1: %Version 1, 10/08/02
2: \documentclass[12pt,preprint]{aastex}
3: \received{}
4: \revised{}
5: \accepted{}
6:
7: \shorttitle{3C 273}
8: %\shortauthors{Fang et al.}
9:
10: \begin{document}
11: \title{{\sl Chandra} Detection of Local \ion{O}{7} He$\alpha$
12: Absorption Along the Sight Line Toward 3C~273}
13:
14: \author{Taotao~Fang\altaffilmark{1,2},
15: Kenneth~R.~Sembach\altaffilmark{3}, AND
16: Claude~R.~Canizares\altaffilmark{1}}
17: \altaffiltext{1}{Department of Physics and Center for Space
18: Research, MIT, 77 Mass. Ave., Cambridge, MA 02139, }
19: \altaffiltext{2}{Department of Physics, Carnegie Mellon University,
20: 5000 Forbes Ave., Pittsburgh, PA 15213}
21: \altaffiltext{3}{Space Telescope Science
22: Institute, 3700 San Martin Dr., Baltimore, MD 21218}
23:
24:
25: \begin{abstract}
26:
27: With the {\sl Chandra} X-ray Telescope we have detected a zero-redshift
28: \ion{O}{7}
29: He$\alpha$ absorption line along the sight line toward 3C~273.
30: This line detection is highly significant, with a signal-to-noise
31: ratio (S/N) of 6.4. We explore two models, which associate this line
32: with (1) the intragroup gas in the Local Group, and (2) the hot halo gas in the vicinity of our Milky Way. In the first model,
33: we find that for a standard $\beta$-model of the gas distribution in the
34: Local Group, the temperature is constrained to $\rm 2.3\times 10^5 < T < 1.2\times 10^6$ K and the baryon overdensity is $\delta_b \sim 100$; both results are
35: consistent with the properties of the so-called warm-hot intergalactic
36: medium (WHIM) predicted by cosmological simulations. We also find the
37: core radius of the gas distribution should be $> 100$ kpc. In the
38: second model we discuss several possible Galactic origins for the
39: absorption, and we comment on the possibility that the \ion{O}{7} is
40: associated with the \ion{O}{6} absorption observed in this direction by
41: the Far Ultraviolet Spectroscopic Explorer ($\sl FUSE$). We find that there is
42: a strong indication that collisional ionization is the dominant ionization
43: source for the observed absorption.
44:
45: \end{abstract}
46:
47: \keywords{ISM: cloud --- Local Group --- intergalactic medium --- quasars: absorption lines --- quasars: individual (3C~273) --- X-rays: individual (3C~273)}
48: \section{Introduction}
49:
50: Highly ionized resonance absorption lines from heavy elements (such
51: as \ion{C}{4}, \ion{Si}{4}, \ion{N}{5}, \ion{O}{6}) at low redshift
52: have been detected along many lines of sight by the Far
53: Ultraviolet Spectroscopic Explorer ($\sl FUSE$) and the Hubble Space
54: Telescope ($\sl HST$) (e.g., \citealp{tsj00,sem00,sav00}). These
55: species arise in the interstellar medium of the Milky Way and the
56: low redshift intergalactic medium (IGM). Among these ions, \ion{O}{6} is
57: of particular interest because it is an excellent tracer of gas at
58: $T \geq 10^5$ K. The detection of
59: \ion{O}{6} in many directions (\citealp{sav02}; \citealp{sem02};
60: \citealp{wak02}) indicates the existence of hot gas in the halo of the
61: Milky Way, in high velocity clouds surrounding the Milky Way, or even in the Local Group \citep{nze02}.
62:
63: Recently, with the unprecedented high resolution of the {\sl Chandra}
64: X-ray Telescope, \citet{nic02} and \citet{fan02}
65: reported the first detection of X-ray absorption lines from higher ionization
66: species, \ion{O}{7} and
67: \ion{O}{8}, at $z\approx 0$ along the sight line toward the quasar
68: PKS~2155-304. \citet{mwc02} reported possible detection of such X-ray absorption feature in the spectrum of H~1821+643. \citet{kas02} also
69: reported the detection of zero-redshift \ion{O}{7} in a long exposure
70: ($\sim$ 900 ksec) of the bright
71: Seyfert galaxy NGC~3783 with {\sl Chandra}. Both \ion{O}{7} and
72: \ion{O}{8} trace
73: gases with temperatures $\gtrsim10^6$\,K. Therefore, it is of
74: great interest to see whether the higher temperature gas is related
75: to the gas that produces the UV absorption lines and whether there is
76: a multi-phase distribution of hot gas in the vicinity of the Milky
77: Way. It is also possible that the X-ray absorption is associated
78: with hot intragroup gas in the Local Group. If so, this might
79: provide important information about the ``missing baryons'' at low
80: redshift. Cosmological simulations indicate that a large fraction of
81: neutral gas at high redshift is heated to high temperatures
82: ($ 10^5 < T < 10^7$ K) at $z\sim 0$
83: (\citealp{cko95,oce96,fhp98,cos99}); gas in the higher end of this
84: temperature range has so far escaped detection
85: at low redshift. Simulations predict that these ``missing baryons''
86: are distributed in intergalactic
87: space in the form of diffused filamentary structures (the warm-hot intergalactic medium, or WHIM) that connect virialized clusters and/or groups of
88: galaxies with ovdensities of $\delta_b \sim 5 - 200$.
89:
90: In this {\sl Letter} we report the detection of a $z\sim0$ \ion{O}{7}
91: He$\alpha$ resonance absorption line in the {\sl Chandra} spectra of
92: 3C~273. We explore two possible sources of this absorption:
93: (1) intragroup gas in the Local Group, and (2) hot gas in the
94: Milky Way halo and interstellar medium along the sight line.
95:
96: \section{Data Analysis}
97:
98: 3C~273 ($l=289.95^{\circ}$, $b=64.36^{\circ}$, $z_{QSO}=0.158$) is one
99: of the brightest quasars in the sky at X-ray wavelengths
100: and has been repeatedly observed as a
101: {\sl Chandra} calibration target. It was observed with the {\sl
102: Chandra} Low Energy Transmission Grating Spectrometer (LETGS) and the
103: Advanced CCD Imaging Spectrometer (ACIS)
104: on 31 October 2000 (obs. \#1198), 13 June 2001 (obs \#2464), and 15 June 2001
105: (obs \#2471), with an exposure time of 40, 30 and 30 ksec, respectively. See the {\sl Chandra} Proposers' Observatory Guide at
106: http://asc.harvard.edu/ for more information about {\sl Chandra} and
107: its instrumentation. In the analysis and discussions that follow, we adopt
108: $\rm H_0 = 70\,h_{70}~ km~s^{-1}~Mpc^{-1}$. Unless
109: otherwise stated, errors are quoted at 90\% confidence.
110:
111: We reduced the Level 1 data using standard CIAO tools\footnotemark
112: \footnotetext{Chandra Interactive Analysis of Observations (CIAO)", see http://asc.harvard.edu/ciao/}, and
113: restricted all spectra to the 2--42\,\AA\ wavelength range to avoid
114: complexities at longer wavelengths from the instrumental Carbon K edge.
115: We summed the events in 0.025\,\AA\ bins, which are half the width of
116: the instrumental line response function (FWHM $\sim
117: 0.05$\,\AA\ across the bandpass).
118: To remove the underlying broad-band continuum
119: behavior for each dataset, we divided each spectrum by the best fit
120: power-law, with a correction for absorption by the foreground Galactic
121: interstellar medium.
122: We then fit the remaining residuals with a sixth-order
123: polynomial. These procedures remove spectral features larger than
124: $\sim 8$\,\AA\ in width (e.g., broad 5-10\% calibration uncertainties) but
125: preserve narrow line features. We combined the data from the three observations
126: as a weighted sum accounting for differences in exposure times
127: and effective areas.
128:
129: All of the continua are well described by a single power law with a
130: similar power-law index. The power-law
131: photon indices are $\Gamma_{1198} \sim 2.08 \pm 0.36, \Gamma_{2464}\sim
132: 1.86 \pm 0.40$,\ and\ $\Gamma_{2471}\sim 1.82 \pm 0.48$, assuming
133: absorption by a Galactic neutral hydrogen column of
134: $N_{H\,I} = 9\times 10^{19}~{\rm cm^{-2}}$ (Lockman \&
135: Savage~1995). The October 2000 observation exhibits the strongest
136: soft X-ray flux; $F_{\rm 0.1-2.4\,keV} = (9.3, 7.9, 7.8 )\times 10^{-11}~{\rm
137: ergs~cm^{-2}s^{-1}}$ for observations \#1198, \#2464, and \#2471, respectively. The total counts per bin in the continuum decline smoothly from $\sim$ 150 at 18\,\AA\ to $\sim$ 40 at 23\,\AA, with $\sim$ 70 at 21.6\,\AA.
138:
139: After a blind search for any statistically significant absorption
140: features in the 2--42\,\AA\ region of the LETGS spectral bandpass
141: (Figure~1), we identified an absorption feature with S/N~$\sim 6.4$ at
142: $21.60\pm0.01$\AA, which is almost identical to the rest wavelength of \ion{O}{7} He$\alpha$ (see Table~1). We fit this feature with single-component
143: Gaussian models using ISIS\footnotemark, taking into account the
144: instrumental line response function (Table~1).
145: \footnotetext{Interactive Spectral Interpretation System, see \citealp{hde00}.}
146: We also examined other {\sl
147: Chandra} LETGS archival data for 3C\,273 and found a similar feature
148: at the same wavelength with a different focal plane detector - the
149: High Resolution Camera (HRC).
150:
151: \section{Local \ion{O}{7} He$\alpha$ Absorption}
152:
153: Based on the detected line equivalent width ($W_{\lambda}$), we can
154: estimate the implied column density of \ion{O}{7}
155: \citep{spi78}. Since the line is unresolved, the Doppler
156: $b$-parameter must be $b=\sqrt{2}\sigma \rm < 400~km~s^{-1}$ (90\%
157: confidence). If $b > 200\rm~km~s^{-1}$, then the line is unsaturated
158: and N(O~{\sc vii})$\rm = 1.8_{-0.7}^{+2.7} \times 10^{16}~cm^{-2}$.
159: The line becomes saturated at $b \sim 100\rm~km~s^{-1}$, giving
160: N(O~{\sc vii})$\rm \sim 10^{17}~cm^{-2}$. At $b < 100\rm~km~s^{-1}$ the
161: line is heavily saturated, but this can be ruled out by the lack of
162: higher Lyman series lines in our spectrum. From the non-detection of
163: local \ion{O}{7} He$\beta$ line at $18.6288\AA$, we estimate a
164: $4\sigma$ equivalent width upper limit of 11.9$\rm~m\AA$. Based
165: on a curve-of-growth calculation for \ion{O}{7} He$\beta$, we find that
166: even in the case for which the line is heavily saturated
167: ($b < 100\rm~km~s^{-1}$),
168: the He$\beta$ upper limit restricts the \ion{O}{7} column density to
169: $< 2\times10^{16}~\rm cm^{-2}$. We therefore
170: adopt N(O~{\sc vii})$\rm=1.8_{-0.7}^{+0.2}
171: \times 10^{16}~cm^{-2}$ in the following discussion.
172:
173: Highly ionized metals such as \ion{C}{4} and \ion{O}{6} at $z\approx
174: 0$ have been detected repeatedly through their absorption-line
175: signatures along many lines of sight (see \citealp{sav02},
176: \citealp{sem02}, and references therein).
177: Unlike these lower ionization species, which typically trace transition
178: temperature gas with $T\sim (1 - 5) \times 10^{5}$ K, the \ion{O}{7}
179: detected with {\sl Chandra} indicates the existence of the gas with
180: temperatures $T\sim 10^{6}$ K. This high temperature gas traced by \ion{O}{7}
181: is unlikely to be produced in the nearby interstellar medium (ISM),
182: although we cannot rule out the possibility of a nearby supernova remnant
183: origin. It is more
184: plausible that the He-like Oxygen is produced in hot Galactic halo
185: gas, or even in the Local Group. It is to these two possibilities that
186: the rest of this ${\sl Letter}$ is devoted.
187:
188:
189: \subsection{Comments on a Possible Local Group Origin}
190:
191: The detection of high column density \ion{O}{7} implies the existence
192: of a large amount of hot gas with a temperature around $10^{6}$ K. In the
193: case of pure collisional ionization, temperature is the only parameter
194: of importance over a wide range of density so long as the gas is
195: optically thin. The \ion{O}{7} ionization fraction peaks at nearly 1,
196: and exceeds 0.5, for temperatures $T \sim (0.5-1.3) \times10^{6}$~K
197: \citep{sdo93}.
198: However, the IGM is also photoionized by the background UV/X-ray
199: radiation, which increases the ionization fraction at low temperatures
200: and densities. We explored these regimes using a grid of CLOUDY models
201: \citep{fer98} for $10^{2} < T < 10^{9}$ K and $10^{-8} < n ({\rm
202: cm^{-3}}) < 10^{-2}$ with 0.2$Z_{\odot}$. We used the UV background
203: from \citet{shu99}, and a power law with an exponential cutoff
204: at 50 keV and $\alpha=0.29$ in the X-ray spectral region
205: (see, e.g. \citealp{fba92}). We combined the ionizing UV and X-ray spectra
206: using
207: normalizations of $J_{13.6\,{\rm eV}} = 1.8\times10^{-23}\ {\rm ergs\
208: cm^{-2}~s^{-1}~Hz^{-1}~sr^{-1}}$ and $J_{1\,{\rm keV}} = 10\ {\rm ph\
209: cm^{-2}~s^{-1}~keV^{-1}~sr^{-1}}$. CLOUDY calculations indicate that the
210: ionization fraction versus $T$ relation begins to deviate appreciably from
211: the pure collisional case for low density regions. At
212: a baryon density $n_{b} \leq 10^{-4}\rm~ cm^{-3}$, photoionized low temperature
213: gas ($ 4 \leq \log T \leq 5.5$) can produce significant amount of
214: \ion{O}{7}. We also considered several other background radiation models,
215: such as the X-ray background from \citet{miy98} and the UV background from
216: \citet{hma96}, but as shown by \citet{che02} we found that
217: the resulting differences in the predicted \ion{O}{7} ionization fraction
218: are negligible.
219:
220: The upper limit on the line width of $\rm \sigma \sim 400~km~s^{-1}$
221: sets an maximum to the path length of $\sim
222: 5$\,h$^{-1}_{70}$\,Mpc, since otherwise the differential Hubble flow
223: would excessively broaden the line. We can further constrain the
224: density of the absorbing gas, assuming it is associated with the Local
225: Group. The path length can be set equal to the distance to the
226: boundary of the Local Group, where the gas begins to participate in
227: the Hubble flow. Assuming a simple spherical geometry for the Local
228: Group, the path length is $\sim 1$ Mpc (see the following
229: text). Adopting the 90\% lower limit of the \ion{O}{7} column density,
230: this gives $n_{b} > (2.2\times10^{-5}\ {\rm cm}^{-3})\
231: Z_{0.2}^{-1}f_{1}^{-1}l_{1}^{-1}$, where $Z_{0.2}$ is the metallicity
232: in units of 0.2 solar abundance, $f_{1}$ is the ionization fraction,
233: and $l_{1}$ is the path length in units of 1 Mpc. This implies a
234: baryon overdensity of $\delta_{b} \sim$ 100, which is roughly
235: consistent with the overdensity of the WHIM gas predicted from
236: cosmological simulations (\citealp{cko95,oce96,cos99,dav01}).
237:
238: A model characterizing the distribution of the Local Group gas is given by the standard
239: $\beta$-model \citep{cfu76}:
240: \begin{equation}
241: n(r) =
242: n_{0}\left[1+\left(\frac{r}{r_{c}}\right)^2\right]^{-\frac{3}{2}\beta}
243: \end{equation}
244: where $n_{0}$ is the central gas number density, $r_{c}$ is the core
245: radius, $\beta$ is the ratio of the specific kinetic energy to
246: thermal energy used to characterize the depth of the potential
247: well. Poor groups like the Local Group typically have shallow
248: potentials with $\beta \sim 0.5$ \citep{mdm96}, so we adopt
249: this value. We adopt a simplified geometry model of the Local Group,
250: where the barycenter is located along the line connecting M31 and the Milky
251: Way in the direction towards $(l,b) \approx (121.7^{\circ},
252: -21.3^{\circ})$ at a distance of about 450 kpc from the Milky Way
253: \citep{rpe01}.
254: At about $r \sim 1200$ kpc, the gravitational
255: contraction of the Local Group starts to dominate the Hubble flow, so this
256: is typically defined as the boundary of the Local Group \citep{cva99}.
257:
258: Based on this simple model, we estimate the column density of
259: \ion{O}{7} along the 3C~273 sight line through the Local Group
260: by integrating the \ion{O}{7} number density $n_{O\,VII}$ out to the
261: boundary. Figure~2 shows the \ion{O}{7} column density ($\rm N_{O\,VII}$)
262: versus baryon density at the center of the Local Group
263: ($n_{0}$). By
264: applying a conservative upper limit of the Local Group gravitational
265: mass and arguing that the cooling timescale should be longer than half the
266: Hubble time, \citet{rpe01} obtained an upper limit of
267: the central density $n_{0} \leq 5\times10^{-4}\ {\rm cm^{-3}}$.
268: From Figure~2 we find a tight upper limit of
269: temperature $T \leq 1.2\times10^6$ K; at temperatures higher than $1.2\times10^6$ K, the
270: \ion{O}{7} ionization fraction drops quickly and \ion{O}{8} starts to
271: dominate. We also find that the temperature of the Local Group should
272: be higher than $2.3\times10^5$ K. To satisfy the observed \ion{O}{7} column
273: density, the gas distribution should have a rather flat
274: core with $r_{c} \geq 100$ kpc.
275:
276: \subsection{Comments on a Hot Halo Gas Origin}
277:
278: Strong \ion{O}{6} absorption along the sight line towards 3C~273 was
279: detected with {\sl FUSE} between -100 and +100 $\rm km\ s^{-1}$
280: \citep{sem01a}. This absorption probably occurs in the
281: interstellar medium of the Milky Way disk and halo. The column
282: density, $\rm \log N(O~VI) = 14.73\pm0.04$, is the second highest
283: O~{\sc vi} column recorded along any sight line through the Milky Way
284: halo (see \citealp{sav02}) . Absorption features of lower ionization species are also
285: present at these velocities. Most notably, the \ion{C}{4}
286: $\lambda\lambda1548.195, 1550.770$ and \ion{N}{5}
287: $\lambda\lambda1238.821, 1242.804$ lines are strong and track
288: the O~{\sc vi} absorption closely. Between -100 and +100 $\rm km\
289: s^{-1}$, N(\ion{O}{6})/N(\ion{C}{4}) and N(\ion{O}{6})/N(\ion{N}{5})
290: are nearly constant, indicating that the species arise in
291: the same gas along the sight line.
292:
293: \ion{O}{6} absorption is also detected between +100 and +240 $\rm km\
294: s^{-1}$ in the form of a broad, shallow absorption wing extending
295: redward of the primary Galactic absorption feature. This \ion{O}{6}
296: absorption wing has $\rm \log N(O~VI) = 13.71$, or $\sim$ 10\% that of
297: the primary absorption feature at lower velocities . This wing is not observed in other species observable with
298: either {\sl HST} or {\sl FUSE}. The \ion{O}{6} absorption wing has
299: been attributed to hot gas flowing out of the Galactic disk as part of
300: a ``Galactic chimney'' or ``fountain'' in the Loop IV and North Polar
301: Spur regions of the sky. Alternatively, the wing might be remnant
302: tidal debris from interactions of the Milky Way and smaller Local
303: Group galaxies \citep{sem01a}. Similar \ion{O}{6}
304: absorption wings are seen in other directions in this general region
305: of the sky \citep{sem02}.
306:
307: It is possible to associate the \ion{O}{7}
308: absorption detected by {\sl Chandra}
309: with these highly ionized metals detected by {\sl
310: FUSE} and {\sl HST}. Here, we briefly discuss several scenarios:
311:
312: {\it (1). The \ion{O}{7} is related to the primary \ion{O}{6}
313: feature.}
314: In this case, $\rm \log$ N(O~{\sc vi}) = 14.73, and $\rm
315: \log$ [N(O~{\sc vii})/N(O~{\sc vi})]$\sim1.5$, assuming N(O~{\sc
316: vii})$\rm = 1.8\times10^{16}\ cm^{-2}$. This is within about a factor of
317: 2 of the O~{\sc vii}/O~{\sc vi} ratio observed for the PKS 2155-304
318: absorber at $z\approx0$ (\citealp{nic02,fan02}) and is consistent with the idea that
319: the gas is radiatively cooling from a high temperature \citep{hns02}.
320: This possibility is appealing since the centroids of the
321: O~{\sc vi} and O~{\sc vii} absorption features are similar ($\sim
322: 6\pm10\ \rm km\ s^{-1}$ versus $-26\pm140\ \rm km\ s^{-1}$) and the
323: width of the resolved O~{\sc vi} line (FWHM $\sim$ 100 $\rm km\
324: s^{-1}$) is consistent with a broad O~{\sc vii} feature.
325:
326: However, this possibility also has drawbacks. First, the amount of \ion{O}{8}
327: predicted by the \citet{hns02} cooling gas model would be $\sim$3
328: times the observed \ion{O}{7} column density. However, we did not detect any
329: \ion{O}{8} Ly$\alpha$ absorption line at the corresponded wavelength
330: in the {\sl Chandra} spectrum. We estimate that the 4$\sigma$ upper
331: limit on the \ion{O}{8} Ly$\alpha$ equivalent width is $\sim 14.20\ \rm
332: m\AA$, or $\sim 10^{16}\ {\rm cm^{-2}}$ if the line is unsaturated. Second, the
333: amount of C~{\sc iv} predicted is a factor of $\sim$10 times less than
334: the O~{\sc vi} column density, which is a factor of 5 lower than
335: observed. It is possible that the C~{\sc iv} column could be
336: increased by including photoionization from the cooling flow itself
337: \citep{bsh00} to bring this ratio closer to the observed
338: value. The predicted N(O~{\sc vi})/N(N~{\sc v}) ratio of $\sim$20 is
339: about a factor of 3 higher than observed. If there are turbulent
340: mixing layers (TMLs)
341: occurring between the hot gas and cooler ISM material (a likely
342: possibility), then the discrepancies in the O~{\sc vi}/C~{\sc iv}
343: ratio might be alleviated since the standard
344: TML models described by \citep{ssb93} predict considerably more C~{\sc iv} (but not more N~{\sc v}) than O~{\sc vi}.
345:
346: {\it (2). The \ion{O}{7} is related to the high velocity \ion{O}{6} wing}.
347: In this
348: case, the \ion{O}{7} is associated only with the \ion{O}{6}
349: absorption ``wing'' and not with the main \ion{O}{6} absorption at
350: lower velocities. Although perhaps less likely than case (1),
351: this may be plausible since the average velocity of the \ion{O}{6} absorption
352: wing ($\sim 150\rm\ km\ s^{-1}$) is still in reasonable agreement with
353: the \ion{O}{7} centroid. In such a situation,
354: $\rm \log$ [N(O~{\sc vii})/N(O~{\sc vi})]$\sim2.5$. In collisional
355: ionization equilibrium, this would imply a temperature of $>10^6$ K
356: (\citealp{smo76,sdo93}). In Figure~3 we
357: show the 90\% limit on N(\ion{O}{7})/N(\ion{O}{6}) (the blue
358: region), which gives a tight constraint on the temperature of the
359: absorber: $6.0 < \log T < 6.9$. A further constraint can be
360: achieved by considering the non-detection of \ion{O}{8} Ly$\alpha$
361: absorption. The 4$\sigma$ upper limit of column density requires
362: $T\lesssim 10^{6.3}$ K; both N(\ion{O}{7})/N({\ion{O}{6}) and
363: N(\ion{O}{8})/N({\ion{O}{7}) give about the
364: same constraint.
365:
366: Another possibility is that
367: the \ion{O}{7} is not related the \ion{O}{6}. In this case,
368: the temperature must be high enough ($> 10^6$ K) to prevent
369: the production of \ion{O}{6}. The upper limit of the temperature can
370: still be obtained by the N(\ion{O}{8})/N({\ion{O}{7}), which gives
371: $T < 10^{6.3}$ K. This case is similar to that described in situation (2).
372:
373: \subsection{Comments on Other Locations}
374:
375: Can other sources contribute to the observed \ion{O}{7} column
376: density? The Local Bubble, a large volume of hot, rarefied gas
377: around the Sun, could be a potential source. The hot Local Bubble
378: gas has a temperature of $\sim10^6$ K \citep{sno98}, and contains highly
379: ionized metal-line species (see \citealp{seim98}).
380: However, given the
381: rather small size of the Local Bubble ($\lesssim 250$ pc), even if
382: the electron density is high ($\sim 0.005\rm\ cm^{-3}$; \citealp{slc99}) and
383: assuming all oxygen ions are in \ion{O}{7}, the overall contribution
384: to the \ion{O}{7} column density is $\lesssim 3\times
385: 10^{15}\rm\ cm^{-3}$, or $\lesssim17$\% of the total \ion{O}{7} column density.
386: Another potential source is the hot gas in the Galactic thick disk / low halo
387: observed in \ion{O}{6} absorption \citep{sav02}.
388: Given the high latitude ($b=64.36^{\circ}$) of 3C~273, the path length
389: through a scale height of this gas is $\sim 2$ kpc pc. Assuming a
390: density of $\sim 0.001\rm\ cm^{-3}$, solar abundances, and unity ionization
391: fraction (an upper bound), the overall contribution to the \ion{O}{7} column
392: density is still small ($\lesssim 30\%$) compared to the value observed.
393:
394: The sight line to 3C~273 extends into the Galactic halo through the edges of Radio Loop I and IV, which have been attributed to supernova remnants (see, e.g., \citealp{eas95}). X-ray observations in this direction reveal a hot, X-ray emitting gas with a temperature of T $\sim 3\times 10^6$ K and emission measure ($n_e^2d$) $\sim (1-2) \times 10^{-2}\ \rm cm^{-6}pc$, where $d$ is the distance through Loops I and IV \citep{iwa80}. This gives a rough estimate of electron density $n_e \approx 0.01\ \rm cm^{-3}$. Using a path length of $< 350$ pc through Loops I and IV \citep{ber73} and an \ion{O}{7} ionization fraction of 0.01 at $\rm T\sim 3\times 10^6$ K in collisional ionization equilibrium, the estimated column density for solar-abundance gas is $\lesssim 8\times10^{15}\ \rm cm^{-2}$, which could be as much as $\sim50$\% of the observed \ion{O}{7} column density if the loops really are this large. (Iwan 1980 estimates a size
395: $\lesssim 100$ pc.) However, since the \ion{O}{7} ionization fraction raises quickly to 1 as the temperature approaches $10^6$ K, a small uncertainty on $T$ could cause a large variation in the predicted column density.
396:
397: \section{Summary}
398:
399: Our {\sl Chandra} observations indicate that a large amount of hot gas exists
400: in the Local Group and/or in the Milky Way. The \ion{O}{7} column density in
401: this direction is a factor of 2-2.5 higher than toward PKS~2155-304 (\citealp{nic02,fan02}) and a factor
402: of 2-5 higher than toward H\,1821+643 \citep{mwc02}, which combined with the
403: strong X-ray emission from the North Polar Spur in the direction of
404: 3C~273, suggests that the absorption we see may well be from a combination
405: of Galactic and Local Group sources. We constrain the temperature to $\rm 2.3\times 10^5 < T < 1.2\times 10^6$ K and find the baryon overdensity $\delta_b$ is $\sim 100$, assuming a standard $\beta$ model for the intragroup medium. The derived properties are
406: consistent with those of the WHIM gas which are predicted from simulations.
407: For the case of Galactic absorption, we have outlined several scenarios to
408: explain the observed \ion{O}{7} absorption, which may be associated with
409: \ion{O}{6} observed at UV wavelengths. We find that in both the Galactic or
410: Local Group cases there is a
411: strong indication that collisional ionization is the dominant ionization
412: source, at least for the gas which produces \ion{O}{7} absorption.
413: We expect future X-ray observations may provide more information on the
414: distribution of \ion{O}{7} at $z \sim 0$ and its strength in different
415: directions. Such observations would provide valuable insight needed
416: to determine the the location of the hot gas.
417:
418: We obtained the data from the {\sl Chandra} Data Archive built by the {\sl Chandra} X-ray Center. We also thank the members of the MIT/CXC team for their support. This work is supported in part by contracts NAS 8-38249 and SAO SV1-61010.
419: KRS acknowledges financial support through NASA contract NAS5-32985
420: and Long Term Space Astrophysics grant NAG5-3485.
421:
422: [Note: Recently, we learned from a conference presentation that \citet{ras02} reported the detection of a similar feature in the XMM spectrum of 3C 273, which confirms our detection.]
423:
424:
425:
426: \clearpage
427: \begin{thebibliography}{}
428: \bibitem[Aldcroft et al.(1994)]{aem94} Aldcroft,~T., Elvis,~M.,
429: McDowell,~J \& Fiore,~F. 94, \apj, 437, 584
430: \bibitem[Benjamin \& Shapiro(2000)]{bsh00} Benjamin, R.~A. \& Shapiro, P.~R.\ 2000, \apjs, submitted
431: \bibitem[Berkhuijsen(1973)]{ber73} Berkhuijsen, E.~M.\ 1973,
432: \aap, 24, 143
433: \bibitem[Cavaliere \& Fusco-Femiano(1976)]{cfu76} Cavaliere,
434: A.~\& Fusco-Femiano, R.\ 1976, \aap, 49, 137
435: \bibitem[Cen et al.(1995)]{cko95} Cen, R.,
436: Kang, H., Ostriker, J.~P., \& Ryu, D.\ 1995, \apj, 451, 436
437: \bibitem[Cen \& Ostriker(1999)]{cos99} Cen, R. \&
438: Ostriker,~J.~P. 1999, \apj, 514, 1
439: \bibitem[Chen et al.(2002)]{che02} Chen, ~X., Weinberg, ~D.~H., Katz, ~N. \& Dav\'{e}, ~R.~2002, \apj, submitted (astro-ph/0203319)
440: \bibitem[Courteau \& van den Bergh(1999)]{cva99} Courteau,
441: S.~\& van den Bergh, S.\ 1999, \aj, 118, 337
442: \bibitem[Dav{\' e} et al.(2001)]{dav01} Dav{\' e}, R.~et al.\
443: 2001, \apj, 552, 473
444: \bibitem[Egger \& Aschenbach(1995)]{eas95} Egger, R.~J.~\&
445: Aschenbach, B.\ 1995, \aap, 294, L25
446: \bibitem[Fabian \& Barcons(1992)]{fba92} Fabian, A.~C.~\& Barcons, X.\ 1992, \araa, 30, 429
447: \bibitem[Fang \& Canizares(2000)]{fca00} Fang, T.~\& Canizares, C.~R.\
448: 2000, \apj, 539, 532
449: \bibitem[Fang et al.(2002)]{fan02} Fang, T., Marshall, H.~L.,
450: Lee, J.~C., Davis, D.~S., \& Canizares, C.~R.\ 2002, \apjl, 572, L127
451: \bibitem[Ferland et al.(1998)]{fer98} Ferland, G.~J.,
452: Korista, K.~T., Verner, D.~A., Ferguson, J.~W., Kingdon, J.~B., \& Verner,
453: E.~M.\ 1998, \pasp, 110, 761
454: \bibitem[Fukugita, Hogan, \& Peebles(1998)]{fhp98} Fukugita, M., Hogan,
455: C.~J. \& Peebles, J.~P.~E. 1998, \apj, 503, 518
456: \bibitem[Haardt \& Madau(1996)]{hma96} Haardt, F.~\& Madau,
457: P.\ 1996, \apj, 461, 20
458: \bibitem[Heckman et al.(2002)]{hns02} Heckman, T.~M., Norman, C.~A., Strickland, D.~K., \& Sembach, K.~R.~2002, \apj, submitted (astro-ph/0205556)
459: \bibitem[Hellsten, Gnedin, \& Miralda-Escud\'{e}(1998)]{hgm98}
460: Hellsten,~U., Gnedin, N.~Y. \& Miralda-Escud\'{e}, J. 1998, \apj,
461: 509, 56
462: \bibitem[Houck \& Denicola(2000)]{hde00} Houck, J.~C.~\& Denicola, L.~A.\ 2000, ASP Conf.~Ser.~216: Astronomical Data Analysis
463: Software and Systems IX, 9, 591
464: \bibitem[Iwan(1980)]{iwa80} Iwan, D.\ 1980, \apj, 239, 316
465: \bibitem[Kaspi et al.(2002)]{kas02} Kaspi, S.~et al.\ 2002,
466: \apj, 574, 643
467: \bibitem[Lockman \& Savage(1995)]{lsa95} Lockman, F.~J.~\& Savage, B.~D.\ 1995, \apjs, 97, 1
468: \bibitem[Mathur, Weinberg, \& Chen(2002)]{mwc02} Mathur, S., Weinberg, D., \& Chen, X.\ 2002, \apj, submitted (astro-ph/0206121)
469: \bibitem[Miyaji et al.(1998)]{miy98} Miyaji, T., Ishisaki,
470: Y., Ogasaka, Y., Ueda, Y., Freyberg, M.~J., Hasinger, G., \& Tanaka, Y.\
471: 1998, \aap, 334, L13
472: \bibitem[Mulchaey et al.(1996)]{mdm96} Mulchaey, J.~S., Davis, D.~S., Mushotzky, R.~F., \& Burstein, D.\ 1996, \apj, 456, 80
473: \bibitem[Nicastro et al.(2002)]{nic02} Nicastro, F.~et al.\
474: 2002, \apj, 573, 157
475: \bibitem[Nicastro et al.(2002)]{nze02} Nicastro, F.~et al.\ 2002, \nat, submitted (astro-ph/0208012)
476: \bibitem[Ostriker \& Cen(1996)]{oce96} Ostriker, J.~P.~\& Cen, R.\ 1996, \apj, 464, 27
477: \bibitem[Perna \& Loeb(1998)]{plo98} Perna, P. \& Loeb, A. 1998,
478: \apjl, 503, L135
479: \bibitem[Rasmussen \& Pedersen(2001)]{rpe01} Rasmussen, J.~\& Pedersen, K.\ 2001, \apj, 559, 892
480: \bibitem[Rasmussen et al.(2002)]{ras02} Rasmussen, A.\ et al.\ 2002, presentation in ``the IGM/Galaxy Connection - The Distribution of Baryons at z=0'', Colorado
481: \bibitem[Sanders et al. (1998)]{seim98}Sanders, W.T., Edgar, R.J., Kraushaar, W. L., McCammon, D.,
482: \& Morgenthaler, J.P. 2001, \apj, 554, 694
483: \bibitem[Sarazin(1989)]{sar89} Sarazin,~C.~L. 1989, \apj, 345, 12
484: \bibitem[Savage et al.(2000)]{sav00} Savage, B.~D.~et al.\
485: 2000, \apjl, 538, L27
486: \bibitem[Sembach et al.(2000)]{sem00} Sembach, K.~R.~et al.\
487: 2000, \apjl, 538, L31
488: \bibitem[Savage et al.(2002)]{sav02} Savage, B.D., et al. 2002, \apjs, submitted (astro-ph/0208140)
489: \bibitem[Sembach(2001)]{sem01} Sembach, K.~R.\ 2001, Extragalactic Gas at Low Redshift, eds. Mulchaey \& Stocke, ASP Conference Series, in press (astro-ph/0108088)
490: \bibitem[Sembach et al.(2001)]{sem01a} Sembach, K.~R., Howk,
491: J.~C., Savage, B.~D., Shull, J.~M., \& Oegerle, W.~R.\ 2001, \apj, 561, 573
492: \bibitem[Sembach et al.(2002)]{sem02} Sembach, K.~R.~et al. 2002, \apjs, submitted (astro-ph/0207562)
493: \bibitem[Sfeir et al.(1999)]{slc99}
494: Sfeir, D.~M., Lallement, R., Crifo, F., \& Welsh, B.~Y.\ 1999, \aap, 346,
495: 785
496: \bibitem[Shapiro \& Bahcall(1980)]{sba80} Shapiro, P.~R. \& Bahcall,
497: J.~N. 1980, \apj, 241, 1
498: \bibitem[Shapiro \& Moore(1976)]{smo76} Shapiro, P.~R.~\& Moore, R.~T.\ 1976, \apj, 207, 460
499: \bibitem[Shull et al.(1999)]{shu99} Shull, J.~M., Roberts,
500: D., Giroux, M.~L., Penton, S.~V., \& Fardal, M.~A.\ 1999, \aj, 118, 1450
501: \bibitem[Slavin, Shull, \& Begelman(1993)]{ssb93} Slavin,
502: J.~D., Shull, J.~M., \& Begelman, M.~C.\ 1993, \apj, 407, 83
503: \bibitem[Snowden et al.(1998)]{sno98} Snowden, S.~L., Egger,
504: R., Finkbeiner, D.~P., Freyberg, M.~J., \& Plucinsky, P.~P.\ 1998, \apj,
505: 493, 715
506: \bibitem[Spitzer(1978)]{spi78} Spitzer, L.~1978, Physical Processes in the Interstellar Medium (Princeton: Princeton University Press)
507: \bibitem[Sutherland \& Dopita(1993)]{sdo93} Sutherland,
508: R.~S.~\& Dopita, M.~A.\ 1993, \apjs, 88, 253
509: \bibitem[Tripp, Savage, \& Jenkins(2000)]{tsj00} Tripp,
510: T.~M., Savage, B.~D., \& Jenkins, E.~B.\ 2000, \apjl, 534, L1
511: \bibitem[Verner, Verner, \& Ferland(1996)]{vvf96} Verner,
512: D.~A., Verner, E.~M., \& Ferland, G.~J.\ 1996, Atomic Data and Nuclear Data
513: Tables, 64, 1
514: \bibitem[Wakker et al.(2002)]{wak02} Wakker, ~B.~P.~et al. 2002, \apjs, submitted (astro-ph/0208009)
515: \end{thebibliography}{}
516:
517: \clearpage
518: \vbox{
519: \begin{center}
520:
521: \begin{tabular}{llcc}
522: \multicolumn{2}{c}{~~~~~~~~Table 1: Line Fitting Parameters~~~~~~} \\
523: \hline \hline
524: & O~{\sc vii} & O~{\sc vii} & O~{\sc viii}\\
525: & He$\alpha$ & He$\beta$ & Ly$\alpha$ \\ \hline
526: Wavelength (\AA) & $21.60\pm0.01$ & ... & ... \\
527: Rest Wavelength (\AA)$^{a}$ & 21.6019 & 18.6288 & 18.9689 \\
528: $cz$ (km~s$^{-1}$) & $-26 \pm 140$ & ... & ... \\
529: Line Width (\AA)$^{b}$ & $<0.02$ & ... & ... \\
530: $\rm Line\ Flux^{c}$ & $4.15_{-0.90}^{+1.83}$ & ... & ... \\
531: $W_{\lambda}$ (m${\rm \AA}$) & $28.4_{-6.2}^{+12.5}$ & $<11.9^d$ & $<14.2^d$\\
532: S/N & 6.4 & ... & ... \\ \hline
533: \end{tabular}
534:
535: \parbox{3.5in}{
536: \vspace{0.1in} \small\baselineskip 9pt \footnotesize \indent
537: a. \citet{vvf96}.\\ b. 90\% upper limit of the line width $\sigma$.\\
538: c. Absorbed line flux, in units of $\rm
539: 10^{-5}~photons~cm^{-2}s^{-1}$.\\ d. 4-$\sigma$ detection upper limit.
540: }
541: \end{center}
542: \normalsize \centerline{} }
543:
544: \clearpage
545: \figcaption[f1.eps]{The {\sl Chandra} LETG-ACIS spectrum of 3C273 between 19
546: and 23 $\rm\AA$. The red solid line is the fitted continuum (Galactic
547: absorbed power law plus the polynomial) plus the Gaussian model. The
548: average $1\sigma$ error bar plotted on the right is based on
549: statistics only. The inset shows the wavelength region where the rest-frame
550: \ion{O}{8} Ly$\alpha$ and \ion{O}{7} He$\beta$ are located. \label{fig1}}
551:
552: \figcaption[f2.eps]{O~{\sc vii} column density versus central baryon density.
553: The four different colors denote four different Local Group temperatures
554: [$T = 2.3\time10^5$ K (black), $5.8\times10^5$ K (red), $1.2\times10^6$ K (green) and $3.5\times10^6$ K (blue)].
555: The shadowed area indicates the 90\% confidence range of
556: the detected \ion{O}{7} column density. The two vertical dashed lines
557: indicate the central density lower limit (from path length constraints)
558: and upper limit (from $\rm N_{O\,VII}$).\label{fig2}}
559:
560: \figcaption[f3.eps]{The column density of \ion{O}{7} versus \ion{O}{6} (solid
561: line) and \ion{O}{8} versus \ion{O}{6} (dashed line). The blue
562: region indicates the allowed range from the measured
563: N(\ion{O}{7})/N(\ion{O}{6}) ratio. The red line is the lower limit of the
564: N(\ion{O}{8})/N(\ion{O}{6}) from the non-detection of
565: \ion{O}{8} in the {\sl Chandra} spectra. \label{fig3}}
566:
567: \clearpage
568: \begin{figure}
569: \epsscale{0.8}
570: \plotone{f1.eps}
571: \end{figure}
572:
573: \clearpage
574: \begin{figure}
575: \epsscale{0.8}
576: \plotone{f2.eps}
577: \end{figure}
578:
579: \clearpage
580: \begin{figure}
581: \epsscale{1.0}
582: \plotone{f3.eps}
583: \end{figure}
584:
585: \end{document}
586:
587: