146, 1
521: \bibitem[Wang et al.(2005)]{wang05} Wang, Q.~D., et al.~2005, \apj,
522: 635, 386
523: \bibitem[Williams et al.(2005)]{williams05} Williams, R.~J., et al.~2005,
524: \apj, 631, 856
525: \bibitem[Williams et al.(2006a)]{williams06a} Williams, R.~J., Mathur, S.,
526: Nicastro, F., \& Elvis, M.~2006a, \apj, 642, L95
527: \bibitem[Williams et al.(2006b)]{williams06b} Williams, R.~J., Mathur, S.,
528: \& Nicastro, F.~2006b, \apj, 645, 179
529: \bibitem[Williams et al.(2007)]{williams07} Williams, R.~J., Mathur, S.,
530: Nicastro, F., \& Elvis, M.~2007, \apj, 665, 247
531: \end{thebibliography}
532:
533:
534: \clearpage
535: \begin{figure*}
536: \begin{center}
537: \includegraphics[scale=0.5]{f1.eps}
538: \end{center}
539: \caption{\chandra--LETG spectrum of Mrk 421 at the wavelengths of three
540: strong absorption lines expected at various gas temperatures. In all
541: cases a strong line appears at $z=0$ but nothing is seen at the host
542: galaxy redshift (marked with the arrows; the horizontal arrow base
543: shows the systematic LETG wavelength uncertainty).
544: \label{fig_m421res}}
545: \end{figure*}
546:
547: \begin{figure*}
548: \begin{center}
549: \includegraphics[scale=0.5]{f2.eps}
550: \end{center}
551: \caption{Comparison of the measured equivalent widths of five $z=0$ absorption
552: lines seen toward the four AGNs ({\it circles with error bars}) along with
553: the $2\sigma$ upper limits measured at three AGN host galaxy redshifts
554: ({\it squares with arrows}). Measurements and limits for
555: Mrk 421 are denoted by red points, PKS2155--304 by green, 3C 273 by blue,
556: and Mrk 279 by black. Only the $z=0$ \ovii\ point is plotted for Mrk 279.
557: \label{fig_limits}}
558: \end{figure*}
559:
560: \end{document}
561: