astro-ph0511621/ms.tex
1: %%
2: %% This is the file `XrU2005.tex'
3: %%
4: %%
5: %% ====================================================
6: %% It describes use of the LaTeX class file XrU2005.cls
7: %% ====================================================
8: %% -------------------------------------------------
9: 
10: % single-column figure:
11: % \begin{figure}
12: % \centering
13: % \includegraphics[width=0.8\linewidth]{fig.eps}
14: % \caption{blah blah blah \label{fig_blah}}
15: % \end{figure}
16: %
17: % double-column: use figure* instead of figure environment
18: % -- can also use epsfig command, check chandra/xmm proposal for details
19: %
20: 
21: 
22: \documentclass{XrU2005}
23: 
24: \title{Galactic Corona or Local Group Intergalactic Medium?}
25: \author{Rik J.~Williams}
26: \author{Smita Mathur}
27: \affil{Department of Astronomy, The Ohio State University, 140 W. 18th Ave., Columbus, OH 43210 USA}
28: \author{Fabrizio Nicastro}
29: \affil{Harvard--Smithsonian Center for Astrophysics, Cambridge, MA  USA}
30: 
31: \newcommand{\kms}{\,km\,s$^{-1}$}
32: \newcommand{\btx}{\textsc{Bib}\TeX}
33: \newcommand{\filename}{ms}
34: \newcommand{\chandra}{\emph{Chandra}}
35: \newcommand{\fuse}{\emph{FUSE}}
36: \newcommand{\hi}{H\,{\sc i}}
37: \newcommand{\novii}{N_{\rm OVII}}
38: \newcommand{\novi}{N_{\rm OVI}}
39: \newcommand{\noviii}{N_{\rm OVIII}}
40: \newcommand{\ovilv}{O\,{\sc vi}$_{\rm LV}$}
41: \newcommand{\ovi}{O\,{\sc vi}}
42: \newcommand{\ovii}{O\,{\sc vii}}
43: \newcommand{\oviii}{O\,{\sc viii}}
44: \newcommand{\nvii}{N\,{\sc vii}}
45: \newcommand{\neix}{Ne\,{\sc ix}}
46: 
47: \usepackage{graphicx}
48: 
49: \begin{document}
50: 
51: \keywords{Intergalactic Medium; Local Group; Galactic Halo}
52: 
53: \maketitle
54: 
55: \begin{abstract}
56: Cosmological hydrodynamic simulations predict that the low redshift universe 
57: comprises a web of warm--hot intergalactic gas and galaxies, with groups of 
58: galaxies and clusters forming at dense  knots in these filaments. Our own 
59: Galaxy 
60: being no exception is also expected to be  surrounded by the warm--hot 
61: intergalactic medium, filling the Local Group. Some theoretical  models 
62: also predict the existence of a hot Galactic corona.  With X-ray and 
63: FUV  observations of extragalactic sources, we can probe the warm--hot 
64: gas through absorption lines  of highly ionized elements. Indeed, \chandra, 
65: \emph{XMM} and \fuse\ observations have detected $z=0$ absorption lines toward 
66: many sightlines. The debate that has emerged is over the  interpretation 
67: of these observations: are the $z=0$ absorption systems from the halo of our  
68: Galaxy or from the extended Local Group environment? This has important 
69: implications for  our understanding of the mass of the Local Group, the 
70: physical conditions in the intergalactic  medium, the structure of the 
71: Galaxy and galaxy formation in general. We will present the  current status 
72: of the debate and discuss our ongoing observing program aimed at  
73: understanding the $z=0$ absorption systems, with an emphasis on the 
74: high quality \chandra\ spectra of the Mrk~421 and Mrk~279 sightlines.   
75: \end{abstract}
76: 
77: \section{Introduction}
78: The intergalactic medium (IGM) is expected to contain most of the baryonic
79: matter in the universe, a tenuous filamentary ``web'' of gas bridging
80: the gaps between collapsed objects such as galaxies and clusters.
81: At high redshifts ($z>2$) this web appears in quasar spectra as a multitude
82: of Lyman alpha forest absorption lines.  In the nearby universe, on the
83: other hand, hydrodynamic simulations show that most of the IGM has
84: been shock--heated to a warm--hot (WHIM) phase with temperatures of
85: $\sim 10^6$\,K \citep{cen99,dave01}.  At these temperatures scant neutral 
86: hydrogen
87: remains, and the IGM is thus best detected through absorption lines
88: from highly ionized metals \citep{hellsten98}, particularly \ovi, \ovii, and
89: \oviii.  Recent \chandra\ observations have indeed confirmed that this 
90: low--redshift WHIM exists and comprises a baryon content consistent with 
91: expectations \citep[Nicastro et al., in preparation][]{nicastro05a,nicastro05b}.
92: 
93: Just as other galaxies are expected to form in the densest ``knots''
94: of the cosmic web, we also expect to see WHIM adjacent to, perhaps 
95: surrounding, our own Milky Way.   Indeed, X-ray spectra of several quasars
96: show likely $z=0$ \ovii\ absorption.  The upper limit on the \ovii\ 
97: emission toward Mrk 421 found by \citet{rasmussen03} indicates that this
98: absorption system probably has an extremely low density and is thus
99: likely to be extragalactic.  Additionally, other nearby low-- and 
100: high--ionization components may be associated with either the WHIM itself
101: or may represent gas from the WHIM that has cooled and is now in the
102: process of accreting onto the Galaxy; for example, the high--velocity
103: \ovi\ absorption seen with \fuse\ along many quasar lines of sight
104: \citep{wakker03} and neutral hydrogen high--velocity clouds (\hi\ HVCs).
105: 
106: \section{The Debate: WHIM or Corona?}
107: The origin of the local \ovii\ absorption, and in particular its relation
108: to the observed \ovi, is still to a large degree unknown.  There is some
109: evidence that these ions could originate in a warm--hot Galactic corona:
110: for example, likely \ovii\ absorption has been detected within 50
111: kpc of the Galaxy by \citet{wang05}.  The observed deflection and
112: stripping of the Magellanic clouds also lends credence to the existence
113: of a low--density corona.  Lower--ionization absorption, such as that
114: from Si\,{\sc iv} and C\,{\sc iv}, is also seen at the same velocities
115: as some \ovi\ HVCs, indicating that these \ovi\ clouds, at least, may
116: have lower temperatures and higher densities than expected from the WHIM.
117: 
118: However, there are also reasons to believe this absorption is tracing
119: extended WHIM gas.  High column densities of \ovi, \ovii, and \oviii\ 
120: from the local IGM are predicted by simulations to lie in certain 
121: directions \citep{kravtsov02}.  Furthermore, the mean velocity vector of the
122: \ovi\ HVCs is highest in the local standard of rest and lowest in the
123: Local Group rest frame, indicating that their origin could indeed
124: be extragalactic \citep{nicastro03}.  The presence of \ovii\ between
125: the Galaxy and Large Magellanic Cloud does not rule out an extragalactic
126: origin for the absorption in some directions since neither the Galactic
127: absorption or WHIM is necessarily homogeneous; additionally, the WHIM
128: is known to be homogeneous and consist of a variety of temperature and
129: density phases, so some lower--ionization lines may be expected as well.
130: 
131: Thus, the questions -- how are the local X-ray and UV absorption
132: components related to each other, which are of Galactic origin, and
133: which arise in the local WHIM?-- are still unanswered.  
134: The answers to these questions have profound implications for both
135: studies of galaxy formation and cosmology.
136: We are now undertaking a program to determine the distribution and
137: properties of this local hot gas and its ties to lower--ionization
138: components, with a focus on new and archival \chandra\ and \fuse\ data. 
139: Here we present the first results of this study, an analysis of the
140: particularly high--quality \chandra\ and \fuse\ spectra of the 
141: bright AGN Mrk~421 and Mrk~279.
142: 
143: \section{The Mrk~421 Sightline}
144: A full discussion of the Mrk~421 \chandra\ and FUSE spectra, and the
145: analysis thereof, can be found in \citet{williams05}; the following is
146: a summary of the main results.
147: 
148: \subsection{Observations and measurements}
149: The bright $z=0.03$ blazar 
150: Mkn~421 was observed during two exceptionally high
151: outburst phases for 100 ks each as part of our \chandra --AO4
152: observing program:
153: one at $f_{\rm 0.5-2 keV}=1.2\times 10^{-9}$\,erg\,s$^{-1}$\,cm$^{-2}$
154: with the Low Energy Transmission Grating (LETG) combined with the
155: ACIS-S array, and another
156: at $f_{\rm 0.5-2 keV}=0.8\times 10^{-9}$\,erg\,s$^{-1}$\,cm$^{-2}$
157: with the HRC-S array and LETG.  Each of these observations
158: contains $\sim 2500$ counts per resolution element at 21.6\,\AA.  Additionally,
159: another short observation of Mkn~421 was taken with HRC/LETG
160: (29 May 2004), providing another 170 counts per resolution element.  These
161: three spectra were combined over the 10--60\,\AA\ range to improve the
162: signal--to--noise ratio (S/N$\sim 55$ at 21\,\AA\ with 0.0125\,\AA\ binning).
163: The final coadded spectrum of Mkn~421 is one of the best ever taken
164: with \chandra: it contains
165: over $10^6$ total counts with $\sim 6000$ counts per resolution element at
166: 21.6\,\AA, providing a $3\sigma$ detection threshold of
167: $W_\lambda\sim 2$\,m\AA\
168: ($N_{\rm OVII}=8\times 10^{14}\rm{cm}^{-2}$ for an unsaturated line).
169: 
170: Using the CIAO fitting package Sherpa we initially modeled the continuum
171: of Mkn~421 as a simple power law with Galactic foreground absorption,
172: excluding the 48--57\,\AA\ HRC chip gap region.  Metal abundances for
173: the Galactic gas were then artificially adjusted to provide a better fit
174: around the \ion{O}{1} and \ion{C}{1} K--edges near 23\,\AA\ and 43\,\AA\
175: respectively.  This is \emph{not} intended to represent actual
176: changes to the absorber composition, but rather to correct uncertainties
177: in the instrument calibration.  
178: After this fit there were still some systematic uncertainties in
179: the best--fit continuum model; these were corrected with
180: broad (${\rm FWHM}=0.15-5$\,\AA) Gaussian emission and absorption components
181: until the modeled continuum appeared to match the data upon inspection.
182: Indeed, the residuals of the spectrum to the final continuum model have a
183: nearly Gaussian distribution, with a negative tail indicating the presence
184: of narrow absorption lines \citep[see][Figure 8]{nicastro05a}.
185: We searched for narrow, unresolved (FWHM$<0.05$\,\AA) absorption lines 
186: at known C, N, O, and Ne transition wavelengths and used Gaussians
187: to measure the equivalent widths (or upper limits thereupon) for all
188: lines found.  All in all, equivalent widths for 9 significantly detected
189: absorption lines were measured (including
190: \ovii\ K$\alpha$, $\beta$, and $\gamma$) and 4 upper limits calculated.
191: 
192: Mrk~421 was also observed for a total of 84.6\,ks with FUSE, providing
193: a signal--to--noise ratio of 17 near the \ovi\ wavelength once
194: all spectra are combined and binned to $\sim 10$\kms.  This spectrum
195: shows strong, broad \ovi\ absorption at $v \sim 0$, most likely originating
196: in the Galactic thick disk, as well as a possible \ovi\ HVC at
197: $v\sim 110$\kms\ (Figure~\ref{fig_mkn421fuse}).
198: \begin{figure}
199: \centering
200: \includegraphics[width=0.8\linewidth]{mkn421_fuse.eps}
201: \caption{FUSE spectrum of Mrk~421 near the \ovi\ doublet.  The 1032\,\AA\ 
202: line is well--fit by a strong absorber at $v\sim 0$\kms\ and a much weaker
203: component at $v\sim 100$\kms\ (inset).\label{fig_mkn421fuse}}
204: \end{figure}
205: 
206: \subsection{Doppler parameters}
207: To convert the measured equivalent widths to ionic column densities,
208: we calculated curves of growth for each absorption line
209: over a grid of Doppler parameters ($b=10-100$\kms) and column densities
210: ($\log N_H/{\rm cm}^{-2} = 12.0-18.0$), assuming a Voigt line profile.  
211: Since the X-ray lines are unresolved, $b$ cannot be measured directly.
212: It can, however, be inferred from the relative strengths of the
213: three measured \ovii\ K--series lines.  
214: These line ratios by themselves are insufficient to determine the physical
215: state of the \ovii--absorbing medium since $b$ and $\novii$ are
216: degenerate: the K$\alpha$ line saturation could be due to high column
217: density, low $b$, or a combination of both.  However, given an absorption 
218: line with a measured equivalent width and known
219: oscillator strength, the inferred column density as a function of the
220: Doppler parameter can be calculated.  The measured equivalent width 
221: (and errors) for
222: each transition thus defines a region in the $\novii-b$ plane.
223: Since the actual value of $\novii$ is fixed, $b$ and $\novii$ can be
224: determined by the region over which the contours ``overlap;''  i.e.
225: the range of Doppler parameters for which the different transitions
226: provide consistent $\novii$ measurements.
227: 
228: \begin{figure}
229: \centering
230: \includegraphics[width=0.8\linewidth]{mkn421_nhb.eps}
231: \includegraphics[width=0.8\linewidth]{mkn421_ovi.eps}
232: \caption{Mrk 421 column density and Doppler parameter diagnostics for 
233: the \ovii\ (top) and \ovi\ (bottom) absorption, with each transition
234: labeled.  In the \ovi\ panel, the ``a'' and ``b'' curves are derived from
235: the 1032\,\AA\ absorption equivalent width and FWHM, respectively. 
236: \label{fig_nb_mkn421}}
237: \end{figure}
238: 
239: Figure~\ref{fig_nb_mkn421} (top panel) shows such $1\sigma$ contours for the 
240: three measured
241: \ovii\ transitions.  The K$\alpha$ and K$\beta$ tracks appear consistent
242: at the $2\sigma$ level for $13<b<55$\kms,
243: while the overlap between the K$\alpha$ and K$\gamma$ tracks provides
244: approximate 2$\sigma$ limits of $24<b<76$\kms.  
245: We thus assume a $2\sigma$ range
246: of $24<b<55$\kms.  It should be noted that Figure~\ref{fig_nb_mkn421} also
247: shows some overlap between the $K_\alpha$ and $K_\gamma$ at $b\le 12$\kms; 
248: however, this solution is unlikely given the 
249: lower limit provided by the $K_\beta$ line.  Moreover, $b=12$\kms\ implies
250: a maximum temperature (assuming purely thermal motion) of 
251: $T_{\rm max}=1.3\times 10^5$\,K; such a low temperature is unlikely to 
252: produce the observed strong high--ionization lines.
253: 
254: A similar analysis is not as effective when applied to the strong \ovilv\ 
255: UV doublet (from the thick disk), since these lines are only slightly 
256: saturated.  The \ovilv\ 1032\,\AA\ line is fully 
257: resolved by \fuse\ 
258: and relatively unblended, so its Doppler parameter can be estimated much more
259: accurately using the measured line width and strength.  In an unsaturated
260: absorption line, ${\rm FWHM}=2(\ln 2)^{1/2}b$; however, the measured FWHM
261: increases if the line is saturated.  We compensated for this by 
262: calculating Voigt profile FWHMs on a grid of $\novi$ and $b$,
263: and determining the region consistent with the \ovilv\ 1032\,\AA\ FWHM 
264: measurement of $152\pm 7$\kms.
265: 
266: When the FWHM--derived contour is overlaid on the $\novi-b$ contour inferred
267: from the equivalent width measurement of the LV--\ovi\ 1032\,\AA\ line, 
268: the two regions overlap nearly orthogonally 
269: (Figure~\ref{fig_nb_mkn421}, bottom panel) leading to a constraint of 
270: $b$(\ovilv)$=80.6\pm 4.2$\kms.  This is more than $3\sigma$ higher than
271: the Doppler parameter calculated for the \ovii\ absorption, indicating
272: that the \ovii\ and thick--disk \ovilv\ cannot arise in the same gaseous
273: phase.  Also, at no value
274: of the Doppler parameter do the 1032\,\AA, 1037\,\AA, and \ovi\ K$\alpha$
275: lines all produce a consistent $\novi$ measurement; in fact, the
276: \ovi K$\alpha$ column density is a factor of $\sim 4$ higher than that
277: inferred from the UV data.  It is thus possible that the \ovi\ UV
278: transition is being suppressed, perhaps by atomic physics effects, in the 
279: absorbing medium and the
280: $K\alpha$ line produces a more accurate representation of the true
281: \ovi\ column density; in the following analysis we consider both
282: possibilities.
283: 
284: \subsection{Temperature and Density Constraints}
285: The abundance ratios of metal ions (for example, \ovii/\oviii) are
286: expected to vary with temperature as a result of collisional ionization;
287: additionally, as the density decreases to typical WHIM values
288: ($\sim 10^{-4}$\,cm$^{-3}$), photoionization from the extragalactic
289: ionizing background plays an increasingly important role.  We used Cloudy
290: version 90.04 \citep{ferland96} to calculate relative ionic abundances
291: for all measured elements over a grid of $\log T$ and $\log n_e$.
292: As with the
293: Doppler parameter and column density diagnostics described above,
294: any measured ion column density ratio then produces a ``track'' of consistency
295: in the $\log T - \log n_e$ plane, and (assuming the ions arise in the 
296: same gaseous phase) the overlap between such ``tracks'' can place constraints
297: on the gas temperature and density.
298: 
299: Column density ratios between different ions of the same element 
300: (e.g. \ovi, \ovii, and \oviii) produce the
301: strongest constraints since these ratios are independent of the relative
302: metal abundance in the gas  Figure~\ref{fig_mkn421_tne} shows such constraints
303: derived from these oxygen abundance ratios.  If the \ovi\ K$\alpha$ line
304: accurately traces the \ovi\ column density of this medium, then the
305: contours overlap at $n_e\sim 10^{-4}$\,cm$^{-3}$. Coupled with
306: the total oxygen column density and assuming O/H of 0.3 times Solar, 
307: this implies that the absorber has a radial extent of $\sim 1$\,Mpc
308: and a mass consistent with the expected baryonic mass of the Local 
309: Group.  
310: 
311: \begin{figure}
312: \centering
313: \includegraphics[width=0.8\linewidth]{oxplot.eps}
314: \caption{$2\sigma$ constraints on the gas temperature and density toward 
315: Mrk~421,
316: derived from oxygen ion ratios as labeled.  The dashed line indicates the
317: upper limit on the temperature from the \ovii\ Doppler parameter.
318: \label{fig_mkn421_tne}}
319: \end{figure}
320: 
321: On the other hand, if the \ovi\ K$\alpha$ line does \emph{not} correctly
322: measure $\novi$, then this absorption can in principle arise in a 
323: higher--density Galactic medium with $\log T\sim 6.1$ (as derived from
324: the \oviii/\ovii\ ratio with collisional ionization as the dominant 
325: process; Figure~\ref{fig_mkn421_tne}).  Even in this case, a low--density
326: (extragalactic) medium is fully consistent with the data.  Furthermore,
327: the contour derived from the measured Ne/O abundance (not shown in the
328: figure) is consistent in the low--density regime if the Ne/O ratio
329: is significantly supersolar, as has been observed in other 
330: Galactic and extragalactic absorption systems.
331: 
332: 
333: \section{The Mrk~279 Sightline}
334: A full discussion of our analysis of the Mrk~279 \chandra\ and \fuse\
335: spectra will appear in a forthcoming paper \citep{williams06}.
336: 
337: \subsection{Observations and Measurements}
338: While Mrk~421 was observed for relatively short periods during bright
339: outburst phases, few other bright sources flare this dramatically.
340: We thus searched the \chandra\ archive for long--duration LETG observations
341: of relatively bright background quasars.  One such source, Mrk~279
342: (an AGN at $z=0.03$) was observed during seven periods in May 2003
343: for a total exposure time of 340 ks.
344: These seven observations were coadded for a final (unbinned)
345: signal--to--noise ratio of ${\rm S/N}\sim 6.5$ near 22\AA.
346: As with the Mrk~421 spectrum, we again used Sherpa to fit a power law and
347: foreground Galactic absorption to the spectrum over $10-100$~\AA\ band 
348: (excluding the $49-57.5$~\AA\ and $60.5-67.5$~\AA\ chip gap regions), leaving
349: the relative Galactic metal abundances 
350: as free parameters in order to produce a better fit around the absorption
351: edges.  The remaining broad residuals were
352: were corrected by including four broad Gaussians in the source model. 
353: 
354: \begin{figure}
355: \centering
356: \includegraphics[width=0.8\linewidth]{mkn279vel.eps}
357: \caption{Velocity plot of the \ovii\ and \ovi\ absorption seen in the
358: Mrk~279 \chandra\ and \fuse\ spectra respectively.  The centroids of the
359: \ovii\ and high--velocity \ovi\ lines differ by $\sim 2.5\sigma$.
360: \label{fig_mkn279_spec}}
361: \end{figure}
362: 
363: Although several strong lines such as C\,{\sc vi}, \ovii, and \nvii\ 
364: are apparent at the blazar redshift ($z=0.03$), only \ovii~K$\alpha$ 
365: $\lambda 21.602$ is 
366: unambiguously detected at $21.623\pm 0.012$\AA\ ($v=290\pm 170$\kms) 
367: with an equivalent width of 
368: $25.7\pm 5.3$~m\AA.  Upper limits are measured for the \ovii~K$\beta$ 
369: and \oviii\ lines.
370: The \chandra\ LETG wavelength scale contains intrinsic random errors 
371: of approximately
372: $0.01$\AA, on the order of the statistical error on the \ovii\ position 
373: measurement, but these errors should not vary with time (J.~J.~Drake, private
374: communication).  To check the absolute wavelength
375: scale near the \ovii\ line, we retrieved the nearest HRC--S/LETG calibration 
376: observation of the X-ray bright star Capella (observation 3675, taken on 2003
377: September 28) from the \chandra\ archive and 
378: reprocessed the data in exactly the same manner as the Mrk 279 data.
379: The wavelength of the strong \ovii\ emission line was found to be 
380: $21.606\pm 0.002$\AA\ or $56\pm 28$\kms, which is consistent with the 
381: $+30$\kms\ radial velocity of Capella as listed in the
382: SIMBAD database.
383: Thus, any systematic effects on the measured velocity of the \ovii\ absorption
384: are likely to be insignificant compared to the statistical error.
385: 
386: Mrk~279 was also observed with \fuse\ for a total exposure time 
387: of 224~ks (though only 177~ks of these data were usable); all calibrated 
388: data from these observations were retrieved and coadded in the same manner 
389: as the Mrk~421 \fuse\ data.  This spectrum also shows strong Galactic
390: thick--disk \ovi\ absorption at $v\sim 0$, but unlike the Mrk~421 spectrum,
391: the high--velocity \ovi\ is strong and clearly separated from the 
392: thick--disk \ovi\ at $v= -160\pm 2.6$\kms.  Figure~\ref{fig_mkn279_spec} shows
393: the \ovii\ and \ovi\ absorption systems plotted against velocity.
394: Although the error on the \ovii\ velocity is large, it is nonetheless
395: significantly higher than the \ovi\ HVC velocity by about $2.5\sigma$.
396: Thus, a direct association between the \ovi\ HVC and \ovii\ can
397: be ruled out with reasonably high confidence.
398: 
399: \subsection{Doppler parameters}
400: \begin{figure}
401: \centering
402: \includegraphics[width=0.8\linewidth]{mkn279nhb_ovii.eps}
403: \caption{Column density and velocity dispersion diagnostics for the Mrk~279
404: \ovii\ absorption.  The shaded region is derived from the $1\sigma$  
405: \ovii\ K$\alpha$ equivalent width limits, the dashed line is from the \ovii\ 
406: K$\beta$ $2\sigma$
407: upper limit, and the 95\% joint confidence interval is depicted by the
408: dark hatched region.
409: \label{fig_mkn279_nhb}}
410: \end{figure}
411: 
412: In the Mrk~421 \chandra\ spectrum, three absorption lines from the \ovii\ 
413: K-series were strongly detected.  While a similar curve--of--growth analysis 
414: can be performed for Mrk~279, placing limits on $\novii$ and $b$ is more
415: difficult because only the \ovii\ K$\alpha$ line is strongly detected;
416: only an upper limit can be measured for the K$\beta$ line.
417: Since the absorption line properties for various column densities
418: and Doppler parameters are known, limits on these quantities 
419: can be determined using the \chandra\ spectrum itself.  For each point
420: in the $\novii-b$ plane, \ovii\ K$\alpha$ and K$\beta$ absorption
421: lines with the calculated $W_\lambda$ and FWHM values were added to 
422: the best--fit continuum model, and the $\chi^2$ statistic calculated using
423: the ``goodness'' command in Sherpa.  Since best--fit K$\beta$ line amplitude
424: is zero, the unsaturated case (assuming the \ovii\ line ratio constraint 
425: $W_\lambda(K\beta)=0.15*W_\lambda(K\alpha)$) produces the best fit to the
426: data.  The minimum $\chi^2$ value was taken from such a fit, and 
427: $\Delta\chi^2=\chi^2(\novii,b)-\chi_{\rm min}^2$ calculated for every point.  
428: The 95\% confidence interval ($\Delta\chi^2 < 6$) calculated with this 
429: method is shown in
430: Figure~\ref{fig_mkn279_nhb} (as well as the $\novii-b$ contours derived
431: from each transition); at this confidence
432: level all Doppler parameters between $20<b<77$\kms\ are ruled out.
433: 
434: Doppler parameters for the \ovi\ absorption were calculated in the same
435: manner as for Mrk~421.  In this case, however, two Gaussians are necessary
436: to fit the low--velocity (thick--disk) \ovi.  The derived velocity
437: dispersions are $b=61.5\pm 3.5$, $38.8\pm 2.8$, and $32.0\pm 4.6$ for the
438: broad Galactic, narrow Galactic, and high--velocity \ovi\ components
439: respectively (producing the best--fit model shown in 
440: Figure~\ref{fig_mkn279_spec}).  These are all strongly inconsistent with the
441: limits found for the \ovii\ velocity dispersion, indicating that the
442: \ovii\ is not related to any of the \ovi\ components.  Even if the
443: low--velocity \ovi\ is considered to be one non--Gaussian component
444: and its width is measured directly from the spectrum, its Doppler
445: parameter is $b\sim 80$\kms, barely consistent with the 95\% lower
446: limit on the \ovii\ $b$ value.  
447: 
448: \begin{figure}
449: \centering
450: \includegraphics[width=0.8\linewidth]{b100plot.eps}
451: \includegraphics[width=0.8\linewidth]{b200plot.eps}
452: \caption{Mrk~279 temperature and density constraints from the \oviii/\ovii\ 
453: and \ovi/\ovii\ upper limits, assuming a putative \ovi\ absorption
454: line with $b=100$\kms (top) and $b=200$\kms (bottom).
455: \label{fig_mkn279_tne}}
456: \end{figure}
457: 
458: \subsection{Temperature and Density Constraints}
459: An upper limit on the temperature of the absorbing medium of $\log T < 6.3$
460: can be derived from the \oviii/\ovii\ column density.  Since the
461: \ovii\ absorption does not appear to be associated with any of the
462: \ovi\ components, finding a lower limit on the temperature is more
463: difficult.  We thus assume that the \ovi\ absorption from the
464: \ovii --bearing gas is undetected in the \fuse\ spectrum, and exists
465: as a very broad ($b>80$\kms) absorption line superposed on the narrower
466: detected components.  Limits on the \ovi\ column density associated with
467: the \ovii\ were thus calculated by placing such an absorption line in the
468: \fuse\ spectrum model, one with $b=100$\kms\ and one with $b=200$\kms,
469: and calculating the corresponding \ovi/\ovii\ $T-n_e$ constraint
470: (shown in Figure~\ref{fig_mkn279_tne}.
471: 
472: The lower limit on temperature derived from the \ovi/\ovii\ upper limit
473: is highly dependent on $b$; in fact, for $b=100$\kms\ the two oxygen
474: line ratios are inconsistent with each other for \emph{all} temperatures
475: and densities.  Thus, if the \ovii\ is associated with an undetected
476: broad \ovi\ line, the velocity dispersion of the absorption must
477: be very high (at least $b\sim 200$\kms).  Such a high velocity dispersion,
478: if purely thermal, implies temperatures of $T\sim 10^7$\,K, which is 
479: ruled out by the non--detection of \oviii\ absorption.  This absorber
480: must therefore be broadened primarily by nonthermal processes, perhaps
481: due to the velocity shear of infalling hot gas associated with HVC
482: Complex C.
483: 
484: \section{Conclusions}
485: We have detected strong $z=0$ X-ray absorption toward both Mrk~421 and
486: Mrk~279.  Both absorption systems appear to exhibit similar column
487: densities and consistent temperature and density limits (albeit with
488: large errors).  Additionally, in both cases the detected X-ray absorption
489: does \emph{not} appear to arise in the same phase as the low-- or 
490: high--velocity \ovi\ absorption seen in the \fuse\ spectra, indicating
491: that the \ovii\ absorption likely comes from either the local WHIM or
492: a heretofore--undiscovered hot Galactic component.  Additionally, the
493: Doppler parameters of the Mrk~421 and Mrk~279 absorption are highly
494: inconsistent with each other, perhaps indicating that these two systems
495: originate from entirely different physical processes.  
496: 
497: \section*{Acknowledgments}
498: We thank the conference organizers for an enjoyable and informative
499: week, and the \chandra\ and \fuse\ teams for their exceptional efforts on these
500: missions.  This work has been supported by \chandra\ award AR5-6017X issued
501: by the \chandra\ X-ray Observatory Center, which is operated by the
502: Smithsonian Astrophysical Observatory for and on behalf of the NASA 
503: under contract NAS8-39073.  RJW derives additional support from an Ohio
504: State University Presidential Fellowship.
505: 
506: \begin{thebibliography}{}
507: \bibitem[Cen \& Ostriker(1999)]{cen99} Cen, R., \& Ostriker, J.~P. 1999,
508:          ApJ, 514, 1
509: \bibitem[Dav\'e et al.(2001)]{dave01} Dav\'e, R., et al. 2001, ApJ, 552, 473
510: \bibitem[Ferland(1996)]{ferland96} Ferland, G.~J., 1996, Hazy, a Brief
511:          Introduction to Cloudy 90, University of Kentucky, Physics
512:          Department Internal Report
513: \bibitem[Hellsten et al.(1998)]{hellsten98} Hellsten, U., Gnedin, N.~Y., 
514:          \& Miralda--Escud\'e, J. 1998, ApJ, 509, 56
515: \bibitem[Kravtsov et al.(2002)]{kravtsov02} Kravtsov, A.~V., Klypin, A.,
516:          \& Hoffman, Y. 2002, ApJ, 571, 563
517: \bibitem[Nicastro et al.(2003)]{nicastro03} Nicastro, F., et al. 2003,
518:          Nature, 421, 719
519: \bibitem[Nicastro et al.(2005a)]{nicastro05a} Nicastro, F., et al. 2005,
520:          ApJ, 629, 700
521: \bibitem[Nicastro et al.(2005b)]{nicastro05b} Nicastro, F., et al. 2005,
522:          Nature, 433, 495
523: \bibitem[Rasmussen et al.(2003)]{rasmussen03} Rasmussen, A., Kahn, S.~M.,
524:          \& Paerels, F. 2003, ASSL Conference Proceedings, 208, 109
525: \bibitem[Wakker et al.(2003)]{wakker03} Wakker, B.~P., et al. 2003,
526:          ApJS, 146, 1
527: \bibitem[Wang et al.(2005)]{wang05} Wang, Q.~D., et al. 2005, ApJ, accepted
528:          (astro-ph/0508661)
529: \bibitem[Williams et al.(2005)]{williams05} Williams, R., et al. 2005,
530:          ApJ, 631, 856 
531: \bibitem[Williams et al.(2006)]{williams06} Williams, R., et al. 2006,
532:          ApJ, submitted
533: \end{thebibliography}
534: 
535: \end{document}
536: %%
537: %% End of file `XrU2005.tex'.
538: