0809.3495/ms.tex
1: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2: %%% PASJ LaTeX template for draft(body)<2007/01/19>
3: %%% 
4: %%% IMPORTANT NOTICE FOR AUTHORS
5: %%% 1. ``\draft'' creates single column and double spaces format.
6: %%% 2. If you comment out ``\draft'', the output will be double column
7: %%%    and single space.
8: %%% 3. For cross-references, the use of \label/\ref/\cite and the 
9: %%%    thebibliography environment is strongly recommended
10: %%% 4. Do NOT use \def/\renewcommand.
11: %%% 5. Do NOT redifine commands provided by PASJ00.cls.
12: %%% 
13: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
14: %%
15: %% REVISION HISTORY 
16: %%
17: %% 2008/05/04 Submitted to PASJ
18: %% 2008/07/15 Reflect Referee comments (PASJ 3374_1), e-mailed to KH
19: %% 2008/07/25 Reflect KH's revisions
20: %% 2008/08/02 Reflect KH's revisions2
21: %% 2008/08/08 Reflect YS's revisions
22: %% 2008/09/18 Reflect Referee's last comment and KH's comments
23: %% 
24: \documentclass{pasj00}
25: \draft
26: 
27: \usepackage{color}
28: \usepackage{ulem}
29: 
30: \begin{document}
31: \SetRunningHead{Y. Ezoe et al.}{
32: Suzaku and XMM-Newton Observations of the Eastern Tip 
33: Region of the Carina Nebula}
34: \Received{}%{yyyy/mm/dd}
35: \Accepted{}%{yyyy/mm/dd}
36: 
37: \title{
38:   Suzaku and XMM-Newton Observations of Diffuse X-ray Emission 
39:   from the Eastern Tip Region of the Carina Nebula
40: }
41: 
42: \author{
43:   Yuichiro \textsc{Ezoe}\altaffilmark{1},
44:   Kenji \textsc{Hamaguchi}\altaffilmark{2,3},
45:   Robert A. \textsc{Gruendl}\altaffilmark{4},
46:   You-Hua \textsc{Chu}\altaffilmark{4},\\
47:   Robert \textsc{Petre}\altaffilmark{5}, and
48:   Michael F. \textsc{Corcoran}\altaffilmark{2,3}
49: %%  and  Leisa K. \textsc{Townsley}\altaffilmark{6}
50: }
51: 
52: \altaffiltext{1}{
53:   Tokyo Metropolitan University, 1-1, Minami-Osawa, Hachioji,
54:   Tokyo, 192-0397, JAPAN
55: %  RIKEN (The Institute of Physical and Chemical Research), 
56: %  2-1, Hirosawa, Wako, Saitama 351-0198, JAPAN
57: }
58: 
59: %\altaffiltext{2}{The Institute of Space and Astronautical Science,
60: %                 3-1-1 Yoshinodai, Sagamihara, Kanagawa 229-8510, JAPAN
61: %}
62: 
63: \altaffiltext{2}{
64:   CRESST and X-ray Astrophysics Laboratory NASA/GSFC, Greenbelt, MD 20771, USA
65: }
66: 
67: \altaffiltext{3}{
68:   Universities Space Research Association,
69:   10211 Wincopin Circle, Suite 500, Columbia, MD 21044-3432, USA
70: }
71: 
72: \altaffiltext{4}{
73:   Department of Astrophysics, University of Illinois, Urbana, IL 61801, USA
74: }
75: 
76: \altaffiltext{5}{
77: Astrophysics Science Division, NASA Goddard Space Flight Center, Greenbelt, MD 20771
78: }
79: 
80: %\altaffiltext{6}{
81: %Department of Astronomy and Astrophysics, 525 Davey Laboratory, Pennsylvania State
82: %University, University Park, PA 16802
83: %}
84: 
85: \email{ezoe@phys.metro-u.ac.jp}
86: 
87: \KeyWords{X-ray: ISM --- ISM: abundances --- ISM: individual (Carina nebula) --- 
88: stars: winds --- stars: supernovae: general} 
89: 
90: \maketitle
91: 
92: \begin{abstract}
93: The eastern tip region of the Carina Nebula was observed with the Suzaku XIS 
94: for 77 ks
95: to conduct a high-precision spectral study of extended X-ray emission.
96: %%{\bf
97: XMM-Newton EPIC data of this region were also utilized 
98: to detect point sources.
99: %%}
100: The XIS detected strong extended X-ray emission from the entire field-of-view 
101: with a 0.2--5 keV flux of $0.7\sim4\times10^{-14}$ erg s$^{-1}$ arcmin$^{-2}$.
102: The emission has a blob-like structure that coincides with an ionized gas filament observed
103: in mid-infrared images.
104: %%{\bf
105: Contributions of astrophysical backgrounds and 
106: the detected point sources were insignificant.
107: Thus the emission is diffuse in nature. 
108: %%}
109: %%
110: The X-ray spectrum of the diffuse emission was represented by a two-temperature
111: plasma model with temperatures of 0.3 and 0.6 keV and an absorption column 
112: density of 2$\times10^{21}$ cm$^{-1}$. 
113: %%
114: The X-ray emission showed normal nitrogen-to-oxygen abundance ratios 
115: and a high iron-to-oxygen abundance ratio.
116: %%
117: The spectrally deduced parameters, such as temperatures and column densities, 
118: are common to the diffuse X-ray emission near $\eta$ Car. 
119: Thus, the diffuse X-ray emission in these two fields may have the same origin.
120: %%
121: The spectral fitting results are discussed to constrain the origin in the context 
122: of stellar winds and supernovae.
123: \end{abstract}
124: 
125: \section{Introduction}
126: \label{sec:intro}
127: 
128: Diffuse X-ray emission extending over several to tens pc 
129: has been reported in many massive star-forming regions, such as
130: NGC 2024 \citep{Ezoe2006a}, 
131: %%{\bf
132: the Orion Nebula \citep{Guedel2008},
133: %%}
134: the Rosette Nebula, \citep{Townsley2003},
135: M17 \citep{Townsley2003,Hyodo2008},
136: RCW 38 \citep{Wolk2002},
137: NGC 6334 \citep{Ezoe2006b},
138: the Carina Nebula \citep{Hamaguchi2007},
139: %%\textcolor{red}{Put a Leisa's proceeding as well}
140: W49 \citep{Tsujimoto2006},
141: NGC 3603 \citep{Moffat2002},
142: the Arches Cluster \citep{Yusef-Zadeh2002,Tsujimoto2007},
143: and the Quintuplet Cluster \citep{Wang2002}.
144: %%
145: This diffuse component contributes a considerable fraction of the 
146: total X-ray emission and shows different spectral characteristics 
147: among these regions.
148: Diffuse X-ray emission can be roughly classified into three types:
149: thin-thermal plasma emission with a temperature $kT\sim$ 0.1--1 keV, 
150: higher-temperature plasma emission with $kT\sim$ 2--10 keV,
151: and possibly non-thermal emission with a photon index of 1-1.5.
152: %%
153: These phenomena can be explained by plasma heating and particle 
154: acceleration in strong shocks by fast stellar winds from young 
155: OB stars \citep{Townsley2003,Ezoe2006a,Ezoe2006b,Guedel2008}
156: and/or past supernova remnants (SNRs) \citep{Wolk2002,Hamaguchi2007}.
157: The precise origin of diffuse X-ray emission, however, is often
158: unclear.
159: 
160: Recently \citet{Hamaguchi2007} suggest that the origin of diffuse 
161: X-ray emission can be constrained by plasma diagnostics or measurements
162: of elemental abundances.
163: %%
164: %%{\bf 2
165: While main-sequence late-O to early B stars have nearly solar abundances
166: (e.g., \cite{Cunha1994,Daflon2004}), evolved stars show non-solar elemental 
167: compositions due to the CNO cycle.
168: %%}
169: %%Evolved stars show non-solar elemental compositions due to the 
170: %%CNO cycle. 
171: For instance, the plasma will be overabundant in nitrogen if its origin 
172: is the wind from a nitrogen-rich Wolf-Rayet 
173: star. 
174: %%
175: On the other hand, the plasma will be overabundant in oxygen, neon, and 
176: silicon if it is produced by a Type II SNR (e.g., \cite{Tsujimoto1995}).
177: %%
178: The low-temperature ($kT\sim$0.1--1 keV) type of diffuse X-ray 
179: emission is ideal for such diagnostic studies, because a variety 
180: of K-shell lines exist in the 0.2--2 keV range.
181: 
182: The Carina Nebula is an excellent site to investigate 
183: plasma diagnostics of diffuse X-ray emission.
184: %%
185: At a distance of 2.3 kpc, it is one 
186: of the most active massive star forming regions in the Galaxy.
187: %%
188: It contains eight massive stellar clusters: Trumpler 14, 15, 16, 
189: Collinder 228, Bochum 10, 11, NGC 3293, and NGC 3324.
190: In total, there exist more than 64 O stars \citep{Feinstein1995,
191: Smith2006}, including the extreme-type luminous blue variable
192: $\eta$ Car and four Wolf-Rayet stars.
193: %
194: % YS's comment
195: %
196: % I checked van der Hucht's catalog of Galactic WR stars and found
197: % at least 4 WR stars within the boundary of the Carina Nebula!
198: %
199: % that include two rate main-sequence O3 stars 
200: %($\eta$ Car and HD 93129A) and two Wolf-Rayet stars.
201: %
202: %%
203: The age of the nebula is estimated to be $\sim3$ Myr based 
204: on the most evolved stars and the size of the HII region 
205: \citep{Smith2000}.
206: %%
207: The young massive stars that are still enshrouded in gas and dust
208: have been observed in optical, infrared, and radio wavelengths 
209: (e.g., \cite{Harvey1979, Smith2000, Yonekura2005}).
210: %%
211: The number counts of the most massive O stars, e.g., O3 stars,
212: suggest that the star-formation activity of the Carina Nebula 
213: rivals those of the most active regions, such as NGC 3603 
214: at $D=6.9$ kpc and W49 at $D=11.4$ kpc.
215: %%
216: The proximity of the Carina Nebula, compared with NGC 3603 and W49,
217: makes it the best target to study diffuse X-ray emission
218: resulting from star-forming activities.
219: 
220: Seward et al.\ (1979) first suggested the existence of possible
221: diffuse soft X-ray emission in the Carina Nebula with a luminosity of
222: $\sim$10$^{35}$ erg s$^{-1}$ and a spatial extent of several pc,  
223: using {Einstein} observations.
224: %%
225: Although the limited energy and spatial resolution of Einstein 
226: hindered the determination of precise plasma parameters and 
227: the contribution from point sources, \citet{Seward1982}
228: postulated that the extended X-rays were from hot gas with 
229: $T\sim10^{7}$ K. 
230: %%
231: With {Chandra}, Evans et al.\ (2003) confirmed the existence
232: of diffuse X-ray emission near $\eta$ Car in addition to 
233: point sources; however, the limited photon statistics and 
234: high background prevented detailed spectral analysis.
235: %%
236: Hamaguchi et al.\ (2007) obtained spectra
237: of the diffuse X-ray emission around $\eta$ Car with {Suzaku}.
238: %%
239: Owing to the good low-energy response and the low background 
240: of the X-ray CCD onboard {Suzaku}, % (XIS: X-ray Imaging Spectrometer), 
241: the spectra extracted from the regions north and south of 
242: $\eta$ Car can be modeled with high precision; both spectra
243: are best represented by three-temperature plasma models with 
244: $kT\sim0.2$, $\sim0.6$, and $\sim5$ keV.
245: %%
246: Analyzing the {Suzaku} data in conjunction with XMM-Newton and 
247: Chandra data, Hamaguchi et al.\ (2007) concluded that the 0.2 keV
248: and 0.6 keV components most likely originated from diffuse plasma, 
249: but the 5 keV component could not be easily distinguished from
250: the unresolved point sources.
251: %%
252: They found that the iron and silicon abundances were significantly 
253: different in the north and south regions, and that the nitrogen-to-oxygen 
254: abundance ratios in both regions were far lower than those of stellar winds 
255: from evolved massive stars such as $\eta$ Car and WR25 in this field.
256: %%
257: From these results, they concluded that the diffuse X-ray emission
258: near $\eta$ Car originated from one or multiple SNRs.
259: 
260: We have studied extended X-ray emission 
261: from an eastern tip region of the Carina Nebula.
262: %%
263: Located $\sim$ 30$'$ ($\sim$ 20 pc) from $\eta$ Car, this region is 
264: less contaminated by X-ray emission from OB stars than the regions 
265: near $\eta$ Car.
266: %%
267: Previous {Einstein} observations have revealed strong extended X-ray 
268: emission in this region, although no massive stars earlier than B3 
269: are known here \citep{Seward1982}.
270: %%
271: It is not known whether this emission is truly diffuse and, 
272: if so, whether its origin is similar to that of the regions near $\eta$ Car.
273: %%
274: In this paper we report the first detailed spectral analysis of 
275: the extended X-ray emission in this eastern tip region of the 
276: Carina Nebula using Suzaku observation.  To augment the limited 
277: angular resolution of Suzaku, the analysis also made use of
278: {XMM-Newton} data.
279: 
280: \section{Suzaku Observation}
281: \label{sec:obs}
282: 
283: Suzaku is the 5th Japanese X-ray observatory \citep{Mitsuda2007}.
284: It carries four scientific instruments; 
285: X-ray optics or the X-Ray Telescope (XRT: \cite{Serlemitsos2007}); 
286: an X-ray calorimeter (XRS: \cite{Kelly2007});
287: four X-ray CCDs (XIS: \cite{Koyama2007});
288: and a hard X-ray detector (HXD: \cite{Takahashi2007,Kokubun2007}).
289: %%
290: The XIS consists of three front-illuminated (FI) CCDs 
291: (XIS0, 2, and 3) and one back-illuminated (BI) CCD (XIS1).
292: %%
293: Due to the low-earth orbit of Suzaku and the large effective 
294: area, the XIS has the lowest particle backgrounds 
295: among all X-ray CCDs in currently available  X-ray observatories.
296: %%
297: Furthermore, the XIS has good energy resolution
298: and superior low-energy response with negligible low-energy tails,
299: compared to the other X-ray CCDs onboard XMM-Newton and Chandra.
300: 
301: We observed the eastern tip region of the Carina Nebula 
302: with Suzaku on 2006 June 5. 
303: %%
304: During the observation, the XIS and HXD were operated in the normal 
305: mode. In the present paper, we use only the XIS data because we are 
306: interested in the spectral analysis of the soft extended X-ray emission.
307: 
308: The data reduction was performed on the version 1.2.2.3 
309: screened data provided by the Suzaku processing facility,
310: using the HEAsoft analysis package ver 6.1.1.
311: %%
312: No background flares were seen in the data.
313: The net exposure of each FI and BI chip was 77 ks.
314: %%
315: For spectral fits, we generated response matrices and auxiliary 
316: files with {\tt xisrmfgen} and {\tt xissimarfgen} released on 
317: 2006 October 26.
318: %%
319: The in-flight gradual degradation of energy resolution and absorption 
320: due to XIS contamination were considered in these softwares.
321: 
322: \section{Extended X-ray Emission}
323: \label{sec:diff}
324: 
325: Figure \ref{fig:overview}a shows the location of our observation
326: (top left box) on an MSX 8.28 $\mu$m image of the Carina Nebula 
327: retrieved from the NASA/IPAC Infrared Science Archive 
328: \footnote{http://irsa.ipac.caltech.edu/applications/MSX/MSX/}.
329: %%
330: This mid-IR image is dominated by cationic polycyclic aromatic 
331: hydrocarbon emission in photodissociation regions \citep{Smith2000}
332: and traces the surface of molecular clouds that are ionized by 
333: stellar winds and ultra-violet radiations from OB stars. 
334: A curved mid-infrared filament runs across the XIS field-of-view (FOV).
335: %%
336: Figure \ref{fig:overview}b shows an X-ray overview of the same
337: area with XMM-Newton MOS\footnote{http://xmm.esac.esa.int/external/xmm\_science/gallery/}.
338: In X-rays, there exists patchy extended soft X-ray emission whose
339: distribution roughly follows the mid-infrared filament.
340: %%
341: This suggests that the X-ray emission contributes to the ionization
342: of the molecular cloud and that the emitting hot gas is in contact 
343: with the surface of the molecular clouds.
344: 
345: We created Suzaku XIS images in 0.2--2 and 2--10 keV bands, 
346: as shown in figure \ref{fig:xisimage}.
347: %%
348: The vignetting in the images has been corrected by dividing
349: the observed images by model XIS images produced with the XRT$+$XIS 
350: simulator {\bf xissim} for a uniform surface brightness.
351: %%
352: In simulations, we assumed monochromatic X-rays of energy 
353: 1.49 keV and 8.05 keV, since the vignetting is best studied 
354: in these energies \citep{Serlemitsos2007}.
355: %%
356: A blob-like extended X-ray emission is clearly detected 
357: in the soft X-ray band.
358: %%
359: No significant X-ray emission is seen in the hard X-ray band, which
360: is consistent with the XMM-Newton image (figure~\ref{fig:overview} b).
361: 
362: To evaluate the significance of the extended X-ray emission, 
363: we defined a region named blob, as shown in figure~\ref{fig:xisimage}.
364: %%
365: The total area of the blob region is 234 arcmin$^2$, or 105 pc$^2$. 
366: We extracted the XIS0-3 spectra from this region. 
367: %%
368: To evaluate the background, we utilized spectra accumulated 
369: from observations of the night side of the Earth, as the night
370: Earth backgrounds reproduced well all the observed spectra above 
371: $\sim$6 keV, where the instrumental background dominated.
372: %%
373: The background-subtracted XIS1 (BI) and XIS$0+2+3$ (FI) 
374: spectra are shown in figure \ref{fig:xisdiffspec}a, 
375: and their 0.2--5 keV count rates with 1$\sigma$ uncertainties 
376: are $1.073\pm0.004$ (BI) and  $0.609\pm0.002$ (FI) counts s$^{-1}$, 
377: respectively.
378: %%
379: The emission is highly significant and shows a 
380: number of emission lines such as O VII, O VIII, 
381: Ne IX, Ne X, Mg XI, Si XIII, and S XV.
382: %%
383: Lines from these ions in different ionization states,
384: such as O VII and Si XIII, suggest that the extended X-rays 
385: are from not a single-temperature but multi-temperature plasma.
386: 
387: We extracted spectra from two surrounding regions named 
388: east and nw, as shown in figure~\ref{fig:xisimage}. 
389: The areas of the east and nw regions are 47 arcmin$^2$ (or 21 pc$^2$)
390: and 21 arcmin$^2$ (or 9.3 pc$^2$), respectively.
391: %%
392: We again used the night Earth spectra at the respective detector 
393: positions as backgrounds. The background-subtracted spectra 
394: are shown in figures \ref{fig:xisdiffspec}b and c.
395: %%
396: Although their surface brightnesses are an order-of-magnitude 
397: lower than that in the blob region, there exist signs of 
398: emission lines from O VII, Ne IX, Ne X, Mg XI, and Si XIII.
399: Thus extended X-ray emission appears to be present over 
400: the entire FOV.
401: %%
402: The surface brightness below 0.3 keV and above 2 keV 
403: is similar in all three regions.
404: %%
405: Therefore, in addition to the plasma emission that is 
406: dominant between 0.3 and 2 keV, there must be 
407: additional X-ray sources.
408: %%
409: Plausible candidates are the local hot bubble (LHB), 
410: the cosmic X-ray background (CXB), the galactic ridge 
411: X-ray emission (GRXE), and point sources.
412: %%
413: LHB and CXB are uniform and common 
414: background sources existing in all X-ray observations.
415: %%
416: GRXE is also a uniform X-ray background and must 
417: contribute to the emission because the observational
418: FOV is located in the Galactic plane 
419: ($l=288^\circ$, $b=-0^\circ\hspace*{-1mm}.4$,
420: see figure~\ref{fig:overview}a).
421: %%
422: The point sources, on the other hand, are position-dependent.
423: %%
424: Contamination from these astrophysical backgrounds
425: and point sources needs to be carefully considered in order to
426: characterize the possible diffuse emission from the Carina
427: Nebula.
428: 
429: \section{Point Sources}
430: \label{sec:ps}
431: 
432: To quantify the contribution from point sources, 
433: we analyzed the XMM-Newton data.
434: %%
435: XMM-Newton observed this region on 2004 December 7 for 27 ks.
436: %% 
437: The European Photon Imaging Camera (EPIC) provided CCD imaging
438: spectroscopy with one pn  camera \citep{Struder2001} and two MOS
439: cameras \citep{Turner2001}.
440: The medium optical blocking filter was used. 
441: %%
442: As shown in figure \ref{fig:overview}b, the FOVs of MOS cover 
443: the entire FOV of the XIS.
444: %%
445: We analyzed the archival processed data using SAS (Science Analysis 
446: Software) version 7.0.0, following the SAS user's 
447: guide\footnote{http://xmm.esac.esa.int/external/xmm\_user\_support/documentation/sas\_usg/USG/}.
448: %%
449: The event files were time-filtered to exclude periods of high 
450: background, during which the count rate from the entire CCD area 
451: at energies $>$10 keV is more than 1.2 times the average rate for
452: each MOS and pn observation.
453: %%
454: This removed $\sim$1 ks from each observation, and yielded 22 and 
455: 26 ks\sout{ec} of usable exposure for pn and each MOS, respectively.
456: %%
457: Source detection was accomplished with the SAS program 
458: {\tt edetect\_chain}.
459: Images in 8 bands (0.2--0.5, 0.5--2.0, 2.0--4.5,
460: 4.5--7.5, 7.5--12.0, and 0.2--12.0 keV) were utilized, 
461: in order not to miss very soft or very hard sources.
462: %%
463: The resulting source list was checked manually for spurious 
464: detections and missed sources.
465: %%
466: As a result, 10 sources have been detected, among which
467: 5 sources were within the FOV of the XIS.
468: %%
469: The locations of these X-ray point sources are marked by
470: circles in figure \ref{fig:overview}b.
471: 
472: For each individual point source, we extracted counts using a circular 
473: region of a 30 arcsec radius centered at the source.
474: %%
475: Background counts were extracted from an adjacent annular region 
476: with an outer radius of 1 arcmin.
477: In the case that the annular region included other sources, the 
478: background was then extracted from a nearby circular source-free region.
479: %%
480: To identify counterparts of these X-ray point sources in other 
481: wavelengths, we searched the 2$\mu$m all sky survey catalog (2MASS)\footnote{http://irsa.ipac.caltech.edu/index.html}
482: and the AXAF Guide and Acquisition Star Catalog (AGAST)\footnote{http://cxc.harvard.edu/cgi-gen/cda/agasc/agascInterface.pl}
483: for candidates within 10 arcsec, the angular resolution of XMM-Newton.
484: %%
485: In cases where multiple candidates exist, we chose the closest one as the
486: most plausible counterpart.
487: %%
488: Among the 10 X-ray sources, 9 have counterparts in 2MASS 
489: and 1 has a counterpart in AGAST.
490: %%
491: The properties of these individual sources are summarized in table \ref{tbl:srclist}.
492: 
493: %Because significant X-ray time variation is one of the 
494: %characteristics of young low-mass stars, 
495: We examined the 10 sources for possible temporal variations. 
496: %%
497: %%For each source, we produced 0.4--10 keV X-ray light curves 
498: %%of pn and MOS using a binning size of 512 s bin$^{-1}$. 
499: For each source, we produced pn and MOS X-ray light curves
500: in the 0.4--10 keV band using a binning size of 512 s bin$^{-1}$. 
501: %%
502: The light curves were examined against a constant hypothesis 
503: in terms of $\chi^2$ statistics. 
504: %%
505: If the $\chi^{2}$ probability of constancy became less than 4\%, 
506: at least in one detector, we regarded the source as variable.
507: %%
508: Only the source No.3 has been found to be variable.
509: %%
510: It showed a factor of about 5 increase in the first 5 ks of the 
511: observation, decreased in the next 5 ks, and stayed constant in the rest.
512: %%
513: Such rapid temporal variations strongly suggest that No.3 
514: is a young low-mass star.
515: 
516: We also conducted spectral analysis for 4 bright sources 
517: (No.1, 2, 4, and 7) that had $>100$ counts in the pn observation.
518: %%
519: We analyzed only the pn spectra because of the limited statistics 
520: of MOS. 
521: %%
522: The SAS tasks {\tt rmfgen} and {\tt arfgen} were utilized to 
523: generate response matrix files and auxiliary files for each source.
524: %%
525: We fitted the spectra using a thin-thermal plasma emission model 
526: in collisional equilibrium (the APEC mode; \cite{Smith2001}) 
527: convolved with the interstellar absorption.
528: Such models, with abundances fixed at 0.3 solar, are commonly 
529: used in X-ray spectral analyses of star-forming regions 
530: (e.g., \cite{Getman2005}).
531: %%
532: We found that all the spectra except that of No.1 were well represented 
533: by this simple model.
534: %%
535: For the No.1 spectrum, we tried a two-temperature plasma emission model 
536: with a common absorption, and were able to find acceptable fits.
537: %%
538: The fitting results are summarized in table \ref{tbl:srcfit} and 
539: figure \ref{fig:xmmsrcspec}.
540: 
541: Source No.1 has a very hard continuum without any sign of emission 
542: lines, thus it may be a background active galactic nucleus.
543: %% Hama 080726 
544: %% \textcolor{red}{Can you try a thermal + power-law model?}
545: %%
546: Source No.2 may be an embedded young low-mass star because 
547: it has a large absorption column density and a moderate temperature,
548: while source No.4 may be a foreground star because of its
549: small absorption column and moderate temperature.
550: %%
551: Source No.7 is peculiar with a low temperature and possibly high luminosity.
552: %%
553: Its position is within 8 arcsec from the X-ray source No.78 in an
554: XMM-Newton observation of $\eta$ Car reported by \citet{Colombo2003}.
555: %%
556: %%If they correspond to the same source, its luminosity varied by a 
557: %%factor of 4 between these two observations.
558: %%
559: %%Such a large temporal variability suggests a young low-mass star but its 
560: %%temperature and luminosity are {\bf similar to} those of evolved OB stars.
561: %%
562: Further optical spectroscopic study is needed to identify the nature 
563: of this source.
564: 
565: We estimated X-ray fluxes of the other 6 sources assuming a 
566: thin-thermal plasma model and convolving it with an estimated interstellar 
567: absorption of $N_{\rm H}=1.3\times10^{22}$ cm$^{-2}$.
568: %%
569: %% Hama 080726
570: %% \textcolor{red}{Question: 
571: %% Why is this value? The absorption to the Carina nebula is 2-3$\times$10$^{22}$, right?
572: %% Do you assume these are extragalactic sources?}
573: %%
574: We adopted a temperature of 3 keV and a metal abundance of 0.3 solar
575: that are typical of emission from young stars (e.g., \cite{Imanishi2001}).
576: %%
577: WebPIMMS\footnote{http://heasarc.gsfc.nasa.gov/Tools/w3pimms.html}
578: was used to convert the pn count rates to X-ray fluxes. The results are shown 
579: in table \ref{tbl:srclist}.
580: %%
581: The fluxes range from 2$\times$10$^{-14}$ to 4$\times$10$^{-13}$
582: erg s$^{-1}$ cm$^{-2}$.
583: 
584: \section{Contamination from Astrophysical Backgrounds
585: and Point Sources}
586: \label{sec:contami}
587: 
588: We proceeded to estimate contribution 
589: to the observed X-ray emission by 
590: LHB, CXB, GRXE, and point sources.
591: %%
592: For the LHB, CXB, and GRXE,
593: we assumed the same models in \citet{Hamaguchi2007}:
594: the Raymond-Smith thin-thermal plasma model with 
595: $kT\sim0.1$ keV and a surface brightness of $\sim4\times10^{4}$
596: counts s$^{-1}$ arcmin$^{-2}$ for LHB based on \citet{Snowden1998},
597: the model Id1 in table 2 of \citet{Miyaji1998} for CXB, and
598: the free abundance model in table 8 of \citet{Ebisawa2005}
599: with the X-ray flux of $1.4\times10^{-11}$ erg cm$^{-2}$ 
600: s$^{-1}$ deg$^{-2}$ (3--20 keV) for GRXE.
601: %%
602: We used the same spectral parameters except 
603: for the normalization to fit, which was adjusted.
604: %%
605: To estimate the X-ray fluxes from these components with
606: different spatial distribution, 
607: we prepared an arf file for
608: the uniform emission (LHB, CXB, and GRXE)
609: by using the {\tt xissimarfgen} program, while
610: we created an arf file for each point source.
611: %%
612: We used the best-fit models in table \ref{tbl:srcfit}
613: for the bright point sources and assumed the typical
614: spectral model of young low-mass stars for the other
615: faint point sources ($kT=3$ keV and $N_{\rm H}=1.3\times10^{22}$ cm$^{-2}$)
616: using the APEC emission code.
617: 
618: In figure \ref{fig:bgdest}a, we plot the 
619: estimated contamination from LHB, CXB, GRXE, and 
620: point sources in the blob region.
621: %%
622: Below 0.3 keV, LHB accounts for the X-ray emission, while
623: X-rays above 2 keV can be explained by the sum of CXB, 
624: GRXE and point sources.
625: This makes a sharp contrast to the $\eta$ Car region 
626: where a residual hard X-ray emission is seen above 2 keV.
627: %%
628: The excellent fit of these sources to the spectrum 
629: below 0.3 keV and above 2 keV supports the validity 
630: of our estimation.
631: %%
632: Thus, almost all the excess counts in 0.3--2 keV can be 
633: considered to be truly diffuse plasma emission.
634: %%
635: This conclusion is also supported by the clear spectral 
636: differences between the diffuse plasma and point sources.
637: 
638: In the same way, we estimated the X-ray contamination 
639: in the east and nw regions as shown in figures 
640: \ref{fig:bgdest}b and c.
641: %%
642: In both regions, LHB, CXB, GRXE, and point sources 
643: explain all the emission in $<0.3$ and $>2$ keV 
644: as well as in the blob region, but there are still 
645: excesses in 0.3--2 keV.
646: %%
647: Therefore, diffuse soft X-ray emission exists in these fields, too.
648: 
649: \section{Spectral Analysis}
650: \label{sec:spec}
651: 
652: \subsection{The blob Region}
653: \label{sec:spec:blob}
654: 
655: The nature of the diffuse X-ray emission is investigated
656: through spectral analysis.
657: We simultaneously fitted the 0.2--5 keV XIS 
658: BI and FI spectra of the blob region.
659: %%
660: %%{\bf
661: We created XIS arf files using {\tt xissimarfgen},
662: assuming the 0.4--2 keV XMM MOS image in figure \ref{fig:overview}b
663: as spatial distribution of the diffuse X-ray emission.
664: %%}
665: %%
666: In single or two-temperature plasma models utilized below, 
667: we allow the abundances of the noticeable elements (O, Ne, Mg, Si, 
668: S, and Fe) to vary, while those of the other elements were 
669: fixed at 0.3 solar value, which is generally seen in low-resolution
670: CCD spectra of young stars.
671: 
672: In addition to the plasma model for the diffuse X-ray emission, 
673: we introduced a thin-thermal plasma model and a power-law 
674: model to reproduce the LHB, CXB, GRXE, and point source
675: contributions (\S \ref{sec:contami}).
676: %%
677: For the plasma model of LHB, we fixed the temperature
678: at 0.1 keV and the abundances at 1 solar, but allowed
679: the normalization to vary.
680: %%
681: For simplicity, the CXB, GRXE, and point sources were together
682: approximated by a single power-law model with the photon
683: index fixed at 1.5 and convolved with the same absorption 
684: of the diffuse X-ray emission.
685: %%
686: To take into account possible uncertainties in the energy 
687: scale calibration, we introduced two additional fitting
688: parameters, gain and offset.
689: %%
690: Throughout the fittings below, the best-fit energy scale 
691: and offset values were $<$1 \% and $<$6 eV, respectively, 
692: consistent with the current calibration uncertainties
693: \footnote{http://www.astro.isas.jaxa.jp/suzaku/process/caveats/}.
694: 
695: We first tested a single-temperature thin-thermal 
696: plasma model convolved with an absorption.
697: %%
698: This simple model yielded an unacceptable best fit 
699: for the spectra with $\chi^2$/d.o.f. of $2.4$.
700: The best-fit temperature, $kT$ = 0.59 keV, was too 
701: high to reproduce the significant OVII and NeIX lines.
702: %%
703: We then tried a commonly-absorbed 
704: two-temperature plasma model as shown in figure 
705: \ref{fig:fit12}a and table \ref{tbl:fit1} (model 1).
706: %%
707: This model represents the data far better ($\chi^2$/d.o.f. of $1.2$). 
708: %% 
709: A small discrepancy of the fit to the data at 0.5--0.8 keV
710: and 1.1--1.2 keV may be caused by inaccuracy of the Fe 
711: L-shell emission line model and calibration uncertainty 
712: near the mirror Au L edges, respectively.   
713: %%
714: Indeed, similar discrepancies at 0.5--0.8 keV can be seen in
715: other Suzaku XIS spectral fits (e.g., \cite{Hamaguchi2007}).
716: %%
717: Therefore, we think the best-fit two-temperature model 
718: represents the data well, although the $\chi^2$ is still 
719: not above the 90\% confidence level.
720: 
721: \subsection{The east and nw Regions}
722: \label{sec:spec:eastnw}
723: 
724: Similar to the analysis of the blob region, we fitted the east 
725: and nw spectra with a two-temperature plasma model.
726: Since no evident spatial structures were seen in these regions 
727: (figure \ref{fig:xisimage}),
728: we created arf files assuming a uniform emission.
729: %%
730: We did not use the additional fitting parameters gain and offset
731: because  the photon statistics were limited.
732: The results are shown in figures \ref{fig:fit12}b and c, and table \ref{tbl:fit1}. 
733: %%
734: In both regions, the fits are acceptable and all resulting 
735: parameters except the surface brightness are consistent with 
736: those in the blob region, although the uncertainties are large.
737: %%
738: As expected from figure \ref{fig:xisimage}, the surface brightness 
739: of the east and nw regions is a factor of $\sim$5 lower than 
740: that in the blob region.
741: 
742: We also tried a single-temperature plasma model and 
743: obtained acceptable fits in both regions, but the best-fit 
744: column densities were small, $\sim8\times10^{20}$ cm$^{-2}$,
745: and the best-fit temperatures were high, $\sim$0.6 keV.
746: %%
747: Such a large variation in the column density within the XIS FOV
748: is inconsistent with the CO map (figure \ref{fig:overview}a), 
749: although the CO gas can lie behind the X-ray emitting gas.
750: %%
751: There is also a hint of OVII K emission in both spectra 
752: (figures \ref{fig:xisdiffspec} b and c) that cannot be reproduced
753: by this best-fit single-temperature model.
754: %%
755: Therefore, it is likely that the east and nw spectra are also
756: best fitted by a two-temperature plasma model.
757: 
758: \subsection{XMM-Newton spectra}
759: \label{sec:spec:xmm}
760: 
761: We also analyzed the XMM-Newton observations of the blob region
762: with the same data used in the point source analysis (\S \ref{sec:ps}).
763: %%
764: We used only the MOS data with relatively low particle background events,
765: whose background spectrum during the observation can be easily estimated
766: from the ESAS package\footnote{http://heasarc.gsfc.nasa.gov/docs/xmm/xmmhp\_xmmesas.html}.
767: %%
768: We used SAS version 7.0.0 and ESAS version 1.0 for the analysis, and
769: generated response matrices using the {\tt rmfgen} and {\tt arfgen}.
770: %%
771: We found that the two-temperature plasma model well represents 
772: the MOS spectra, as was the case for the Suzaku XIS spectra.
773: Figure \ref{fig:fit3} and table \ref{tbl:fit3} show the fitting result. 
774: Since there can be C K$_\alpha$ emission around 0.4 keV,
775: we allowed the C abundance to vary.
776: %%
777: The energy band of 1.16--1.28, 1.4--1.6, and 1.7--1.8 keV 
778: are omitted in order to exclude the instrumental emission lines
779: from Mg K$_\alpha$, Al K$_\alpha$, and Si K$_\alpha$.
780: %%
781: Because the low energy tail of the MOS response 
782: prevents the distinction between the diffuse 
783: low temperature plasma component and the LHB, 
784: only an upper limit is obtained for the LHB. 
785: %%
786: Alternatively the best-fit column density becomes somewhat 
787: lower than that in table \ref{tbl:fit1} (model 1),
788: to compensate the decreased low energy counts by the LHB component.
789: %%
790: The other two-temperature plasma parameters such as 
791: temperatures and abundances are surprisingly similar
792: to those in table \ref{tbl:fit1} (model 1).
793: %%
794: The only difference is the higher power-law component  flux.
795: This may be caused by the relatively higher background of 
796: XMM-Newton observations and their higher uncertainty.
797: %%
798: We thus conclude that the XIS and MOS data are consistent 
799: with each other and the fitting result of the XIS data is reliable.
800: 
801: \section{Discussion}
802: \label{sec:diss}
803: 
804: We investigated extended X-ray emission in the eastern tip region of the 
805: Carina Nebula with Suzaku XIS. 
806: %
807: For the first time, we conducted detailed spectral analysis of the 
808: X-ray emission and found that there is indeed diffuse X-ray emission,
809: even considering LHB, CXB, GRXE, and point sources.
810: %
811: The diffuse X-ray emission is well represented by a two-temperature 
812: plasma model with $kT$ $\sim$0.25 and $\sim$0.55 keV. 
813: %%
814: Emission measures and abundances of O, Ne, Mg, Si, S, and Fe 
815: are well constrained owing to the good photon statistics and 
816: the excellent energy response of XIS.
817: %%
818: %%Below we compare the plasma parameters to those in the $\eta$ 
819: %%Car region. We further estimate the plasma properties and 
820: %%examine the abundance pattern, to assess the origin of the diffuse
821: %%plasma.
822: %%
823: Below we estimate plasma properties based on the 
824: spectral fitting and then compare the spectral parameters 
825: such as absorption column density, temperature and abundance
826: to those in the $\eta$ Car region, in order to assess the 
827: origin of the diffuse plasma.
828: 
829: \subsection{Physical Properties of the Plasmas}
830: \label{sec:diss:prop}
831: 
832: %%{\bf
833: High signal-to-noise XIS spectra enable us to accurately 
834: constrain the parameters of the diffuse plasma in the eastern 
835: tip region of the Carina Nebula.
836: %%
837: The best-fit column density of $N_{\rm H}\sim2$--$3\times10^{21}$ cm$^{-2}$
838: and also the temperatures of $kT\sim0.2$--0.3 keV and 0.4--0.6 keV 
839: in the blob, east and nw regions agree very well with those in the $\eta$ 
840: Car regions (see the medium-temperature components of the $N_{\rm H}$, $kT$ 
841: tied model in table 2 of \cite{Hamaguchi2007} or hereafter H2007 model; 
842: $N_{\rm H}\sim2\times10^{21}$ cm$^{-2}$, $kT\sim0.2$ keV and $0.6$ keV).
843: %%
844: Although the high-temperature component represented by a hot plasma model
845: with $kT\sim5$ keV in the H2007 model is not seen in the eastern
846: tip region, this is thought to be a composite of diffuse hard X-ray 
847: emission and contaminations from CXB, GRXE, and point source, and hence
848: can be ignored.  
849: %%
850: This similarity in basic plasma parameters provides strong evidence that 
851: the diffuse plasma in the vicinity of the Carina Nebula has the same origin.
852: %%}
853: 
854: %%{\bf
855: The continuous distribution of the diffuse X-ray emission over the 
856: Carina Nebula, as shown in figure \ref{fig:overview}, supports this hypothesis. 
857: %%
858: Thus, we can draw a scenario that the X-ray emitting diffuse plasma generated 
859: by stellar winds from OB stars and/or SNRs forms hot bubbles with the size of 
860: several tens of pc and ionizes ambient molecular clouds which can be seen in 
861: the mid-infrared emission.
862: %%}
863: 
864: To investigate the origin of the diffuse plasma,
865: we estimate physical properties of the plasma in the blob region. 
866: %%
867: We assume that the plasma in the blob region have a prolate ellipsoidal 
868: shape with the major and minor axis lengths of 8 and 3 pc, respectively.
869: %%
870: Following the plasma analysis by \citet{Townsley2003}, we estimate the 
871: electron density, the pressure, the total energy content, the cooling 
872: time and the mass of the plasma from the observed X-ray luminosity, 
873: temperatures, and the assumed volume. 
874: %%
875: We derived two sets of plasma parameters for the two-temperature 
876: components from model 1 in table \ref{tbl:fit1}.
877: %%using the cooling function in \citet{Raymond1976}.
878: The derived parameters are summarized in table \ref{tbl:plasma}.
879: %%
880: Below we examine the two interpretations, i.e., stellar winds
881: from OB stars and SNRs, based on these derived parameters.
882: %%
883: Because there are no massive OB stars in the eastern tip region, 
884: we must consider the massive stellar clusters in the central part 
885: of the Carina Nebula for the former scenario.
886: 
887: The estimated plasma pressure $P$ is on the order of $\sim$10$^{6}$ 
888: K cm$^{-3}$  and should be larger than that of the surrounding gas 
889: in the eastern tip region since both the CO and radio continuum 
890: intensities are weak in the vicinity of the blob region 
891: \citep{Yonekura2005, Huchtmeier1975}.
892: %%{\bf 
893: This means that diffuse plasma in the eastern tip region
894: could flow from its neighbor.
895: %%}
896: If we consider that the plasma originated in stellar winds from OB stars
897: near $\eta$ Car and has been 
898: propagated to this region at the plasma sound velocity, the crossing 
899: time will be 0.1 Myr. 
900: %%
901: Because this time scale is much shorter than that of the radiative 
902: cooling timescale, $t_{\rm cool}$ $>$1 Myr, the plasma temperatures
903: can be held constant.
904: %%
905: %%If we consider the stellar-wind interpretation, 
906: OB stars are able to continuously produce the plasma during 0.1 Myr 
907: since their typical life time is at least ten times longer.
908: %%
909: The SNR interpretation is also possible in terms of the pressure
910: if one or multiple SNRs occurred in these regions.
911: 
912: The total thermal energy of the plasma, $U$ = $1\times10^{48}$ ergs,
913: is marginally lower than the total kinetic energy supplied by the 
914: stellar wind from a single massive star within $\sim1$ Myr, 
915: $\sim3\times10^{48}$ ergs \citep{Ezoe2006a}.
916: %%
917: As the Carina Nebula contains $>$64 OB stars, the observed thermal
918: energy can be easily supplied by the mechanical energy of the
919: stellar winds from the $>$64 OB stars, $>2\times10^{50}$ ergs.
920: %%
921: If all the diffuse X-ray emission observed with {Einstein} has 
922: the same origin, the total plasma energy will be about 10 times
923: larger, i.e., $1\times10^{49}$ ergs. 
924: %%
925: %% 0.5- 2 keV blob luminosity 1.5e34 erg/s -> x 10 2e35 erg/s
926: %% ??? Hamaguchi et al. ?? 2e50 ergs ???????H@???
927: %% ??ATonwsley et al. 2003 ? T2 ?????????vB
928: %%
929: Assuming that the stellar winds from these $>64$ OB stars 
930: ($>2\times10^{50}$ ergs) are responsible for the hot plasma, 
931: the kinetic-to-thermal energy conversion efficiency will be $<$5\%.
932: %%
933: According to \citet{Weaver1977}, the thermal energy in the 
934: shocked stellar wind is 5/11 of the total stellar wind 
935: kinetic energy. Thus, this conversion efficiency may be
936: doubled, $<$10 \%, which is comparable to that in M17 ($\sim$10\%, 
937: \cite{Townsley2003}) 
938: and larger than that in the Orion nebula ($\sim0.01$\%, 
939: \cite{Guedel2008}). 
940: %%
941: In the SNR case, we can also explain the energy by 
942: only one canonical supernova ($\sim1\times10^{51}$ ergs) even if 
943: we must explain all the diffuse X-ray emission in the entire 
944: Carina Nebula.
945: %% 
946: %%Thus, SNR(s) can explain the emission more easily.
947: %%
948: 
949: The mass of the plasma, $M_{\rm plasma}$ $\sim0.4M_{\odot}$, 
950: needs at least four typical OB stars assuming a typical mass 
951: loss rate of stellar winds of 10$^{-7}M_{\odot}$yr$^{-1}$ in 1 Myr. 
952: %%
953: If we consider the whole Carina Nebula,  
954: about $>$80 OB stars are necessary for the entire diffuse 
955: plasma mass. The known number of OB stars is thus marginal.
956: %%
957: On the other hand, one SNR again can supply this mass 
958: (e.g., \cite{Willingale2003}).
959: 
960: Therefore, although both the stellar-wind and SNR interpretations 
961: are possible, SNR(s) can explain the derived plasma parameters
962: such as the total plasma energy and the plasma mass, more easily.
963: %% 
964: As suggested by \citet{Hamaguchi2007}, the existence of the Carina 
965: flare supershell \citep{Fukui1999} validates the SNR scenario.
966: %%
967: If true, the observed plasma temperatures of 0.3 and 0.6 keV 
968: limit the SNR age to less than $\sim$10$^4$ yr, since an older SNR 
969: would be in the radiative phase and efficiently cool down to 
970: less than 0.1 keV.
971: 
972: %\subsection{Comparison with the $\eta$ Car Region}
973: %\label{sec:diss:comp}
974: 
975: \subsection{Abundance}
976: \label{sec:diss:abd}
977: 
978: The abundance pattern of the X-ray emitting plasma provides 
979: another key piece of information to constrain its origin.
980: %%
981: We showed abundance patterns of the blob region and that
982: of the $\eta$ Car regions in figures \ref{fig:abd}~a and b, 
983: respectively.
984: %%
985: Since \citet{Hamaguchi2007} divided the $\eta$ Car region 
986: into the north and south regions and fitted the two spectra 
987: simultaneously with tied column density and temperatures
988: in the H2007 model, two sets of abundances are shown.
989: 
990: All the metal abundances in the blob region (model 1, black) 
991: are significantly higher than those in the $\eta$ Car regions.
992: %%
993: This may strengthen the result of \citet{Hamaguchi2007}
994: that the metal abundances of the diffuse X-ray emission 
995: show spatial variations.
996: %%
997: However, there is a possibility that the fixed abundances 
998: of C, N, Al, Ar, Ca and Ni in table \ref{tbl:fit1} 
999: influence the other metal abundances.
1000: %%
1001: Thus, we refitted the blob region spectra with different 
1002: fixed abundance sets.
1003: 
1004: %%{\bf 
1005: Firstly we tested one solar for the fixed abundances.
1006: %%
1007: In model 1, we implicitly assumed 0.3 solar for them 
1008: because the value is generally used for young stars in star-forming regions.
1009: %%
1010: However, since any mixing of the plasma generated by either
1011: stellar winds or SNR(s) interacting with the ambient molecular clouds 
1012: contains processed stellar material and may have higher abundances,
1013: it is thus worth considering the solar abundances in modeling the 
1014: diffuse emission.
1015: %%
1016: The results are summarized in figure \ref{fig:abd}~a and 
1017: table \ref{tbl:fit1} (model 2).
1018: %%
1019: The $\chi^2$ was comparable to that of model 1 and all 
1020: the abundances increased by a factor of $\sim$2, while 
1021: the emission measures of the 
1022: two plasma components decreased by about the same amount,
1023: to balance the increased line intensities.
1024: %%
1025: Since the SNRs with an age around 10$^4$ yrs, like the Cygnus 
1026: Loop or Vela, globally show no strong deviation from 
1027: solar abundances \citep{McEntaffer2008},
1028: this fitting result allows the SNR interpretation.
1029: %%}
1030: 
1031: Next we assumed the abundances of the H2007 model 
1032: in which all the abundances were allowed to vary and 
1033: constrained with good photon statistics. We used two sets of fixed 
1034: abundances corresponding to the north and the south fits. 
1035: %%
1036: The results are summarized in figure \ref{fig:abd}~a and 
1037: table \ref{tbl:fit1} (models 3 and 4).
1038: %%
1039: In both cases, the fittings were acceptable and all the 
1040: free abundances were significantly decreased.
1041: %%
1042: These changes were caused by the decreased N and Ni fixed abundances 
1043: that influence the others via the NV K and Ni L emission lines.
1044: %%
1045: For instance, when we used the abundances for the north region (model 3), 
1046: the N abundance changes from 0.3 to 0 solar and to compensate for the decreased 
1047: photon counts in the 0.5--0.6 keV band, the normalization of the lower-temperature 
1048: component ($kT\sim0.25$ keV) increases and the abundances of the other 
1049: elements are suppressed.
1050: %%
1051: When the abundances for the south region (model 4) are utilized,
1052: the Ni L lines in the 0.8--1.4 keV range increases and the 
1053: abundances of the other elements related to this energy band 
1054: decreases.
1055: %%
1056: The abundances of the blob region approaches to those 
1057: of the $\eta$ Car region (figure \ref{fig:abd}~b) 
1058: if we use these abundance sets.
1059: 
1060: %%{\bf 
1061: The abundance values are therefore strongly influenced 
1062: by the fixed abundances.
1063: %%
1064: In model 2, all the abundances are around one solar, while
1065: in model 1, 3 and 4, the best-fit metal abundances are far less 
1066: than one.
1067: %%
1068: This is due to the fact that the metal abundance and
1069: emission measure are coupled with each other.
1070: %%
1071: To decouple these parameters, we need more precise spectral 
1072: measurements with X-ray microcalorimeters in future missions. 
1073: %%}
1074: 
1075: %%{\bf
1076: In spite of the difficulty to determine the absolute abundances, 
1077: abundance patterns of the blob region in models 1-4 are 
1078: similar to those in the $\eta$ Car regions.
1079: %%}
1080: %%
1081: This similarity is another line of evidence 
1082: that the diffuse X-ray emission in the eastern tip region and the 
1083: $\eta$ Car region has the same origin.
1084: 
1085: %%{\bf 
1086: Because the abundance patterns are rather independent of the fixed abundances, 
1087: we can also evaluate the abundance ratio of different elements.
1088: %%}
1089: Importantly, there is no significant overabundance 
1090: in nitrogen to oxygen that is expected from the optical, UV and X-ray
1091: spectroscopy of $\eta$ Car and WR 25 \citep{Davidson1982, vanderHucht1981, Tsuboi1997}.
1092: %%
1093: %%{\bf 2
1094: This result agrees with that in the $\eta$ Car region \citep{Hamaguchi2007} and
1095: suggests that, if stellar winds produced the diffuse X-ray plasma,
1096: the main drivers of the winds are main-sequence OB stars, and not
1097: evolved massive stars.
1098: %%}
1099: %%
1100: %%This result agrees with that in the $\eta$ Car region \citep{Hamaguchi2007} 
1101: %%and may be inconsistent with the stellar-wind interpretation 
1102: %%
1103: Also the observed Fe/O ratio of 1.3--1.9 is too high 
1104: compared to that of a type-II supernova, 0.5 \citep{Tsujimoto1995}. 
1105: %%
1106: Although the Fe/O ratio increases up to 1 in type-II SNRs 
1107: for less massive stars ($\sim13M_{\odot}$, table 1 in 
1108: \cite{Tsujimoto1995}), this contradicts the fact that
1109: the more massive stars evolve faster and explode earlier.
1110: 
1111: We note that similar situations exist in other massive star 
1112: forming regions. For example, detailed spectral study of 
1113: the diffuse X-ray emission has been conducted with Suzaku 
1114: in M17 \citep{Hyodo2008}. 
1115: %%
1116: We plot the abundance pattern of the diffuse
1117: plasma in M17 in figure \ref{fig:abd}~c.
1118: %%
1119: As is the case for the Carina Nebula, it shows a 
1120: high Fe/O ratio and not enhanced N-to-O ratio, 
1121: %%{\bf 2
1122: the latter of which may be natural because M17 contains 
1123: no WR stars \citep{Townsley2003}.
1124: %%}
1125: %%
1126: %%{\bf
1127: Its subsolar abundances may be affected by the
1128: fixed values at 0.3 solar.
1129: %%}
1130: 
1131: %%{\bf
1132: Thus, both the stellar-wind and SNR interpretations are 
1133: possible in terms of the absolute abundances.
1134: %%
1135: %%{\bf 2
1136: The abundance pattern may favor plasma heating by winds from main-sequence OB stars.
1137: %%} 
1138: %%
1139: %%These abundance patterns may suggest that we need additional 
1140: %%mechanisms to explain the diffuse X-ray emission in massive 
1141: %%star forming regions in general. 
1142: %%
1143: %%{\bf We can consider other possibilities, for example, 
1144: %%a mixture of stellar winds and SNRs.}
1145: Another possibility is a mixture of stellar winds and SNRs.
1146: %%
1147: Also, since hot shocked gas by stellar winds and/or SNRs mix up 
1148: interstellar gas, we may only see the interstellar abundances 
1149: rather than stellar and/or SNR abundances.
1150: %%
1151: Further studies are necessary from both observational 
1152: and theoretical aspects.
1153: 
1154: \subsection{Hard X-ray component}
1155: \label{sec:diss:hardX}
1156: 
1157: The X-ray spectrum above 2~keV can be 
1158: explained by contribution of CXB, GRXE and known point sources.
1159: %%
1160: This means that the residual hard X-ray emission,
1161: seen around $\eta$~Carinae 
1162: \citep{Hamaguchi2007}, localizes within $\sim$30$'$ from $\eta$~Car.
1163: %%
1164: The scale size is consistent with an apparently extended hard X-ray emission
1165: discovered with the GINGA satellite \citep{Koyama1990}
1166: though the GINGA result did not count out emission from known point sources.
1167: %%
1168: Because soft ($<$2~keV) diffuse X-ray spectra of the eastern tip region and the 
1169: southern part of the $\eta$~Car region are almost identical,
1170: the diffuse plasma would not relate to the residual hard X-ray emission.
1171: %%
1172: Probably, the residual emission originates from a large number of low 
1173: mass young stars embedded in the cloud around $\eta$~Car,
1174: which are too faint to be detected individually.
1175: 
1176: \section{Conclusion}
1177: 
1178: %%{\bf 
1179: In the present paper, we have investigated the properties of diffuse 
1180: X-ray emission associated with the eastern tip region of the Carina 
1181: Nebula using the Suzaku and XMM-Newton data.
1182: Our conclusion is as follows.
1183: 
1184: (1) Strong extended X-ray emission was detected from the entire
1185: field-of-view of Suzaku with a 0.2--5 keV surface brightness of 
1186: 0.7$\sim$4$\times10^{-14}$ erg s$^{-1}$ arcmin$^{-2}$.
1187: Comparisons with the estimated contamination from astrophysical 
1188: backgrounds and point sources suggest that most of the emission 
1189: is diffuse in nature.
1190: 
1191: (2) The observed absorption column density and temperature are
1192: consistent with those in the $\eta$Car region, suggesting the same origin as
1193: the diffuse X-ray emission in the vicinity of the Carina Nebula.
1194: 
1195: (3) Estimated physical properties of the plasma such as pressure,
1196: total energy, and mass can be explained by stellar winds from 
1197: OB stars in the Carina Nebula or young SNR(s) with the age less 
1198: than $\sim$10$^4$ yr. The SNR interpretation can provide the 
1199: necessary energy and mass more easily.
1200: 
1201: (4) Absolute abundance values are strongly affected by abundances of 
1202: metals fixed in spectral fits, allowing both the stellar-wind and 
1203: SNR interpretations. The low nitrogen-to-oxygen and high 
1204: iron-to-oxygen ratios derived from the spectral fits 
1205: %%
1206: %%{\bf 2
1207: may support that the diffuse plasma heated up by stellar winds 
1208: from main-sequence OB stars. The abundance ratios can be produced 
1209: by a mixture of stellar winds and SNRs, as well.
1210: %%}
1211: 
1212: The authors acknowledge discussion with Y. Hydo.
1213: K.H. is supported by the NASA Astrobiology Program under RTOP 344-53-51.
1214: 
1215: \clearpage
1216: 
1217: %% Fig. 1 Overview
1218: 
1219: \begin{figure}[p]
1220:   \begin{center}
1221:     \FigureFile(0.7\textwidth,){fig1av1p.eps}
1222:     \FigureFile(0.7\textwidth,){fig1bv1.eps}
1223:   \end{center}
1224:   \caption{(a) %An MSX band-A (6.8-10.8 $\mu$m) image of the Carina nebula
1225:                    An MSX band-A 8.28 $\mu$m image of the Carina nebula
1226:                     taken from the NASA/IPAC Infrared Science Archieve.
1227:                     Yellow contours show the 12CO (J$=$2-1) map 
1228:                     \citep{Yonekura2005}.
1229:                     Two white boxes represent field-of-views of this 
1230: 	           (upper left) and previous Suzaku observations (lower right, 
1231: 	            \cite{Hamaguchi2007}).
1232: 	           (b) An EPIC MOS mosaic of the same nebula 
1233: 	            with XMM-Newton taken from the XMM-Newton Image Gallery.
1234: 	           The red, green and blue colors show soft (0.4--0.7 keV), 
1235: 	           medium (0.7--1.3 keV) and hard (2--7 keV) X-ray emission, 
1236: 	           respectively.
1237: 	   	  Circles and numbers indicate point sources detected in our 
1238: 	           analysis (see \S \ref{sec:ps}). 
1239:   }\label{fig:overview}
1240: \end{figure}
1241: 
1242: \clearpage
1243: 
1244: %% Fig. 2 Suzaku Images
1245: 
1246: \begin{figure}[p]
1247:   \begin{center}
1248:     \FigureFile(0.8\textwidth,){fig2v2.eps}
1249:   \end{center}
1250:   \caption{Suzaku BI images of the eastern tip region of the Carina 
1251:            nebula in the (a) 0.2--2 keV and (b) 2--10 keV bands, 
1252: 	   displayed on the J2000.0 coordinates.
1253: 	   For clarity, images are binned by a factor of 8 and smoothed 
1254: 	   by a Gaussian of $\sigma$ = 3 pixels. %% ds9 default kernel size = 3 pix
1255: 	   Vignettings are corrected (see \S \ref{sec:obs}).
1256: 	   The unit of intensity in the greyscale wedge is arbitrary.
1257: 	   Solid black lines mark regions utilized in the 
1258: 	   spectral analysis. 
1259: 	   The strong emission at the bottom-left and -right parts of
1260:            the panel (b) corresponds to calibration 55Fe sources.
1261:   }\label{fig:xisimage}
1262: \end{figure}
1263: 
1264: \clearpage
1265: 
1266: %% Fig. 3 Suzaku Spectra of Three regions
1267: 
1268: \begin{figure}[p]
1269:   \begin{center}
1270:     \FigureFile(0.55\textwidth,){fig3v2.eps}
1271:   \end{center}
1272:   \caption{Background-subtracted BI (black) 
1273:     and FI (red) spectra of (a) the blob, 
1274:     (b) the east, and (c) the nw regions. Center energies of 
1275:     emission lines are shown in the panel (a). 
1276:     For comparison, the vertical axis is normalized by the 
1277:     angular size of each region.
1278:   }\label{fig:xisdiffspec}
1279: \end{figure}
1280: 
1281: \clearpage
1282: 
1283: %% Fig. 4 XMM point source spectra
1284: 
1285: \begin{figure}[p]
1286:   \begin{center}
1287:     \FigureFile(0.8\textwidth,){fig4v0.eps}
1288:   \end{center}
1289:   \caption{XMM-Newton EPIC-pn spectra of the bright point sources
1290:     (No.1, 2, 4, and 7).
1291:     The solid lines show the best-fit absorbed plasma models. 
1292:     The dotted lines in the panel (a) show two plasma components. 
1293:     The bottom panels exhibit residuals from the best-fit models.
1294:   }\label{fig:xmmsrcspec}
1295: \end{figure}
1296: 
1297: \clearpage
1298: 
1299: %% Fig. 5 Suzaku X1 spectra & Background Estimation
1300: 
1301: \begin{figure}[p]
1302:   \begin{center}
1303:     \FigureFile(0.55\textwidth,){fig5v2.eps}
1304:   \end{center}
1305:   \caption{Background-subtracted BI spectra of 
1306:     (a) the blob, (b) the east, and (c) the nw regions.
1307:     Solid lines show estimated contamination from 
1308:     X-ray sources.
1309:   }\label{fig:bgdest}
1310: \end{figure}
1311: 
1312: %% Fig. 6 Suzaku X0123 spectral fit of the diffuse X-ray emission 
1313: 
1314: \begin{figure}[p]
1315:   \begin{center}
1316:     \FigureFile(0.55\textwidth,){fig6v2.eps}
1317:   \end{center}
1318:   \caption{Best-fit spectral results of the diffuse X-ray emission
1319:       in (a) the blob, (b) the east, and (c) the nw regions. 
1320:       The best-fit model is shown in a solid black line. The model 
1321:       components for the XIS1 spectrum are shown in solid colored lines 
1322:       (cyan: the LHB component, magenta and red: the two-temperature 
1323:       plasma component, blue: the power-law component).
1324:       See tables \ref{tbl:fit1} and \ref{tbl:fit2} 
1325:       for the obtained parameters.
1326:   }\label{fig:fit12}
1327: \end{figure}
1328: 
1329: \clearpage
1330: 
1331: %% Fig. 7 MOS12 spectral fit of the diffuse X-ray emission 
1332: 
1333: \begin{figure}[p]
1334:   \begin{center}
1335:     \FigureFile(0.55\textwidth,){fig7v0.eps}
1336:   \end{center}
1337:   \caption{XMM-Newton MOS fitting result of the diffuse X-ray emission
1338:       in the blob region. 
1339:       Line colors are the same as figure \ref{fig:fit12} but for 
1340:       the MOS1 and MOS2 data.
1341:       See table \ref{tbl:fit3} for the obtained parameters.
1342:   }\label{fig:fit3}
1343: \end{figure}
1344: 
1345: %\clearpage
1346: \bigskip
1347: 
1348: %% Fig. 8 Abundance distribution
1349: 
1350: \begin{figure}[p]
1351: %  \begin{center}
1352: \hspace*{-10mm}
1353:     \FigureFile(1.1\textwidth,){fig8v3.eps}
1354: %  \end{center}
1355:   \bigskip
1356:   \caption{Abundance distributions in (a) the blob region in the
1357:        eastern tip region (this paper), 
1358:        (b) the north and south regions in the 
1359:        $\eta$ Car region \citep{Hamaguchi2007}, 
1360:        and (c) sub-regions in M17 \citep{Hyodo2008}.
1361:        The filled marks represent the fixed values.
1362:        }\label{fig:abd}
1363: \end{figure}
1364: 
1365: \clearpage
1366: 
1367: %% Tbl. 1 X-ray Sources
1368: 
1369: \begin{table}[p]
1370:   \caption{Properties of XMM-Newton point sources.}\label{tbl:srclist}
1371:   \begin{center}
1372:     \begin{tabular}{lcccccccc}
1373: 
1374:       \hline
1375:       No. & R.A.$^{\rm a}$  & Decl.$^{\rm a}$ 
1376:       & $C_{\rm MOS}$$^{\rm b}$ & $C_{\rm pn}$$^{\rm b}$
1377:       & Flux$^{\rm c}$  & Var$^{\rm d}$ & Counterpart$^{\rm e}$ \\
1378:       \hline
1379:       
1380:       1   & 10:47:31.68  & -59:32:33.7  
1381:       & 199$\pm13$       & 411$\pm30$  
1382:       &   3.8     &  No & 2MASS J 10473148-5932345       \\
1383: 
1384:       2	  & 10:49:28.08  & -59:34:55.2  
1385:       &  68$\pm8$        & 124$\pm$17  
1386:       &   1.1     &  No & 2MASS J 10492780-5935000       \\
1387: 
1388:       3$^{\rm f}$	  & 10:48:52.80  & -59:40:39.4  
1389:       &  62$\pm9$        &  22$\pm13$   
1390:       &   0.07    &  Yes & 2MASS J 10485248-5940392       \\ 
1391: 
1392:       4	  & 10:47:22.80  & -59:23:37.3  
1393:       &   66$\pm8$       &  174$\pm$19
1394:       &    0.38   &  No & 2MASS J 10472233-5923392       \\
1395: 
1396:       5	  & 10:46:45.84  & -59:30:13.0  % old 6
1397:       &   30$\pm7$       &  83$\pm15$
1398:       &    0.27   &  No & \begin{tabular}{c}
1399: 	                2MASS J 10464595-5930130       \\
1400: 			GSC 0862602325\\
1401:                         \end{tabular}\\
1402: 
1403:       6	  & 10:47:05.28  & -59:30:27.0  % old 7
1404:       &    15$\pm7$      &  61$\pm17$   
1405:       &   0.20    &  No & 2MASS J 10470505-5930247       \\
1406: 
1407:       7$^{\rm g}  $& 10:46:09.12& -59:43:08.0  % old 11
1408:       &    ---           &  336$\pm28$
1409:       &   0.79   &  No & 2MASS J 10460901-5943025       \\
1410: 
1411:       8  & 10:48:10.80  & -59:41:44.9  % old 13
1412:       &   47$\pm10$      &  73$\pm22$   
1413:       &   0.23   &  No & 2MASS J 10480990-5941489       \\
1414: 
1415:       9  & 10:48:30.96  & -59:42:14.8  % old 14
1416:       &   26$\pm8$       &   78$\pm19$   
1417:       &   0.25   &  No & 2MASS J 10483064-5942144       \\
1418: 
1419:       10  & 10:48:14.16  & -59:43:32.2 % old 15
1420:       &   15$\pm9$      &  81$\pm19$   
1421:       &   0.26   &  No & ---      \\
1422: 
1423: %      old 5	  & 10:49:06.72  & -59:33:55.4  
1424: %      &   26$\pm8$       &  8$\pm12$   
1425: %      &         &  No & 2MASS J 10490603-5933529       \\
1426: 
1427: %      old 8$^e$& 10:46:42.96  & -59:38:56.0  
1428: %      &    3$\pm9$        &  -26$\pm18$   
1429: %      &         &  No & 2MASS J 10464286-5938565       \\
1430: 
1431: %      old 9$^{e,f}$& 10:46:21.36 & -59:26:20.8  
1432: %      &    ---           &  -3$\pm10$  
1433: %      &         &  No & 2MASS J 10462175-5926220       \\
1434: 
1435: %      old 10$^e$& 10:46:35.52& -59:39:45.0  
1436: %      &    -1$\pm9$      &  -31$\pm22$    
1437: %      &         &  No & 2MASS J 10463483-5939417       \\
1438: 
1439: %      old 12$^e$& 10:49:40.56  & -59:34:22.4 
1440: %      &    -8$\pm6$      &  10$\pm10$   
1441: %      &         &  No & 2MASS J 10494122-5934237       \\
1442: 
1443: %      old 16  & 10:49:12.24  & -59:32:47.8  
1444: %      &   9$\pm6$       &   31$\pm15$   
1445: %      &         &  No & 2MASS J 10491204-5932467       \\
1446: 
1447:         \hline
1448:     \end{tabular}
1449:   \end{center}
1450: {\noindent
1451:   $^{\rm a}$ Source positions in J2000 coordinates.\\
1452:   $^{\rm b}$ Background-subtracted photon counts 
1453:   detected with EPIC MOS and pn in 0.4--10 keV. 
1454:   $C_{\rm MOS}$ is the average counts of MOS1 and MOS2. 
1455:   Errors are 1$\sigma$.\\
1456:   $^{\rm c}$ The 0.4--10 keV X-ray flux in 10$^{-13}$ erg s$^{-1}$ cm$^{-2}$ 
1457:   which corresponds to 6$\times10^{31}$ erg s$^{-1}$ at 2.3 kpc.
1458:   Fluxes of No.1, 2, 4, and 7 are derived from the spectral fittings,
1459:   while the other are estimated from $C_{\rm pn}$ assuming the thin-thermal
1460:   plasma model (see text).\\
1461:   $^{\rm d}$ Time variability based on the $\chi^2$ statistics.\\ 
1462:   $^{\rm e}$ Counterpart candidates within 10 arcsec of the X-ray position,
1463:   based on searches of the 2MASS and AGAST catalogs.\\
1464:   $^{\rm f}$ This source falls in the CCD gap of pn.\\
1465:   $^{\rm g}$ This source is outside the FOV of MOS.\\
1466: }
1467: \end{table}
1468: 
1469: %% Tbl. 2 Sources Spectral Fits
1470: 
1471: \begin{table}[p]
1472:   \caption{Results of the spectral fits to the point sources$^{\rm a}$.}
1473:   \label{tbl:srcfit}
1474: 
1475:   \begin{center}
1476:     \begin{tabular}{lccccc}
1477: 
1478:       \hline
1479:       No. & $N_{\rm H}$$^{\rm b}$ 
1480:       & $kT$$^{\rm c}$ 
1481:       & Normalization$^{\rm d}$ 
1482:       & $L_{\rm X}$$^{\rm e}$ 
1483:       & $\chi^2$/d.o.f. \\ \hline
1484:       
1485:       1 & 0.17$_{-0.12}^{+0.57}$    
1486:       & 0.42$_{-0.18}^{+0.45}$, $>32$
1487:       & 4.8$_{-2.9}^{+4.1}$$\times10^{-5}$, 2.5$_{-0.6}^{+0.5}$$\times10^{-4}$
1488:       & 3
1489:       & 19.5/25 \\
1490: 
1491:       2 & 9.6$_{-3.8}^{+6.3}$    
1492:       & 1.8$_{-0.8}^{+2.3}$
1493:       & 1.0$\pm{0.1}$$\times10^{-3}$ 
1494:       & 6
1495:       & 3.5/7 \\
1496: 
1497:       4 & $<0.95$
1498:       & 0.70$_{-0.44}^{+0.10}$
1499:       & 4.4$\pm0.7$$\times10^{-5}$ 
1500:       & 0.3
1501:       & 7.1/9 \\
1502:  
1503: 
1504:       7 & $0.72_{-0.16}^{+0.17}$ 
1505:       & $0.12_{-0.03}^{+0.04}$ 
1506:       & 8.4$_{-4.5}^{+4.6}$$\times10^{-2}$  
1507:       & 80
1508:       & 37.6/23 \\
1509: 
1510:         \hline
1511:     \end{tabular}
1512:   \end{center}
1513: {\noindent
1514: 
1515:   $^{\rm a}$ A single plasma model is assumed for No. 2, 4, and 7, while a two temperature 
1516:   model is used for No.1. A metal abundance is fixed at 0.3 solar value. 
1517:   Errors refer to the 90\% confidence range. \\
1518:   $^{\rm b}$ Hydrogen column density in 10$^{22}$ cm$^{-2}$.\\
1519:   $^{\rm c}$ Plasma temperature in keV.\\
1520:   $^{\rm d}$ Normalization factor of the APEC model, representing 
1521:   10$^{-14}$/4$\pi D^2$ $EM$, where $D$ is
1522:   a distance to the Carina nebula and $EM$ is an emission measure in 
1523:   cm$^{-3}$.\\
1524:   $^{\rm e}$ Absorption-corrected 0.4--10 keV luminosity 
1525:   in 10$^{32}$ erg s$^{-1}$ assuming a distance of 2.3 kpc.\\
1526: 
1527: }
1528: \end{table}
1529: 
1530: %% Tbl. 3  Blob - Two-temp plasma model
1531: 
1532: \begin{table}[p]
1533:   \caption{Results of the two-temperature plasma model fit to the diffuse X-ray emission in the blob region.}
1534:   \label{tbl:fit1}
1535: 
1536:   \begin{center}
1537:     \begin{tabular}{lccccc}
1538: 
1539: \hline\hline
1540: Model$^{\rm a}$                               &  1           &   2          & 3            & 4            & Typical error$^{\rm b}$ \\
1541: \hline									       	        								   
1542: 									       	        								   
1543: Two-temperature plasma component$^{\rm c}$\\				       	        			   
1544: 									       	        								   
1545: $N_{\rm H}$ (10$^{22}$ cm$^{-2}$)              & 0.23         & 0.25         & 0.22         & 0.26         & 0.01\\ % par1
1546: $kT_1$  (keV)                                & 0.25         & 0.24         & 0.25         & 0.24         & 0.01  \\ % par2
1547: $kT_2$  (keV)                                & 0.55         & 0.56         & 0.55         & 0.54         & 0.01  \\ % par18
1548: C       (solar)                              & 0.3 (fixed)  & 1.0 (fixed)  & 0.0 (fixed)  & 0.0 (fixed)  & $-$   \\
1549: N       (solar)                              & 0.3 (fixed)  & 1.0 (fixed)  & 0.0 (fixed)  & 0.0 (fixed)  & $-$   \\
1550: O       (solar)                              & 0.24         & 0.55         & 0.16         & 0.10         & 0.01  \\ % par6
1551: Ne      (solar)                              & 0.46         & 0.93         & 0.36         & 0.21         & 0.01  \\ % par7
1552: Mg      (solar)                              & 0.44         & 0.94         & 0.32         & 0.24         & 0.01  \\ % par8
1553: Al      (solar)                              & 0.3 (fixed)  & 1.0   (fixed)& 0.075 (fixed)& 0.029 (fixed)& $-$   \\
1554: Si      (solar)                              & 0.54         & 1.1          & 0.40         & 0.37         & 0.02  \\ % par10
1555: S       (solar)                              & 0.74         & 1.5          & 0.56         & 0.57         & 0.1   \\ % par11
1556: Ar      (solar)                              & 0.3 (fixed)  & 1.0 (fixed)  & 0.0 (fixed)  & 0.13 (fixed) & $-$   \\ 
1557: Ca      (solar)                              & 0.3 (fixed)  & 1.0 (fixed)  & 0.0 (fixed)  & 0.0 (fixed)  & $-$   \\ 
1558: Fe      (solar)                              & 0.32         & 0.71         & 0.23         & 0.19         & 0.01  \\ % par14
1559: Ni      (solar)                              & 0.3 (fixed)  & 1.0 (fixed)  & 0.089 (fixed)& 0.78 (fixed) & $-$   \\ 
1560: log$EM_1$ (cm$^{-3}$ arcmin$^{-2}$)           & 54.9         & 54.6         & 55.0         & 55.4         & 0.02 \\ % par17 
1561: log$EM_2$ (cm$^{-3}$ arcmin$^{-2}$)           & 54.5         & 54.2         & 54.6         & 54.7         & 0.02 \\ % par33
1562: Flux1 (10$^{-14}$ erg s$^{-1}$ arcmin$^{-2}$)  & 1.8          & 1.8          & 1.8          & 2.3          & 0.1 \\ % par17 
1563: Flux2 (10$^{-14}$ erg s$^{-1}$ arcmin$^{-2}$)  & 2.2          & 2.2          & 2.2          & 1.8          & 0.1 \\ % par33
1564: \hline									   									   
1565: 									   								   
1566: Power-law component$^{\rm d}$        \\  				   				   
1567: Flux (10$^{-14}$ erg s$^{-1}$ arcmin$^{-2}$) & 0.56         & 0.57         & 0.56         & 0.56         & 0.02  \\ % par35
1568: \hline									   									   
1569: 									   									   
1570: LHB component$^{\rm e}$              \\					   					   
1571: Flux (10$^{-14}$ erg s$^{-1}$ arcmin$^{-2}$) & 0.10         &  0.11        &  0.10        &  0.10        & 0.02  \\ % par39
1572: \hline									   									   
1573: 									   									   
1574: $\chi^2$/d.o.f.                              &  1.24        & 1.31         & 1.21         & 1.22         &       \\
1575:          d.o.f.                              &  1231        & 1231         & 1231         & 1231         &       \\
1576: \hline
1577: 
1578: %Gain component$^{\rm e}$              \\
1579: %Slope (XIS1)                                 &              &              &             & \\
1580: %Offset (XIS1)                                &              &              &             & \\
1581: %Slope  (XIS0$+$2$+$3)                        &              &              &             & \\
1582: %Offset (XIS0$+$2$+$3)                        &              &              &             & \\
1583: %\hline
1584: 
1585:     \end{tabular}
1586: 
1587:   \end{center}
1588: {\noindent
1589: 
1590:   $^{\rm a}$ Fitting models with different fixed abundances.  \\
1591: %  In model 1, the fixed abundances are typical values in 
1592: %  star-forming regions. In models 2 and 3, they are the derived values 
1593: %  in the north and south regions of the
1594: %  $\eta$ Car region \citep{Hamaguchi2007}.\\ 
1595:   $^{\rm b}$ Typical fitting errors at the 90\% confidence level.\\ 
1596:   $^{\rm c}$ A commonly-absorbed plasma model.
1597:   Arabic numbers 1 and 2 denote the two temperature components.
1598:   Parameter definitions are the same as those in table \ref{tbl:srcfit}.
1599:   Fluxes are calculated in 0.2--5 keV.\\
1600:   % The abundances of the elements not described in this table are fixed at 0.3 solar. 
1601:   $^{\rm d}$ A power-law model representing CXB, GRXE, and point sources. 
1602:   A photon index is fixed at 1.5. The same
1603:   absorption for the two-temperature plasma is assumed. Normalization
1604:   is photon keV$^{-1}$ cm$^{-2}$ at 1 keV.\\
1605:   $^{\rm e}$ A single-temperature plasma model representing LHB.
1606:   A plasma temperature is fixed at 0.1 keV.\\
1607: 
1608: }
1609: \end{table}
1610: 
1611: %% Tbl. 4  East & NW - Two-temp plasma model
1612: 
1613: \begin{table}[p]
1614:   \caption{Results of the two-temperature plasma model fit to the diffuse X-ray emission in the east and nw regions$^{\rm a}$.}
1615:   \label{tbl:fit2}
1616: 
1617:   \begin{center}
1618:     \begin{tabular}{lccccc}
1619: 
1620: \hline\hline
1621: Region                                         &  east                  &   nw                      \\
1622: \hline
1623: 
1624: Two-temperature plasma component\\
1625: 
1626: $N_{\rm H}$ (10$^{22}$ cm$^{-2}$)              & 0.21$_{-0.07}^{+0.09}$ & 0.32$_{-0.07}^{+0.11}$     \\ % par1 
1627: $kT_1$  (keV)                                  & 0.20$_{-0.02}^{+0.04}$ & 0.19$_{-0.03}^{+0.02}$     \\ % par2 
1628: $kT_2$  (keV)                                  & 0.54$_{-0.07}^{+0.04}$ & 0.41$_{-0.08}^{+0.10}$     \\ % par18
1629: C       (solar)                                & 0.3 (fixed)            & 0.3 (fixed)                \\ 	    
1630: N       (solar)                                & 0.3 (fixed)            & 0.3 (fixed)                \\ 	    
1631: O       (solar)                                & 0.15$_{-0.06}^{+0.12}$ & 0.07$_{-0.02}^{+0.04}$     \\ % par6 
1632: Ne      (solar)                                & 0.33$_{-0.14}^{+0.31}$ & 0.27$_{-0.05}^{+0.14}$     \\ % par7 
1633: Mg      (solar)                                & 0.30$_{-0.14}^{+0.29}$ & 0.25$_{-0.10}^{+0.16}$     \\ % par8 
1634: Al      (solar)                                & 0.3 (fixed)            & 0.3 (fixed)                \\   	    
1635: Si      (solar)                                & 0.43$_{-0.20}^{+0.37}$ & 0.96$_{-0.43}^{+0.57}$     \\ % par10
1636: S       (solar)                                & 0.74 (fixed)           & 0.74 (fixed)               \\ % par11
1637: Ar      (solar)                                & 0.3 (fixed)            & 0.3 (fixed)                \\ 	    
1638: Ca      (solar)                                & 0.3 (fixed)            & 0.3 (fixed)                \\ 	    
1639: Fe      (solar)                                & 0.16$_{-0.05}^{+0.10}$ & 0.17$_{-0.06}^{+0.03}$     \\ % par14
1640: Ni      (solar)                                & 0.3 (fixed)            & 0.3 (fixed)                \\   	    
1641: log$EM_1$ (cm$^{-3}$ arcmin$^{-2}$)            & 54.3$_{-0.5}^{+0.4}$   & 55.0$\pm{0.7}$       \\ % par17
1642: log$EM_2$ (cm$^{-3}$ arcmin$^{-2}$)            & 54.0$\pm{0.2}$         & 54.3$_{-0.3}^{+0.8}$       \\ % par33
1643: Flux1 (10$^{-14}$ erg s$^{-1}$ arcmin$^{-2}$)  & 0.23$_{-0.15}^{+0.41}$ & 0.41$\pm{0.08}$     \\ % par17 
1644: Flux2 (10$^{-14}$ erg s$^{-1}$ arcmin$^{-2}$)  & 0.48$_{-0.20}^{+0.31}$ & 0.45$_{-0.20}^{+0.11}$     \\ % par33
1645: \hline
1646: 	
1647: Power-law component        \\    
1648: Flux  (10$^{-14}$ erg s$^{-1}$ arcmin$^{-2}$)  & 0.56$\pm{0.06}$          & 0.55$\pm{0.09}$          \\ % par17 
1649: \hline
1650: 
1651: LHB component              \\
1652: Flux  (10$^{-14}$ erg s$^{-1}$ arcmin$^{-2}$)  & 0.087$_{-0.032}^{+0.031}$& 0.15$_{-0.06}^{+0.05}$  \\ % par17 
1653: \hline
1654: 
1655: $\chi^2$/d.o.f.                                & 0.77                     & 0.60                    \\
1656:          d.o.f.                                & 245                      & 111                     \\
1657: \hline
1658:     \end{tabular}
1659: 
1660:   \end{center}
1661: {\noindent
1662: 
1663:   $^{\rm a}$ Notations and symbols are the same as table \ref{tbl:fit1}.
1664: 
1665: }
1666: \end{table}
1667: 
1668: \clearpage
1669: 
1670: %% Tbl. 5  XMM MOS Two-temp plasma model
1671: \begin{table}[p]
1672:   \caption{Result of the two-temperature plasma model fit to the XMM MOS spectra of the blob region$^{\rm a}$.}
1673:   \label{tbl:fit3}
1674: 
1675:   \begin{center}
1676:     \begin{tabular}{lccccc}
1677: 
1678: \hline\hline
1679: Model                                         &  1                   \\
1680: \hline
1681: 
1682: Two-temperature plasma component\\
1683: 
1684: $N_{\rm H}$ (10$^{22}$ cm$^{-2}$)             & 0.19$_{-0.02}^{+0.03}$  \\ % par1 
1685: $kT_1$  (keV)                                                    & 0.24$\pm{0.01}$              \\ % par2 
1686: $kT_2$  (keV)                                                    & 0.58$\pm{0.01}$              \\ % par18
1687: 
1688: C       (solar)                                                        & 1.2$_{-0.9}^{+0.4}$            \\
1689: N       (solar)                                                        & 0.3 (fixed)                          \\ 	    
1690: O       (solar)                                  & 0.23$_{-0.03}^{+0.02}$      \\ % par6 
1691: Ne      (solar)                                 & 0.44$\pm{0.07}$                  \\ % par7 
1692: Mg      (solar)                                & 0.46$\pm{0.07}$                   \\ % par8 
1693: Al      (solar)                                  & 0.3 (fixed)                               \\   	    
1694: Si      (solar)                                  & 0.48$_{-0.08}^{+0.07}$      \\ % par10
1695: S       (solar)                                  & 0.48$_{-0.18}^{+0.20}$      \\ % par11
1696: Ar      (solar)                                  & 0.3 (fixed)                              \\ 	    
1697: Ca      (solar)                                & 0.3 (fixed)                               \\ 	    
1698: Fe      (solar)                                 & 0.32$_{-0.04}^{+0.05}$        \\ % par14
1699: Ni      (solar)                                 & 0.3 (fixed)                                 \\   	    
1700: log$EM_1$ (cm$^{-3}$ arcmin$^{-2}$)               & 54.8$\pm{0.1}$                    \\ % par17
1701: log$EM_2$ (cm$^{-3}$ arcmin$^{-2}$)               & 54.5$\pm{0.1}$                    \\ % par33
1702: Flux1 (10$^{-14}$ erg s$^{-1}$ arcmin$^{-2}$)  & 1.7$\pm{0.2}$               \\ % par17 
1703: Flux2 (10$^{-14}$ erg s$^{-1}$ arcmin$^{-2}$)  & 2.7$_{-0.4}^{+0.2}$     \\ % par33
1704: \hline
1705: 	
1706: Power-law component        \\    
1707: Flux  (10$^{-14}$ erg s$^{-1}$ arcmin$^{-2}$)   & 0.84$\pm{0.05}$                    \\ % par17 
1708: \hline
1709: 
1710: LHB component              \\
1711: Flux  (10$^{-14}$ erg s$^{-1}$ arcmin$^{-2}$)  & $<0.06$  \\ % par17 
1712: \hline
1713: 
1714: $\chi^2$/d.o.f.                                & 1.20                       \\
1715:          d.o.f.                                       & 337                        \\
1716: \hline
1717:     \end{tabular}
1718: 
1719:   \end{center}
1720: {\noindent
1721: 
1722:   $^{\rm a}$ Notations and symbols are the same as table \ref{tbl:fit1}.
1723: 
1724: }
1725: \end{table}
1726: 
1727: \clearpage
1728: 
1729: %% Tbl. Physical Properties of Plasmas
1730: 
1731: \begin{table}[p]
1732:   \caption{Physical properties of the diffuse plasma in the blob region$^a$.}
1733:   \label{tbl:plasma}
1734: 
1735:   \begin{center}
1736:     \begin{tabular}{lccccc}
1737: 
1738: \hline\hline
1739: Parameter                                    &  Scale Factor                &       $T_1$                     &      $T_2$\\
1740: \hline
1741: 
1742: \multicolumn{4}{c}{Observed X-ray Properties}\\
1743: \hline
1744: 
1745: $kT$ (keV)                                    &   $-$                               &        0.3                          &      0.6 \\
1746: $L_{\rm X}$ (ergs s$^{-1}$)       &   $-$                               &   2$\times10^{34}$     &     1$\times10^{34}$\\
1747: $V$ (cm$^{3}$)                            &   $\eta$                          &   1$\times10^{57}$     &     1$\times10^{57}$\\
1748: % print pi * 4 /3  * 4 * 1.5 * 1.5 * (3.085e18)**3
1749: % Fe (solar)                                       &   $-$                               &    0.3                              &     0.3  \\
1750: \hline
1751: 
1752: \multicolumn{4}{c}{Estimated X-ray Plasma Properties}\\
1753: \hline
1754: 
1755: % cooling 15e-23 erg/sec/cc @ 0.2 keV = 2.3e6 K
1756: %                  5e-23 erg/sec/cc @ 0.6 keV = 7.0e6 K
1757: 
1758: $n_{\rm e}$ (cm$^{-3}$)             &   $\eta^{-1/2}$            &        0.3                          &      0.4 \\
1759: % print (1.5e34/15e-23/1.1e57)**0.5
1760: % print (0.8e34/5e-23/1.1e57)**0.5
1761: $P/k$ (K cm$^{-3}$)                    &   $\eta^{-1/2}$            &   2$\times10^{6}$       &     5$\times10^{6}$\\
1762: % print 2 * (1.5e34/15e-23/1.1e57)**0.5 * (0.25e3*1.60e-19/1.38e-23)
1763: % print 2 * (0.8e34/5e-23/1.1e57)**0.5 * (0.55e3*1.60e-19/1.38e-23)
1764: $U$ (ergs)                                     &   $\eta^{1/2}$             &   4$\times10^{47}$     &     1$\times10^{48}$\\
1765: % print 3 * (1.5e34/15e-23/1.1e57)**0.5 * (0.25e3*1.60e-12) * 1.1e57
1766: % print 3 * (0.8e34/5e-23/1.1e57)**0.5 * (0.55e3*1.60e-12) * 1.1e57
1767: $t_{\rm cool}$ (yr)                         &  $\eta^{1/2}$             &   1$\times10^{6}$       &     4$\times10^{6}$\\
1768: % print 3 * (1.5e34/15e-23/1.1e57)**0.5 * (0.25e3*1.60e-12) * 1.1e57 / (1.5e34 * 24 * 3600 * 365)
1769: % print 3 * (0.8e34/5e-23/1.1e57)**0.5 * (0.55e3*1.60e-12) * 1.1e57/(0.8e34 * 24 * 3600 * 365) 
1770: $M_{\rm plasma}$ ($M_\odot$) &  $\eta^{1/2}$             &   0.2                               &     0.2 \\
1771: % print 0.62 * 1.673e-24 * (1.5e34/15e-23/1.1e57)**0.5 * 1.1e57 / 1.99e33
1772: % print 0.62 * 1.673e-24 * (0.8e34/5e-23/1.1e57)**0.5 * 1.1e57 / 1.99e33
1773: 
1774: %$M_{\rm Fe}$ ($M_\odot$)         &  $\eta^{1/2}$             &   2$\times10^{-3}$       &     2$\times10^{-3}$ \\
1775: % print 52 * 4.68e-5 * 1.673e-24 * (1.5e34/15e-23/8.9e57)**0.5 * 8.9e57 / 1.99e33
1776: % print 52 * 4.68e-5 * 1.673e-24 * (0.8e34/5e-23/8.9e57)**0.5 * 8.9e57 / 1.99e33
1777: %http://heasarc.gsfc.nasa.gov/docs/xanadu/xspec/manual/XSabund.html
1778: \hline
1779: 
1780:     \end{tabular}
1781: 
1782:   \end{center}
1783: {\noindent
1784: 
1785:   $^{\rm a}$ $\eta$ is a filling factor for the volume of the 
1786:              plasma. $T_1$ and $T_2$ indicate the two temperature plasma
1787:              component in table \ref{tbl:fit1} model 1.
1788: 
1789: }
1790: \end{table}
1791: 
1792: \clearpage
1793: 
1794: \begin{thebibliography}{}
1795: 
1796: % ???????????AP?????A??CjV??i???CjV??????????????jA?????????J?}????????A???????O??&?????B??A???9l?????A1l????????et al.????????B????A??CjV??. ?i?????t????????j???????J?}?????????B
1797: 
1798: \bibitem[Colombo et al.(2003)]{Colombo2003}
1799:   Colombo, J.F.A., Mendez, M., \& Morrell, N. I.\ 2003, \mnras, 346, 704
1800: 
1801: \bibitem[Cunha et al.(1994)]{Cunha1994}
1802:   Cunha, K., \& Lambert, D. L.\ 1994, \apj, 426, 170
1803: 
1804: %%\bibitem[Daflon et al.(1999)]{Daflon1999}
1805: %%  Daflon, S., CunHa, A, \& Becker, S. R.\ 1999, \apj, 522, 950
1806: 
1807: \bibitem[Daflon et al.(2004)]{Daflon2004}
1808:   Daflon, S., CunHa, A, \& Butler, K.\ 2004, \apj, 604, 326
1809: 
1810: \bibitem[Davidson et al.(1982)]{Davidson1982}
1811:   Davidson, K., Walborn, N. R., \& Gull, T. R.\ 1982, \apj, 254, L47
1812: 
1813: \bibitem[Davidson \& Humphreys(1997)]{Davidson1997}
1814:   Davidson, K., \& Humphreys, R. M.\ 1997, 
1815:   \araa, 35, 1
1816: 
1817: \bibitem[Dicky \& Lockman(1990)]{Dickey1990}
1818:   Dicky, J.M. ., \& Lockman, F.J.\ 1990, \araa, 28, 215
1819: 
1820: \bibitem[Ebisawa et al.(2005)]{Ebisawa2005}
1821:   Ebisawa, K., et al.\ 2005, \apj, 635, 214
1822: 
1823: \bibitem[Evans et al.(2003)]{Evans2003}
1824:   Evans, N.R., Seward, F.D., Krauss, M.I., Isobe, T., Nichols, J., 
1825:   Schlegel, E.M., \& Wolk, S.J.\ 2003, \apj, 589, 509
1826: 
1827: \bibitem[Ezoe et al.(2006a)]{Ezoe2006a}
1828:   Ezoe, Y., Kokubun, M., Makishima, K., Sekimoto, Y., \& Matsuzaki, K.\
1829:   2006a, \apj, 638, 860
1830: 
1831: \bibitem[Ezoe et al.(2006b)]{Ezoe2006b}  
1832:   Ezoe, Y., Kokubun, M., Makishima, K., Sekimoto, Y., \& Matsuzaki, K.\ 
1833:   2006b, \apj, 649, L123
1834: 
1835: \bibitem[Feinstein(1995)]{Feinstein1995}
1836:   Feinstein, A.\ 1995, RMxAC, 2, 57
1837:   
1838: \bibitem[Fukui et al.(1999)]{Fukui1999}
1839:   Fukui, Y., Onishi, T., Abe, R., Kawamura, A., Tachihara, K., Yamaguchi, R., 
1840:   Mizuno, A., \& Ogawa, H.\ 1999, \pasj, 51, 751
1841: 
1842: \bibitem[Fujimoto et al.(2007)]{Fujimoto2007}
1843:   Fujimoto, R., et al.\ 2007, \pasj, 59, 133
1844: 
1845: \bibitem[Getman et al.(2005)]{Getman2005}
1846:   Getman, K. V., et al.\ 2005, \apjs, 160, 319
1847: 
1848: \bibitem[G{\" u}edel et al.(2008)]{Guedel2008}
1849:   G{\" u}del, M., Briggs, K. R., Montmerle, T., Audard, M., Rebull, L.,
1850:   \& Skinner, S. L.\ 2008, Science, 319, 309
1851: 
1852: \bibitem[Hamaguchi et al.(2007)]{Hamaguchi2007}
1853:   Hamaguchi, K., et al.\ 2007, \pasj, 59, S151
1854:   
1855: %\bibitem[Hamaguchi et al.(2007b)]{Hamaguchi2007b}
1856: %  Hamaguchi, K., the Suzaku Eta Carinae team,  \& the Carinae D1 team, 
1857: %  2007b, astro-ph, 0704.346
1858: 
1859: \bibitem[Harvey et al.(1979)]{Harvey1979}
1860:   Harvey, P. M., Hoffmann, W. F., \& Campbell, M. F.\ 1979, \apj, 227, 114
1861: 
1862: \bibitem[Huchtmeier \& Day(1975)]{Huchtmeier1975}
1863:   Huchtmeier, W. K., \& Day G. A.\ 1975, A\&A, 41, 153
1864: 
1865: \bibitem[Hyodo et al.(2008)]{Hyodo2008}
1866:   Hyodo, Y., Tsujimoto, M., Hamaguchi, K., Koyama, K., 
1867:   Kitamoto, S., Maeda, Y., Tsuboi, Y., \& Ezoe, Y.\ 2008, \pasj, 60, 85
1868:     
1869: \bibitem[Imanishi et al.(2001)]{Imanishi2001}
1870:   Imanishi, K., Koyama, K., \& Tsuboi, Y.\ 2001, \apj, 557, 747
1871: 
1872: \bibitem[Kelly et al.(2007)]{Kelly2007}
1873:   Kelly, R., et al.\ 2007, \pasj, 59, S77
1874: 
1875: \bibitem[Kokubun et al.(2007)]{Kokubun2007}
1876:   Kokubun, M., et al.\ 2007, \pasj, 59, S53
1877: 
1878: \bibitem[Koyama et al.(1990)]{Koyama1990}
1879:   Koyama, K., Asaoka, I., Ushimaru, N., Yamauchi, S., \& Corbert, R.H.,\ 
1880:   1990, \apj, 362, 215
1881: 
1882: \bibitem[Koyama et al.(2007)]{Koyama2007}
1883:   Koyama, K., et al.\ 2007, \pasj, 59, S23
1884: 
1885: \bibitem[McEntaffer \& Cash (2008)]{McEntaffer2008}
1886:   McEntaffer, R. L., \& Cash, W. \ 2008, \apj, 680, 328 
1887: 
1888: \bibitem[Miyaji et al.(1998)]{Miyaji1998}
1889:   Miyaji, T., Ishisaki, Y., Ogasaka, Y., Ueda, Y., Freyberg, M.J.,
1890:   Hasinger, G., \& Tanaka, Y.\ 1998, \aa, 334, L13 
1891: 
1892: %\bibitem[Miyata et al.(2007)]{Miyata2007}
1893: %  Miyata, E., Katsuda, S., Tsunemi, H., Hughes, J. P., 
1894: %  Kokubun, M., \& Porter, F. S.\ 2007, \pasj, 59, S163 
1895: 
1896: \bibitem[Mitsuda et al.(2007)]{Mitsuda2007}
1897:   Mitsuda, K., et al.\ 2007, \pasj, 59, S1
1898: 
1899: \bibitem[Moffat et al.(2002)]{Moffat2002}
1900:   Moffat, A. F. J., et al.\ 2002, \apj, 573, 191
1901: 
1902: %%\bibitem[Raymond et al.(1976)]{Raymond1976}
1903: %%  Raymond, J. C., Cox, D. P., \& Smith, B. W. 1976, \apj, 204, 290
1904: 
1905: \bibitem[Serlemitsos et al.(2007)]{Serlemitsos2007}
1906:   Serlemitsos, P. J., et al.\ 2007, \pasj, 59, S9
1907: 
1908: \bibitem[Seward et al.(1979)]{Seward1979}
1909:   Seward, F.D., Forman, W.R., Giacconi, R., Griffith, R.E., Harnden, F.R Jr.,
1910:   Jones, C., \& Pye, J.P.\ 1979, \apj, 234, L55
1911: 
1912: \bibitem[Seward \& Chebowski(1982)]{Seward1982}
1913:   Seward, F.D., \& Chebowski, T.\ 1982, \apj, 256, 530
1914: 
1915: \bibitem[Smith et al.(2000)]{Smith2000}
1916:   Smith, N., Egan, M. P.,  Carey, S., Price, S. D., Morse, P. J., \& Price, P. A.\ 
1917:   2000, \apjl, 532, L145
1918: 
1919: \bibitem[Smith et al.(2001)]{Smith2001}
1920:   Smith, R. K., Brickhouse, N. S., Liedahl, D. A., \& Raymond, J. C.\ 
1921:   2001, \apjl, 556, L91
1922: 
1923: \bibitem[Smith et al.(2006)]{Smith2006}
1924:   Smith, N.\ 
1925:   2006, \mnras, 367, 763
1926: 
1927: \bibitem[Smith et al.(2007)]{Smith2007}
1928:   Smith, N., \& Brooks, K. J.\ 
1929:   2007, \mnras, 379, 1279
1930: 
1931: \bibitem[Snowden et al.(1998)]{Snowden1998}
1932:   Snowden, S. L., Egger, R., Finkbeiner, D. P., Freyberg, M. J., 
1933:   \&  Plucinsky, P. P.\ 1998, \apj, 493, 715
1934: 
1935: \bibitem[Str{\"u}der et al.(2001)]{Struder2001}
1936:   Str{\"u}der, L., et al.\ (2001), A\&A, 365, 18
1937: 
1938: \bibitem[Takahashi et al.(2007)]{Takahashi2007}
1939:   Takahashi, T., et al.\ 2007, \pasj, 59, S35
1940: 
1941: \bibitem[Townsley et al.(2003)]{Townsley2003}
1942:   Townsley, L. K., Feigelson, E. D., Montmerle, T., Broos, Patrick S.,
1943:   Chu, You-Hua., \& Garmire, G. P.\ 2003, \apj, 593, 874
1944: 
1945: \bibitem[Tsuboi et al.(1997)]{Tsuboi1997}
1946:   Tsuboi, Y., Koyama, K., Sakano, M., \& Petre, R.\ 1997, \pasj, 49, 85
1947: 
1948: \bibitem[Tsujimoto et al.(1995)]{Tsujimoto1995}
1949:   Tsujimoto, T., Nomoto, K., Yoshii, Y., Hashimoto. M., Yanagida. S., 
1950:   \& Thielemann, F. -K.\ 1995, \mnras, 277, 945
1951: 
1952: %\bibitem[Tsujimoto et al.(2002)]{Tsujimoto2002}
1953: %  Tsujimoto, T., Koyama, K., Tsuboi, Y., Miwa, G., \& Kobayashi, N.\
1954: %  2002, \apj, 566, 974
1955:   
1956: \bibitem[Tsujimoto et al.(2006)]{Tsujimoto2006}
1957:   Tsujimoto, M., Hosokawa, T., Feigelson, E. D., Getman, K. V., 
1958:   \& Broos, P. S.\ 2006, \apj, 653, 409
1959: 
1960: \bibitem[Tsujimoto et al.(2007)]{Tsujimoto2007}
1961:   Tsujimoto, M., Hyodo, Y., \& Koyama, K.\ 2007, \pasj, 59, S229
1962: 
1963: \bibitem[Turner et al.(2001)]{Turner2001}
1964:   Turner, M.J.L., et al.\ \aa, 365, 27
1965: 
1966: \bibitem[Yonekura et al.(2005)]{Yonekura2005}
1967:   Yonekura, Y., et al.\ 2005, \apj, 634, 476
1968: 
1969: \bibitem[Yusef-Zadeh et al.(2002)]{Yusef-Zadeh2002}
1970:   Yusef-Zadeh, F., Law, C., Wardle, M., Wang, Q. D., Fruscione, A.,
1971:   Lang, C. C., \& Cotera, A.\ 2002, \apj, 570, 665
1972: 
1973: \bibitem[van der Hucht et al.(1981)]{vanderHucht1981}
1974:   van der Hucht, K. A., Conti, P. S., Lundstrom, I., \& Stenholm, B.\ 
1975:   1981, \ssr, 28, 227
1976:   
1977: \bibitem[Wang et al.(2002)]{Wang2002}
1978:   Wang, Q. D., Gotthelf, E. V., \& Lang, C. C.\ 2002, \nat, 415, 148
1979: 
1980: \bibitem[Weaver et al.(1977)]{Weaver1977}
1981:   Weaver, R., McCray, R., Castor, RJ., Shapiro, P., \& Moore, R.\ 1997, \apj, 218, 377
1982:  
1983: \bibitem[Willingale et al.(2003)]{Willingale2003}
1984:   Willingale, R.,  Bleeker, J. A. M., van der Heyden, K. J., 
1985:    \& Kaastra, J. S.\ 2003, \aa, 398, 1021
1986: 
1987: \bibitem[Wolk et al.(2002)]{Wolk2002}
1988:   Wolk, S. J., Bourke, T. L., Smith, R. K., Spitzbart, B., \& Alves, J.\
1989:   2002, \apj, 850, L161
1990: 
1991: %\bibitem[Aauthor et al.(2001)]{key-1}
1992: %   Aauthor, A., Bauthor, B., Cauthor, C.\ 2001, PASJ, vol, page
1993: %
1994: %\bibitem[Aauthor \& Author(2001a)]{key-2}
1995: %   Aauthor, A., Author, B.\ 2001, Name of Book(Publisher, Tokyo) ch0
1996: %
1997: %\bibitem[Aauthor \& Bauthor(2001b)]{key-3}
1998: %  Aauthor, A., Bauthor, B.\ 2001, Name of Book(Publisher, Tokyo) page0
1999: %
2000: %\bibitem[Dauthor(2001)]{key-n}
2001: %  Dauthor A. A.\ 2001, in Name of Book,
2002: %   ed Editor D.\ Editor(Publisher, Tokyo) page0
2003: 
2004: \end{thebibliography}
2005: 
2006: \end{document}
2007: