1: \documentclass[12pt,preprint]{aastex}
2: \received{}
3: \revised{}
4: \accepted{}
5:
6: \shorttitle{\ion{O}{8} Absorption}
7: \shortauthors{Fang et al.}
8:
9:
10: \begin{document}
11:
12: \title{{\sl Chandra} Detection of \ion{O}{8} Ly$\alpha$ Absorption from an Overdense Region in the Intergalactic Medium}
13:
14: \author{Taotao Fang, Herman L.~Marshall, Julia C.~Lee, David S.~Davis AND\\Claude R.~Canizares}
15: \affil{Department of Physics and Center for Space Research, MIT\\NE80-6081, 77 Mass. Ave., Cambridge, MA 02139, fangt@space.mit.edu}
16:
17: \begin{abstract}
18: We report the first detection of an \ion{O}{8} Ly$\alpha$ absorption
19: line associated with an overdense region in the intergalactic medium
20: (IGM) along the sightline towards PKS~2155-304 with the {\sl Chandra}
21: Low Energy Transmission Grating Spectrometer (LETGS). The absorption line is detected at $4.5\sigma$ level with $cz
22: \approx 16,600~\rm km\ s^{-1}$. At the same velocity \citet{sps98} detected a small group of spiral galaxies (with an
23: overdensity of $\delta_{gal} \sim 100$) and low metallicity \ion{H}{1}
24: Ly$\alpha$ clouds. We constrain the intragroup gas that gives rise
25: to the \ion{O}{8} Ly$\alpha$ line to a baryon density in the range
26: $1.0 \times 10^{-5} < n_{b} < 7.5\times 10^{-5}~\rm cm^{-3}$ ($50 <
27: \delta_{b} < 350$) and a temperature of $4-5\times10^{6}$ K, assuming 0.1 solar abundance. These estimates are in accordance
28: with those of the warm/hot intergalactic medium (WHIM) that are
29: predicted from hydrodynamic simulations. Extrapolating from this
30: single detection implies a large fraction of the ``missing baryons''
31: ($\sim$ 10\%, or $\sim$ 30-40\% of the WHIM) are probed by the \ion{O}{8} absorber.
32:
33: \end{abstract}
34:
35: \keywords{intergalactic medium --- quasars: absorption lines --- large-scale structure of universe --- BL Lacertae objects: individual (PKS~2155-304)}
36:
37:
38: \section{Introduction}
39:
40: The cosmic baryon budget at low and high redshift indicates that a large fraction of baryons in the local universe have so far escaped detection (e.g., \citealp{fhp98}). While there is clear evidence that a significant fraction of these ``missing baryons'' (between 20-40\% of total baryons) lie in photoionized, low-redshift Ly$\alpha$ clouds \citep{pss00}, the remainder could be located in intergalactic space with temperatures of $10^{5}-10^{7}$ K (warm-hot intergalactic medium, or WHIM). Resonant absorption from highly-ionized ions located in the WHIM gas has been predicted based on both analytic
41: studies of structure formation and evolution
42: (\citealp{sba81,aem94,plo98,fca00}) and cosmic hydrodynamic
43: simulations (\citealp{hgm98,cos99a,dco01,fbc02}). Recent discovery of \ion{O}{6}
44: absorption lines by the Hubble Space Telescope ({\sl HST}) and the Far
45: Ultraviolet Spectroscopic Explorer ({\sl FUSE}) (\citealp{tsa00,
46: tsj00, tgs01}) indicates that there may be a significant reservoir of
47: baryons in \ion{O}{6} absorbers. While Li-like \ion{O}{6} probes about
48: $\sim 30-40\%$ of the WHIM gas (\citealp{cto01, fbr01}),
49: the remaining $\sim 60-70\%$ is hotter and can only be probed by
50: ions with higher ionization potentials, such as H- and He-like Oxygen.
51:
52: The WHIM gas, predicted by hydrodynamic simulations, typically has an
53: overdensity of $\delta_{b} = 5-200$ and is distributed in small groups
54: of galaxies or in large scale filamentary structures that connect
55: virialized halos. Given the expected physical conditions, current
56: instruments such as {\sl Chandra} and {\sl XMM} should be capable of
57: detecting resonance features from the WHIM gas \citep{fbc02}. A first
58: attempt by \citet{fmb01} with the {\sl Chandra} High Energy
59: Transmission Grating Spectrometer (HETGS) yielded an upper limit of ${\rm N(O~VIII) < 10^{17}\
60: cm^{-2}}$. Recently, with the {\sl Chandra} LETGS-HRC, \citet{nic02}
61: discovered X-ray resonant absorption features from warm/hot local gas
62: along the line of sight towards PKS~2155-304, giving the first X-ray
63: evidence for possible WHIM gas. These absorbers are near zero
64: redshift and therefore most likely associated with the Galaxy or
65: local group (see \S~2).
66:
67: We report the first detection of an \ion{O}{8}
68: Ly$\alpha$ absorption line along the sightline towards PKS~2155-304 at
69: $cz \sim 16,600~\rm km~s^{-1}$ plausibly associated with an {\it
70: intervening}, modestly overdense region in the IGM. The observations were made with the {\sl
71: Chandra} LETG-ACIS (Advanced CCD Imaging Spectrometer). Physical diagnostics show the gas that gives rise
72: to this resonance line has the typical properties of the WHIM,
73: indicating the detection of a significant fraction of the ``missing
74: baryons''.
75:
76: \section{Data Analysis}
77:
78: PKS~2155-304 at $z \sim 0.116$ \citep{bbc84} is one of the
79: brightest soft X-ray sources (e.g. \citealp{umu82} and
80: references therein) and has been used as a {\sl Chandra} calibration
81: target. It was observed with the {\sl Chandra} LETG-ACIS (see, e.g., {\sl Chandra} Proposers' Observatory Guide at http://asc.harvard.edu/) on May 31,
82: 2000 , December 6, 2000 and November 31, 2001. We restrict all spectra to the range 2---42 $\rm\AA$ to avoid complexities at longer wavelengths from the instrumental C-K edge. The events were summed in 0.025$\rm\AA$ bins, half of the instrumental line response function, which has nearly constant FWHM $\sim 0.05\,$~\AA\, across its bandpass (for data processing procedures, see \citealp{mcc01}). We corrected the wavelength scale by calibrating the spectrum with two known features (atomic \ion{O}{1} from the Galaxy and solid-state \ion{O}{1} from the instruments).
83: The three data sets were separately
84: divided by the best fit Galactic absorbed power-law to remove the
85: underlying broad-band continuum behavior. The remaining
86: residuals were then fit with a five-order polynomial. Such a technique removes spectral features larger than $\sim
87: 8{\rm \AA}$ (e.g. 5-10\% calibration uncertainties) but preserves narrow line
88: features. The data
89: from the three observations were then combined as a weighted sum that
90: accounts for the differences in exposures and effective area.
91:
92: The observation during November 2001 has the strongest soft flux (
93: respectively for May 2000, December 2000 and November 2001, $F_{\rm 0.1-2.4} = 1.35, 0.87~{\rm and}~2.92 \times 10^{-10}~{\rm ergs~cm^{-2}s^{-1}}$). All continua are well described
94: by a single power law ($\Gamma \sim 2.31 \pm 0.25, 2.60 \pm 0.20$,\ and\ $2.72 \pm 0.47$, respectively) absorbed by a Galactic hydrogen column of $N_{H}
95: = 1.36\times 10^{20} {\rm cm^{-2}}$ \citep{lsa95} (errors are quoted at 90\% confidence).
96:
97: After a blind search for any statistically significant absorption features, two absorption features with S/N~$>$~4 were detected in the
98: 2--42~\AA\, region of the LETGS spectral bandpass (Figure~1). These
99: features were subsequently fit in ISIS (Interactive Spectral
100: Interpretation System : http://space.mit.edu/ASC/ISIS/)
101: with Gaussian models (Table~1). The absorption
102: feature at $\sim 21.6$~\AA\, was reported by
103: \citet{nic02} in the LETGS-HRC archival data, which they attribute to \ion{O}{7}~He$\alpha$ resonance absorption from warm/hot local gas. We also show in Table~1 our results for two other features claimed by \citet{nic02}. The feature attributed to \ion{O}{8}~Ly$\alpha$ is marginally detected ($3\sigma$) while \ion{O}{7}~He$\beta$ is not significant ($1\sigma$). We defer the discussion of these three features to a later paper, noting only that they are at $cz\sim 0$ and therefore probably associated with the Galaxy or local group. We searched for the 20\AA~feature in the LETG-HRC data of this source, but find that this region is complicated by instrumental feature(s) that we also found in the LETG-HRC observations of 3C~273.
104:
105: \section{An absorption feature at $\sim 20$~\AA\, -- evidence of the WHIM}
106:
107: \subsection{Line Identification}
108:
109: We concentrate on the absorption feature which appears at $20.02~{\rm \AA}$ (619~eV). This falls at about the same energy where \citet{ckr84} detected a broad ($\sim$ 50-100 eV) absorption feature from an observation with the {\sl Einstein Observatory} Objective Grating Spectrometer. A similar broad feature was detected by \citet{mad91}. However, the line we detect is unresolved and has a width of $\rm <0.039\AA$ (or 1.2 eV) at 90\% confidence. The Poisson significance of that this feature is due to a chance fluctuation is $2.8\times10^{-6}$, equivalent to $4.5\sigma$ for Gaussian statistics. The probability of finding one such feature by chance in the 1520 bins from 2--40 \AA~is 0.35\%. The 21.61 \AA~feature has a similar confidence level, while the next most prominent feature has 20\% probability of occuring by chance (these values were confirmed using monte carlo simulations).
110:
111: Considering cosmic abundances and oscillator strengths (e.g. \citealp{vya95}) for different ions, \ion{O}{8} Ly$\alpha$ is the only strong candidate line between 18 and 20~\AA\,, the measured wavelength de-redshifted to the source. One scenario which may reconcile \ion{O}{8} Ly$\alpha$ at $18.97{\rm \AA}$ with the $20.02$~\AA\, of the detected feature is that this is
112: the signature of a high velocity $\rm \sim 18,000~km~s^{-1}$ wind/jet blueshifted from the
113: BL Lac rest frame. In this case the
114: narrowness of the line would be surprising, especially when compared to
115: the very broad ($\approx~30,000~\rm km s^{-1}$) feature seen by \citet{ckr84} and \citet{mad91}, although we note that the relativistic jet in the galactic binary SS433 does
116: have roughly constant velocity \citep{mcs01}. The implied mass outflow would be hundreds to thousands
117: of $\rm M_{\odot} \, \rm yr^{-1}$, depending on assumptions about
118: ionization conditions and beaming factors, as shown in \citet{kkf85}.
119: They also discuss some of the resulting difficulties with such an outflow.
120:
121: It is more plausible that the 20~\AA\ absorption is due to \ion{O}{8}
122: Ly$\alpha$ in a known intervening system at $cz\approx 16,734\rm\
123: km~s^{-1}$. With the {\sl Hubble Space Telescope} ({\sl HST}), \citet{sps98} discovered a cluster of low metallicity \ion{H}{1}
124: Ly$\alpha$ clouds along the line-of-sight (LOS) towards PKS~2155-304,
125: most of which have redshift between $cz = 16,100\ {\rm km\ s^{-1}}$
126: and $18,500\ {\rm km\ s^{-1}}$. Using 21 cm images from the {\sl Very
127: Large Array} ({\sl VLA}), they detected a small group of four
128: \ion{H}{1} galaxies offset by $\sim 400-800\ {\rm h_{70}^{-1}}$ kpc \footnote{ We use $\rm H_0
129: = 70h_{70}~km~s^{-1}Mpc^{-1}$, and a standard $\Lambda$CDM model with $\Omega_{m}=0.3$ and $\Omega_{\Lambda} = 0.7$ throughout the paper.}
130: from the LOS, and suggested that the \ion{H}{1} Ly$\alpha$ clouds
131: could arise from gas associated with the group. The velocity centroid
132: of the four galaxies is $\left<V_{gal}\right> = 16,853\rm~km~s^{-1}$,
133: which within the uncertainties is identical to that of our absorption
134: feature. Therefore the hot gas giving rise to the \ion{O}{8}
135: absorption can plausibly be associated with the same overdensity
136: indicated by the \ion{H}{1} Ly$\alpha$ absorbers and the small galaxy
137: group.
138:
139: Taking the absorption line to be \ion{O}{8} Ly$\alpha$, the Doppler $b$-parameter must be $b=\sqrt{2}\sigma \rm < 830~km~s^{-1}$ (90\% confidence) since the line is unresolved. If $b > 200\rm~km~s^{-1}$, then the line is unsaturated and $\rm N(O~VIII) \sim 9.5 \times 10^{15}~cm^{-2}$. At $b < 100\rm~km~s^{-1}$ the line is heavily
140: saturated, but this can be ruled out by the lack of higher Lyman series
141: lines in our spectrum. Associating the absorber with the galaxy group
142: detected by \citet{sps98}, which has a radial velocity dispersion
143: of $\left<\sigma_{gal}\right> = 325\rm~km~s^{-1}$, suggests that the unsaturated case is most appropriate, and we adopt that for the subsequent discussion.
144: However, simulations assuming complete viralization of the hot gas predict $b < 100\rm~km~s^{-1}$ for lines such as this one
145: \citep{fbc02}, so saturation cannot be ruled out.
146:
147: \subsection{Density and Temperature Constraints}
148:
149: We can constrain the density of the absorbing gas, assuming it is
150: associated with the intervening galaxy group. First, the upper limit on
151: the line width of $\rm \sigma \sim 580~km~s^{-1}$ sets an upper limit to the path length of ${\rm\sim 8h_{70}^{-1}Mpc}$, since otherwise the differential Hubble flow would excessively broaden the line. For $\rm N(O~VIII) \sim 9.5 \times 10^{15}~cm^{-2}$, this gives $n_{b} > (1.0\times10^{-5}\ {\rm cm}^{-3})\
152: Z_{0.1}^{-1}f_{0.5}^{-1}l_{8}^{-1}$ where $Z_{0.1}$ is the
153: metallicity in units of 0.1 solar abundance, $f_{0.5}$ is the
154: ionization fraction in units of 0.5 and $l_{8}$ is the path
155: length in units of ${\rm 8h_{70}^{-1}Mpc}$. We adopt a metallicity of $\rm 0.1~Z_{\odot}$ \citep{agr89}, which
156: is the mean value predicted from hydrodynamic simulations for overdensities of $\sim$ 30, with a dispersion of a factor of two \citep{cos99b}. A more reasonable estimate of the
157: path length comes from the mean projected separation of $\sim$1 Mpc for the
158: galaxies in the group \citep{sps98}, which gives $n_{b} \approx 7.5 \times 10^{-5}\ {\rm cm}^{-3}~Z_{0.1}^{-1}f_{0.5}^{-1}$. In fact, this could be taken as
159: a plausible upper limit, since the WHIM associated with this overdensity could well be more extended than the galaxies, as are the \ion{H}{1} Ly$\alpha$ clouds. This implies a range of baryon overdensity over the cosmic mean $\left<n_{b}\right> = 2.14\times 10^{-7}~\rm cm^{-3}$ \citep{tos00} of $\delta_{b} \sim$ 50 - 350. Interestingly, \citet{sps98} estimate an overdensity for the galaxy group of $\delta_{gal}\sim$100.
160:
161: In the case of pure collisional ionization, temperature is the only parameter
162: of importance over a wide range of density so long as the gas is optically
163: thin. The \ion{O}{8} ionization fraction peaks at 0.5, and exceeds 0.1
164: for temperatures $T \sim 2-5 \times10^{6}$ K. Using CLOUDY \citep{fkv98} we find that photoionzation by the cosmic UV/X-ray background is not important for $n_{b} > 10^{-5}\ {\rm cm^{-3}}$.
165:
166: We can compare this estimate for T to what we might expect from the
167: dynamics of the group of galaxies, if it forms a bound system. Then its
168: virial mass is $\rm M_{vir} \approx 3\times10^{13}h_{70}^{-1}\ M_{\odot}$. Scaling the mass-temperature relationship
169: down from richer clusters of galaxies \citep{bno98}, gives a temperature of $\sim 9\times 10^{6}$ K. Similarly, using the
170: relation between velocity dispersion and temperature obtained from fitting a sample of galaxy groups gives a value
171: of $1-2 \times 10^{7}$ K \citep{xwu00}. These values are somewhat
172: higher than what we find from the ionization fraction. This could suggest that
173: the system is not bound and virialized. However, \citet{mmb96} suggested that the
174: intragroup gas in spiral-rich groups, such as this one, may be cooler than
175: that of groups and clusters of early-type galaxies. They suggested that these cooler systems may imprint absorption lines from
176: highly-ionized species (e.g. \ion{O}{6}) on quasar spectra, which may
177: be what we observe.
178:
179: CLOUDY calculations of the column density ratios between other ions
180: and \ion{O}{8} give another tight constraint on gas temperature. Based on
181: the non-detection of other ion species, we calculate the $4\sigma$
182: upper limits of the column densities of H- and He-like C, N, Ne and He-like
183: oxygen relative to \ion{O}{8}. N(O~VII)/N(O~VIII) provide the strongest constraints on gas temperature. At roughly $T \ga 10^{6.4}$ K, all the column density ratios satisfy the observed constraints.
184:
185: Adopting the value of $T = 4-5\times10^{6}$ K, we can compare the
186: cooling timescale for the gas to the Hubble time $t_{H}$. The cooling
187: time scale $t_{cool} \approx (3/2)(kT/n_{b}\Lambda(T))$, here
188: ${\rm \Lambda(T)}$ is the cooling function \citep{sdo93}, which is a function of temperature and metallicity (in
189: the temperature range of interest, line emission from iron is the most
190: effective coolant). For 0.1$Z_{\odot}$, the baryon density should be $n_{b} \la
191: 2\times10^{-4}~(t_{H}/t_{cool})~{\rm cm^{-3}}$ so that the gas remains hot. This density limit, however, would be lower by a factor of $\sim
192: 3$ if the abundance is solar. Alternatively, we can use the cooling
193: argument to set an upper limit on the metallicity. Adopting a
194: baryon density of ${\rm 7.5\times10^{-5}\ cm^{-3}}$ and requiring $t_{cool} > t_{H}$, the cooling function $\Lambda$ must be less than ${\rm 2\times 10^{-23}\ ergs\ cm^{3}s^{-1}}$, which implies a metallicity of $\la 0.3Z_{\odot}$.
195:
196: Following \citet{lmw91} and \citet{tsa00} we
197: estimate the baryonic content $\rm \Omega_{b}(O~VIII)$ that can be probed by our
198: \ion{O}{8} resonance absorption line, expressed in units of $\left<n_{b}\right>$. We assume conservative upper limits of $\rm Z
199: \la 0.5Z_{\odot}$ and \ion{O}{8} ionization fraction of $f \la
200: 0.5$. Given the path length of $\Delta z \la 0.116$, we estimate
201: $\rm \Omega_{b}(O~VIII) \ga 0.005h_{70}^{-1}$. This is about 10\% of the
202: total baryon fraction, or about 30-40\% of the WHIM gas, if the WHIM gas contains about 30-40\% of total baryonic matter \citep{dco01}. This baryon fraction is consistent with the prediction from \citet{plo98} based on a simple analytic model; however, these numbers are very rough estimates, due to the large uncertainties in
203: the abundance and ionization fraction.
204:
205: \section{Discussion \& Summary}
206:
207: Our detection of \ion{O}{8} Ly$\alpha$ from an intervening overdense region in the
208: IGM at $cz \sim 16,600\rm~km~s^{-1}$, together with the detection by
209: \citet{nic02} of
210: absorption from systems at $cz \sim 0$ associated with the Galaxy or local group, have begun
211: to reveal the much-anticipated warm-hot component of intergalactic matter
212: (\citealp{hgm98,cos99a,dco01,fbc02}).
213:
214: The \ion{O}{8} Ly$\alpha$ feature we observe is plausibly associated with the same
215: cosmic overdensity that gives rise to the \ion{H}{1} Ly$\alpha$ clouds
216: and the four spiral galaxies found by \citet{sps98}. Three of the
217: galaxies have radial velocities within $\sim100~\rm km~s^{-1}$ of one another,
218: while the fourth is within $\sim400~\rm km~s^{-1}$. In contrast, the velocities
219: of the \ion{H}1 Ly$\alpha$ clouds span $\sim1800~\rm km~s^{-1}$, with two
220: falling very close to the velocities of the closest projected galaxies. The
221: galaxies represent an overdensity of $\delta_{gal} \sim 100$ relative to the mean. Our
222: estimates for the overdensity of the hot absorbing gas are $\delta_{b} \sim$50-350. The
223: \ion{H}{1} Ly$\alpha$ clouds have very low metallicity, indicating that
224: they are primarily composed of primordial material, whereas the hot
225: gas should have an oxygen abundance of $\sim 0.1$ solar to give sensible
226: values for the column density. It is unlikely
227: that the same gas is responsible for both \ion{H}{1} and \ion{O}{8} absorption. Following \citet{sps98}, we
228: identify this as a region of modest overdensity (the galaxy group) connected
229: to filamentary or sheet-like structures demarcated by the \ion{H}{1} Ly$\alpha$ clouds, such as those seen in numerous
230: cosmological simulations.
231:
232: We note that the detection of a system such as this is roughly consistent with
233: previous simulations. Based on the models of \citet{plo98} and \citet{fbc02}, we estimate the cumulative absorption line number along a random sightline (Fig.~2). We adopt the numerical relation between baryon overdensity and metallicity from \citet{cos99b} and assume collisional ionization. This observation is $\sim 1.5\sigma$ above the predicted value at the measured column density. According to Fig.~3, the probability of detecting such an \ion{O}{8} Ly$\alpha$ line or stronger to $z\sim 0.1$ is $\sim$ 10\%. The consistency, if borne out by further studies, suggests that the simulations
234: are valid descriptors of the warm/hot component of the IGM, at least
235: for overdensities of $\sim$100.
236: Using those simulations, we estimate that we can probe about 10\% of
237: total baryons and about 30-40\% of the WHIM gas in the local universe.
238:
239: Other observations of bright AGN
240: with the {\sl Chandra} and {\sl XMM-Newton} grating spectrometers are likely
241: to reveal additional WHIM absorption features, and may turn the present
242: few samplings into a real forest of high ionization Ly$\alpha$ or He$\alpha$ lines. However, the
243: sensitivity of these instruments permit us only to probe the high density
244: tail of the distribution (e.g. \citealp{fca00}). It will be left to future missions like {\sl Constellation-X} and {\sl XEUS}, assuming they have both high throughput and
245: spectral resolving powers of 1000 or more at energies 0.1-1 keV, to fully reveal the X-ray forest.
246:
247: We are grateful to F.~Nicastro and his colleagues for sharing an advance
248: copy of their paper. We thank the other members of the MIT/CXC team for its support. We also thank the referee, J.~M.~Shull, for helpful comments. This work is supported in part by contracts NAS 8-38249 and SAO SV1-61010.
249:
250: \clearpage
251: \begin{thebibliography}{}
252: \bibitem[Aldcroft, Elvis, McDowell, \& Fiore(1994)]{aem94} Aldcroft, T., Elvis, M., McDowell, J., \& Fiore, F.\ 1994, \apj, 437, 584
253: \bibitem[Anders \& Grevesse(1989)]{agr89} Anders, E.\ \& Grevesse, N.\ 1989, \gca, 53, 197
254: \bibitem[Bowyer, Brodie, Clarke, \& Henry(1984)]{bbc84} Bowyer, S., Brodie, J., Clarke, J.~T., \& Henry, J.~P.\ 1984, \apjl, 278, L103
255: \bibitem[Bryan \& Norman(1998)]{bno98} Bryan, G.~L.~\& Norman, M.~L.\ 1998, \apj, 495, 80
256: \bibitem[Canizares \& Kruper(1984)]{ckr84} Canizares, C.~R.~\& Kruper, J.\ 1984, \apjl, 278, L99
257: \bibitem[Cen \& Ostriker(1999a)]{cos99a} Cen, R. \& Ostriker, J.P. 1999a, \apj, 514, 1
258: \bibitem[Cen \& Ostriker(1999b)]{cos99b} Cen, R.\ \& Ostriker, J.\ P.\ 1999b, \apjl, 519, L109
259: \bibitem[Cen et al.(2001)]{cto01} Cen, R., Tripp, T.M., Ostriker, J.P. \& Jenkins, E.B. 2001, \apjl, 559, L5
260: \bibitem[Dav{\' e} et al.(2001)]{dco01} Dav{\' e}, R.\ et al.\ 2001, \apj, 552, 473
261: \bibitem[Fang \& Bryan(2001)]{fbr01} Fang, T.~\& Bryan, G.~L.\ 2001, \apjl, 561, L31
262: \bibitem[Fang, Bryan \& Canizares(2002)]{fbc02} Fang, T., Bryan, G.L. \& Canizares, C.R. 2002, \apj, 564, 604
263: \bibitem[Fang \& Canizares(2000)]{fca00} Fang, T.~\& Canizares, C.~R.\ 2000, \apj, 539, 532
264: \bibitem[Fang et al.(2002)]{fdl02} Fang, T., Davis, D.S., Lee, J.L., Marshall, H.L., Bryan, G.L. \& Canizares, C.R. 2002, \apj, 565, 86
265: \bibitem[Fang, Marshall, Bryan, \& Canizares(2001)]{fmb01} Fang, T., Marshall, H.~L., Bryan, G.~L., \& Canizares, C.~R.\ 2001, \apj, 555, 356
266: \bibitem[Ferland et al.(1998)]{fkv98} Ferland, G.J., Korista, K.T., Verner, D.A., Ferguson, J.W., Kingdon, J.B., Verner, E.M. 1998, \pasp, 110, 761
267: \bibitem[Fukugita, Hogan, \& Peebles(1998)]{fhp98} Fukugita, M., Hogan, C.~J., \& Peebles, P.~J.~E.\ 1998, \apj, 503, 518.
268: \bibitem[Gehrels(1986)]{geh86} Gehrels, N.\ 1986, \apj, 303, 336
269: \bibitem[Hellsten, Gnedin, \& Miralda-Escud{\' e}(1998)]{hgm98} Hellsten, U., Gnedin, N.~Y., \& Miralda-Escud{\' e}, J.\ 1998, \apj, 509, 56
270: \bibitem[Krolik, Kallman, Fabian, \& Rees(1985)]{kkf85} Krolik, J.~H., Kallman, T.~R., Fabian, A.~C., \& Rees, M.~J.\ 1985, \apj, 295, 104
271: \bibitem[Lanzetta et al.(1991)]{lmw91} Lanzetta, K.~M., McMahon, R.~G., Wolfe, A.~M., Turnshek, D.~A., Hazard, C., \& Lu, L.\ 1991, \apjs, 77, 1
272: \bibitem[Lockman \& Savage(1995)]{lsa95} Lockman, F.\ J.\ \& Savage, B.\ D.\ 1995, \apjs, 97, 1
273: \bibitem[Marshall, Canizares, \& Schulz(2001)]{mcs01} Marshall, H.\ M., Canizares, C.\ R., \& Schulz, N.\ S.\ 2001, \apj, in press (astro-ph/0108206)
274: \bibitem[Madejski et al.(1991)]{mad91} Madejski, G.~M., Mushotzky, R.~F., Weaver, K.~A., Arnaud, K.~A., \& Urry, C.~M.\ 1991, \apj, 370, 198
275: \bibitem[McClintock et al.(2001)]{mcc01} McClintock, J.~E.~et al.\ 2001, \apj, 555, 477
276: \bibitem[Mulchaey, Mushotzky, Burstein, \& Davis(1996)]{mmb96} Mulchaey, J.~S., Mushotzky, R.~F., Burstein, D., \& Davis, D.~S.\ 1996, \apjl, 456, L5
277: \bibitem[Mushotzky, Cowie, Barger, \& Arnaud(2000)]{mcb00} Mushotzky, R.~F., Cowie, L.~L., Barger, A.~J., \& Arnaud, K.~A.\ 2000, \nat, 404, 459
278: \bibitem[Nicastro et al.(2002)]{nic02} Nicastro, F. et al.~2002, \apj, submitted (astro-ph/0201058)
279: \bibitem[Perna \& Loeb(1998)]{plo98} Perna, P. \& Loeb, A. 1998, \apjl,503, L135
280: \bibitem[Penton, Shull, \& Stocke(2000)]{pss00} Penton, S.~V., Shull, J.~M., \& Stocke, J.~T.\ 2000, \apj, 544, 150.
281: \bibitem[Shapiro \& Bahcall(1981)]{sba81} Shapiro, P.~R.~\& Bahcall, J.~N.\ 1981, \apj, 245, 335
282: \bibitem[Shull et al.(1998)]{sps98} Shull, J.~M., Penton, S.~V., Stocke, J.~T., Giroux, M.~L., van Gorkom, J.~H., Lee, Y.~H., \& Carilli, C.\ 1998, \aj, 116, 2094
283: \bibitem[Sutherland \& Dopita(1993)]{sdo93} Sutherland, R.S. \& Dopita, M.A. 1993, \apjs, 88, 253
284: \bibitem[Tripp \& Savage(2000)]{tsa00} Tripp, T.M. \& Savage, B.D. 2000, \apj, 542, 42
285: \bibitem[Tripp, Savage \& Jenkins(2000)]{tsj00} Tripp, T.M., Savage, B.D. \& Jenkins, E.B. 2000, \apj, 534, L1
286: \bibitem[Tripp et al.(2001)]{tgs01} Tripp, T.~M., Giroux, M.~L., Stocke, J.~T., Tumlinson, J., \& Oegerle, W.~R.\ 2001, \apj, 563, 724
287: \bibitem[Tytler, O'Meara, Suzuki, \& Lubin(2000)]{tos00} Tytler, D., O'Meara, J.~M., Suzuki, N., \& Lubin, D.\ 2000, Physica Scripta Volume T, 85, 12
288: \bibitem[Urry \& Mushotzky(1982)]{umu82} Urry, C.~M.~\& Mushotzky, R.~F.\ 1982, \apj, 253, 38
289: \bibitem[Verner \& Yakovlev(1995)]{vya95} Verner, D.A. \& Yakovlev, D.G. 1995, \aaps, 109, 125
290: \bibitem[Xue \& Wu(2000)]{xwu00} Xue, Y.~\& Wu, X.\ 2000, \apj, 538, 65
291: \end{thebibliography}
292:
293: \clearpage
294:
295: \vbox{
296: \scriptsize
297: \begin{center}
298:
299: \begin{tabular}{lllll}
300: \multicolumn{5}{c}{Table 1: Fitting parameters of the X-ray absorption Lines~~~} \\
301: \hline
302: \hline
303: & O~{\sc viii} Ly$\alpha$ & O~{\sc viii} Ly$\alpha$ & O~{\sc vii} He$\alpha$ & O~{\sc vii} He$\beta$ \\
304: \hline
305: $\lambda_{rest}$$^{a}$ & 18.9689 & 18.9689 & 21.6019 & 18.6288 \\
306: $\lambda_{obs}$ & $18.95_{-0.02}^{+0.02}$ & $20.02_{-0.015}^{+0.015}$ & $21.61_{-0.01}^{+0.01}$ & $18.63_{-0.02}^{+0.02}$ \\ $cz({\rm km\ s^{-1}})$ & $-300_{-315}^{+317}$ & $16624\pm237$ & $112_{-138}^{+140}$ & $20_{-323}^{+321}$ \\
307: Line Width$^{b}$ & $ ... $ & $< 0.039$ & $<0.027$ & $ ... $\\
308: $\rm Line\ Flux^{c}$ & $3.0_{-2.3}^{+1.2}$ & $4.8_{-1.9}^{+2.5}$ & $5.5_{-1.7}^{+3.0}$ & $1.1_{-0.6}^{+1.8}$ \\
309: EW (m${\rm \AA}$) & $8.8_{-6.9}^{+3.5}$ & $14.0_{-5.6}^{+7.3}$ & $15.6_{-4.9}^{+8.6}$ & $3.3_{-1.8}^{+5.5}$ \\
310: SNR$^{d}$ & 3.0 & 4.5 & 4.6 & 1.1 \\
311: \hline
312: \end{tabular}
313:
314: \parbox{3.5in}{
315: \vspace{0.1in}
316: \small\baselineskip 9pt
317: \footnotesize
318: \indent
319: a. Rest-frame wavelengths in units of $\rm \AA$ \citep{vya95}.\\
320: b. 90\% upper limit of the line width $\sigma$, in units of $\rm\AA$.\\
321: c. Absorbed line flux in units of $\rm 10^{-5}~photons~cm^{-2}s^{-1}$.\\
322: d. In units of equivalent sigma of a Gaussian distribution with the same confidence level as the Poisson significance gives.\\
323: }
324: \end{center}
325: \normalsize
326: \centerline{}
327: }
328:
329:
330: \clearpage
331: \figcaption[f1.eps]{The {\sl Chandra} LETG-ACIS spectrum of PKS~2155-304 between 16 and 23 $\rm\AA$. (a): The red solid line is the fitted continuum (Galactic absorbed power law plus the polynomial), the averaged $1\sigma$ error bar plotted on the right is based on statistics only; (b) the signal-to-noise ratio; (c) enlarged portion of the spectrum between 18 and 22 $\rm\AA$, the red solid line is the fitted continuum plus three Gaussian models, the ion species are labeled in blue.\label{fig1}}
332:
333: \figcaption[f2.eps]{Cumulative \ion{O}{8} absorption lines per unit redshift vs. column density. Three data points are: PKS~2155 (this paper), H~1821 \citep{fdl02}, S5~0836 \citep{fmb01}. The errors are estimated at $2\sigma$ level \citep{geh86}. For non-detection cases, $2\sigma$ upper limits are given.}
334:
335: \clearpage
336: \begin{figure}
337: \epsscale{0.5}
338: \plotone{f1.eps}
339: \end{figure}
340: \clearpage
341:
342: \begin{figure}
343: \epsscale{1.0}
344: \plotone{f2.eps}
345: \end{figure}
346:
347: \end{document}