1: % ----- AASTex 5.0.2 ------------------------------------
2: %%
3: %% Uncomment next 2 lines to emulate ApJ style
4: %%\documentclass{aastex}
5: %%\usepackage{emulateapj5}
6: %%
7: % ----------------------------------------------------------
8: %% Uncomment the following line (only) for final ApJ submittal
9: \documentclass[12pt,preprint]{aastex}
10: %%
11: %% --------------------------------------------------------
12: %%
13: %% Full page width, double-spaced
14: %%\documentstyle[12pt,aasms4]{article}
15: %%
16: % Full page width, single-spaced
17: %%\documentstyle[11pt,aaspp4]{article}
18: %%
19: %% 2-column preprint style
20: %%\documentstyle[aas2pp4]{article}
21:
22: %\documentstyle[11pt,eqsecnum,aaspp4]{article}
23: %\documentstyle[12pt,amssym,aasms4]{article}
24:
25: % Here's some slug-line data. The receipt and acceptance dates will be
26: % filled in by the editorial staff with the appropriate dates. Rules will
27: % appear on the title page of the manuscript until these are uncommented
28: % out by the editorial staff.
29:
30: %\received{23 May 2006}
31: %\accepted{23 May 2006}
32: %\journalid{337}{15 January 1989}
33: %\articleid{11}{14}
34:
35: %\slugcomment{Not to appear in Nonlearned J., 45.}
36:
37: % Authors may supply running head information, if they wish to do so, although
38: % this may be modified by the editorial offices. The left head contains a
39: % list of authors, usually three allowed---otherwise use et al. The right
40: % head is a modified title of up to roughly 44 characters. Running heads
41: % are not printed.
42:
43: \lefthead{Skinner et al.}
44: \righthead{Sigma Orionis}
45:
46: % This is the end of the "preamble". Now we wish to start with th
47: % real material for the paper, which we indicate with \begin{document}.
48: % Following the \begin{document} command is the front matter for the
49: % paper, viz., the title, author and address data, the abstract, and
50: % any keywords or subject headings that are relevant.
51:
52: \begin{document}
53:
54: \title{High-Resolution {\em Chandra} X-ray Imaging and Spectroscopy of the
55: $\sigma$ Orionis Cluster}
56:
57: \author{Stephen L. Skinner and Kimberly R. Sokal}
58: \affil{CASA, Univ. of Colorado, Boulder, CO 80309-0389 }
59:
60: \author{David H. Cohen}
61: \affil{Dept. of Physics and Astronomy, Swarthmore College,
62: Swarthmore, PA 19081}
63:
64: \author{Marc Gagn\'{e}}
65: \affil{Dept. of Geology and Astronomy, West Chester Univ.,
66: West Chester, PA 19383-2130}
67:
68: %%\author{Manuel G\"{u}del}
69: %%\affil{Paul Scherrer Inst., W\"{u}renlingen and Villigen,
70: %% CH-5232 Villigen, Switzerland}
71:
72:
73: \author{Stanley P. Owocki and Richard D. Townsend}
74: \affil{Bartol Research Inst., Univ. of Deleware,
75: Newark, DE 19716}
76:
77:
78:
79: % The abstract environment prints out the receipt and acceptance dates
80: % if they are relevant for the journal style. For the aasms style, they
81: % will print out as horizontal rules for the editorial staff to type
82: % on, so long as the author does not include \received and \accepted
83: % commands. This should not be done, since \received and \accepted dates
84: % are not known to the author.
85: %
86: % Define symbol \ltsimeq
87: \newcommand{\ltsimeq}{\raisebox{-0.6ex}{$\,\stackrel{\raisebox{-.2ex}%
88: {$\textstyle<$}}{\sim}\,$}}
89: %
90: % Define symbol \gtsimeq
91: \newcommand{\gtsimeq}{\raisebox{-0.6ex}{$\,\stackrel{\raisebox{-.2ex}%
92: {$\textstyle>$}}{\sim}\,$}}
93: %
94:
95: \begin{abstract}
96: We present results of a 90 ksec {\em Chandra} X-ray observation
97: of the young $\sigma$ Orionis cluster (age $\sim$3 Myr)
98: obtained with the High Energy Transmission Grating Spectrometer.
99: We use the high resolution grating spectrum and
100: moderate resolution CCD spectrum of the massive central star
101: $\sigma$ Ori AB (O9.5V $+$ B0.5V) to test wind shock theories of
102: X-ray emission and also analyze the high spatial resolution
103: zero-order ACIS-S image of the central cluster region.
104:
105: {\em Chandra} detected 42 X-ray sources on the primary CCD
106: (ACIS-S3). All but five have near-IR or
107: optical counterparts and about one-fourth
108: are variable. Notable high-mass stellar detections are
109: $\sigma$ Ori AB, the magnetic B star $\sigma$ Ori E, and
110: the B5V binary HD 37525. Most of the other detections have
111: properties consistent with lower mass K or M-type stars.
112: We present the first
113: X-ray spectrum of the unusual infrared source IRS1, located
114: $\approx$3$''$ north of $\sigma$ Ori AB. Its X-ray properties
115: and elongated mid-IR morphology suggest it is an embedded low mass
116: T Tauri star whose disk/envelope is being photo-evaporated
117: by $\sigma$ Ori AB.
118:
119: We focus on the radiative wind shock interpretation of the
120: soft luminous X-ray emission from $\sigma$ Ori AB, but also
121: consider possible alternatives including magnetically-confined
122: wind shocks and colliding wind shocks. Its emission lines
123: show no significant asymmetries or centroid shifts
124: and are moderately broadened to HWHM $\approx$ 264 km s$^{-1}$,
125: or one-fourth the terminal wind speed.
126: Forbidden lines in He-like ions are formally
127: undetected, implying strong UV suppression. The Mg XI triplet forms
128: in the wind acceleration zone within one stellar radius
129: above the surface. These X-ray properties are consistent
130: in several respects with the predictions of radiative wind shock
131: theory for an optically thin wind, but explaining the narrow
132: line widths presents a challenge to the theory.
133:
134: \end{abstract}
135:
136: % The different journals have different requirements for keywords. The
137: % keywords.apj file, found on aas.org in the pubs/aastex-misc directory,
138: % contains a list of keywords used with the ApJ and Letters. These are
139: % usually assigned by the editor, but authors may include them in their
140: % manuscripts if they wish.
141:
142: \keywords{open clusters and associations: individual ($\sigma$ Orionis) ---
143: stars: formation --- stars: individual ($\sigma$ Orionis AB)
144: --- stars: pre-main-sequence ---
145: X-rays: stars}
146:
147:
148: % In the first two sections, you should notice the use of the LaTeX \cite
149: % command to identify citations. The citations are tied to the
150: % reference list via symbolic KEYs. We have chosen the first three
151: % characters of the first author's name plus the last two numeral of the
152: % year of publication. The corresponding reference has a \bibitem
153: % command in the reference list below.
154: %
155: % Please see the AASTeX manual for a more complete discussion on how to make
156: % \cite-\bibitem work for you.
157:
158: \newpage
159: \clearpage
160: \section{Introduction}
161: Sensitive X-ray grating observations of massive OB stars
162: with {\em Chandra} and {\em XMM-Newton} during the past several
163: years have provided a wealth of new data that is improving our
164: understanding of the physical processes that produce their
165: X-ray emission. Luminous OB stars
166: are not expected to have the outer convection zones needed to
167: sustain internally-generated magnetic fields and solar-like
168: (coronal) X-ray emission. Thus, their X-ray emission has generally
169: been attributed to wind shocks.
170:
171: Theoretical
172: models have attempted to explain the X-ray emission of {\em single}
173: massive OB stars in terms of X-ray emitting shocks distributed
174: throughout their powerful winds that
175: form as a result of line-driven flow instabilities (Lucy \&
176: White 1980; Lucy 1982; ; Feldmeier et al. 1997;
177: Owocki, Castor, \& Rybicki 1988). The radiative wind shock emission
178: is predicted to be soft (kT $<<$ 1 keV), in contrast to
179: the harder emission (kT $\gtsimeq$ 2 keV) that has been detected
180: in some massive colliding wind binaries and in a few OB stars with
181: strong $\sim$kG magnetic fields whose winds are
182: magnetically-confined (Secs. 3.4 and 3.7.2).
183: X-ray grating observations provide crucial information
184: on line shapes, widths, centroids, and fluxes needed
185: to test wind shock models.
186:
187: Grating observations of a limited sample of O and early B-type stars
188: have provided some support for the radiative wind shock picture.
189: Perhaps the best support so far comes from the O4f star $\zeta$ Pup,
190: which has a high mass-loss rate log $\dot{M}$ $\sim$ $-$5.3
191: M$_{\odot}$ yr$^{-1}$ (Pauldrach et al. 1994). Its X-ray
192: emission lines are asymmetric with blueshifted centroids and are
193: quite broad with half-width half-maximum
194: values HWHM $\sim$ 900 km s$^{-1}$, equivalent to $\approx$40\%
195: of the terminal wind speed
196: (Cassinelli et al. 2001; Kahn et al. 2001; Kramer et al. 2003).
197: An analysis of a {\em Chandra} grating spectrum of the
198: O9 supergiant $\zeta$ Ori by Cohen et al. (2006) also
199: revealed broad blueshifted asymmetric emission lines,
200: a different conclusion than was reached by Waldron \& Cassinelli (2001).
201: The blueshifted asymmetry
202: is predicted by radiative wind shock models of
203: optically thick winds where the redward part of the line is
204: attenuated from receding wind material on the far side of the
205: star (Owocki \& Cohen 2001).
206:
207:
208:
209: For late O and early B-type stars with lower mass-loss rates than
210: O supergiants, clear confirmation of radiative wind shock theory has been more
211: difficult to obtain. Grating spectra of the Orion belt eclipsing binary system
212: $\delta$ Ori (O9.5 II $+$ B0.5 III;
213: log $\dot{M}$ $\sim$ $-$6. M$_{\odot}$ yr$^{-1}$;
214: Miller et al. 2002) reveal cool plasma at kT $\sim$ 0.3 keV,
215: and symmetric unshifted emission lines with
216: HWHM $\sim$ 430 km s$^{-1}$, or $\approx$22\% of the terminal
217: wind speed. The dominant cool plasma is compatible with a shock
218: origin and symmetric unshifted lines are predicted by radiative
219: wind shock simulations for an optically thin wind
220: (MacFarlane et al. 1991; Owocki \& Cohen 2001). However, Miller et al.
221: (2002) raised some questions about the surprising narrowness and symmetry
222: of the lines in the framework of radiative wind shock models.
223: An even more extreme case is the early B giant $\beta$ Cru (B0.5 III;
224: log $\dot{M}$ $\sim$ $-$8. M$_{\odot}$ yr$^{-1}$; Cohen et al. 2008).
225: It also has a low plasma temperature and symmetric unshifted lines
226: which are quite narrow (HWHM $\sim$ 150 km s$^{-1}$).
227: Cohen et al. (2008) concluded that the narrow lines were difficult
228: to understand in the framework of radiative wind shock
229: theory for fast winds.
230:
231: Only a handful of X-ray grating observations of OB stars
232: have been obtained and observations of a broader sample of
233: objects are needed to clarify how X-ray properties vary
234: with spectral type, luminosity
235: class, and mass-loss parameters.
236: The Orion complex is particularly well-suited to observational
237: tests of X-ray emission models in massive young stars because
238: of its relative proximity and large concentration of very
239: young OB stars. It is the closest giant molecular cloud
240: containing young massive stars. The density of
241: young stars in Orion is high and the deep
242: {\em Chandra} observation of the Orion Nebula obtained
243: in the COUP project detected more than 1400 X-ray sources
244: (Getman et al. 2005). {\em Chandra's} superb angular
245: resolution is clearly an advantage in deep X-ray
246: observations of Orion in order to avoid source misidentification.
247:
248:
249: A promising target
250: for exploring X-ray production in young high-mass stars is
251: $\sigma$ Orionis AB (Table 1), the focus of this study.
252: It is located in the Ori OB1b association and is quite young with
253: an estimated age of 3$\pm$2 Myr (Caballero 2007a and references
254: therein; Caballero 2008). The
255: {\em Hipparcos} distance is 352$^{+166}_{-85}$ pc,
256: which we adopt here.
257: A slightly smaller value of 334$^{+25}_{-22}$ pc
258: was recently obtained by Caballero (2008) who also
259: noted that d $\sim$ 385 pc is possible if $\sigma$ Ori AB
260: is a triple system.
261: For comparison, recent distance measurements of the Orion Nebula cluster
262: give 389$^{+24}_{-21}$ pc (Sandstrom et al. 2007),
263: 392$^{+32}_{-32}$ pc (Jeffries 2007),
264: 414 $\pm$ 7 pc (Menten et al. 2007), and
265: 391$^{+12}_{-9}$ pc (Mayne \& Naylor 2008).
266:
267:
268:
269: $\sigma$ Ori AB is a visual binary consisting of a O9.5V primary and
270: a B0.5V companion separated by 0.$''$25 (Mason et al. 1998)
271: %%Tokovinin \& Ismailov 1988)
272: and is known to be a bright
273: X-ray source (Bergh\"{o}fer \& Schmitt 1994; Bergh\"{o}fer, Schmitt,
274: \& Cassinelli 1996). The A and B components cannot be separated
275: by existing X-ray telescopes. At a distance of 352 pc, the
276: angular separation equates to a projected linear separation of
277: 88 AU, or 2101 R$_{O-star}$ (Table 1).
278: The extinction toward $\sigma$ Ori AB
279: is low with E(B-V) = 0.05 - 0.07
280: (Table 1), thus minimizing interstellar absorption of any
281: soft X-rays that might be produced in stellar wind shocks.
282: In terms of wind strength, $\sigma$ Ori A provides a good
283: comparison with observations of other OB stars. Its wind
284: is much weaker than the O4f star $\zeta$ Pup, and its
285: mass-loss rate falls roughly in between that of the moderate
286: wind source $\delta$ Ori and the weak wind source $\beta$ Cru.
287:
288:
289: There is additional interest in $\sigma$ Ori because it is
290: surrounded by a cluster of several hundred young stars,
291: originally discovered in {\em ROSAT} X-ray observations
292: (Walter et al. 1997).
293: {\em Spitzer} mid-IR observations have identified more than 300
294: cluster members, with a disk fraction of $\approx$35\% in
295: low-mass stars (Hern\'{a}ndez et al. 2007). A similar
296: disk fraction was found by Oliveira et al. (2006) from
297: ground-based K and L$'$-band images. A recent analysis of
298: {\em TYCHO} photometry by Mayne \& Naylor (2008) gives
299: a nominal cluster age of $\sim$3 Myr and a distance
300: of 389$^{+34}_{-24}$ pc. Some stars are still in the
301: accretion phase but the estimated accretion rates are
302: low, with typical values $\dot{M}$$_{acc}$ $<$
303: 10$^{-8}$ M$_{\odot}$ yr$^{-1}$ (Gatti et al. 2008).
304: This may have some effect on X-ray emission, which is
305: of primary interest here, given that several studies
306: suggest that the X-ray luminosities of accreting classical
307: T Tauri stars are on average about a factor of $\sim$2
308: lower than their non-accreting counterparts, the weak-lined
309: T Tauri stars (Preibisch et al. 2005; Franciosini,
310: Pallavicini, \& Sanz-Forcada 2006, hereafter FPS06;
311: Telleschi et al. 2007).
312:
313:
314: X-ray observations of the $\sigma$ Ori region have
315: also been obtained with {\em XMM-Newton}
316: and 174 X-ray sources were detected, 76 of which were identified
317: with cluster members
318: (Sanz-Forcada, Franciosini,\& Pallavicini 2004; FPS06).
319: A 98 ksec {\em Chandra} HRC-I
320: observation (ObsId 2560) has also been acquired,
321: detecting more than 140 X-ray sources (Wolk et al. 2006)
322: including faint X-ray emission from the infrared source IRS 1
323: located just 3.$''$3 north of $\sigma$ Ori AB
324: (van Loon \& Oliveira 2003; Caballero 2007b).
325:
326:
327: We present results of a sensitive {\em Chandra} X-ray
328: observation of the $\sigma$ Ori cluster obtained
329: with the High Energy Transmission Grating (HETG)
330: and the ACIS-S detector. This observation provides
331: higher spatial resolution than the previous
332: {\em XMM-Newton} observation, but over a smaller
333: field-of-view. The spatial resolution acquired here
334: is comparable to that of the {\em Chandra}
335: HRC-I observation, but the HETG/ACIS-S combination
336: provides spectral information whereas HRC-I does not.
337: On the other hand, the HRC-I observation is more sensitive,
338: covers a larger field-of-view, and gives superior image
339: quality. Thus, the existing
340: {\em Chandra} and {\em XMM-Newton} data sets are complementary.
341:
342:
343: Our primary objective was to obtain a high-resolution
344: X-ray spectrum of $\sigma$ Ori AB to test wind shock
345: models of X-ray emission in massive stars. We focus
346: on spectral properties of $\sigma$ Ori AB but also
347: analyze zeroth order CCD images, spectra, and light
348: curves of 41 other X-ray sources near $\sigma$ Ori AB,
349: including the magnetic helium-strong B2Vpe star
350: $\sigma$ Ori E (HD 37479) and the unusual infrared source
351: IRS1.
352:
353:
354:
355: \section{Chandra Observations and Data Reduction}
356:
357: The {\em Chandra} observation (ObsId 3738) began on 2003
358: August 12 at 20:59:02 UT and ended on August 13 at 23:29:25 UT,
359: yielding 90,967 s of usable exposure. Pointing was
360: centered near $\sigma$ Ori AB (HD 37468) at nominal
361: pointing coordinates (J2000.0)
362: R.A. = 05$^{\rm h}$38$^{\rm m}$44.$^{\rm s}$43,
363: decl. = $-$02$^{\circ}$35$'$25.$''$9. Grating spectra
364: of $\sigma$ Ori AB were
365: obtained with the High-Energy Transmission Grating (HETG)
366: using the ACIS-S detector in faint timed event mode
367: with 3.2 s frame times.
368:
369: The Level 1 events file provided by the {\em Chandra} X-ray
370: Center (CXC) was processed using CIAO v. 3.0
371: \footnote{Further information on {\em Chandra} Interactive
372: Analysis of Observations (CIAO) software can be found at
373: http://asc.harvard.edu/ciao.} using standard science
374: threads. The CIAO processing applied calibration updates
375: (CALDB v. 2.26), selected good event patterns, removed streak
376: artifacts from CCD-8,
377: determined zero-order source centroids, defined
378: grating arm masks and extracted grating spectra,
379: and created response matrix files (RMFs) and auxiliary
380: response files (ARFs) for the extracted spectra.
381: The spectral extraction provides separate dispersed spectra for the
382: medium-energy grating (MEG) and high-energy grating (HEG).
383: The HEG provides higher spectral
384: resolution ($\Delta$$\lambda$ = 12 m\AA~ FWHM, 1.2 - 15 \AA~) than the
385: MEG ($\Delta$$\lambda$ = 23 m\AA~, 2.5 - 31 \AA~) but the MEG
386: gives a higher signal-to-noise ratio (S/N) in
387: 1st order spectra for $\sigma$ Ori AB. We focus here on
388: the higher S/N data obtained in MEG 1st order spectra,
389: combining the $+1$ and $-1$ orders for spectral analysis.
390:
391:
392: Additionally, we used the CIAO {\em wavdetect} tool to
393: identify X-ray sources on the ACIS-S3 chip
394: (CCD-7) which covers a $\approx$8.3$'$ $\times$ 8.3$'$ region
395: near the primary target $\sigma$ Ori AB.
396: We ran {\em wavdetect} on full-resolution
397: (0.$''$49 pixels) images using events in the 0.5 - 7 keV
398: range to reduce the background. The {\em wavdetect}
399: threshold was set at $sigthresh$ = 1.5 $\times$ 10$^{-5}$
400: and scale sizes of 1,2,4,8, and 16 were used.
401: We visually compared the {\em wavdetect} source list
402: against the ACIS-S image and the archived
403: HRC-I image to check for spurious and missed detections.
404:
405:
406: CIAO {\em psextract}
407: was used to extract zero-order source and background spectra
408: for brighter sources ($>$100 net counts).
409: The nominal extraction radius used was
410: R$_{e}$ = 4 pixels (= 1.$''$97). Zero order light curves were also
411: extracted for brighter sources using CIAO tools.
412: Spectral and timing analysis were undertaken
413: with the HEASOFT {\em Xanadu}
414: \footnote{http://heasarc.gsfc.nasa.gov/docs/xanadu/xanadu.html.}
415: software package including XSPEC v. 12.3.0. We also used
416: a modified version of the ISIS v. 1.4.8
417: \footnote{http://space.mit.edu/csc/isis/}
418: spectral analysis package to complement the XSPEC analysis.
419: The ISIS implementation allows the flux ratios of
420: the resonance (r), intercombination (i), and forbidden (f)
421: lines in He-like ions to be varied as free parameters
422: when fitting grating spectra. This is an advantage
423: for OB stars since their f/i line flux ratios can be affected by the
424: strong UV radiation field. The modified ISIS package
425: is discussed in more detail by
426: Cohen et al. (2008).
427:
428:
429: \section{X-ray Source Identification and Properties}
430:
431: \subsection{Source Identification}
432:
433: Our analysis of zero-order data focuses on sources
434: detected on the ACIS-S3 chip (CCD7) which contains
435: the primary target $\sigma$ Ori AB. Sources detected
436: on the other chips in the ACIS-S array lie
437: far off-axis and are heavily vignetted and will
438: not be discussed here. Figure 1
439: shows the positions of the 42 sources detected on
440: ACIS-S3 and their properties are summarized in Table 2.
441: All of the sources in Table 2 are also visible in the
442: archived HRC-I image (obsId 2560).
443:
444:
445:
446: Figure 2 is a zoomed view of the ACIS-S3 image
447: in the vicinity of $\sigma$ Ori AB, the
448: brightest X-ray source on ACIS-S3.
449: The infrared source IRS1 lying 3.$''$3
450: north of $\sigma$Ori AB is also detected (CXO 20).
451: Other noteworthy detections are the
452: magnetic B2Vpe star $\sigma$ Ori E (CXO 23; Sec. 3.7.2)
453: and the B5V star HD 37525 (CXO 42), the latter being
454: a known binary (Sec. 3.7.3). Interestingly,
455: we do not detect the B2V star $\sigma$ Ori D
456: even though its spectral type is the same as
457: $\sigma$ Ori E, a clue that $\sigma$ Ori E
458: is anomalously bright in X-rays for its spectral
459: type. We do note however that
460: an X-ray source at or near the position
461: of $\sigma$Ori D is visible in the more
462: sensitive HRC-I image. The A-type star
463: $\sigma$ Ori C was not detected.
464:
465:
466:
467:
468: All but five of the X-ray sources in Table 2 have optical or
469: near-IR counterparts as determined from a search of
470: the HEASARC object catalogs
471: \footnote{http://heasarc.gsfc.nasa.gov/docs/archive.html}
472: using a 2$''$ search radius.
473: As Table 2 shows, the offsets between the X-ray and
474: IR/optical positions are in all cases $<$1$''$
475: and in most cases $<$0.$''$5. Interestingly, no
476: IR counterpart was found for CXO 8, which is one
477: of the brighter X-ray sources. However, a faint
478: V = 20.88 mag optical source was found in
479: Mayne et al. (2007) at an offset of 0.$''$18 from
480: the X-ray position. The X-ray source has an unusually hard
481: spectrum (discussed below) and is variable.
482: Further deep optical and infrared observations would
483: be useful to elucidate its nature.
484:
485:
486: \subsection{Faint Detections and Detection Limit}
487:
488: The faintest ACIS-S X-ray detection (CXO 32) has
489: 5 $\pm$ 2 counts and is confirmed in the HRC-I image.
490: Assuming a slightly more conservative 7 count
491: detection threshold, the corresponding X-ray
492: luminosity threshold determined from PIMMS
493: \footnote{For information on PIMMS (Portable Interactive
494: Multi-Mission Simulator) see
495: http://asc.harvard.edu/ciao/ahelp/pimms.html.}
496: at a cluster distance d $\approx$ 390 pc is
497: log L$_{x}$ = 28.81 ergs s$^{-1}$ (0.5 - 7 keV). Here
498: we have assumed a typical late-type pre-main-sequence star
499: modeled as an isothermal X-ray source with a
500: Raymond-Smith plasma at kT = 2 keV and an absorption
501: column density
502: N$_{\rm H}$ = 3 $\times$ 10$^{20}$ cm$^{-2}$
503: (E(B-V) $\approx$ 0.05; Gorenstein 1975 conversion).
504:
505:
506: At a luminosity threshold log L$_{x}$ = 28.81 ergs s$^{-1}$
507: we would expect to detect X-ray emission from pre-main
508: sequence stars in Orion down to masses of $\sim$0.2 - 0.3 M$_{\odot}$,
509: based on previous deep {\em Chandra} Orion observations
510: (Figs. 3 and 7 of Preibisch et al. 2005). A comparison of
511: the X-ray detections in Table 2 with the optical catalog
512: of Sherry, Walter, \& Wolk (2004, = S04) confirms
513: this. Two low-mass detections are S04 15 = CXO 17 and
514: S04 18 = CXO 31, both of which have high cluster membership
515: probability (92\% - 93\%). Their mass estimates
516: from S04 are 0.22 M$_{\odot}$ (S04 15) and
517: 0.18 M$_{\odot}$ (S04 18) and their respective
518: V - R$_{c}$ colors imply equivalent main-sequence
519: spectral types of $\sim$M4V - M5V (Kenyon \& Hartmann 1995).
520: Also noteworthy is {\em Chandra} source CXO 22, which was
521: classified as M5 by FPS06 but was listed in their Table A.2
522: (object name B3.01-67) as undetected by {\em XMM-Newton}.
523:
524: Some of the faint sources in Table 2 are likely extragalactic
525: background objects. In particular, the five sources without
526: optical or near-IR identifications are extragalactic candidates
527: since all are faint ($\leq$25 counts), non-variable,
528: and have above-average mean photon energies. If we assume
529: a typical extragalactic source
530: power-law X-ray spectrum with a photon power-law index $\Gamma$ = 1.4,
531: N$_{\rm H}$ = 3 $\times$ 10$^{20}$ cm$^{-2}$,
532: and a 7-count detection threshold, then the hard-band (2 - 8 keV)
533: number counts from the {\em Chandra} Deep Field North (DFN)
534: observations (Cowie et al. 2002) predict $\approx$15 extragalactic
535: sources in the ACIS-S3 CCD field-of-view above our detection limit.
536: The expression for hard-band (2 - 8 keV) number counts obtained from
537: DFN data by Brandt et al. (2001) gives a similar result.
538: However, the accuracy
539: of the log $N$ - log $S$ distribution for extragalactic sources
540: from the Deep Field observations toward lower galactic latitude
541: star-forming regions such as $\sigma$ Ori ($b$ $\approx$ $-$17$^{\circ}$)
542: is not well-known. For the deep {\em Chandra} COUP observation of
543: the Orion Nebula Cluster ($b$ $\approx$ $-$19$^{\circ}$),
544: Getman et al. (2005) concluded that the number of extragalactic
545: sources in the X-ray sample was a factor of $\sim$3 - 4 less
546: than predicted by {\em Chandra} Deep Field number counts.
547:
548:
549:
550: \subsection{Hardness}
551:
552: The typical
553: mean photon energy (Table 2) of detected sources is
554: $<$E$>$ = 1.83 keV and the distribution of $<$E$>$ values
555: is shown in Figure 3. The brightest X-ray source
556: $\sigma$ Ori AB also has the lowest
557: mean energy $<$E$>$ = 0.86 keV. The hardest emission is from
558: the very faint source CXO 32 with $<$E$>$ = 4.06 keV, for
559: whicn no optical or IR counterpart was found. These properties
560: suggest that it may be an extragalactic background source.
561:
562: \subsection{Variability}
563:
564: X-ray variability is frequently detected in T Tauri stars.
565: The most extensive data on X-ray variability for young stars in
566: Orion come from the {\em Chandra} Orion Ultradeep Project (COUP),
567: which obtained a long $\sim$13 day observation of the Orion
568: Nebula Cluster (ONC). Rapid X-ray variability ($\sim$minutes
569: to hours) associated with magnetic-reconnection flares was
570: detected in T Tauri stars (Favata et al. 2005), as well as
571: slower X-ray modulation linked to stellar rotation at periods of
572: $\sim$2 - 12 days (Flaccomio et al. 2005). A variability analysis
573: of 20 massive O, B, and A stars detected in the COUP observation
574: was undertaken by Stelzer et al. (2005). Significant variability
575: was detected in at least 6 of the 9 earliest O7 - B3 stars having
576: strong winds. These include known multiple systems which appear to
577: show magnetic behavior such as $\theta^1$ Ori A (B0),
578: $\theta^2$ Ori A (O9.5), and $\theta^1$ Ori C (O7). But, no
579: variability was seen in stars showing the soft X-ray spectra
580: typcially associated with radiative wind shocks such as
581: $\theta^1$ Ori D (B0.5) and NU Ori (B1). {\em Chandra} observations
582: of other massive stars in Orion have also revealed objects with
583: soft X-ray spectra and no detectable variability, such as
584: $\delta$ Ori (Miller et al. 2002).
585:
586:
587: Nine sources detected in our $\sigma$ Ori observation
588: have high probability of variability as gauged by
589: P$_{var}$ $\geq$ 0.95 based
590: on the Kolmogorov-Smirnov (KS) test (Press et al. 1992)
591: using unbinned photon arrival times. These variable sources
592: are identified in column 4 of Table 2. The X-ray light curve
593: of $\sigma$ Ori AB (Fig. 4) shows no evidence
594: of large-amplitude variability. A KS test
595: on unbinned events gives a variability probability
596: P$_{var}$ = 0.935,
597: slightly below the limit P$_{var}$ $\geq$ 0.95
598: that we adopt for variability.
599: A $\chi^2$ test on the light curve binned at
600: 4000 s intervals ($\approx$108 counts/bin) gives
601: P$_{var}$ $<$ 0.001. Thus, we find no
602: evidence for significant ($\geq$95\% confidence)
603: X-ray variability in $\sigma$ Ori AB.
604:
605: The magnetic B2Vpe star $\sigma$Ori E is variable (Fig. 5;
606: see also Sec. 3.7.2). However, no large flares were
607: detected, in contrast to the large outbursts seen
608: in the {\em XMM-Newton} observation (Sanz-Forcada et al. 2004)
609: and in the {\em Chandra} HRC-I exposure.
610: Slow low-amplitude
611: X-ray variability is seen in the unusual
612: hard source CXO 8 (Fig. 6).
613: X-ray flares are clearly visible in CXO 9 (Fig. 7)
614: and CXO 25 (Fig. 8). The counterpart of CXO 9
615: has been classified as K8 and has properties of
616: a weak-lined TTS (Zapatero Osorio et al. 2002). CXO 25 is
617: associated with a M4 star that has a near-IR
618: excess (Oliveira et al. 2006) indicative of
619: a residual disk or envelope. The second brightest
620: X-ray source in Table 2 (CXO 37) showed a slow
621: decline in count rate by a factor of two during
622: the observation, but no large flares.
623:
624: \subsection{Near-IR Colors}
625:
626: Figure 9 shows J-K$_{s}$ colors of those
627: X-ray detections with 2MASS identifications.
628: IRS1 was not resolved from $\sigma$ Ori AB
629: by 2MASS and no published JHK magnitudes are available.
630: Most of the sources lie in the range
631: 0.7 $\leq$ J-K$_{s}$ $\leq$ 1.3, typical of
632: early K to late M dwarfs or giants
633: (Bessell \& Brett 1988). The three massive
634: stars $\sigma$ Ori AB, $\sigma$ Ori E, and
635: the B5V star HD 37525 are
636: clearly separated in color-magnitude space
637: by their smaller J-K$_{s}$ and K$_{s}$ values.
638: An interesting outlier is the faint hard variable
639: X-ray source CXO 12 which has
640: a large value J-K$_{s}$ = 2.33 and is the faintest
641: near-IR identification with K$_{s}$ = 14.04.
642: It is associated with a faint V = 18.8 mag optical
643: source (Mayne et al. 2007).
644:
645:
646: Figure 10 is a color-color diagram based on
647: 2MASS JHK$_{s}$ photometry of X-ray detections
648: with 2MASS identifications. Excluding the three
649: known high-mass stars, nineteen of the X-ray sources
650: in Figure 10 have assigned spectral types (FPS06)
651: and all of these are K or M stars. This substantiates
652: the above conclusion that most of the low-mass X-ray detections
653: are K or M stars. Of the nineteen X-ray sources with
654: known spectral types, slightly more than half lie
655: to the right of their respective reddening lines as
656: determined from the intrinsic colors of
657: Bessell \& Brett (1988) and the extinction law of
658: Rieke \& Lebofsky (1985). Thus, in this limited
659: sample of 19 stars, roughly half show evidence of
660: near-IR excesses based on existing spectral
661: type assignments and {\em 2MASS} colors.
662:
663:
664:
665: \subsection{CCD Spectra}
666:
667: We have fitted the zero-order CCD spectra of brighter sources
668: ($\geq$100 counts) with one-temperature (1T) or two-temperature (2T)
669: {\em apec} solar-abundance optically thin plasma models
670: (Smith et al. 2001) in XSPEC, as summarized
671: in Table 3. We have also fitted the spectrum of the high-interest
672: object IRS1 (CXO 20), but the fit is not well-constrained because
673: this source is faint (43 counts).
674: Photon pileup is low ($\approx$5\%) in the zero-order
675: spectrum of $\sigma$ Ori AB (neglibigle in first order) and
676: zero-order pileup is negligible ($<$1.7\%) in all the other
677: sources.
678:
679: In all but a few sources there were insufficient counts to constrain both
680: the temperature and absorption column density, so we held the
681: absorption fixed at N$_{\rm H}$ = 3 $\times$ 10$^{20}$ cm$^{-2}$,
682: consistent with E(B$-$V) $\approx$ 0.05 mag for the
683: $\sigma$ Ori region. The best-fit value of N$_{\rm H}$ was
684: within a factor of $\approx$2 of the above value for the brighter
685: sources where N$_{\rm H}$ was allowed to vary during the fit.
686:
687:
688: Two-temperature models provide an acceptable approximation
689: to the spectra in most cases. With the notable exception
690: of $\sigma$ Ori AB (discussed below), the derived temperatures from 2T models
691: are typical of magnetically-active pre-main sequence stars
692: in the Orion region (Preibisch \& Feigelson 2005), showing
693: both a cool component (kT$_{1}$ $<$ 1 keV) and
694: hotter plasma that is usually in the range
695: kT$_{2}$ $\approx$ 2 - 3 keV.
696:
697:
698: %%We held the metal abundances fixed at solar values
699: %%in fitting the zero-order spectra due to limited counts.
700: %%If the global metal abundance is allowed to vary in
701: %%the $\sigma$ Ori AB fit then the fit is improved
702: %%and converges to Z = 0.25 [0.15 - 0.38; 90\% conf.]
703: %%Z$_{\odot}$ (Table 3). More detailed information on abundances
704: %%is given below as derived from higher resolution
705: %%grating spectra.
706:
707: \subsection{Comments on Specific Stars}
708:
709: \subsubsection{HD 37468 ($\sigma$ Ori AB)}
710:
711: Fits of the CCD spectrum of $\sigma$ Ori AB
712: excluded events with energies below 0.5 keV to avoid
713: possible contamination from optical light leaks
714: \footnote{The ACIS CCDs are sensitive at optical frequencies
715: but most optical light is diffracted away from the CCDs
716: by the optical blocking filter. Also, the optical sensitivity
717: is further reduced when the MEG/HEG gratings are in place.
718: Further details can be found at
719: http://www.astro.psu.edu/xray/docs/cal\_report/node188.html.}.
720: A simple 1T $apec$ optically thin plasma model (1T $apec$)
721: was not acceptable.
722: But, a 2T $apec$ model does give an acceptable fit (Fig. 11)
723: with two cool components at
724: kT$_{1}$ = 0.21 keV
725: and kT$_{2}$ = 0.47 keV whose emission-measure weighted average
726: is kT = 0.3 keV. A fit with the differential emission measure
727: model $c6pmekl$ in XSPEC using solar abundances gives
728: good results ($\chi^2$/dof = 43.4/44) and shows a single broad
729: peak in the emission measure between 0.2 - 0.3 keV
730: with significant emission measure over the 0.15 - 0.45 keV
731: range (Fig. 12). Thus, the 2T model is probably an approximation to
732: this broad cool emission measure distribution.
733: The spectrum of $\sigma$ Ori AB is clearly dominated by cool
734: plasma, consistent with expectations for radiative wind shocks.
735:
736: As noted above, {\em Chandra} does not spatially resolve the
737: A and B components. Thus, our observation does not exclude
738: the possibility that the X-ray emission is the superposition of
739: soft emission from the O9.5V primary and the B0.5V secondary.
740: Soft X-ray emission from some B-type stars in Orion has previously
741: been reported, including the B0.5 star $\theta^1$ Ori D (Stelzer
742: et al. 2005).
743:
744:
745:
746: \subsubsection{HD 37479 ($\sigma$ Ori E)}
747:
748: The helium-strong star $\sigma$ Ori E has been extensively
749: studied and is known to exhibit optical, UV, and radio
750: modulations at a period P = 1.1908 days (Reiners et al.
751: 2000), which has been interpreted as the stellar rotation period.
752: A strong magnetic field that varies from $-$2.2 to $+$2.8 kG
753: at the 1.19 day period was detected by
754: Landstreet \& Borra (1978) and discussed further by
755: Borra \& Landstreet (1979).
756: A distance of 640 pc was obtained by
757: Hunger, Heber, \& Groote (1989) which, if correct,
758: would place it behind the $\sigma$ Ori cluster.
759:
760: An X-ray flare detected by {\em ROSAT} was reported
761: by Groote \& Schmitt (2004), who concluded that the
762: flare likely occurred on the magnetic B star and not
763: on a hypothesized late-type companion. A large X-ray
764: flare whose rise and decay spanned at least $\approx$8
765: hours was also detected by {\em XMM} (Sanz-Forcada et al. 2004).
766: The X-ray flare CCD spectrum recorded by the PN camera
767: showed fluorescent Fe emission at $\approx$6.4 keV,
768: a signature of cool neutral or near-neutral material
769: near the star that was irradiated by hard X-rays
770: during the flare. Sanz-Forcada et al. argued that
771: an unseen low-mass companion was likely responsible
772: for the flare. A large flare is also visible in the
773: archived {\em Chandra} HRC-I data.
774:
775:
776: The {\em Chandra} ACIS zero order light curve (Fig. 5) is clearly
777: variable but no large-amplitude flares were detected.
778: A KS test gives a variability probability
779: P$_{var}$ = 0.986. Low-level modulation may
780: be present in the ACIS light curve, as shown in Figure 5.
781: However, the {\em Chandra} observation spanned only
782: 88\% of the 1.19 day rotation period and additional
783: time monitoring over more than one period would be
784: needed to substantiate any periodic X-ray modulation.
785:
786:
787: The {\em Chandra} ACIS spectrum of $\sigma$ Ori E
788: (Fig. 13) is definitely harder than that of
789: $\sigma$ Ori AB, as anticipated from their
790: respective mean photon energies (Table 2).
791: However, the ACIS spectrum is not as hard as
792: seen by {\em XMM} during the flare state and we
793: do not detect the fluorescent Fe line at
794: 6.4 keV. A 2T $apec$ model of the ACIS spectrum
795: gives a marginally acceptable fit with a
796: high-temperature component at kT$_{2}$ = 2.4 keV,
797: which is typical of magnetically-active stars.
798: The fit can be improved by using a 3T model
799: and allowing N$_{\rm H}$ to vary (Table 3 Notes),
800: and this fit is overlaid in Figure 13.
801: The 3T model gives a temperature kT$_{3}$ =
802: 2.6 [2.1 - 3.3; 90\% confidence] keV
803: for the hot component, and the inferred value of
804: N$_{\rm H}$ is consistent with E(B$-$V) = 0.05 (or possibly larger)
805: estimated for this star by Hunger et al. (1989).
806:
807: The stellar luminosity from the blackbody relation
808: is L$_{*}$ =
809: 2.48 $\times$ 10$^{37}$ ergs s$^{-1}$,
810: or log (L$_{*}$/L$_{\odot}$) = 3.81,
811: where we have assumed T$_{eff}$ = 22,500 K and
812: R$_{*}$ = 5.3$^{+1.35}_{-1.1}$ R$_{\odot}$ from Hunger et al. (1989).
813: The unabsorbed flux from the 3T model gives
814: an X-ray luminosity
815: L$_{x}$ (0.5 - 7 keV) = 2.88 $\times$ 10$^{25}$d$_{pc}^2$,
816: where d$_{pc}$ is the distance in parsecs.
817: Adopting d = 640 pc (Hunger et al. 1989) gives
818: log (L$_{x}$/L$_{*}$) = $-$6.32. This ratio is
819: much higher than usually found for B2V stars, for
820: which a catalog of {\em ROSAT} detections
821: gives typical values log (L$_{x}$/L$_{*}$)
822: $\approx$ $-$7.2 to $-$6.8 (Bergh\"{o}fer et al. 1996).
823: In fact, these typical values are too large since the
824: {\em ROSAT} All Sky Survey (RASS) was based on
825: short exposures and thus selected in favor of the detection of
826: X-ray bright B2V stars. A ratio log (L$_{x}$/L$_{*}$)
827: $\approx$ $-$8 may actually be more representative
828: (Cohen, Cassinelli, \& MacFarlane 1997).
829: This strengthens the conclusion
830: that the value of L$_{x}$/L$_{*}$ for $\sigma$
831: Ori E at d = 640 pc is anomalously
832: large for a B2V star. This is a clue that the
833: star may actually lie in the $\sigma$ Ori cluster
834: instead of behind it.
835:
836: If $\sigma$ Ori E lies in the cluster at d $\sim$ 390 pc,
837: then log (L$_{x}$/L$_{*}$) = $-$6.75. This is in better
838: agreement with other B2V stars, but still at the high
839: end of the range. Taking into account the uncertainty in
840: the X-ray flux measurement from the 3T $apec$ fit ($\approx$15\%),
841: the uncertainty in R$_{*}$ (see above), and allowing for a slightly
842: higher temperature T$_{eff}$ = 23,500 adopted in some studies
843: (Nakajima 1985; Groote \& Schmitt 2004), the above ratio
844: could be as low as log (L$_{x}$/L$_{*}$) = $-$7.09.
845: Thus, uncertainties could conspire to decrease the
846: L$_{x}$/L$_{*}$ ratio to a value that is within the
847: range found for X-ray bright {\em ROSAT} B2V star detections.
848:
849: In summary, the L$_{x}$/L$_{*}$ ratio for
850: $\sigma$ Ori E is anomalously high compared to other B2V
851: stars if it lies at d = 640 pc. If the star lies in the cluster
852: at d = 390 pc, the ratio is in better agreement with that of
853: other X-ray luminous B2V stars, but still at the high end of
854: the range if the nominal stellar parameters are adopted.
855: These indications that the L$_{x}$/L$_{*}$ ratio is high
856: and persistent X-ray flaring leave open the possibility that
857: a late-type companion may be contributing to the X-ray emission
858: of $\sigma$ Ori E. However, there have been no reports of
859: the detection of a companion so far.
860:
861:
862: The rather high plasma temperature for this
863: star and its X-ray variability are not consistent
864: with radiative wind shock predictions. The
865: detection of a strong magnetic field gives reason
866: to believe that the magnetically-confined wind shock (MCWS)
867: picture may be relevant. This model is capable of
868: producing higher plasma temperatures than
869: the conventional radiative wind shock scenario
870: and was originally formulated to explain hot
871: plasma in stars with strong $\sim$kG magnetic
872: fields such as the Ap star IQ Aur
873: (Babel \& Montmerle 1997a = BM97a). It has also
874: been invoked to explain hard rotationally-modulated
875: X-ray emission in the young magnetic O star
876: $\theta^1$ Ori C (Babel \& Montmerle 1997b;
877: Donati et al. 2002; Gagn\'{e}
878: et al. 2005; Schulz et al. 2000). The good agreement
879: between model predictions and observed X-ray parameters
880: for $\theta^1$ Ori C found by Gagn\'{e} et al. (2005)
881: provides solid support for the MCWS interpretation.
882:
883: In the MCWS picture, a strong surface magnetic field
884: can channel the wind flow into two oppositely
885: directed streams from each hemisphere, which collide
886: near the magnetic equator and liberate their
887: kinetic energy in shock-heated plasma.
888: The degree to which the wind is confined by the
889: magnetic field is expressed in terms of the
890: confinement parameter
891: $\eta$ = B$_{eq}^2$R$_{*}^2$/$\dot{M}$v$_{\infty}$
892: where B$_{eq}$ is the equatorial magnetic field
893: strength (ud-Doula \& Owocki 2002).
894: For B $\sim$ 2 kG and plausible mass-loss
895: parameters (Groote \& Hunger 1997), one
896: obtains $\eta$ $>>$ 1 for $\sigma$ Ori E.
897: That is, the wind is strongly confined by
898: the magnetic field.
899:
900: Building on earlier work, Townsend, Owocki,
901: \& Groote (2005)
902: extended the MCWS idea into a rigidly rotating
903: magnetosphere model. They showed that
904: by allowing the center of a dipole magnetic
905: field to be displaced away from the center of
906: the star, many aspects of the periodic behavior
907: of $\sigma$ Ori E could be explained. This
908: work has now been expanded into a rigid field
909: hydrodynamics (RFHD) approach by Townsend, Owocki,
910: \& ud-Doula (2007). The RFHD approach is relevant to
911: stars with strong magnetic fields such as $\sigma$ Ori E
912: where the wind has little effect on the field lines.
913:
914: Numerical RFHD simulations of the X-ray emission of
915: a generic star with properties similar to $\sigma$ Ori E
916: by Townsend et al. (2007) predict hot plasma
917: with an emission measure distribution peaking
918: near kT $\approx$ 3.5 keV. This is somewhat
919: hotter than we infer from our best-fit 3T
920: model, which gives kT$_{3}$ = 2.6 [2.1 - 3.3; 90\% conf.]
921: keV. However, as Townsend et al. (2007) have noted,
922: the higher temperature in the RFHD model could be
923: a consequence of uncertainties in the mass loss
924: parameters and neglecting thermal conduction effects.
925: Future refinements of the RFHD
926: model tailored specifically to $\sigma$ Ori E
927: will address these temperature differences
928: and also have the potential to provide
929: theoretical estimates of the X-ray luminosity
930: and X-ray absorption that can be directly compared
931: with observations.
932:
933:
934:
935:
936:
937: \subsubsection{HD 37525}
938:
939: This star is classified as B5V and has a known
940: companion at a separation of 0.$''$45 $\pm$ 0.$''$04
941: (Caballero 2005). The {\em Chandra} ACIS-S spatial resolution
942: is not sufficient to distinguish between the primary
943: and the companion. But, if the companion is a pre-main
944: sequence star (Caballero 2005), then it likely contributes
945: to the X-ray emission.
946:
947: \subsubsection{$\sigma$ Ori IRS1}
948:
949: IRS1 lies 3.$''$3 north-northeast of $\sigma$ Ori AB
950: and was discovered in sub-arcsecond mid-IR observations
951: with the ESO 3.6 m telescope by
952: van Loon \& Oliveira (2003). It shows a clear near to
953: mid-IR excess indicative of a disk. The mid-IR emission
954: was resolved into an elongated structure extending away
955: from $\sigma$ Ori AB that was described as
956: ``fan-shaped'' by van Loon \& Oliveira (2003).
957: The unusual elongated structure and small angular separation
958: between IRS1 and $\sigma$ Ori AB support the conclusion
959: that the two objects are physically associated and that
960: circumstellar material around IRS1 is being
961: influenced by the strong OB system radiation field.
962:
963:
964: Observational evidence for the source now known as IRS1
965: was in fact already present in earlier {\em VLA} radio
966: observations obtained by Drake (1990). No radio source was
967: detected at the optical position of $\sigma$ Ori AB, but he
968: reported a radio source at J2000 position
969: RA = 5$^{\rm h}$ 38$^{\rm m}$ 44.78$^{\rm s}$,
970: decl. = $-$2$^{\circ}$ 35$'$ 57.$''$6. This
971: source lies $\approx$2.$''$7 north-northeast of
972: $\sigma$ Ori AB, but is offset by only $\approx$1$''$ from
973: the IRS1 8.6 $\mu$m peak (Fig. 1 of van Loon \& Oliveira 2003).
974: Thus, an association between the radio source and
975: IRS1 seems likely.
976:
977: The 6 cm flux densities reported by Drake (1990) are
978: much larger than would be expected for free-free wind
979: emission from $\sigma$ Ori AB, and this provides
980: additional support that the radio source is not
981: $\sigma$ Ori AB.
982: Adopting the wind parameters for $\sigma$ Ori AB
983: in Table 1, the predicted 6 cm radio flux density
984: for free-free emission from a spherical fully-ionized
985: wind is S$_{4.86 GHz}$ = 0.1 mJy (eq. [7] of
986: Skinner, Brown, \& Stewart 1993). This value is a
987: factor of 15 - 20 smaller than the observed flux
988: density of the detected radio source (Table II of
989: Drake 1990). The upper limit on the radio flux density
990: measured at the position of $\sigma$ Ori AB in {\em VLA}
991: images is S$_{4.86 GHz}$ $\leq$ 0.3 mJy (S. Drake,
992: private communication), which is consistent
993: with the predicted 0.1 mJy flux density for free-free
994: wind emission.
995:
996:
997: {\em Chandra's} high angular resolution clearly reveals
998: a faint X-ray source (CXO 20) located 2.$''$85 north
999: and 0.$''$93 east of $\sigma$ Ori AB, which we associate
1000: with IRS1 (Fig. 2). Only 43 net counts were detected from
1001: this X-ray source,
1002: but this provides enough information to undertake a rudimentary
1003: spectral analysis. The CCD spectrum (Fig. 14) is absorbed
1004: below $\approx$0.7 keV and peaks at E $\approx$ 1 keV.
1005: Spectral fits imply hotter plasma at kT $>$ 1 keV (Table 3).
1006: The unabsorbed X-ray luminosity of IRS1 is
1007: L$_{x}$ (0.5 - 7 keV) = 2.75 $\times$
1008: 10$^{24}$ d$_{pc}^2$ (ergs s$^{-1}$),
1009: where d$_{pc}$ is the distance in parsecs. Assuming
1010: IRS1 is physically associated with $\sigma$ Ori AB at
1011: a distance of $\sim$352 pc, one obtains
1012: log L$_{x}$ (0.5 - 7 keV) = 29.53 ergs s$^{-1}$.
1013: The KS test gives a high variability probability
1014: P$_{var}$ = 0.95 for IRS1. A coarse light curve
1015: binned at 11000 s intervals (9 bins with $\approx$5 counts/bin)
1016: shows a slow increase in count rate by a factor of $\sim$3
1017: during the last 70 ksec of the observation, but with large
1018: count-rate uncertainties because of the faint emission.
1019: The likely presence of variable X-ray emission from IRS1 at
1020: a plasma temperature of kT $\approx$ 1 keV (and perhaps
1021: higher) demonstrates that the infrared source detected by
1022: van Loon \& Oliveira harbors a magnetically-active young star.
1023:
1024:
1025: The {\em Chandra} COUP study has shown that the X-ray luminosity
1026: of optically-revealed pre-main sequence (PMS) stars in the Orion
1027: nebula correlates with mass, albeit with large scatter
1028: (Preibisch et al. 2005). Assuming a similar correlation
1029: for the $\sigma$ Ori cluster, the above L$_{x}$
1030: along with the mid-IR data would imply that IRS1 is
1031: likely a T Tauri star with a mass of a few tenths of
1032: a solar mass whose disk/envelope is being photo-evaporated
1033: in the harsh UV environment of $\sigma$ Ori AB.
1034:
1035:
1036:
1037:
1038: \subsubsection{Source CXO 8 (J053835.21$-$023438.1)}
1039:
1040: A thermal plasma model requires a very high temperature
1041: for the variable source CXO 8, but a simple absorbed power-law model
1042: also gives an acceptable fit (Table 3). We find no counterpart in
1043: HEASARC extragalactic catalogs but the X-ray source lies at a small
1044: offset from a faint V = 20.88 mag optical source (Mayne et al. 2007).
1045:
1046:
1047:
1048:
1049: \section{Grating Spectrum of $\sigma$ Ori AB}
1050:
1051: \subsection{General Properties}
1052: Figure 15 shows the first-order MEG1 spectrum of
1053: $\sigma$ Ori AB with prominent lines identified.
1054: Line properties are summarized in Table 4.
1055:
1056: In order to test the feasibility of various emission
1057: models (Sec. 5.), we need information on X-ray
1058: temperature, line widths, line centroid shifts
1059: (if any), and flux ratios of specific lines.
1060: Information on line shape is also important,
1061: since radiative wind shock models predict
1062: line asymmetries for optically thick winds.
1063: But at the signal-to-noise ratio
1064: of the MEG1 spectrum, the lines can be
1065: acceptably fitted with Gaussians and we do
1066: not find any conclusive evidence of line asymmetries.
1067:
1068:
1069: The spectrum is clearly dominated by lines
1070: from relatively cool plasma, with
1071: the coolest lines being the O VII He-like
1072: triplet $r$ and $i$ lines and the
1073: O VII line at 18.627 \AA~
1074: (log T$_{max}$ = 6.3 K). The highest
1075: temperature and shortest wavelength line detected was
1076: Si XIII ($\lambda_{lab}$ = 6.6479 \AA;
1077: log T$_{max}$ = 7.0 K).
1078:
1079:
1080: \subsection{Line Widths}
1081: Line fluxes and widths were based on fits with a
1082: weak power-law component plus a Gaussian profile convolved
1083: with the instrumental response. The weak power-law
1084: component accounts for faint emission outside
1085: the line, due either to true continuum or
1086: a pseudo-continuum arising from many faint
1087: unresolved lines.
1088:
1089: The strongest lines
1090: based on observed line flux are O VIII
1091: ($\lambda_{lab}$ = 18.967 \AA) and the O VII
1092: intercombination line ($\lambda_{lab}$ = 21.804 \AA).
1093: In general, all lines for which widths could be
1094: reliably measured are moderately broadened in
1095: excess of thermal (Doppler) widths, the latter being
1096: no greater than $\sim$60 km s$^{-1}$ for
1097: the detected ions at a temperature T $\sim$ 3.5 MK.
1098:
1099: Gaussian fits of nine prominent lines
1100: give a mean half-width at half-maximum
1101: $\overline{\rm HWHM}$ = 264$\pm$80 (1$\sigma$) km s$^{-1}$.
1102: Spectral fits with ISIS give nearly identical results.
1103: Taking the uncertainties
1104: of the line width measurements into account, there is no
1105: clear dependence of line width on wavelength or
1106: maximum line power temperatures T$_{max}$
1107: (Figs. 16 - 17).
1108:
1109:
1110:
1111: The mean HWHM for $\sigma$ Ori AB is much less than observed
1112: for O supergiants (Figure 6 of Cohen et al. 2003),
1113: slightly less than for the O9.5 II star $\delta$ Ori
1114: (HWHM = 430 km s$^{-1}$; Miller et al. 2002),
1115: very similar to that of the B0.2V star $\tau$ Sco
1116: (Cohen et al. 2003), but slightly larger than found
1117: for the B0.5 III star $\beta$ Cru (HWHM $\approx$ 150 km s$^{-1}$;
1118: Cohen et al. 2008). Although there is considerable scatter
1119: in the line widths for any particular star and the sample is
1120: still small, these results suggest tentatively that there
1121: is a trend for broader lines in O stars with stronger winds
1122: versus B stars with weaker winds.
1123: The mean HWHM value for $\sigma$ Ori AB is well below
1124: the terminal wind speed
1125: v$_{\infty}$ $\approx$ 1060 km s$^{-1}$ (Howarth and Prinja 1989).
1126: Expressed as a ratio, the value from individual line fits
1127: gives $\overline{\rm HWHM}$/v$_{\infty}$ = 0.22 - 0.25.
1128: These ratios are comparable to the value
1129: 0.22 found for $\delta$ Ori (Miller et al. 2002).
1130:
1131:
1132: \subsection{Line Centroids}
1133: Gaussian line centroids were measured for 13 stronger lines.
1134: In all cases the
1135: centroids lie within 7 m\AA ~of the laboratory values
1136: and for eight lines the offsets are $\leq$3 m\AA,
1137: as shown in Figure 18.
1138: These offsets are well within the $\pm$11 m\AA ~absolute wavelength
1139: calibration accuracy of MEG1
1140: \footnote{{\em Chandra} Proposer's Observatory Guide:
1141: http://asc.harvard.edu/proposer/POG/ .}.
1142: Both positive and negative offsets were found with
1143: ten lines blueshifted and three redshifted.
1144: The mean centroid offset was
1145: $-$1.6 $\pm$ 3.5 ($\pm$1$\sigma$) m\AA.~
1146: In only two cases does $\lambda_{lab}$ lie outside
1147: the $\pm$1$\sigma$ uncertainty interval of the
1148: measured centroid. The centroid measurements of
1149: these two lines are uncertain due to close blends
1150: (Ne X 12.1321/12.1375 \AA)
1151: and possible contamination by weak nearby Fe lines
1152: (Ne IX 13.553 \AA).
1153:
1154:
1155:
1156: \subsection{Helium-like Ions}
1157: He-like triplets show strong resonance (r) and
1158: intercombination (i) lines, but the forbidden (f)
1159: lines are very faint and formally undetected.
1160: The resonance and intercombination lines were detected in
1161: the Mg XI (Fig. 19), Ne IX (Fig. 20), and O VII (Fig. 21) He-like
1162: triplets. The Si XIII resonance line was detected but
1163: the emission in the triplet is so faint that no reliable
1164: diagnostic information can be obtained apart from an
1165: estimate of the resonance line flux. The strong i lines
1166: and faint or absent f lines are a clear indication that
1167: the f and i line fluxes are being affected by the strong
1168: UV radiation field of $\sigma$ Ori A (T$_{eff}$ =
1169: 33,000 K; Howarth \& Prinja 1989).
1170:
1171: In general, the
1172: line flux ratio $R$ = f/i is sensitive to both
1173: electron density $n$ and the UV radiation field
1174: (Gabriel \& Jordan 1969).
1175: If $\phi$ is the photoexcitation rate
1176: from the $^{3}$S$_{1}$ level to the
1177: $^{3}$P$_{2,1}$ levels, then
1178: $R$ = $R_{0}$/[1 $+$ ($\phi$/$\phi_{c}$) $+$ ($n$/$n_{c}$)], where
1179: $\phi_{c}$ and $n_{c}$ are the critical photoexcitation
1180: rate and critical density. The quantities $R_{0}$, $\phi_{c}$,
1181: and $n_{c}$ (Table 5) depend only on atomic parameters and the electron
1182: temperature of the source, as defined in
1183: Blumenthal, Drake, \& Tucker 1972, hereafter BDT72).
1184:
1185: In hot star winds, densities $n$ $>$ $n_{c}$ high
1186: enough to affect $R$ would only occur very close to the
1187: star and it is thus reasonable to assume that any
1188: observed reduction in $R$ relative to $R_{0}$ is the result
1189: of photoexcitation by the hot star UV radiation
1190: field. Thus, in hot stars the value of $R$ is not a
1191: reliable density diagnostic but is instead a measure of
1192: the photoexcitation rate in the line-emitting region,
1193: and can be used along with atmospheric models to
1194: constrain the distance from the star at which the
1195: He-like triplet line is formed.
1196: In cool stars with negligible UV flux, the value
1197: of $R$ can be used to constrain the density.
1198: In the limit of low photoexcitation
1199: ($\phi$ $<<$ $\phi_{c}$) and low
1200: density ($n$ $<<$ $n_{c}$),
1201: ~$R$ $\rightarrow$ $R_{0}$.
1202:
1203: For the He-like ions detected in the MEG spectrum of
1204: $\sigma$ Ori AB, the most stringent limit on the
1205: line formation radius based on the measured upper
1206: limit of $R$ comes from the Mg XI triplet.
1207: We have used a solar abundance TLUSTY model atmosphere
1208: (Lanz \& Hubeny 2007) with T$_{eff}$ = 32,500 and
1209: log $g$ = 4.0, along with the formalism in BDT72
1210: and Leutenegger et al. (2006), taking
1211: an average of the photospheric fluxes of the two
1212: $^3$S - $^3$P$_{1,2}$ transitions (997.5 \AA~ and
1213: 1034.3 \AA~ for Mg XI). We have averaged these fluxes
1214: over $\approx$3 \AA~ blueward of these wavelengths to
1215: account for the Doppler shift in the wind (Leutenegger et al.
1216: 2006). We find an expression for the ratio $R$(r) = f/i
1217: as a function of the distance r from the center of the
1218: star, which takes into account the dilution factor W(r)
1219: and is plotted in Figure 22.
1220:
1221:
1222: For Mg XI our spectral fits give a 90\% confidence upper
1223: limit $R$ = f/i $\leq$ 0.17 (Table 5).
1224: As can be seen in Figure 22, this upper
1225: limit implies a small line formation
1226: radius r $\leq$ 1.8 R$_{*}$, where as usual, r is measured
1227: from the center of the star. A similar analysis for Ne IX
1228: gives $R$ $\leq$ 0.15 and a formation radius r $\leq$ 4.4 R$_{*}$.
1229: The O VII triplet does not provide a useful constraint
1230: on the line formation radius.
1231: Using the adopted mass-loss parameters in Table 1 and
1232: an estimated wind X-ray opacity $\kappa_{w}$ $\approx$ 80
1233: cm$^{2}$ g$^{-1}$ near E $\approx$ 1 keV
1234: (Waldron et al. 1998; Cohen et al. 1996;
1235: Ignace, Oskinova, \& Foullon 2000),
1236: the wind is expected to be optically thin at E $\approx$ 1 keV
1237: down to the stellar surface. In addition, our modeling of
1238: the Fe XVII line (15.014 \AA~) confirms that the wind
1239: optical depth is consistent with zero.
1240: Thus, X-rays formed at the small
1241: radii given above could escape through the wind.
1242:
1243:
1244: \subsection{Abundances}
1245:
1246: We have simultaneously fitted the zero-order spectrum and
1247: the MEG1 and HEG1 first order spectra of $\sigma$ Ori AB
1248: to search for significant deviations from solar abundances.
1249: These fits were conducted using an absorbed 2T $vapec$ model
1250: with a modified version of
1251: ISIS (Cohen et al. 2008) that allows the R and
1252: G ratios of He-like triplets
1253: to be fitted as free parameters. We allowed the
1254: abundances of elements with prominent lines in the spectrum
1255: to vary, namely O, Ne, Mg, Si, and Fe. We found
1256: no significant deviations from solar abundances. The
1257: largest deviation found was for magnesium, which gave
1258: a best-fit abundance Mg = 0.78 [0.62 - 1.08; 90\% conf.]
1259: $\times$ solar.
1260:
1261:
1262: \subsection{Summary of Grating Spectra Results}
1263:
1264: In summary, the key properties determined from the
1265: spectral analysis of $\sigma$ Ori AB are:
1266: (i) dominant cool plasma (kT $\approx$ 0.3 keV),
1267: (ii) moderate line broadening
1268: (HWHM $\approx$ 264 km s$^{-1}$)
1269: that is a factor of $\sim$4 below the terminal wind speed of $\sigma$ Ori,
1270: (iii) no significant mean line centroid shifts relative to laboratory
1271: values, (iv) strong resonance and intercombination lines and
1272: very faint (formally
1273: undetected) forbidden lines with low f/i flux ratios,
1274: indicative of strong UV field
1275: excitation effects, and (v) Mg XI line formation within a few
1276: radii of the star (r $\ltsimeq$ 1.8 R$_{*}$).
1277:
1278:
1279: \section{Interpretation of the X-ray Emission from $\sigma$ Ori AB}
1280:
1281: \subsection{Radiative Wind Shocks}
1282:
1283: The temperature of the cool plasma that
1284: dominates the X-ray emission of $\sigma$
1285: Ori AB is in good agreement with model predictions for
1286: shocks distributed in a radiatively-driven stellar wind.
1287: Numerical radiation hydrodynamics simulations
1288: of shocks in radiatively-driven winds give typical shock jumps
1289: $\Delta$V = 500 - 1000 km s$^{-1}$ and post-shock temperatures
1290: in the range $\sim$1 - 10 MK (Owocki, Castor, \& Rybicki 1988). This
1291: range can easily accommodate $\sigma$ Ori AB, whose emission-measure
1292: weighted temperature from the best-fit 2T $apec$ model
1293: is T $\sim$ 3.5 MK.
1294:
1295: The predicted post-shock X-ray temperature corresponding to a shock jump
1296: velocity $\Delta$V is T$_{x}$ = 1.4 $\times$ 10$^{5}$
1297: ($\Delta$V/100 km s$^{-1}$)$^2$~K ~(Krolik \& Raymond 1985).
1298: For $\sigma$ Ori AB we have T $\sim$
1299: 3.5 $\times$ 10$^{6}$ K (kT = 0.3 keV) so the required shock jump is
1300: $\Delta$V $\approx$ 500 km s$^{-1}$, or about half the
1301: terminal wind speed.
1302:
1303:
1304: Also, the emission measure (EM) in the wind is sufficient to
1305: account for the observed X-ray emission measure.
1306: The wind emission measure outside a given radius R$_{o}$ is
1307: EM$_{wind}(r > R_{o}) = \int_{R_{o}}^{\infty}n_{e}n_{H} 4 \pi r^2 dr$,
1308: where $n_{e}$ and $n_{H}$ are the electron and hydrogen
1309: number densities. Using the mass loss parameters
1310: for $\sigma$ Ori AB in Table 1 and assuming a solar-abundance
1311: constant-velocity wind with v(r) = v$_{\infty}$ we have
1312: EM$_{wind}$(r$>$R$_{o}$) = 5.5 $\times$ 10$^{55}$/R$_{o}$ cm$^{-3}$,
1313: where R$_{o}$ is in units of R$_{*}$ = 9 R$_{\odot}$.
1314: By comparison, the total X-ray emission
1315: measure determined from the best-fit 2T $apec$ model
1316: is EM$_{X}$ = 1.8 $\times$ 10$^{54}$ cm$^{-3}$.
1317: To compute this value, we have converted the normalization
1318: factors ($norm$) for both temperature components in Table 3 to
1319: equivalent emission measures and summed them, using
1320: the conversion EM = 10$^{14}$4$\pi$d$_{cm}^2$$norm$,
1321: where d$_{cm}$ is the stellar distance in cm. Adopting
1322: R$_{o}$ = 1.5 as a representative example gives an inferred
1323: ratio of X-ray EM to wind EM of $\approx$5\%.
1324: Thus, the emission measure of the X-ray emitting plasma
1325: relative to that of the wind is low.
1326:
1327: The grating spectra provide clues that a significant
1328: fraction of the X-ray emission originates
1329: close to the star in the region where the wind is still
1330: accelerating. The relatively narrow lines and
1331: absence of any significant centroid
1332: shifts are good evidence that the bulk of the X-ray
1333: plasma is slowly moving and does not form at
1334: large radii where the wind
1335: has reached terminal speed. Also, the very weak
1336: (formally undetected) forbidden lines in He-like triplets
1337: imply line formation close to the star.
1338: As already noted, the most stringent constraint on
1339: the location of the X-ray emitting plasma comes from
1340: the Mg XI triplet, for which we infer an upper
1341: limit on the line formation radius r $<$ 1.8 R$_{*}$
1342: (Fig. 22). Assuming a standard wind acceleration
1343: law v(r) = v$_{\infty}$[1 - (R$_{*}$/r)]$^{\beta}$
1344: and v$_{\infty}$ = 1060 km s$^{-1}$ (Table 1),
1345: the above radius is within the acceleration zone
1346: for typical hot star wind values $\beta$ $\approx$
1347: 0.5 - 1.0. A recent analysis of He-like triplet
1348: line ratios in O stars by Leutenegger et al. (2006)
1349: concludes that minimum line formation radii are
1350: typically 1.25 $<$ r/R$_{*}$ $<$ 1.67 and that
1351: these small radii do not preclude the formation of
1352: strong shocks.
1353:
1354: A key question that remains to be answered is whether
1355: numerical simulations of the line-driven instability mechanism
1356: can reproduce the narrow line widths observed for
1357: $\sigma$ Ori AB and some other late O and early B stars.
1358: As already noted (Sec. 4.2) the line widths observed for
1359: $\sigma$ Ori AB have HWHM values equivalent to about one-fourth
1360: v$_{\infty}$. But larger widths would be in better accord
1361: with numerical simulations, which tend to show shock onset
1362: radii of about 1.5 R$_{*}$ and velocities of shock-heated material
1363: that are an appreciable fraction of v$_{\infty}$
1364: (e.g. Feldmeier, Puls, \& Pauldrach 1997; Runacres \& Owocki 2002).
1365: Even so, it should be kept in mind that predicted line properties
1366: from simulations are sensitive to assumptions about wind properties
1367: and the wind velocity profile close to the star, which are
1368: uncertain for OB stars in general. A more detailed discussion of
1369: the issues related to line widths and profiles in non-magnetic OB stars
1370: in the framework of radiative wind shock models will be presented
1371: in a future paper (D. Cohen et al., in preparation).
1372:
1373:
1374:
1375: \subsection{Magnetically-Confined Wind Shocks}
1376:
1377: As already mentioned in the discussion of
1378: $\sigma$ Ori E (Sec. 3.7.2), a sufficiently strong surface
1379: magnetic field can channel the wind of a hot star
1380: into two streams, producing X-ray emission from
1381: a magnetically-confined
1382: wind shock near the magnetic equator. To our knowledge,
1383: there have been no published reports of a magnetic field detection
1384: on $\sigma$ Ori AB. But, an upper limit on its line-strength
1385: weighted magnetic field of
1386: B$_{l}$ $\leq$ $-$140 ($\pm$260) G. was obtained
1387: in the longer of the two observations reported by
1388: Landstreet (1982). Since the upper limit on
1389: B$_{l}$ has a large uncertainty and this star is
1390: young with a rather uncertain age
1391: (Table 1 of Caballero 2007), it could still retain
1392: a weak undetected fossil magnetic field. We thus briefly consider
1393: the MCWS mechanism.
1394:
1395:
1396: Using the relation for the wind confinement
1397: parameter $\eta$ (Sec. 3.7.2)
1398: and the adopted stellar parameters for
1399: $\sigma$ Ori A (Table 1), the field strength
1400: for critical confinement ($\eta$ = 1) is
1401: B$_{eq}$ $\approx$ 70 G. Thus, even
1402: a relatively weak field could perturb the
1403: wind of $\sigma$ Ori A but a field of this strength
1404: would be difficult to detect in a hot star.
1405:
1406:
1407: Analytic expressions for the X-ray temperature
1408: and luminosity of a magnetically-confined
1409: wind shock were derived by BM97a.
1410: The predicted temperature at the shock front
1411: is T$_{shock}$ = 1.13 $\times$ 10$^5$v$_{100}^2$ K,
1412: where v$_{100}$ is the pre-shock wind speed
1413: in units of 100 km s$^{-1}$. In order to
1414: reconcile the MCWS picture with the relatively
1415: cool emission at T $\sim$ 3.5 MK
1416: observed for $\sigma$ Ori AB, one would need to postulate
1417: a pre-shock wind speed of v $\sim$ 550 km s$^{-1}$.
1418: This is about half the terminal wind speed and
1419: would require that the closed field lines exist
1420: only in the wind acceleration zone near the star
1421: (r $\ltsimeq$ 2 R$_{*}$ for values of
1422: $\beta$ typical of hot star winds).
1423: Intuitively, this is what one would expect in the
1424: weak confinement regime.
1425:
1426:
1427: The predicted X-ray luminosity from a
1428: magnetically-confined wind shock in the BM97a
1429: model is L$_{x}$ $\approx$
1430: 2.6 $\times$ 10$^{30}$B$_{kG}^{0.4}$~$\xi$ ergs s$^{-1}$.
1431: Here, B$_{kG}$ is the field strength in kG and
1432: $\xi$ is a free parameter of the theory that depends
1433: on $\dot{M}$ and v$_{\infty}$. For critical
1434: confinement (B $\sim$ 70 G.), this relation
1435: requires $\xi$ $\sim$ 35 to account for the
1436: observed L$_{x}$ (Table 1). This value of
1437: $\xi$ is within the (broad) range allowed by the
1438: theory and the adopted mass-loss parameters
1439: of $\sigma$ Ori AB.
1440:
1441:
1442: Based on the above comparisons, we conclude that the
1443: MCWS model could plausibly account for the observed
1444: X-ray properties of $\sigma$ Ori AB if the magnetic field is
1445: weak and only able to close the field lines near the star.
1446: Detailed MCWS simulations would be needed to test this idea.
1447: The justification for invoking the MCWS model is not strong
1448: in the absence of a magnetic field detection or any documented
1449: H$\alpha$ or UV wind line modulation.
1450:
1451:
1452:
1453: \subsection{Colliding Wind Shocks}
1454:
1455: The presence of a B0.5V companion at a separation
1456: 0f 0.$''$25 from $\sigma$ Ori A raises the possibility
1457: that X-rays could be produced in a colliding wind shock
1458: region between the two stars. However, the predicted
1459: maximum X-ray temperature for a colliding wind
1460: shock in the case of solar abundances is
1461: kT$_{cw}$ $\approx$ 1.17 (v$_{\perp,1000}$)$^2$ keV,
1462: where v$_{\perp,1000}$ is the velocity of the
1463: wind component perpendicular to the shock
1464: interface in units of 1000 km s$^{-1}$
1465: (Stevens, Blondin, \& Pollock 1992). Setting
1466: v$_{\perp}$ equal to the terminal wind
1467: speed of $\sigma$ Ori AB gives
1468: kT$_{cw}$ $\approx$ 1.3 keV, much higher than observed.
1469:
1470: Furthermore, wind momentum balance considerations
1471: predict that the shock contact discontinuity
1472: along the line-of-centers between the binary
1473: components will lie at least
1474: $\sim$1000 R$_{O-star}$ from $\sigma$ Ori A.
1475: This equates to about half the distance between the
1476: A and B components. To arrive at this estimate,
1477: we have assumed that the winds have reached
1478: terminal speeds and have used eq. [1] of
1479: Stevens, Blondin, \& Pollock (1992), along
1480: with the mass-loss parameters in Table 1 and
1481: the assumed B0.5V star wind parameters
1482: log $\dot{M}_{B-star}$ $\sim$ $-$7.4 to $-$8.0
1483: (M$_{\odot}$ yr$^{-1}$) and v$_{\infty,B-star}$
1484: $\sim$ 2000 km s$^{-1}$ (Abbott 1982;
1485: Cassinelli et al. 1994). X-ray formation at
1486: such large distances is ruled out by the
1487: weak or absent forbidden lines in He-like ions,
1488: which for Mg XI demonstrate that the bulk of the X-rays
1489: are formed within a few radii of the O star.
1490:
1491: Based on the above estimates, a colliding wind origin
1492: for the X-rays seems very unlikely. But, before
1493: ruling out this mechanism it is important to note
1494: that there have been several reports that $\sigma$ Ori A
1495: is itself a double-lined spectroscopic binary
1496: (e.g. Bolton 1974; Morrell \& Levato 1991), which if
1497: true would make $\sigma$ Ori AB a triple system.
1498: This has also been
1499: mentioned by Caballero (2008) as a possible means of
1500: bringing the derived distance of $\sigma$ Ori AB into better
1501: agreement with that of the cluster.
1502:
1503:
1504: If the A component
1505: is indeed a spectroscopic binary then the wind of the more
1506: massive primary could be shocking onto the wind of
1507: the spectroscopic companion. From above, the wind speed needed to
1508: produce the kT $\approx$ 0.3 keV plasma detected
1509: by {\em Chandra} is v$_{wind}$ $\approx$
1510: 500 km s$^{-1}$. This would imply that the
1511: winds are shocking at sub-terminal speeds.
1512: For a $\beta$ = 1 velocity law, this speed
1513: could be achieved at r $\approx$ 1.9 R$_{*}$
1514: and for $\beta$ = 0.8 the computed radius
1515: is r $\approx$ 1.6 R$_{*}$. Considering our
1516: imprecise knowledge of the wind velocity
1517: law close to the star, these values are compatible
1518: with the upper limit r $\leq$ 1.8 R$_{*}$ obtained
1519: from Mg XI line flux ratios (Sec. 4.4).
1520:
1521: A close colliding wind system consisting of
1522: $\sigma$ Ori A and a spectroscopic companion
1523: might resemble close spectroscopic binaries
1524: such as HD 150136. This massive O3 $+$ O6V
1525: SB2 system has a short-period 2.66 day orbit
1526: (Niemela \& Gamen 2005). A sensitive {\em Chandra}
1527: HETG observation detected HD 150136 but
1528: the O3 $+$ O6V SB2 system could not be spatially
1529: resolved (Skinner et al. 2005). The X-ray
1530: spectrum of HD 150136 is soft with a
1531: characteristic temperature kT $\approx$ 0.3 keV,
1532: remarkably similar to that detected
1533: here in $\sigma$ Ori AB. Because of the close
1534: separation of the two components in HD 150136,
1535: the winds would be shocking at sub-terminal
1536: speeds and radiative cooling in the shock
1537: region is important (Skinner et al. 2005).
1538: Thus, the predicted temperature for a colliding
1539: wind shock is lower than for an equivalent
1540: adiabatic system at wide separation.
1541: If the analogy between the
1542: O3 $+$ O6V spectroscopic binary HD 150136 and
1543: $\sigma$ Ori A is indeed relevant, then an early B
1544: star may be playing the role of the spectroscopic companion
1545: to $\sigma$ Ori A (Caballero 2008).
1546:
1547:
1548: If $\sigma$ Ori A does have a spectroscopic companion,
1549: then orbitally-modulated X-ray emission could be present
1550: if the X-rays form in a colliding wind shock. But,
1551: the presence of orbital X-ray modulation depends critically
1552: on the orbital eccentricity and inclination, for which no
1553: information is currently available. We see no clear evidence for
1554: X-ray modulation in the $\sim$1 day {\em Chandra} light
1555: curve (Fig. 4). More specific information on the properties of the
1556: putative spectroscopic companion and its orbit would be
1557: needed to make further comparisons between colliding wind
1558: shock theory and the X-ray data.
1559:
1560:
1561:
1562: \subsection{Coronal Models and Late-Type Companions}
1563:
1564: The existence of a thin X-ray emitting coronal zone
1565: at the base of the wind was invoked by Cassinelli \& Olson (1979)
1566: to explain strong O VI, N V, and C IV lines
1567: in some OB supergiants and Of stars. Waldron \& Cassinelli (2001)
1568: have also proposed a composite model involving both wind shocks
1569: and confinement of dense plasma in magnetic loops close to the star
1570: to explain the broad emission lines detected in the O9.7 Ib star
1571: $\zeta$ Ori, but Cohen et al. (2006) arrive at a different
1572: interpretation that does not invoke magnetic confinement.
1573: Magnetic loop confinement close to the star strongly resembles
1574: coronal models for late-type stars.
1575:
1576:
1577: But the
1578: coronal model has not gained wide acceptance
1579: due to the lack of a solid theoretical basis
1580: for explaining how strong outer convection zones
1581: would be produced in hot stars. Such convection
1582: zones are thought to be essential for generating
1583: self-sustained magnetic fields in the Sun and
1584: other late-type stars. Even so, the detection of
1585: strong $\sim$kG magnetic fields in some very young O-type
1586: stars such as $\theta^1$ Ori C (Donati et al. 2002)
1587: does justify a brief
1588: examination of the possibility of coronal emission,
1589: either from $\sigma$ Ori AB or an unseen late-type
1590: companion.
1591:
1592: Because of the young age of $\sigma$ Ori AB
1593: ($\sim$3 Myr), a coeval unseen late-type companion
1594: would likely be a T Tauri star and thus a potential
1595: X-ray emitter. However, adaptive optics H-band images
1596: with FWHM $\approx$ 0.$''$2 - 0.$''$4 (Caballero 2005)
1597: have only detected IRS1, which at a separation of
1598: $\approx$3$''$ is also revealed by {\em Chandra}.
1599: The best adaptive optics resolution of
1600: FWHM $\approx$ 0.$''$2 probes down to a linear
1601: separation of $\sim$70 AU, so any as yet
1602: undetected T Tauri star would presumably lie
1603: very close to $\sigma$ Ori A.
1604:
1605: Most T Tauri stars detected in the deep {\em Chandra}
1606: Orion COUP survey show a hot X-ray
1607: component at kT $>$ 1 keV (Preibisch et al. 2005).
1608: We do not detect such a component in the X-ray
1609: spectrum of $\sigma$ Ori AB, so the observed
1610: temperature structure provides no hint for
1611: a T Tauri companion.
1612: Impulsive X-ray flaring or moderate amplitude
1613: variability on short timescales is often (but not
1614: always) detected in TTS but no such variability
1615: has so far been detected in $\sigma$ Ori AB.
1616: Finally, at the spectral resolution of {\em Chandra's}
1617: HETG, emission lines of coronal late-type stars such
1618: as Capella (Ness et al. 2003) and of TTS
1619: such as SU Aur (Smith et al. 2005) and
1620: HDE 245059 (Baldovin Saavedra et al. 2007) show
1621: narrow lines with no broadening beyond
1622: thermal widths, in contrast to the broadened
1623: lines detected here in $\sigma$ Ori AB.
1624: Thus, we find no strong
1625: reason to invoke a coronal interpretation for
1626: the X-ray emission of $\sigma$ Ori AB.
1627:
1628:
1629:
1630: \section{Conclusions}
1631:
1632: The main conclusions of this study are the following:
1633:
1634: \begin{enumerate}
1635:
1636: \item {\em Chandra} detected 42 X-ray sources on the ACIS-S3
1637: CCD in the central region of the $\sigma$ Ori cluster.
1638: Three are known high-mass OB stars and the remainder have
1639: X-ray and near-IR properties consistent with
1640: magnetically-active K or M stars. Nineteen of the
1641: X-ray sources have K or M spectral types determined
1642: from previous studies, and about half of these
1643: show near-IR excesses based on {\em 2MASS} colors.
1644:
1645: \item The X-ray emission associated with IRS1 is variable
1646: and shows a high-temperature (kT $\gtsimeq$ 1 keV)
1647: component. Its X-ray properties and elongated
1648: mid-IR morphology suggest that it is
1649: a low-mass magnetically-active T Tauri star whose
1650: disk/envelope is being photo-evaporated by
1651: by $\sigma$ Ori AB.
1652:
1653: \item The X-ray emission of $\sigma$ Ori E is variable and
1654: a hard component (kT $\approx$ 2.6 keV) is detected,
1655: as typical of magnetically-active stars.
1656: Its ratio L$_{x}$/L$_{*}$ is unusually high for a B2V
1657: star assuming the nominal distance of 640 pc, suggesting
1658: that it could lie closer. Numerical magnetosphere models
1659: based on the magnetically confined wind picture
1660: predict hot plasma at
1661: temperatures similar to (but slightly higher than) observed
1662: and can also explain many aspects of the variability of
1663: this star (Townsend et al. 2005, 2007). Future numerical
1664: models offer the potential to provide estimates of X-ray
1665: luminosity and X-ray absorption that
1666: can be directly compared with observations.
1667:
1668: \item The X-ray emission of the OB binary system $\sigma$ Ori AB is
1669: exceptional, being the most luminous X-ray source in
1670: the field and the softest.
1671: Its low X-ray temperature (kT $\approx$ 0.3 keV)
1672: and moderately-broadened emission lines (HWHM $\approx$
1673: 264 km s$^{-1}$) clearly argue against a coronal
1674: interpretation and point to a shock origin. Overall,
1675: its X-ray properties are quite similar to some other
1676: late O and early B stars for which X-ray grating spectra
1677: show relatively cool
1678: plasma and slightly broadened unshifted symmetric emission lines.
1679:
1680: \item Radiative wind shock predictions for an optically thin wind
1681: are consistent with the cool plasma and symmetric unshifted
1682: emission lines of $\sigma$ Ori AB. However, the rather narrow
1683: line widths and low f/i ratios imply X-ray formation in slowly-moving
1684: plasma in the wind acceleration zone close to the star.
1685: It remains to be seen whether future numerical simulations of the
1686: line-driven instability mechanism can reproduce the narrow lines.
1687:
1688:
1689:
1690: \item Other possible emission mechanisms for $\sigma$ Ori AB that are
1691: not yet totally ruled out but remain speculative are:
1692: (i) a magnetically confined wind shock in the weak-field
1693: limit, and (ii) a sub-terminal speed colliding wind system if
1694: $\sigma$ Ori A does have a spectroscopic companion.
1695: The MCWS scenario lacks solid observational support in the absence
1696: of any evidence for a magnetic field in $\sigma$ Ori AB.
1697: More complete information about the properties of the putative
1698: spectroscopic companion and its orbit will be needed to make
1699: meaningful comparisons of colliding wind theory with observations.
1700:
1701:
1702:
1703: \end{enumerate}
1704:
1705:
1706: \acknowledgments
1707:
1708: We thank Ms. Victoria Swisher for assistance with data
1709: analysis and Dr. Stephen Drake for providing information on
1710: {\em VLA} observations of $\sigma$ Ori.
1711: This research was supported by grants
1712: GO5-6009X (SAO) and NNG05GE69G (NASA GSFC).
1713:
1714:
1715: \clearpage
1716:
1717:
1718:
1719: % Now comes the reference list. In this document, we used \cite to call
1720: % out citations, so we must use \bibitem in the reference list, which
1721: % means we use the LaTeX thebibliography environment. Please note that
1722: % \begin{thebibliography} is followed by a null argument. If you forget
1723: % this, mayhem ensues, and LaTeX will say "Perhaps a missing item?" when
1724: % you run it. Do not call us, do not send mail when this happens. Put
1725: % the silly {} after the \begin{thebibliography}.
1726: %
1727: % Each reference has a \bibitem command to define the citation format
1728: % to be placed in the text (in []) and the symbolic tag used for
1729: % cross referencing (in {}).
1730: %
1731: % See sample1.tex, or the AASTeX guide, for an alternative to the \cite-
1732: % \bibitem command.
1733:
1734: \begin{thebibliography}{}
1735: \bibitem[]{} Abbott, D.C. 1982, \apj, 259, 282
1736: \bibitem[]{} Anders, E. \& Grevesse, N. 1989, \gca, 53, 197
1737: \bibitem[]{} Baldovin Saavedra, C., Audard, M., Duch\^{e}ne, G.,
1738: G\"{u}del, M., Skinner, S.L., \& Paerels, F.B.S. 2007,
1739: BAAS, 39 (4), 851
1740: \bibitem[]{} Babel, J. \& Montmerle, T. 1997a, \aap, 323, 121 (BM97a)
1741: \bibitem[]{} Babel, J. \& Montmerle, T. 1997b, \aap, 485, L29
1742: \bibitem[]{} Bergh\"{o}fer, T.W. \& Schmitt, J.H.M.M. 1994, \aap, 290, 435
1743: \bibitem[]{} Bergh\"{o}fer, T.W. \& Schmitt, J.H.M.M., \& Cassinelli, J.P.
1744: 1996, A\&AS, 118, 481
1745: \bibitem[]{} Bessell, M.A. \& Brett, J.M. 1988, \pasp, 100, 1134
1746: \bibitem[]{} Blumenthal, G.R., Drake, G.W.F., \& Tucker, W.H. 1972,
1747: \apj, 172, 205 (BDT72)
1748: \bibitem[]{} Bolton, C.T. 1974, \apj, 192, L7
1749: \bibitem[]{} Borra, E.F. \& Landstreet, J.D. 1979, \apj, 228, 809
1750: \bibitem[]{} Brandt, W.N. et al. 2001, \aj, 122, 2810
1751: \bibitem[]{} Caballero, J.A. 2005, Astron. Nachr., 326, 1007
1752: \bibitem[]{} Caballero, J.A. 2007a, \aap, 466, 917
1753: \bibitem[]{} Caballero, J.A. 2007b, Astron. Nachr., 328 (No. 3-4), 1
1754: \bibitem[]{} Caballero, J.A. 2008, \mnras, 383, 750
1755: \bibitem[]{} Cassinelli, J.P., Cohen, D.H., MacFarlane, J.J.,
1756: Sanders, W.T., \& Welsh, B.Y. 1994, \apj, 421, 705
1757: \bibitem[]{} Cassinelli, J.P., Miller, N.A., Waldron, W.A.,
1758: MacFarlane, J.J., \& Cohen, D.H. 2001, \apj, 554, L55
1759: \bibitem[]{} Cassinelli, J.P. \& Olson, G.L. 1979, \apj, 229, 304
1760: \bibitem[]{} Cohen, D.H., Cassinelli, J.P., \& Macfarlane, J.J. 1997,
1761: \apj, 487, 867
1762: \bibitem[]{} Cohen, D.H., Cooper, R.G., Macfarlane, J.J., Owocki, S.P.,
1763: Cassinelli, J.P., \& Wang, P. 1996, \apj, 460, 506
1764: \bibitem[]{} Cohen, D.H., de Messi\`{e}res, G.E., MacFarlane, J.J.,
1765: Miller, N.A., Cassinelli, J.P., Owocki, S.P., \&
1766: Liedahl, D.A. 2003, \apj, 586, 495
1767: \bibitem[]{} Cohen, D.H., Kuhn, M.A., Gagn\'{e}, M., Jensen, E.L.N.,
1768: \& Miller, N.A. 2008, \mnras, in press (arXiv:0802.4084v1)
1769: \bibitem[]{} Cohen, D.H., Leutenegger, M.A., Grizzard, K.T.,
1770: Reed, C.L., Kramer, R.H., \& Owocki, S.P. 2006,
1771: \mnras, 368, 1905
1772: \bibitem[]{} Cowie, L.L., Garmire, G.P., Bautz, M.W., Barger, A.J.,
1773: Brandt, W.N., \& Hornschemeier, A.E. 2002,
1774: \apj, 566, L5
1775: %%\bibitem[]{} Czesla, S. \& Schmitt, J.H.H.M. 2007, \aap, 465, 493
1776: \bibitem[]{} Donati, J.-F., Babel, J., Harries, T.J., Howarth, I.D.,
1777: Petit, P., \& Semel, M. 2002, \mnras, 333, 55
1778: \bibitem[]{} Drake, S.A. 1990, \aj, 100, 572
1779: \bibitem[]{} Edwards, T.W. 1976, \aj, 81, 245
1780: \bibitem[]{} Favata, F. et al. 2005, \apjs, 160, 469
1781: %%\bibitem[]{} Feigelson, E.D., Broos, P., Gaffney, J.A. III,
1782: %% Garmire, G., Hillenbrand, L.A., Pravdo, S.H., Townsley, L.,
1783: %% \& Tsuboi, Y. 2002, \apj, 574, 258
1784: \bibitem[]{} Feldmeier, A., Kudritzki. R.-P., Palsa, R.
1785: Pauldrach, A.W.A., \& Puls, J. 1997, \aap, 320, 899
1786: \bibitem[]{} Feldmeier, A., Puls, J., \& Pauldrach, A.W.A. 1997,
1787: \aap, 322, 878
1788: \bibitem[]{} Flaccomio, E. et al. 2005, \apjs, 160, 450
1789: \bibitem[]{} Franciosini, E., Pallavicini, R., \& Sanz-Forcada, J.
1790: 2006, \aap, 446, 501 (FPS06)
1791: \bibitem[]{} Gabriel, A.H. \& Jordan, C. 1969, \mnras, 145, 241
1792: \bibitem[]{} Gagn\'{e}, M. et al. 2005, \apj, 628, 986
1793: \bibitem[]{} Gatti, T., Natta, A., Randich, S., Testi, L., \& Sacco, G.
1794: 2008, \aap, in press (arXiv:0801.2886)
1795: \bibitem[]{} Getman, K.V. et al. 2005, \apjs, 160, 319
1796: \bibitem[]{} Gorenstein, P. 1975, \apj, 198, 95
1797: \bibitem[]{} Groote, D. \& Hunger, K. 1997, \aap, 319, 250
1798: \bibitem[]{} Groote, D. \& Schmitt, J.H.M.M. 2004, \aap, 418, 235
1799: \bibitem[]{} Hern\'{a}ndez, J. et al. 2007 \apj, 662, 1067
1800: \bibitem[]{} Howarth, I.D. \& Prinja, R.K. 1989, \apjs, 69, 527
1801: %%\bibitem[]{} Howk, J.C., Cassinelli, J.P., Bjorkman, J.E.,
1802: %% \& Lamers, H.J.G.L.M. 2000, \apj, 534, 348
1803: \bibitem[]{} Hunger, K., Heber, U., \& Groote, D. 1989, \aap, 224, 57
1804: \bibitem[]{} Ignace, R., Oskinova, L.M., \& Foullon, C. 2000,
1805: \mnras, 318, 214
1806: \bibitem[]{} Jeffries, R.D. 2007, \mnras, 376, 1109
1807: \bibitem[]{} Kahn, S.M. et al. 2001, \aap, 365, L312
1808: \bibitem[]{} Kenyon, S.J. \& Hartmann, L. 1995, \apjs, 101, 117
1809: \bibitem[]{} Kramer, R.H., Cohen, D.H., \& Owocki, S.P. 2003, \apj, 592, 532
1810: \bibitem[]{} Krolik, J.H. \& Raymond, J.C. 1985, \apj, 298, 660
1811: %%\bibitem[]{} Lamers, H.J.G.L.M. \& Cassinelli, J.P. 1999,
1812: %% Introduction to Stellar Winds (Cambridge: Cambridge U. Press), 232
1813: \bibitem[]{} Landstreet, J.D. 1982, \apj, 258, 639
1814: \bibitem[]{} Landstreet, J.D. \& Borra, E.F. 1978, \apj, 224, L5
1815: \bibitem[]{} Lanz, T. \& Hubeny, I. 2007, \apj, 169, 83
1816: \bibitem[]{} Lee, T.A. 1968, \apj, 152, 913
1817: \bibitem[]{} Leutenegger, M.A., Paerels, F.B.S., Kahn, S.M., \& Cohen, D.C.
1818: 2006, \apj, 650, 1096
1819: \bibitem[]{} Lucy, L.B. 1982, \apj, 255, 286
1820: \bibitem[]{} Lucy, L.B. \& White, R.L. 1980, \apj, 241, 300
1821: %%\bibitem[]{} Luo, D., McCray, R., \& MacLow, M.-M. 1990, \apj, 362, 267
1822: \bibitem[]{} MacFarlane, J.J., Cassinelli, J.P., Welsh, B.Y., Vedder, P.W.,
1823: Vallerga, J.V., \& Waldron, W.L. 1991, \apj, 380, 564
1824: \bibitem[]{} Mason, B.D., Gies, D.R., Hartkopf, W.I., Bagnuolo, W.G.,
1825: Brummelaar, T.T., \& McAlisster, H.A. 1998,
1826: \aj, 115, 821
1827: \bibitem[]{} Mayne, N.J., Naylor, T., Littlefair, S.P., Saunders, E.S.,
1828: \& Jeffries, R.D. 2007, \mnras, 375, 1220
1829: \bibitem[]{} Mayne, N.J., Naylor, T. 2008, \mnras, in press (arXiv:0801.4085)
1830: \bibitem[]{} Miller, N.A., Cassinelli, J.P., Waldron, W.L., MacFarlane, J.J.,
1831: \& Cohen, D.H. 2002, \apj, 577, 951
1832: \bibitem[]{} Menten, K.M., Reid, M.J., Forbrich, J., \&
1833: Brunthaler, A. 2007, \aap, 474, 515
1834: \bibitem[]{} Morrell, N. \& Levato, H. 1991, \apjs, 75, 965
1835: \bibitem[]{} Nakajima, R. 1985, \apss, 116, 285
1836: \bibitem[]{} Ness, J.-U., Brickhouse, N.S., Drake, J.J., \&
1837: Huenemoerder, D.P. 2003, \apj, 598, 1277
1838: \bibitem[]{} Niemela, V.S. \& Gamen, R.C. 2005, \mnras, 356, 974
1839: %%\bibitem[]{} Oliveira, J.M., Jeffries, R.D., \& van Loon, J. Th. 2004,
1840: %% \mnras, 347, 1327
1841: \bibitem[]{} Oliveira, J.M., Jeffries, R.D., van Loon, J. Th.,
1842: \& Rushton, M.T. 2006, \mnras, 369, 272
1843: %%\bibitem[]{} Oliveira, J.M. \& van Loon, J. Th. 2004, \aap, 418, 663
1844: \bibitem[]{} Owocki, S.P., Castor, J.I., \& Rybicki, G.B. 1988,
1845: \apj, 335, 914
1846: \bibitem[]{} Owocki, S.P. \& Cohen, D.H. 2001, \apj, 559, 1108
1847: \bibitem[]{} Pauldrach, A.W.A., Kudritzki, R.P., Puls, J.,
1848: Butler, K., \& Hunsinger, J. 1994, \aap, 283, 525
1849: \bibitem[]{} Perryman, M.A.C. et al. 1997, \aap, 323, L49
1850: \bibitem[]{} Porquet, D. 2005, in X-Ray Diagnostics of Astrophysical
1851: Plasmas, ed. R.K. Smith (American Inst. of Physics), 177
1852: \bibitem[]{} Porquet, D., Mewe, R., Dubau, J., Raassen, A.J.J.,
1853: \& Kaastra, J.S., 2001 \aap, 376, 1113
1854: \bibitem[]{} Pradhan, A.K. 1982, \apj, 263, 477
1855: \bibitem[]{} Preibisch, T. \& Feigelson, E.D. 2005, \apjs, 160, 390
1856: \bibitem[]{} Preibisch, T. et al. 2005, \apjs, 160, 401
1857: \bibitem[]{} Press, W.H., Teukolsky, S.A., Vetterling, W.T., \&
1858: Flannery, B.P. 1992, Numerical Recipes in Fortran:
1859: The Art of Scientific Computing (2nd ed.; New York: Cambridge
1860: Univ. Press), 617
1861: \bibitem[]{} Reiners, A., Stahl, O., Wolf, B., Kaufer, A., \& Rivinius, T.
1862: 2000, \aap, 363, 585
1863: \bibitem[]{} Rieke, G.H. \& Lebofsky, M.J. 1985, \apj, 288, 618
1864: \bibitem[]{} Runacres, M.C. \& Owocki, S.P. 2002, \aap, 381, 1015
1865: \bibitem[]{} Sandstrom, K.M., Peek, J.E.G., Bower, G.C., Bolatto, A.D.,
1866: \& Plambeck, R.L. 2007, \apj, 667, 1161
1867: \bibitem[]{} Sanz-Forcada, J., Franciosini, E., \& Pallavicini, R. 2004,
1868: \aap, 421, 715
1869: \bibitem[]{} Savage, B.D., Massa, D., Meade, M., \& Wesselius, P.R. 1985,
1870: \apjs, 59, 397
1871: \bibitem[]{} Schulz, N.S., Canizares, C.R., Huenemoerder, D., \&
1872: Lee, J.C. 2000, \apj, 545, L135
1873: %%\bibitem[]{} Schulz, N.S., Testa, P., Huenemoerder, D.P., Ishibashi, K.,
1874: %% \& Canizares, C.R. 2006, \apj, 653, 636
1875: \bibitem[]{} Sherry, W.H., Walter, F.M., \& Wolk, S.J. 2004,
1876: \aj, 128, 2316 (S04)
1877: \bibitem[]{} Skinner, S.L., Brown, A., \& Stewart, R.T. 1993,
1878: \apjs, 87, 217
1879: \bibitem[]{} Skinner, S.L., Zhekov, S.A., Palla, F., \&
1880: Barbosa, C.L.D.R. 2005, \mnras, 361, 191
1881: \bibitem[]{} Smith, K., Audard, M., G\"{u}del, M., Skinner, S.,
1882: \& Pallavicini, R. 2005, in Proc. 13th Cambridge
1883: Workshop on Cool Stars, Stellar Systems, and the Sun,
1884: eds. F. Favata et al., ESA SP-560 (astro-ph/0504562)
1885: \bibitem[]{} Smith, R.K., Brickhouse, N.S., Liedahl, D.A., \&
1886: Raymond, J.C. 2001, \apj, 556, L91
1887: \bibitem[]{} Stelzer, B. et al. 2005, \apjs, 160, 557
1888: \bibitem[]{} Stevens, I.R., Blondin, J.M., \& Pollock, A.M.T. 1992,
1889: \apj, 386, 265
1890: \bibitem[]{} Telleschi, A., G\"{u}del, M., Briggs, K.R., Audard, M.,
1891: \& Palla, F. 2007, \aap, 468, 425
1892: \bibitem[]{} Townsend, R.H.D., Owocki, S.P., \& Groote, D. 2005,
1893: \apj, 630, L81
1894: \bibitem[]{} Townsend, R.H.D., Owocki, S.P., \& ud-Doula, A. 2007,
1895: \mnras, 382, 139
1896: %%\bibitem[]{} Tokovinin, A.A. \& Ismailov, R.M. 1988, \aaps, 72, 563
1897: \bibitem[]{} ud-Doula, A. \& Owocki, S.P. 2002, \apj, 576, 413
1898: %%\bibitem[]{} Usov, V.V. \& Melrose, D.B. 1992, \apj, 395, 575 (UM92)
1899: \bibitem[]{} van Loon, J. Th. \& Oliveira, J.M. 2003, \aap, 405, L33
1900: \bibitem[]{} Vogt, N. 1976, \aap, 53, 9
1901: \bibitem[]{} Waldron, W. \& Cassinelli, J.P. 2001, \apj, 548, L45
1902: \bibitem[]{} Waldron, W., Corcoran, M.F., Drake, S.A., \& Smale, A.P. 1998,
1903: \apjs, 118, 217
1904: \bibitem[]{} Walter, F.M., Wolk, S.J., Freyberg, M., \& Schmitt, J.H.M.M.
1905: 1997, Mem. Soc. Astr. Ital., 68, 1081
1906: \bibitem[]{} Wolk, N.R., Wolk, S.J., Walter, F.M., \& Sherry, W.H.
1907: 2006, in Protostars and Planets V, eds. B. Reipurth,
1908: D. Jewitt, \& K. Keil (U. of Arizona Press: Tucson), 8425
1909: \bibitem[]{} Zapatero Osorio, M.R. et al. 2002, \aap, 384, 937
1910: \end{thebibliography}
1911:
1912: \clearpage
1913:
1914: % TABLE1 : Sigma Ori Properties
1915: \begin{deluxetable}{llll}
1916: %%\rotate
1917: \tabletypesize{\scriptsize}
1918: \tablewidth{0pt}
1919: \tablecaption{ $\sigma$ Ori AB Properties\tablenotemark{a}}
1920: \tablehead{
1921: \colhead{} &
1922: \colhead{} &
1923: \colhead{Reference} &
1924: }
1925: \startdata
1926: Spectral Type & O9.5V $+$ B0.5V & (1) \\
1927: Distance (pc)\tablenotemark{b} & 352 $^{+166}_{-85}$ & (2) \\
1928: V (mag) & 3.80 - 3.81 & (3),(4),(5) \\
1929: E(B$-$V) (mag)& 0.05 - 0.07\tablenotemark{c} & (3),(4),(5) \\
1930: T$_{eff}$ (K)& 33,000 & (6) \\
1931: R$_{*}$ (R$_{\odot}$) & 9 & (6) \\
1932: log $\dot{M}$ (M$_{\odot}$ yr$^{-1}$) & $-$7.1 & (6) \\
1933: v$_{\infty}$ (km s$^{-1}$) & 1060 & (6) \\
1934: log [L$_{*}$/L$_{\odot}$] & 4.9 & (6) \\
1935: log L$_{wind}$ (ergs s$^{-1}$)\tablenotemark{d} & 34.4 & (6) \\
1936: log L$_{x}$ (ergs s$^{-1}$) & 31.5 & (7) \\
1937: log [L$_{x}$/L$_{*}$] & $-$7.0 & (7) \\
1938:
1939: \enddata
1940: \tablenotetext{a}{
1941: Notes:~$\sigma$ Ori = HD 37468. L$_{x}$ is the unabsorbed value
1942: in the 0.5 - 7 keV range.
1943: Refs. (1) Edwards 1976;
1944: (2) {\em Hipparcos} (Perryman et al. 1997);
1945: (3) Lee 1968;
1946: (4) Savage et al. 1985;
1947: (5) Vogt 1976;
1948: (6) Howarth \& Prinja 1989;
1949: (7) this work (0.5 - 7 keV).}
1950: \tablenotetext{b}{We adopt d = 352 pc in this work. A
1951: distance d = 334$^{+25}_{-22}$ pc was determined by
1952: Caballero (2008), who also noted that d $\sim$ 385 pc is possible if
1953: the system is a triple.}
1954: \tablenotetext{c}{Adopting A$_{\rm V}$ $\approx$ 3E(B$-$V) =
1955: 0.15 - 0.21 mag gives N$_{\rm H}$ =
1956: (3.3 - 4.7) $\times$ 10$^{20}$ cm$^{-2}$ using the
1957: Gorenstein (1975) conversion.}
1958: \tablenotetext{d}{L$_{wind}$ = $\dot{M}$v$_{\infty}^2$/2.}
1959:
1960: \end{deluxetable}
1961:
1962: \clearpage
1963:
1964: % TABLE2 Chandra X-ray Sources (S3)
1965: \begin{deluxetable}{lllllll}
1966: %%\rotate
1967: \tabletypesize{\scriptsize}
1968: \tablewidth{0pt}
1969: \tablecaption{Chandra X-ray Sources in the Vicinity of $\sigma$ Ori AB\tablenotemark{a} }
1970: \tablehead{
1971: \colhead{CXO} &
1972: \colhead{R.A.} &
1973: \colhead{Decl.} &
1974: \colhead{Net Counts} &
1975: \colhead{$<$E$>$} &
1976: \colhead{K$_{s}$} &
1977: \colhead{Identification(offset)} \\
1978: \colhead{No. } &
1979: \colhead{(J2000) } &
1980: \colhead{(J2000) } &
1981: \colhead{(cts) } &
1982: \colhead{(keV) } &
1983: \colhead{(mag)} &
1984: \colhead{(arcsec) }
1985: }
1986: \startdata
1987: 1 & 05 38 29.12 & -02 36 03.0 & 69$\pm$8(v.99) & 1.58 & 11.69 & 2M053829.12-023602.7(0.30)\\
1988: 2 & 05 38 31.43 & -02 36 33.8 & 7$\pm$3 & 1.29 & 10.99 & 2M053831.41-023633.8(0.24)\\
1989: 3 & 05 38 31.57 & -02 35 14.7 & 32$\pm$6 & 2.07 & 10.35 & 2M053831.58-023514.9(0.24)\\
1990: 4 & 05 38 32.83 & -02 35 39.4 & 147$\pm$12 & 1.35 & 10.73 & 2M053832.84-023539.2(0.30)\\
1991: 5 & 05 38 33.34 & -02 36 18.3 & 21$\pm$5 & 1.41 & 11.11 & 2M053833.36-023617.6(0.72)\\
1992: 6 & 05 38 34.04 & -02 36 37.2 & 8$\pm$3 & 1.10 & 11.08 & 2M053834.06-023637.5(0.42)\\
1993: 7 & 05 38 34.31 & -02 35 00.2 & 50$\pm$7 & 1.29 & 10.35 & 2M053834.31-023500.1(0.12)\\
1994: 8 & 05 38 35.21 & -02 34 38.1 & 369$\pm$19(v.95) & 3.01 & ... & M07:053835.22-023438.0(0.18) \\
1995: 9 & 05 38 38.22 & -02 36 38.6 & 390$\pm$20(v.99) & 1.79 & 10.31 & 2M053838.22-023638.4(0.18)\\
1996: 10 & 05 38 38.48 & -02 34 55.2 & 697$\pm$26(v.97) & 1.60 & 9.12 & 2M053838.49-023455.0(0.24)\\
1997: 11 & 05 38 39.72 & -02 40 19.2 & 12$\pm$4(v.96) & 3.09 & 12.88 & 2M053839.73-024019.7(0.54) \\
1998: 12 & 05 38 41.21 & -02 37 37.4 & 29$\pm$5 & 3.40 & 14.04 & 2M053841.23-023737.7(0.48)\\
1999: 13 & 05 38 41.28 & -02 37 22.7 & 323$\pm$18 & 1.64 & 10.59 & 2M053841.29-023722.6(0.24)\\
2000: 14 & 05 38 41.35 & -02 36 44.6 & 14$\pm$4 & 1.13 & 12.08 & 2M053841.35-023644.5(0.12)\\
2001: 15 & 05 38 42.27 & -02 37 14.5 & 9$\pm$3 & 2.56 & 10.77 & 2M053842.28-023714.7(0.30)\\
2002: 16 & 05 38 43.01 & -02 36 14.7 & 6$\pm$3 & 1.93 & 10.63 & 2M053843.02-023614.6(0.18)\\
2003: 17 & 05 38 43.85 & -02 37 07.3 & 16$\pm$4 & 1.38 & 11.77 & 2M053843.87-023706.9(0.54)\\
2004: 18 & 05 38 44.22 & -02 40 20.0 & 158$\pm$13 & 1.52 & 10.44 & 2M053844.23-024019.7(0.30)\\
2005: 19 & 05 38 44.76 & -02 36 00.3 & 2514$\pm$50 & 0.86 & 4.49 & 2M053844.76-023600.2(0.12); $\sigma$ Ori AB\\
2006: 20 & 05 38 44.83 & -02 35 57.4 & 43$\pm$7(v.95) & 1.37 & ... & $\sigma$ Ori IRS1 \\
2007: 21 & 05 38 45.35 & -02 41 59.6 & 50$\pm$5 & 1.69 & 11.04 & 2M053845.37-024159.4(0.42)\\
2008: 22\tablenotemark{c} & 05 38 46.82 & -02 36 43.5 & 6$\pm$2 & 0.95 & 12.35 & 2M053846.85-023643.5(0.48)\\
2009: 23 & 05 38 47.19 & -02 35 40.5 & 454$\pm$21(v.98) & 1.56 & 6.95 & 2M053847.20-023540.5(0.18); $\sigma$ Ori E\\
2010: 24 & 05 38 47.46 & -02 35 25.4 & 87$\pm$9 & 1.64 & 10.72 & 2M053847.46-023525.3(0.18)\\
2011: 25 & 05 38 47.89 & -02 37 19.6 & 188$\pm$14(v.99) & 2.20 & 10.78 & 2M053847.92-023719.2(0.54)\\
2012: 26 & 05 38 48.27 & -02 36 41.2 & 42$\pm$7 & 1.35 & 11.14 & 2M053848.29-023641.0(0.30)\\
2013: 27 & 05 38 48.68 & -02 36 16.4 & 106$\pm$10(v.95) & 1.35 & 11.17 & 2M053848.68-023616.2(0.18)\\
2014: 28 & 05 38 49.12 & -02 41 23.9 & 22$\pm$6 & 1.62 & ... & S04:053849.14-024124.8 (0.96)\\
2015: 29 & 05 38 49.17 & -02 38 22.5 & 112$\pm$11 & 1.32 & 10.51 & 2M053849.17-023822.2(0.24)\\
2016: 30 & 05 38 49.69 & -02 34 52.6 & 8$\pm$3 & 1.22 & 12.09 & 2M053849.70-023452.6(0.24)\\
2017: 31 & 05 38 50.02 & -02 37 35.5 & 15$\pm$4 & 1.64 & 12.10 & 2M053850.03-023735.5(0.18)\\
2018: 32\tablenotemark{b,c} & 05 38 50.16 & -02 36 54.6 & 5$\pm$2 & 4.06 & ... & ... \\
2019: 33 & 05 38 51.43 & -02 36 20.7 & 53$\pm$7 & 1.86 & 11.55 & 2M053851.45-023620.6(0.36)\\
2020: 34 & 05 38 51.75 & -02 36 03.3 & 6$\pm$3 & 1.56 & 12.03 & 2M053851.74-023603.3(0.18)\\
2021: 35 & 05 38 51.85 & -02 33 33.0 & 17$\pm$4 & 2.94 & ... & ... \\
2022: 36 & 05 38 53.03 & -02 38 53.7 & 12$\pm$4 & 1.50 & 10.83 & 2M053853.07-023853.6(0.61) \\
2023: 37 & 05 38 53.37 & -02 33 23.2 & 863$\pm$29(v.99) & 1.67 & 9.73 & 2M053853.37-023323.0(0.24)\\
2024: 38 & 05 38 58.24 & -02 38 51.5 & 25$\pm$5 & 2.67 & ... & ... \\
2025: 39 & 05 38 59.03 & -02 34 13.2 & 13$\pm$4 & 2.73 & ... & ... \\
2026: 40 & 05 38 59.51 & -02 35 28.6 & 12$\pm$4 & 3.06 & ... & ... \\
2027: 41 & 05 39 00.53 & -02 39 38.7 & 94$\pm$10 & 1.66 & 11.11 & 2M053900.53-023939.0(0.30)\\
2028: 42 & 05 39 01.47 & -02 38 56.5 & 342$\pm$19 & 1.42 & 8.09 & 2M053901.50-023856.4(0.36); HD 37525\tablenotemark{d}\\
2029:
2030: \enddata
2031: \tablenotetext{a}{
2032: Notes:~X-ray data are from CCD7 (ACIS chip S3) using events in the 0.5 - 7 keV range.
2033: Tabulated quantities are: running source number, X-ray position (R.A., decl.), net counts and
2034: net counts error from {\em wavdetect} (accumulated in a 90,967 s exposure, rounded to the nearest integer,
2035: background subtracted and PSF-corrected), mean photon energy $<$E$>$,
2036: K$_{s}$ magnitude of near-IR 2MASS counterpart, and 2MASS (2M) or optical (S04 = Sherry, Walter,
2037: \& Wolk 2004; M07 = Mayne et al. 2007) candidate counterpart identification
2038: within a 2$''$ search radius. The offset (arc seconds) between the X-ray and counterpart position is given
2039: in parentheses. A (v) following Net Counts error indicates that the source is likely variable as indicated
2040: by a variability probability P$_{var}$ $\geq$ 0.95 determined from the Kolmogorov-Smirnov (KS) statisitic.
2041: The number following v is the KS variability probability, i.e. v.99 indicates a variability
2042: probability P$_{var}$ = 0.99.
2043: All sources were confirmed to be present in the archived
2044: 98 ksec {\em Chandra} HRC-I image (Obsid 2560).}
2045:
2046: \tablenotetext{b}{
2047: Probable {\em XMM-Newton} counterpart
2048: is source NX 99 in Table B.1 of Franciosini et al. (2006). High value of $<$E$>$ suggests
2049: possible extragalactic background source. }
2050:
2051: \tablenotetext{c}{
2052: Low-significance {\em wavdetect} detection (2. $<$ significance $<$ 3.).}
2053:
2054: \tablenotetext{d}{
2055: Double star with a B5V primary and a companion at separation
2056: 0.$''$45 (Caballero 2005).}
2057:
2058:
2059: \end{deluxetable}
2060:
2061:
2062:
2063: %%TABLE3 - CCD Spectral fit results: Sig Ori
2064:
2065: \clearpage
2066: \begin{deluxetable}{lllllllc}
2067: %%\rotate
2068: \tabletypesize{\tiny}
2069: \tablewidth{0pt}
2070: \tablecaption{Spectral Fits of Zero-Order ACIS-S3 CCD Spectra\tablenotemark{a} }
2071: \tablehead{
2072: \colhead{CXO} &
2073: \colhead{N$_{\rm H}$} &
2074: \colhead{kT$_{1}$} &
2075: \colhead{kT$_{2}$} &
2076: \colhead{norm$_{1}$} &
2077: \colhead{norm$_{2}$} &
2078: \colhead{$\chi^2$/dof ($\chi^2_{red}$)} &
2079: \colhead{Flux} \\
2080: \colhead{No.} &
2081: \colhead{(10$^{20}$ cm$^{-2}$)} &
2082: \colhead{(keV)} &
2083: \colhead{(keV)} &
2084: \colhead{ (10$^{-5}$)} &
2085: \colhead{(10$^{-5}$)} &
2086: \colhead{ } &
2087: \colhead{(10$^{-13}$ ergs cm$^{-2}$ s$^{-1}$)}
2088: }
2089: \startdata
2090: 4 & \{3.0\} & 0.37 [0.23-0.51] & 1.4 [1.2-1.7] & 1.71 [0.67-2.27] & 3.32 [2.38-3.99] & 5.6/5 (1.12) & 0.70 (0.76) \\
2091: 8 & 5.0 [3.8-8.8] & 0.69 [0.47-2.16] & 19.8 [8.1-...]& 3.49 [0.10-13.6] & 14.1 [12.4-17.4] & 32.3/31 (1.04)\tablenotemark{b} & 1.97 (3.07) \\
2092: 9 & 1.5 [0.0-4.5] & 0.61 [0.35-0.78] & 2.5 [2.1-2.9] & 2.14 [1.44-2.79] & 10.4 [9.00-12.0] & 31.2/31 (1.01) & 1.87 (1.93) \\
2093: 10 & 2.5 [1.2-5.0] & 0.70 [0.61-0.76] & 2.4 [2.1-2.8] & 5.81 [4.83-7.10] & 14.4 [12.0-16.3] & 72.8/52 (1.40) & 3.19 (3.37) \\
2094: 13 & \{3.0\} & 0.74 [0.63-0.84] & 3.1 [2.3-4.8] & 2.58 [1.52-3.37] & 7.19 [5.53-8.52] & 11.4/11 (1.04) & 1.55 (1.65) \\
2095: 18 & \{3.0\} & 0.37 [0.26-0.59] & 2.0 [1.6-2.5] & 1.89 [0.54-2.56] & 3.73 [2.96-4.78] & 13.1/14 (0.94) & 0.78 (0.84) \\
2096: 19\tablenotemark{c} & \{3.0\} & 0.30 [0.29-0.31] & ... & 111. [105.-116.] & ... & 76.5/49 (1.56) & 15.4 (17.8) \\
2097: 19\tablenotemark{c} & \{3.0\} & 0.21 [0.18-0.24] & 0.47 [0.45-0.55] & 79.9 [69.0-90.9] & 41.4 [25.8-50.3] & 45.6/47 (0.97)& 16.6 (19.5) \\
2098: 20 & \{3.0\} & 0.61 [0.27-0.83] & 2.2 [1.2-...] & 0.59 [0.27-0.89] & 0.61 [0.10-1.11] & 1.2/6 (0.20) & 0.21 (0.23) \\
2099: 23\tablenotemark{d} & \{3.0\} & 0.72 [0.61-0.80] & 2.4 [2.0-3.0] & 4.43 [3.46-5.24] & 8.46 [6.91-10.4] & 28.6/24 (1.19) & 2.09 (2.23) \\
2100: 25 & \{3.0\} & 3.20 [2.59-4.09] & ... & 6.96 [6.07-7.86] & ... & 18.1/17 (1.06) & 0.96 (1.00) \\
2101: 27 & \{3.0\} & 0.66 [0.49-0.82] & 2.2 [1.5-3.4] & 1.24 [0.81-1.62] & 1.79 [0.96-2.55] & 6.4/7 (0.91) & 0.50 (0.54) \\
2102: 29 & \{3.0\} & 0.67 [0.47-0.82] & 1.6 [1.1-3.2] & 1.29 [0.67-1.89] & 1.71 [0.71-2.52] & 7.0/8 (0.88) & 0.52 (0.56) \\
2103: 37 & 2.1 [0.9-4.1] & 0.75 [0.65-0.83] & 2.3 [2.1-2.6] & 6.02 [4.42-7.06] & 24.9 [22.0-27.8] & 26.2/32 (0.82) & 4.59 (4.80) \\
2104: 42 & \{3.0\} & 0.76 [0.68-0.83] & 2.6 [2.0-3.5] & 4.69 [3.65-5.49] & 5.56 [4.15-7.37] & 34.4/29 (1.19) & 1.76 (1.89) \\
2105: \enddata
2106: \tablenotetext{a}{
2107: Notes:~The CXO no. in column (1) refers to Table 2.
2108: Fits are based on one-temperature (1T) or two-temperature (2T) optically thin plasma
2109: {\em apec} solar-abundance models of rebinned spectra in XSPEC (v. 12.3).
2110: Solar abundances are referenced to Anders \& Grevesse (1989).
2111: The notation N$_{\rm H}$ = \{3.0\} means the absorption column density was held fixed at that
2112: value during fitting. Brackets enclose 90\% confidence intervals and an
2113: ellipsis means the algorithm used to compute the 90\% confidence limit did not converge.
2114: The {\em norm} is related to the volume emission measure (EM) by
2115: EM $=$ 10$^{14}$4$\pi$d$_{cm}^2${\em norm}, where d$_{cm}$ is the distance (cm) to the source.
2116: The total X-ray fluxes are the absorbed values in the 0.5-7 keV
2117: range, followed in parentheses by unabsorbed values. }
2118: \tablenotetext{b}{An absorbed power-law model with a photon power-law index $\alpha$ = $+$1.6 [1.4 - 1.8]
2119: and N$_{\rm H}$ = 2.5e20 cm$^{-2}$ gives nearly identical fit statistics
2120: with $\chi^2$/dof = 34.7/33 and an unabsorbed flux 2.47e-13 ergs cm$^{-2}$ s$^{-1}$.}
2121: %%\tablenotetext{c}{A variable abundance 1T {\em apec} fit converges to a global
2122: %% abundance Z = 0.20 [0.14-0.28] Z$_{\odot}$ with kT$_{1}$ = 0.30 keV
2123: %% ($\chi^2$/dof = 71.1/52). An analogous 2T {\em apec} fit converges to
2124: %% Z = 0.25 [0.15-0.38] Z$_{\odot}$ with kT$_{1}$ = 0.30 keV and
2125: %% kT$_{2}$ = 1.7 keV ($\chi^2$/dof = 59.2/50).}
2126: \tablenotetext{c}{Estimated zero-order photon pileup is 5\% for $\sigma$ Ori AB. Events with energies
2127: below 0.5 keV were excluded from the fit to avoid any possible contamination by soft UV/optical leaks.}
2128: \tablenotetext{d}{A 3T $apec$ model gives an improved fit with N$_{\rm H}$ = 2.6 [1.1-10.4] $\times$ 10$^{20}$ cm$^{-2}$,
2129: kT$_{1}$ = 0.24 [0.09-0.39] keV, kT$_{2}$ = 0.81 [0.75-1.05] keV, kT$_{3}$ = 2.6 [2.1-3.3] keV,
2130: norm$_{1}$ = 2.91 [1.00-12.6]e-5, norm$_{2}$ = 3.77 [2.05-5.42]e-5, norm$_{3}$ = 8.11 [6.73-9.76]e-5,
2131: $\chi^2$/dof = 18.6/21 ($\chi^2_{red}$ = 0.89), and Flux = 2.24 (2.41) $\times$ 10$^{-13}$
2132: ergs cm$^{-2}$ s$^{-1}$.}
2133: \end{deluxetable}
2134:
2135: \clearpage
2136:
2137:
2138: % TABLE4.TEX
2139: \begin{deluxetable}{llllllc}
2140: %%\rotate
2141: \tabletypesize{\footnotesize}
2142: \tablewidth{0pt}
2143: \tablecaption{$\sigma$ Ori AB Emission Line Properties\tablenotemark{a} }
2144: \tablehead{
2145: \colhead{Ion} &
2146: \colhead{$\lambda_{lab}$} &
2147: \colhead{$\lambda_{obs}$} &
2148: \colhead{$\Delta\lambda$} &
2149: \colhead{Line Flux} &
2150: \colhead{HWHM} &
2151: \colhead{log T$_{max}$} \\
2152: \colhead{ } &
2153: \colhead{(\AA) } &
2154: \colhead{(\AA) } &
2155: \colhead{(m\AA) } &
2156: \colhead{(10$^{-5}$ ph cm$^{-2}$ s$^{-1}$) } &
2157: \colhead{(km s$^{-1}$)} &
2158: \colhead{(K)}
2159: }
2160: \startdata
2161: Si XIII(r) & 6.6479 & ... & ... & 0.12 (0.01 - 0.22) & \{300\} & 7.0 \\
2162: Mg XI(r) & 9.1687 & 9.165 (9.162 - 9.169) & $-$4. & 0.42 (0.30 - 0.55) & 250 (8 - 374) & 6.8 \\
2163: Mg XI(i) & 9.2312 & ... & ... & 0.41 (0.27 - 0.52) & 250\tablenotemark{c} & 6.8 \\
2164: Mg XI(f) & 9.3143 & ... & ... & ...\tablenotemark{d} & ... & 6.8 \\
2165: Ne X &10.239 & ... & ... & 0.25 (0.09 - 0.42) & \{300\} & 6.8 \\
2166: Ne IX &11.544 & ... & ... & 0.48 (0.23 - 0.74) & \{300\} & 6.6 \\
2167: Ne X\tablenotemark{b} &12.132 & 12.127 (12.124 - 12.130) & $-$5. & 1.64 (1.28 - 1.98) & 390 (248 - 542) & 6.8 \\
2168: Ne IX(r) &13.447 & 13.446 (13.441 - 13.449) & $-$1. & 3.36 (2.80 - 4.03) & 314 (216 - 381) & 6.6 \\
2169: Ne IX(i) &13.553 & 13.546 (13.543 - 13.548) & $-$7. & 3.53 (2.87 - 4.09) & 314\tablenotemark{c} & 6.6 \\
2170: Ne IX(f) &13.699 & ... & ... & ...\tablenotemark{d} & ... & 6.6 \\
2171: Fe XVII &15.014 & 15.013 (15.010 - 15.016) & $-$1. & 7.94 (6.69 - 9.14) & 279 (195 - 355) & 6.7 \\
2172: O VIII &15.176 & ... & ... & 1.11 (0.53 - 1.98) & \{300\} & 6.5 \\
2173: Fe XIX &15.198 & ... & ... & 1.01 (0.55 - 2.24) & \{300\} & 6.9 \\
2174: Fe XVII &15.261 & 15.258 (15.251 - 15.266) & $-$3. & 3.09 (2.36 - 3.83) & \{300\} & 6.7 \\
2175: O VIII &16.006 & 16.009 (16.002 - 16.015) & $+$3. & 2.63 (1.57 - 3.60) & 195 (0 - 405) & 6.5 \\
2176: Fe XVII &16.780 & 16.778 (16.772 - 16.785) & $-$2. & 4.98 (3.66 - 6.27) & 155 (21 - 278) & 6.7 \\
2177: Fe XVII &17.051 & 17.049 (17.043 - 17.055) & $-$2. & 5.32 (4.15 - 6.63) & 163 (92 - 240) & 6.7 \\
2178: Fe XVII &17.096 & 17.091 (17.086 - 17.096) & $-$5. & 6.91 (5.58 - 8.29) & \{163\} & 6.7 \\
2179: O VII &18.627 & ... & ... & 1.86 (0.25 - 3.46) & \{300\} & 6.3 \\
2180: O VIII\tablenotemark{b} &18.967 & 18.968 (18.964 - 18.972) & $+$1. & 22.0 (19.6 - 27.8) & 330 (273 - 435) & 6.5 \\
2181: O VII(r) &21.602 & 21.608 (21.598 - 21.617) & $+$6. & 15.5 (10.7 - 21.8) & 298 (97 - 507) & 6.3 \\
2182: O VII(i) &21.804 & 21.803 (21.796 - 21.810) & $-$1. & 22.4 (15.2 - 30.1) & 298\tablenotemark{c} & 6.3 \\
2183: O VII(f) &22.098 & ... & ... & ...\tablenotemark{d} & ... & 6.3 \\
2184: \enddata
2185: \tablenotetext{a}{
2186: Notes:~X-ray data are from the background-subtracted HETG/MEG1 spectrum
2187: ($+$1 and $-1$ orders combined), regrouped to a minimum
2188: of 10 counts per bin. Parentheses enclose 1$\sigma$ intervals.
2189: Tabulated quantities are: ion name (Ion) where r,i,f denote resonance, intercombination, and
2190: forbidden lines, laboratory wavelength of transition ($\lambda_{lab}$),
2191: measured wavelength ($\lambda_{obs}$), $\lambda_{obs}$ $-$ $\lambda_{lab}$ ($\Delta\lambda$),
2192: observed (absorbed) continuum-subtracted line flux (Line Flux), line half-width at half-maximum
2193: (HWHM), and maximum line power temperature (T$_{max}$).
2194: An ellipsis means that $\lambda_{obs}$ was not measured due to faint or closely-spaced lines.
2195: Curly braces enclose quantities that were held fixed during fitting.
2196: The value of $\lambda_{obs}$ is subject to the $\pm$11 m\AA~ absolute wavelength
2197: calibration accuracy of HETG/MEG.}
2198: \tablenotetext{b}{Blend.}
2199: \tablenotetext{c}{The line width was allowed to vary during fitting but was kept equal to the
2200: width of the resonance (r) line.}
2201: \tablenotetext{d}{The forbidden line was undetected.}
2202: \end{deluxetable}
2203:
2204:
2205: \clearpage
2206:
2207:
2208:
2209: % TABLE5.TEX
2210: \begin{deluxetable}{lll}
2211: \tablewidth{0pc}
2212: \tabletypesize{\scriptsize}
2213: \tablecaption{$\sigma$ Ori AB He-like Ion Line Flux Ratios\tablenotemark{a} }
2214: \tablehead{
2215: \colhead{ } &
2216: \colhead{Mg XI } &
2217: \colhead{Ne IX } \\
2218: }
2219: \startdata
2220: $\lambda_{lab,r}$ (\AA~) & 9.1687 & 13.447\tablenotemark{b} \nl
2221: log T$_{max}$ (K) & 6.8 & 6.6 \nl
2222: $G$\tablenotemark{c} & 1.21 [0.76 - 1.92] & 0.99 [0.73 - 1.32] \nl
2223: $R$\tablenotemark{c} & $\leq$0.17 & $\leq$0.15 \nl
2224: $R_{0}$ & 2.7 & 3.6 \nl
2225: $\phi_{c}$ (s$^{-1}$) & 4.86 $\times$ 10$^{4}$ & 7.73 $\times$ 10$^{3}$ \nl
2226: $n_{c}$ (cm$^{-3}$) & 6.2 $\times$ 10$^{12}$ & 6.4 $\times$ 10$^{11}$ \nl
2227: r$_{fi}$/R$_{*}$ & $\leq$1.8 & $\leq$4.4 \nl
2228: \tablenotetext{a}{
2229: The laboratory rest wavelength of the resonance line
2230: $\lambda_{lab,r}$ and the maximum line power temperature
2231: T$_{max}$ are from APED.
2232: The line flux ratios $G$ = (f $+$ i)/r and $R$ = f/i are from ISIS
2233: spectral fits (see below).
2234: $R_{0}$ is the theoretical limit of $R$ at T$_{max}$
2235: for low photoexcitation and low density
2236: (Ne IX from Pradhan 1982; Mg XI from Porquet et al. 2001).
2237: The critical photoexcitation rate $\phi_{c}$ and critical
2238: density $n_{c}$ are from BDT72.
2239: The 90\% confidence upper limit on the line formation radius
2240: is r$_{fi}$/R$_{*}$.
2241: }
2242: %% O VII from Porquet 2005).}
2243: \tablenotetext{b}{The Ne IX resonance line may be weakly contaminated by
2244: emission from Fe XIX (13.462 \AA~).}
2245: \tablenotetext{c}{Determined from simultaneous fits of MEG1$+$HEG1 spectra
2246: with ISIS. The He-like complexes were fitted allowing $R$, $G$, and the metal
2247: abundances to vary. ISIS applies the appropriate instrumental and
2248: thermal broadening to each line. The best-fit abundances
2249: of Mg and Ne are consistent
2250: with solar at the 90\% confidence level. The continuum level was
2251: determined by a global 2T $vapec$ fit of the zero-order spectrum.
2252: Square brackets enclose 90\% confidence intervals on $G$ and upper
2253: limits on $R$ are 90\% confidence.}
2254: \enddata
2255: \end{deluxetable}
2256:
2257:
2258:
2259:
2260: % And finally, we must deal with the figures.
2261:
2262: %% Uncomment the following line to skip the figures
2263: %%\end{document}
2264:
2265:
2266:
2267:
2268: \begin{figure}
2269: \figurenum{1}
2270: \epsscale{1.0}
2271: %%\includegraphics*[width=12.5cm,angle=-90]{f1.ps}
2272: \includegraphics*[width=8.5cm,angle=-90]{f1.eps}
2273: \caption{Circles mark the positions of the 42 X-ray
2274: sources detected on the ACIS-S3 CCD as listed in Table 2.
2275: A cross marks the {\em Chandra} aimpoint and the primary
2276: target $\sigma$ Ori AB is identified.}
2277: \end{figure}
2278:
2279:
2280: \clearpage
2281: \begin{figure}
2282: \figurenum{2}
2283: \epsscale{1.0}
2284: %%\includegraphics*[width=12.5cm,angle=-90]{f2.eps}
2285: \includegraphics*[width=8.5cm,angle=-90]{f2.eps}
2286: \caption{Unsmoothed ACIS-S3 image (0.5 - 7 keV) of
2287: the region near $\sigma$ Ori AB.
2288: Circles mark ACIS-S3 detections and squares mark
2289: the optical positions of undetected stars $\sigma$ Ori C and
2290: D. Numbered sources correspond to Table 2.}
2291: \end{figure}
2292:
2293:
2294: \clearpage
2295: \begin{figure}
2296: \figurenum{3}
2297: \epsscale{1.0}
2298: %%\includegraphics*[width=12.5cm,angle=-90]{f3.eps}
2299: \includegraphics*[width=8.5cm,angle=-90]{f3.eps}
2300: \caption{Histogram of mean photon energies for X-ray sources
2301: on ACIS-S3. The value of $\sigma$ Ori AB is shown and
2302: the downward arrow marks the mean value.}
2303: \end{figure}
2304:
2305: \clearpage
2306: \begin{figure}
2307: \figurenum{4}
2308: \epsscale{1.0}
2309: %%\includegraphics*[width=12.5cm,angle=-90]{f4.eps}
2310: \includegraphics*[width=8.5cm,angle=-90]{f4.eps}
2311: \caption{ACIS-S3 light curve of $\sigma$ Ori AB (CXO 19) binned
2312: at 4000 s intervals using events in the 0.5 - 7 keV
2313: range and extraction radius R$_{e}$ = 1.$''$97.
2314: The solid line is a best-fit constant count rate model.
2315: Error bars are 1$\sigma$. The KS variability probability
2316: is P$_{var}$ = 0.935.
2317: }
2318: \end{figure}
2319:
2320:
2321:
2322: \clearpage
2323: \begin{figure}
2324: \figurenum{5}
2325: \epsscale{1.0}
2326: %%\includegraphics*[width=12.5cm,angle=-90]{f5.eps}
2327: \includegraphics*[width=8.5cm,angle=-90]{f5.eps}
2328: \caption{ACIS-S3 light curve of $\sigma$ Ori E (CXO 23) binned
2329: at 6000 s intervals using events in the 0.5 - 7 keV
2330: range and extraction radius R$_{e}$ = 1.$''$97.
2331: Error bars are 1$\sigma$. A KS test gives a high probability
2332: of variability P$_{var}$ = 0.986. The solid curve is a best fit
2333: sinusoid model with a constant offset of 0.004937 counts/s
2334: (dashed line) and the period fixed at the stellar rotation period
2335: P$_{rot}$ = 1.19084 d = 28.58 hours. The time axis
2336: shows the elapsed time in hours from start of the day.
2337: The {\em Chandra} observation began at 20:59:02 UT on
2338: 2003 August 12 and spanned 1.053 d. }
2339: \end{figure}
2340:
2341:
2342: \clearpage
2343: \begin{figure}
2344: \figurenum{6}
2345: \epsscale{1.0}
2346: %%\includegraphics*[width=12.5cm,angle=-90]{f6.eps}
2347: \includegraphics*[width=8.5cm,angle=-90]{f6.eps}
2348: \caption{ACIS-S3 light curve of the variable source
2349: CXO J053835.21$-$023438.1 (CXO 8)
2350: binned at 3000 s intervals using events in the 0.5 - 7 keV
2351: range and extraction radius R$_{e}$ = 1.$''$97.
2352: Error bars are 1$\sigma$.
2353: No {\em 2MASS} counterpart was found but a faint V = 20.88 mag
2354: optical source lies at an offset of 0.$''$18 from the X-ray
2355: position (Mayne et al. 2007).
2356: }
2357: \end{figure}
2358:
2359:
2360:
2361:
2362: \clearpage
2363: \begin{figure}
2364: \figurenum{7}
2365: \epsscale{1.0}
2366: %%\includegraphics*[width=12.5cm,angle=-90]{f7.eps}
2367: \includegraphics*[width=8.5cm,angle=-90]{f7.eps}
2368: \caption{ACIS-S3 light curve of CXO J053838.22$-$023638.6 (CXO 9)
2369: binned at 2500 s intervals using events in the 0.5 - 7 keV
2370: range and extraction radius R$_{e}$ = 1.$''$97.
2371: Error bars are 1$\sigma$.
2372: This T Tauri star shows weak H$\alpha$ emission and was classified as
2373: K8 by Zapatero Osorio (2002).}
2374: \end{figure}
2375:
2376:
2377:
2378: \clearpage
2379: \begin{figure}
2380: \figurenum{8}
2381: \epsscale{1.0}
2382: %%\includegraphics*[width=12.5cm,angle=-90]{f8.eps}
2383: \includegraphics*[width=8.5cm,angle=-90]{f8.eps}
2384: \caption{ACIS-S3 light curve of CXO J053847.89$-$023719.6 (CXO 25)
2385: binned at 5000 s intervals using events in the 0.5 - 7 keV
2386: range and extraction radius R$_{e}$ = 1.$''$97.
2387: Error bars are 1$\sigma$.
2388: This star has been classified as M4 and shows a near-IR
2389: excess (Oliveira et al. 2006).}
2390: \end{figure}
2391:
2392:
2393:
2394:
2395: \clearpage
2396: \begin{figure}
2397: \figurenum{9}
2398: \epsscale{1.0}
2399: %%\includegraphics*[width=12.5cm,angle=-90]{f9.eps}
2400: \includegraphics*[width=8.5cm,angle=-90]{f9.eps}
2401: \caption{Color-magnitude diagram of X-ray sources
2402: on ACIS-S3 having 2MASS identifications. Colors are
2403: based on 2MASS photometry. Source numbers
2404: correspond to Table 2. Intrinsic colors are shown for
2405: late-type dwarfs and giants from Bessell \& Brett (1988).
2406: Source CXO 42 is the B5V binary star HD 37525. Source CXO 12 is
2407: not shown and lies off the right edge of the plot
2408: at (J-K$_{s}$,K$_{s}$) = (2.33,14.04).}
2409: \end{figure}
2410:
2411: \clearpage
2412:
2413: \begin{figure}
2414: \figurenum{10}
2415: \epsscale{1.0}
2416: %%\includegraphics*[width=12.5cm,angle=-90]{f10.eps}
2417: \includegraphics*[width=8.5cm,angle=-90]{f10.eps}
2418: \caption{Color-color diagram of X-ray sources
2419: on ACIS-S3 having 2MASS identifications.
2420: Colors are based on 2MASS photometry.
2421: Source numbers correspond to Table 2. The dotted line at
2422: left marks the unreddened main-sequence.
2423: The three sloping dashed lines show the loci
2424: of M0 V, K0 V, and A0 V stars
2425: based on intrinsic colors from Bessell \& Brett (1988)
2426: and the extinction law of Rieke \& Lebofsky (1985). }
2427: \end{figure}
2428:
2429:
2430: \clearpage
2431:
2432: \begin{figure}
2433: \figurenum{11}
2434: \epsscale{1.0}
2435: %%\includegraphics*[width=12.5cm,angle=-90]{f11.eps}
2436: \includegraphics*[width=8.5cm,angle=-90]{f11.eps}
2437: \caption{Zero-order CCD spectrum of $\sigma$ Ori AB
2438: rebinned to a minimum of 20 counts per bin. The dotted
2439: line shows the best-fit 2T $apec$ model with fit parameters
2440: as listed in Table 3.}
2441: \end{figure}
2442:
2443:
2444: \clearpage
2445:
2446: \begin{figure}
2447: \figurenum{12}
2448: \epsscale{1.0}
2449: %%\includegraphics*[width=12.5cm,angle=-90]{f12.eps}
2450: \includegraphics*[width=8.5cm,angle=-90]{f12.eps}
2451: \caption{Emission measure distribution of $\sigma$ Ori AB
2452: obtained from a best-fit of the zero-order CCD spectrum
2453: with the XSPEC model $c6pmekl$ using solar abundances
2454: and hydrogen column density N$_{\rm H}$ = 3 $\times$ 10$^{20}$
2455: cm$^{-2}$.}
2456: \end{figure}
2457:
2458:
2459:
2460:
2461: \clearpage
2462:
2463: \begin{figure}
2464: \figurenum{13}
2465: \epsscale{1.0}
2466: %%\includegraphics*[width=12.5cm,angle=-90]{f13.eps}
2467: \includegraphics*[width=8.5cm,angle=-90]{f13.eps}
2468: \caption{Zero-order CCD spectrum of $\sigma$ Ori E
2469: rebinned to a minimum of 15 counts per bin. The dotted
2470: line shows the best-fit 3T $apec$ model with fit parameters
2471: as listed in Table 3 notes.}
2472: \end{figure}
2473:
2474:
2475:
2476:
2477:
2478: \clearpage
2479:
2480: \begin{figure}
2481: \figurenum{14}
2482: \epsscale{1.0}
2483: %%\includegraphics*[width=12.5cm,angle=-90]{f14.eps}
2484: \includegraphics*[width=8.5cm,angle=-90]{f14.eps}
2485: \caption{ACIS-S3 CCD spectrum of $\sigma$ Ori IRS1
2486: (CXO 20) binned to 5 counts per bin.
2487: The dotted line shows the 2T
2488: $apec$ model with fit parameters as listed in Table 3.}
2489: \end{figure}
2490:
2491: \clearpage
2492:
2493: \begin{figure}
2494: \figurenum{15}
2495: \epsscale{1.0}
2496: %%\includegraphics*[width=12.5cm,angle=-90]{f15.eps}
2497: \includegraphics*[width=8.5cm,angle=-90]{f15.eps}
2498: \caption{HETG/MEG1 background-subtracted spectrum
2499: ($+$1 and $-$1 orders combined; 1936 net counts)
2500: of $\sigma$ Ori AB with prominent lines identified.}
2501: \end{figure}
2502:
2503: \clearpage
2504:
2505:
2506: \begin{figure}
2507: \figurenum{16}
2508: \epsscale{1.0}
2509: %%\includegraphics*[width=12.5cm,angle=-90]{f16.eps}
2510: \includegraphics*[width=8.5cm,angle=-90]{f16.eps}
2511: \caption{Measured line widths of $\sigma$ Ori AB
2512: expressed as half-width at half-maximum
2513: (1$\sigma$ error bars) versus line
2514: laboratory wavelength using data from Table 4.
2515: The two lines shown with open squares are blends
2516: and their HWHM values are likely overestimated.
2517: The solid horizontal line is the mean value
2518: $\overline{\rm HWHM}$ = 264 km s$^{-1}$.}
2519: \end{figure}
2520:
2521:
2522: \begin{figure}
2523: \figurenum{17}
2524: \epsscale{1.0}
2525: %%\includegraphics*[width=12.5cm,angle=-90]{f17.eps}
2526: \includegraphics*[width=8.5cm,angle=-90]{f17.eps}
2527: \caption{Measured line widths of $\sigma$ Ori AB
2528: expressed as half-width at half-maximum
2529: (1$\sigma$ error bars) versus maximum
2530: line power temperature T$_{max}$ using data from
2531: Table 4. The plotted values of T$_{max}$ for
2532: O VIII, Fe XVII, and Ne X have been shifted
2533: slightly from the nominal values given in Table 4
2534: to avoid error bar overlap.
2535: The two lines shown with open squares are blends
2536: and their HWHM values are likely overestimated.
2537: The solid horizontal line is the mean value
2538: $\overline{\rm HWHM}$ = 264 km s$^{-1}$.}
2539: \end{figure}
2540:
2541: \clearpage
2542:
2543: \begin{figure}
2544: \figurenum{18}
2545: \epsscale{1.0}
2546: %%\includegraphics*[width=12.5cm,angle=-90]{f18.eps}
2547: \includegraphics*[width=8.5cm,angle=-90]{f18.eps}
2548: \caption{Plot of the centroid offset between the measured
2549: and laboratory line centroid of prominent lines in the MEG1
2550: spectrum, using the offset values
2551: $\Delta$$\lambda$ = $\lambda_{obs}$ $-$
2552: $\lambda_{lab}$ from Table 4. Errors bars are 1$\sigma$.
2553: The mean offset (solid line) and $\pm$1$\sigma$ uncertainty
2554: (dashed lines) for 13 measurements are $-$1.6 ($\pm$3.5) m\AA~.
2555: }
2556: \end{figure}
2557:
2558:
2559:
2560: \clearpage
2561:
2562: \begin{figure}
2563: \figurenum{19}
2564: \epsscale{1.0}
2565: %%\includegraphics*[width=12.5cm,angle=-90]{f19.eps}
2566: \includegraphics*[width=8.5cm,angle=-90]{f19.eps}
2567: \caption{Histogram plot of the MEG1 spectrum ($+$1 and $-$1 orders combined)
2568: of $\sigma$ Ori AB near the Mg XI He-like triplet, binned to a minimum
2569: of 10 counts per bin. Solid squares are data points and the dotted
2570: line is a Gaussian fit with weak power-law continuum.
2571: The resonance (r) and intercombination (i) lines
2572: are detected but the forbidden (f) line is not.}
2573: \end{figure}
2574:
2575:
2576: \clearpage
2577:
2578: \begin{figure}
2579: \figurenum{20}
2580: \epsscale{1.0}
2581: %%\includegraphics*[width=12.5cm,angle=-90]{f20.eps}
2582: \includegraphics*[width=8.5cm,angle=-90]{f20.eps}
2583: \caption{Histogram plot of the MEG1 spectrum ($+$1 and $-$1 orders combined)
2584: of $\sigma$ Ori AB near the Ne IX He-like triplet, binned to a minimum
2585: of 10 counts per bin. Solid squares are data points and the dotted
2586: line is a Gaussian fit with weak power-law continuum.
2587: The resonance (r) and intercombination (i) lines
2588: are detected but the forbidden (f) line is not.}
2589: \end{figure}
2590:
2591:
2592: \clearpage
2593:
2594: \begin{figure}
2595: \figurenum{21}
2596: \epsscale{1.0}
2597: %%\includegraphics*[width=12.5cm,angle=-90]{f21.eps}
2598: \includegraphics*[width=8.5cm,angle=-90]{f21.eps}
2599: \caption{Histogram plot of the MEG1 spectrum ($+$1 and $-$1 orders combined)
2600: of $\sigma$ Ori AB near the O VII He-like triplet, binned to a minimum
2601: of 10 counts per bin. Solid squares are data points and the dotted
2602: line is a Gaussian fit with weak power-law continuum.
2603: The resonance (r) and intercombination (i) lines
2604: are detected but the forbidden (f) line is not.}
2605: \end{figure}
2606:
2607: \clearpage
2608:
2609:
2610: \begin{figure}
2611: \figurenum{22}
2612: \epsscale{1.0}
2613: %%\includegraphics*[width=12.5cm,angle=-90]{f22.eps}
2614: \includegraphics*[width=8.5cm,angle=-90]{f22.eps}
2615: \caption{The ratio of forbidden-to-intercombination Mg XI
2616: line fluxes for $\sigma$ Ori AB as a function of distance
2617: from the center of the star. The horizontal axis expresses
2618: the distance in units of the stellar radius, where
2619: R$_{*}$ = 9 R$_{\odot}$. The solid line shows the predicted
2620: $R$ = f/i flux ratio based on a TLUSTY atmospheric model
2621: with T$_{eff}$ = 32,500 K (see text). The vertical dashed line
2622: marks the 90\% upper limit on the line formation radius
2623: r$_{fi}$ $<$ 1.8 R$_{*}$ (referenced to the center of the star)
2624: based on the limit $R$ $<$ 0.17 determined from
2625: spectral fits (Table 5).
2626: }
2627:
2628: \end{figure}
2629:
2630: \clearpage
2631:
2632:
2633:
2634:
2635:
2636:
2637:
2638:
2639:
2640:
2641: \end{document}
2642:
2643:
2644: %% Optional figure: bvapec fit of meg1
2645:
2646: \begin{figure}
2647: \figurenum{17}
2648: \epsscale{1.0}
2649: %%\includegraphics*[width=12.5cm,angle=-90]{f17.eps}
2650: \includegraphics*[width=8.5cm,angle=-90]{f17.eps}
2651: \caption{HETG/MEG1 spectrum of $\sigma$ Ori AB showing
2652: a best fit absorbed velocity-broadened optically thin plasma
2653: model ({\em bvapec}) using solar abundances. The best-fit parameters
2654: (Table XX) are kT = 0.32 keV and a velocity broadening with Gaussian
2655: $\sigma$ equivalent to $v$ = 279 km s$^{-1}$. The absorption column
2656: density was held fixed at N$_{\rm H}$ = 3e20 cm$^{-2}$. }
2657: \end{figure}
2658:
2659: \clearpage
2660:
2661:
2662: \end{document}
2663:
2664:
2665:
2666:
2667:
2668:
2669:
2670:
2671:
2672:
2673:
2674:
2675:
2676:
2677:
2678:
2679:
2680:
2681:
2682:
2683:
2684:
2685:
2686:
2687:
2688:
2689:
2690:
2691:
2692:
2693: