1: \documentclass[12pt,preprint]{aastex}
2: %\documentclass{emulateapj}
3:
4: \newcommand{\perval}[2]{{#1\mbox{$^{#2}$}}}
5: \newcommand{\erg}{\mbox{$\rm\,erg$}\/}
6: \newcommand{\cm}{\mbox{$\rm\,cm$}}
7: \newcommand{\persec}{\perval{s}{-1}\/}
8: \newcommand{\percm}{\mbox{$\cm^{-2}$}}
9: \newcommand{\peryear}{{yr$^{-1}$}}
10: \newcommand{\cgsflux}{\erg~\percm~\persec}
11: \newcommand{\cgslum}{\erg~\persec}
12: \def\gsev {G70.7$+$1.2}
13: \def\ptsrc{CXO~J200423.4$+$333907}
14: \def\lgs {LGS-AO}
15: \def\smicron {\mbox{ }\mu\mbox{{\ss m}}}
16: \def\micron {\mbox{ }\mu\mbox{{m}}}
17: \def\kms {\mbox{ km s}^ {-1}}
18: \def\ms {\mbox{ m s}^ {-1}}
19: \def\us {\mbox{ }\mu\mbox{s}} %microseconds
20: \def\jy {\mbox{ Jy}}
21: \def\simlt{\mathrel{\hbox{\rlap{\hbox{\lower4pt\hbox{$\sim$}}}\hbox{$<$}}}}
22: \def\simgt{\mathrel{\hbox{\rlap{\hbox{\lower4pt\hbox{$\sim$}}}\hbox{$>$}}}}
23: \def\ra#1#2#3{#1$^{\rm h}$#2$^{\rm m}$#3$^{\rm s}$}
24: \def\dec#1#2#3{$#1^\circ#2'#3$}
25: \def\sci#1#2{$#1 \times 10^{#2}$}
26: \newcommand{\chandra}{{\em Chandra\/}}
27:
28: \begin{document}
29: \title{Near-Infrared and X-ray Observations of the Enigmatic \gsev}
30: \author{P. B. Cameron and S. R. Kulkarni}
31: \affil{California Institute of Technology, Division Physics,
32: Math and Astronomy, MC 105-24, Pasadena, CA 91125 \\
33: Electronic mail: pbc@astro.caltech.edu}
34:
35: \begin{abstract}
36: We present high resolution imaging of the puzzling radio and optical
37: nebula \gsev\ with the Keck Observatory's laser guide star adaptive
38: optics (\lgs) system and the {\em Chandra X-ray Observatory}. The
39: archival X-ray observations show a hard ($\Gamma \approx 1.8$), low
40: luminosity ($L_X \approx $ \sci{4}{31} \cgslum) point source at the
41: center of the nebula. Follow-up \lgs\ near-infrared imaging of the
42: \chandra\ error circle reveals a relatively bright ($K^{\prime}$
43: $\approx$ 14 magnitude) counterpart. Both its color and brightness are
44: consistent with a heavily obscured B-star or possibly a late-G/early-K
45: giant. The most plausible explanation is that this newly discovered
46: X-ray source is a non-accreting B-star/pulsar binary powering the
47: radio and optical nebula. If so, the luminous Be-star discussed in the
48: literature seemingly embedded in the nebula is not the dominant force
49: responsible for shaping \gsev. Thus, we suggest that \gsev\ is the
50: result of two unrelated objects (a B-star X-ray binary and a Be star)
51: interacting with a dense molecular cloud. With this explanation we
52: believe we have solved the mystery of the origin of \gsev.
53: \end{abstract}
54: \keywords{X-rays: binaries --- ISM: individual (\gsev)}
55:
56: \section{Introduction}
57: \gsev\ is a compact ($\approx 20$\arcsec) radio and optical nebula in
58: the Galactic Plane whose origin is controversial
59: \citep{reich+85,green86,demuizon+88,bally+89}. The shell-like radio
60: morphology is accompanied by broad, blue-shifted [O {\sc I}] and [S
61: {\sc II}] emission lines indicative of an interstellar shock
62: \citep{demuizon+88,kulkarni92}. Millimeter CO emission traces this
63: radio and optical emission, suggesting the shock is interacting with
64: molecular material \citep{bally+89,phillips93,onello95}. In addition,
65: a bright near-infrared (NIR) star appears to be embedded in \gsev, and
66: it is surrounded by a strong H$\alpha$ reflection of spectral type Be
67: \citep{becker88,kulkarni92}.
68:
69: Ironically, it is the plethora of clues that make \gsev\ a perplexing
70: object, despite its discovery more than two decades ago
71: \citep{reich+85}. \gsev\ is unique in that nearly every Galactic
72: prototype has been proposed to explain it: young supernova remnant,
73: nova shell, stellar wind bubble, H~{\sc II} region and
74: Herbig-Haro-like outflow
75: \citep{reich+85,demuizon+88,green86,becker88}. However, none of these
76: can explain the low expansion velocities and the non-thermal radio
77: emission. The only currently proposed consistent theory to explain
78: these properties is one in which the bright NIR star is paired with an
79: unseen neutron star to form a Be-radio pulsar binary moving
80: supersonically through the dense gas \citep{kulkarni92}. In this
81: model, significant mass loss from the luminous Be-star inflates a
82: bubble which is filled by a mixture of the stellar wind with energetic
83: particles and magnetic field from the pulsar. This combination
84: creates the non-thermal radio emission coincident with the optical bow
85: shock of the medium surrounding the system. This model makes the
86: prediction that an X-ray source or pulsar should be seen coincident
87: with the embedded Be-star.
88:
89: Here, we report on archival X-ray and new Keck \lgs\ observations that
90: reveal an X-ray source with a NIR counterpart in the center of \gsev\
91: which is unassociated with the Be-star. We suggest that the object
92: known as \gsev\ is the result of the interaction of a luminous Be-star
93: and an X-ray emitting B-star/pulsar binary with dense molecular
94: material. If correct the resulting study of this object will be an
95: interesting laboratory for the study of plasma processes. The
96: observations and results are presented in \S\ref{sec:obs}. In
97: \S\ref{sec:dis} we discuss the implications of this source as it
98: relates to resolving the mystery surrounding \gsev.
99:
100: \section{Observations and Analysis}
101: \label{sec:obs}
102: \subsection{X-ray}
103: \gsev\ was observed 2003 October 11.33 UT with the ACIS-S detector on
104: {\em Chandra} in the standard, timed exposure mode. The archival data
105: were analyzed with CIAO version
106: 3.2\footnote{http://www.cxc.harvard.edu/ciao/}. We reprocessed the
107: level 1 events from the {\em Chandra} X-ray Center (CXC) in order to
108: make use of the latest calibration and removed pixel randomization.
109: The level 2 event file was created by filtering grades 0,2,3,4,6 and
110: good-time intervals. The total exposure time after filtering periods
111: higher than 3-$\sigma$ above the mean background level was 37.6\,ksec.
112:
113: Diffuse emission and a point source (hereafter \ptsrc;
114: Figure~\ref{fig:combo}) are detected at the position of \gsev\ (as
115: first noted by \citealt{arz04}). We compared the positions of 27 X-ray
116: sources on the S3 chip with counterparts in the 2MASS point source
117: catalog to correct the native astrometry \citep{cutri03}. This
118: comparison showed evidence for a small systematic shift,
119: $\Delta\alpha_{\rm 2MASS-CXO} =$ -0\farcs13 $\pm$ 0\farcs11,
120: $\Delta\delta_{\rm 2MASS-CXO} =$ -0\farcs08 $\pm$ 0\farcs11. The
121: best-fit position of \ptsrc\ including this offset is
122: $\alpha$(J2000)$=$\ra{20}{04}{23.430} and
123: $\delta$(J2000)$=$\dec{33}{39}{06\farcs73} with measurement uncertainty of
124: 0\farcs03 and 0\farcs08 in each coordinate, respectively. Combining
125: the measurement and transformation errors in quadrature gives an
126: uncertainty of 0\farcs18 (1-$\sigma$) for the X-ray position of
127: \ptsrc\ relative to 2MASS. This position lies 3\farcs6 from the nearby
128: luminous NIR star, which implies the two are not associated
129: \citep{kulkarni92,arz04}.
130:
131: We applied the adaptive smoothing algorithm CSMOOTH to highlight
132: the diffuse emission after subtraction of \ptsrc\ and produced a
133: flux-calibrated image by applying an exposure
134: map as outlined in the CIAO threads. The contours of this emission are
135: overlaid on an NIR image (see \S\ref{sec:lgsao}) of the nebula in
136: Figure~\ref{fig:combo}. Evidently, most of the diffuse X-ray emission
137: is not coincident with the diffuse NIR emission.
138:
139: We extracted photons within a 1\farcs5 circle (corresponding to 90\%
140: of the expected counts at 1.4\,keV) around \ptsrc\ to perform spectral
141: and variability analyses. The source contains only 33$^{+7}_{-6}$
142: counts. Upon examination of regions both inside the diffuse emission
143: and in a source-free area, we expect only two of these to be
144: background photons. A Kolmogorov-Smirnov test shows that the arrival
145: times of the source photons differed from a constant rate at only the
146: $\approx$ 1-$\sigma$ level, thus the source cannot be considered
147: variable.
148:
149: We begin our spectral analysis by noting that all the photons from
150: \ptsrc\ fall in the range 1.0--4.5\,keV, suggesting a hard
151: spectrum. After calculating the response matrix and effective area of
152: this portion of the CCD, we fit an absorbed power-law model to the
153: spectrum using Cash statistics (due to the limited number of counts;
154: \citealt{cash79}). The best-fit parameters in Table~\ref{tab:fit} show
155: a relatively hard photon index, $\Gamma \approx 1.8$, and low
156: luminosity, $L_X \approx 4 \times 10^{31}$\,erg\,s$^{-1}$
157: (2.0--10.0\,keV) for an assumed distance of 4.5\,kpc
158: \citep{bally+89}. These values are consistent with known accreting
159: neutron stars in quiescence (e.g. \citealt{rutledge01,campana05}).
160: In addition, we fit two absorbed power-law models with fixed
161: parameters (see Table~\ref{tab:fit}). The first has the photon index
162: set to a typical value for quiescent neutron stars, $\Gamma=2$, while
163: the second has the column density fixed to the best fit value of the
164: diffuse emission, $N_H$ = $1.0 \times 10^{22}$\,cm$^{-2}$ (see below).
165:
166: The probability of finding a source as bright or brighter than \ptsrc\
167: within the extent of \gsev\ can be determined from the local source
168: density. A WAVDET analysis of the active CCDs (ACIS-I2,3 and
169: ACIS-S1,2,3,4) finds that 12 sources are as bright or brighter than
170: \ptsrc. The inferred density is then $\approx$ 112
171: sources/deg$^{2}$. This density is consistent with observations taken
172: as part of the ChaMPlane Survey \citep{grindlay05}, which predicts
173: $\approx$ 100 sources/deg$^{2}$ with fluxes as bright or brighter than
174: \ptsrc\ \citep{hong05}. Consequently, there is a 0.3\% probability
175: that such a source would be found within \gsev\ by chance.
176:
177: The diffuse emission presented enough counts for basic spectroscopy
178: with $\chi^2$ statistics. We extracted events from a region of
179: dimension $\approx30$\arcsec$\times 30$\arcsec\ surrounding the
180: diffuse emission (excluding the point source) and a source-free
181: background region immediately east of the nebula with the same shape.
182: This yielded $690 \pm 26$ source counts, of which $\approx$ 320 are
183: expected to be due to the background. The resulting source plus
184: background photons were grouped such that each bin contained at least
185: 25 counts.
186:
187: The background subtracted spectrum was analyzed using
188: XSPECv11\footnote{See http://heasarc.nasa.gov/docs/xanadu/xspec/}. We
189: fit two models modified by absorption to the spectrum: a power-law and
190: a Raymond-Smith plasma (see Table~\ref{tab:fit}). The unphysically
191: steep photon index of the power-law model and the lower $\chi^2_\nu$
192: value lead us to adopt the Raymond-Smith model for the remainder of
193: our analysis. The derived value of $N_H$ is reasonably consistent
194: with that of \ptsrc\ and estimated value of
195: $1.25\times10^{22}$\,cm$^{-2}$ from \citet{dickey90}. Integrating
196: this model over the 0.5--2.5\,keV bandpass implies a luminosity of 8.1
197: $\times$ 10$^{32}$\cgslum\ at the distance of \gsev, although it is
198: not clear that this emission is associated with the nebula (see
199: \S\ref{sec:dis}).
200:
201:
202: \subsection{Near-Infrared LGS-AO}
203: \label{sec:lgsao}
204: \gsev\ was observed under photometric conditions on 2005 April 30 UT
205: with Laser Guide Star Adaptive Optics (\lgs; \citealt{lgs1,lgs2}) on
206: the Keck II telescope and the Near-Infrared Camera 2 (NIRC2). We
207: imaged the field in the $J$, $H$ and $K^{\prime}$-bands with the wide
208: camera of NIRC2, which provides a $\approx$40\arcsec$\times$40\arcsec\
209: field of view and a $\approx$ 0\farcs04 pixel scale. The $H$ and
210: $K^{\prime}$-band data sets consisted of five frames in each
211: band. Each frame was exposed for 5\,sec with 10 additions performed on
212: the chip at five dither positions separated by $\approx$
213: 30\arcsec. The $J$-band data consisted of two images at the center of
214: the chip.
215:
216: Each frame was flat-fielded, background subtracted, and repaired for
217: bad pixels using custom PyRAF software\footnote{PyRAF is a product of
218: Space Telescope Science Institute, which is operated by AURA for
219: NASA.}. We then performed a second round of sky subtraction using a
220: median combination of similarly processed frames of a nearby field. We
221: used these processed images of \gsev\ for photometric analysis, but
222: produced a separate set of images for astrometry due to optical
223: distortion in the NIRC2 camera. The distortion in the second set was
224: corrected using algorithms derived from the preshipment review
225: documents\footnote{available at
226: http://www2.keck.hawaii.edu/inst/nirc2/} with the IDL procedure
227: provided by the Keck Observatory\footnote{See
228: http://www2.keck.hawaii.edu/optics/lgsao/software/}. The correction
229: does not conserve flux, and thus is not suitable for photometry.
230:
231: We registered a median combination of the distortion corrected
232: $H$-band frames to the 2MASS point source catalog using 8 stars that
233: were not over-exposed. We find residuals of 0\farcs04 and 0\farcs09
234: in right ascension and declination, respectively. Registering the $J$
235: and $K^{\prime}$-band frames to this image yielded negligible
236: residuals. Combining these errors with those in the X-ray position of
237: \ptsrc\ yields an uncertainty of 0\farcs19 (1-$\sigma$) of the X-ray
238: image with respect to the NIR images. Figure~\ref{fig:combo} shows
239: the registered $H$-band frame with the Chandra error circle (99\%
240: confidence). We clearly identify a single bright NIR counterpart in
241: all filters within the X-ray error circle. The best fit position of
242: this source is $\alpha$(J2000)$=$\ra{20}{04}{23.446} and
243: $\delta$(J2000)$=$\dec{33}{39}{06\farcs62} with an uncertainty of
244: 0\farcs04 and 0\farcs09 (relative to 2MASS), respectively. The
245: centroiding errors are negligible. This position lies 0\farcs23 from
246: the {\em Chandra} position.
247:
248: We performed aperture photometry of the source in each band on each
249: individual frame relative to 2MASS stars in the field with the IRAF
250: package APPHOT. We assume that the color term used to transform from
251: the 2MASS $K_s$ filter to the $K^{\prime}$ is negligible for our
252: purposes. The uncertainties were determined with the 2MASS photometric
253: uncertainty, the standard deviation of the zero-point determinations
254: from the same 2MASS star in multiple frames and the photometric error
255: of the NIR source itself added in quadrature. We find magnitudes of $J
256: = 15.56 \pm 0.09$, $H = 14.51 \pm 0.11$ and $K^{\prime} = 13.97 \pm
257: 0.06$.
258:
259: The probability of finding a star with $K^{\prime} \approx 14$ magnitude in
260: our \chandra\ error circle by chance is very low. To quantify this we
261: extracted all sources present in the 2MASS catalog within 20\arcmin\
262: of \gsev. We find that the differential number of sources per
263: magnitude per square arcsecond is well described by a single power-law
264: with index 0.35 over the magnitude range 3 $< K_{\rm s} < $ 15. We can
265: conservatively assume (based on Galactic star count models by
266: \citealt{nakajima00}) that this can be extrapolated to our 5-$\sigma$
267: detection limit of $m_{K^{\prime}} \approx 20.0$ magnitude. From this
268: we calculate that there is a $\approx$ 25\% percent chance of finding
269: a source brighter than our detection limit in a circular region with a
270: 0\farcs49 (99\% confidence) radius. However, the probability of
271: finding a source with $K_{s} = 14.0$ magnitude or brighter is $\simlt$
272: 0.1\%. Thus it is unlikely that our NIR counterpart is drawn from the
273: background population, and we assume that it associated with \ptsrc.
274:
275: The key issue in determining the nature of this source is the assumed
276: extinction. Based on the colors of the luminous NIR star,
277: \citet{becker88} estimate $A_V \approx 5.6$. This agrees well with the
278: value of $A_V \approx 5.4$ obtained by taking $N_H$ as determined from
279: the spectrum of the diffuse X-ray emission and translating it into
280: extinction \citep{predehl95}. If we plot the NIR counterpart on a
281: color-magnitude diagram (see Figure~\ref{fig:cmd}) using this
282: reddening we find that the star is consistent with a late G/early K
283: giant spectral type at a distance of $\approx 11$\,kpc. Consequently,
284: the star is under luminous if it is associated with \gsev\ at a
285: distance of 4.5\,kpc.
286:
287: The spectral fitting of the point source spectrum itself, albeit with
288: poor statistics, implies a higher extinction of $A_V =
289: 8.4^{+6.4}_{-5.5}$. This allows for the possibility that the NIR
290: counterpart is a heavily obscured main sequence B-star with $A_V
291: \approx 10.0$ at the distance of \gsev. We prefer this interpretation
292: when we consider the probabilistic arguments and existing
293: multifrequency observations of \gsev\ (see \S\ref{sec:dis}).
294:
295: \section{Discussion and Conclusions}
296: \label{sec:dis}
297: We identify a low-luminosity, hard X-ray point source with a NIR
298: counterpart at the center of \gsev\ using high resolution imaging.
299: Both the measured X-ray luminosity, $L_X \approx 4 \times
300: 10^{31}$\,\cgslum, and the photon index, $\Gamma \approx 1.8$, of
301: \ptsrc\ are consistent with quiescent neutron star systems
302: (e.g. \citealt{rutledge01,campana05}). The magnitude and
303: $J$-$K^{\prime}$ color in combination with the X-ray column density
304: suggests the NIR counterpart is either an evolved background star or a
305: heavily extincted B-star. However, an isolated background
306: late-G/early-K giant cannot explain the observed X-ray flux. These
307: stars have deep convective zones that power coronal X-ray emission,
308: but it is typically $\simlt 10^{31}$\cgslum \citep{gudel04}. This is
309: an order of magnitude below the required $L_X \approx $
310: \sci{2.5}{32}\cgslum calculated assuming the observed X-ray flux at a
311: distance of 11\,kpc. In addition, spectral types later than B2 have
312: have observed X-ray luminosities $\simlt 10^{31}$\cgslum
313: \citep{berghofer97}. This suggests that the NIR source and \ptsrc\
314: constitute an X-ray binary, and probabilistic arguments suggest that
315: this binary is associated with \gsev.
316:
317: A simple geometric model can explain the existing multifrequency data
318: (Figure~\ref{fig:diagram}). The velocity of the molecular gas as
319: measured by CO observations is $5\kms$ with respect to the local
320: standard of rest \citep{bally+89}. The stellar H$\alpha$ line profile
321: from the bright Be-star is redshifted with respect to the CO with a
322: velocity of 20--60$\kms$, while H$\alpha$ reflected by dust in the
323: eastern region is also redshifted with respect to the CO, but is
324: blueward of the stellar H$\alpha$ by 10--50$\kms$. This implies that
325: the bright NIR star is moving into the cloud, away from the
326: Earth. However, the [O I] and H$\alpha$ throughout the rest of nebula
327: traces the non-thermal radio emission and is uniformly blue-shifted by
328: 10--120$\kms$ with respect to CO, suggesting that the source
329: responsible for the shock is moving into the cloud, toward the Earth.
330:
331: The cloud size, as inferred from CO, is 3$D_{4.5}$\,pc on the sky,
332: where $D_{4.5}$ is the distance to \gsev\ in units of 4.5\,kpc. If
333: the cloud is roughly spherical and has $n_H \sim 10^3$\,cm$^{-3}$,
334: then objects will have an additional $\approx$ 5 magnitudes of
335: extinction with respect to objects on the near side. Thus, a natural
336: explanation for the geometry of \gsev\ is that the bright Be-star is
337: moving into the near side of the cloud creating a reflection nebula,
338: while on the far side, a heavily extincted B-star/pulsar binary is
339: moving into the cloud creating a bow shock and powering the nebula
340: (Figure~\ref{fig:diagram}).
341:
342: One remaining puzzle is the origin and impact of the hot gas powering
343: the diffuse X-ray emission. Figure~\ref{fig:combo} shows that the
344: radio/optical and diffuse X-ray morphologies are substantially
345: different, and the peak of the diffuse X-ray emission is separated
346: $\approx 20$\arcsec from the center of the radio/optical emission
347: (which contains the Be-star and X-ray binary). Thus, it is apparent
348: that this hot gas does not play an important dynamic nor, given the
349: its luminosity is $\sim 10^{32}$\cgslum, energetic role in shaping
350: \gsev. Two viable explanations for the origin of the hot gas are,
351: given the quasi-shell like morphology, that it is the result of a
352: previous explosive event that the X-ray binary is overtaking or it may
353: be unassociated with the \gsev. In any case, the origin of this plasma
354: --- either related or unrelated to \gsev\ --- is unknown.
355:
356: The definitive proof of the proposed model (Figure~\ref{fig:diagram})
357: would be the discovery of a pulsar associated with \gsev. A search for
358: pulsations with the Green Bank Telescope at 2.2\,GHz is underway. If a
359: pulsar is found, \gsev\ will be an important laboratory for studying
360: plasma processes taking place between the pulsar/B-star wind and the
361: interaction of that mixture with the cold molecular gas.
362:
363: \acknowledgements We thank A. Kraus for useful discussions. This work
364: is supported in part by grants from the National Science Foundation
365: and NASA. The W. M. Keck Observatory is operated as a scientific
366: partnership among the California Institute of Technology, the
367: University of California, and the National Aeronautics and Space
368: Administration. The Observatory was made possible by the generous
369: financial support of the W. M. Keck Foundation. The authors wish to
370: recognize and acknowledge the very significant cultural role and
371: reverence that the summit of Mauna Kea has always had within the
372: indigenous Hawaiian community. We are most fortunate to have the
373: opportunity to conduct observations from this mountain.
374:
375: \bibliographystyle{apj}
376: %\bibliography{journals,ms}
377:
378: \begin{thebibliography}{20}
379: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
380:
381: \bibitem[{{Arzoumanian} {et~al.}(2004){Arzoumanian}, {Cordes}, {Van Buren},
382: {Corcoran}, {Safi-Harb}, \& {Petre}}]{arz04}
383: {Arzoumanian}, Z., {Cordes}, J., {Van Buren}, D., {Corcoran}, M., {Safi-Harb},
384: S., \& {Petre}, R. 2004, AAS/High Energy Astrophysics Division, 8,
385:
386: \bibitem[{{Bally} {et~al.}(1989){Bally}, {Pound}, {Stark}, {Israel}, {Hirano},
387: {Kameya}, {Sunada}, {Hayashi}, {Thronson}, \& {Hereld}}]{bally+89}
388: {Bally}, J., {Pound}, M.~W., {Stark}, A.~A., {Israel}, F., {Hirano}, N.,
389: {Kameya}, O., {Sunada}, K., {Hayashi}, M., {Thronson}, H.~J., \& {Hereld}, M.
390: 1989, \apjl, 338, L65
391:
392: \bibitem[{{Becker} \& {Fesen}(1988)}]{becker88}
393: {Becker}, R.~H. \& {Fesen}, R.~A. 1988, \apjl, 334, L35
394:
395: \bibitem[{{Bessell} \& {Brett}(1988)}]{bessell88}
396: {Bessell}, M.~S. \& {Brett}, J.~M. 1988, \pasp, 100, 1134
397:
398: \bibitem[Berghoefer et al.(1997)]{berghofer97} Berghoefer, T.~W.,
399: Schmitt, J.~H.~M.~M., Danner, R., \& Cassinelli, J.~P.\ 1997, \aap, 322,
400: 167
401:
402: \bibitem[Campana et al.(2005)]{campana05}
403: {Campana}, S., {Ferrari}, N., {Stella}, L., \& {Israel}, G.~L. 2005,
404: \aap, 434, L9
405:
406: \bibitem[{{Cash}(1979)}]{cash79}
407: {Cash}, W. 1979, \apj, 228, 939
408:
409: \bibitem[{{Cutri} {et~al.}(2003){Cutri}, {Skrutskie}, {van Dyk}, {Beichman},
410: {Carpenter}, {Chester}, {Cambresy}, {Evans}, {Fowler}, {Gizis}, {Howard},
411: {Huchra}, {Jarrett}, {Kopan}, {Kirkpatrick}, {Light}, {Marsh}, {McCallon},
412: {Schneider}, {Stiening}, {Sykes}, {Weinberg}, {Wheaton}, {Wheelock}, \&
413: {Zacarias}}]{cutri03}
414: {Cutri}, R.~M., {Skrutskie}, M.~F., {van Dyk}, S., {Beichman}, C.~A.,
415: {Carpenter}, J.~M., {Chester}, T., {Cambresy}, L., {Evans}, T., {Fowler}, J.,
416: {Gizis}, J., {Howard}, E., {Huchra}, J., {Jarrett}, T., {Kopan}, E.~L.,
417: {Kirkpatrick}, J.~D., {Light}, R.~M., {Marsh}, K.~A., {McCallon}, H.,
418: {Schneider}, S., {Stiening}, R., {Sykes}, M., {Weinberg}, M., {Wheaton},
419: W.~A., {Wheelock}, S., \& {Zacarias}, N. 2003, {2MASS All Sky Catalog of
420: point sources.} (The IRSA 2MASS All-Sky Point Source Catalog, NASA/IPAC
421: Infrared Science Archive.~http://irsa.ipac.caltech.edu/applications/Gator/)
422:
423: \bibitem[{{de Muizon} {et~al.}(1988){de Muizon}, {Strom}, {Oort}, {Claas}, \&
424: {Braun}}]{demuizon+88}
425: {de Muizon}, M., {Strom}, R.~G., {Oort}, M.~J.~A., {Claas}, J.~J., \& {Braun},
426: R. 1988, \aap, 193, 248
427:
428: \bibitem[{{Dickey} \& {Lockman}(1990)}]{dickey90}
429: {Dickey}, J.~M. \& {Lockman}, F.~J. 1990, \araa, 28, 215
430:
431: \bibitem[{{Green}(1986)}]{green86}
432: {Green}, D.~A. 1986, \mnras, 219, 39P
433:
434: \bibitem[{{Grindlay} {et~al.}(2005){Grindlay}, {Hong}, {Zhao}, {Laycock}, {van
435: den Berg}, {Koenig}, {Schlegel}, {Cohn}, {Lugger}, \& {Rogel}}]{grindlay05}
436: {Grindlay}, J.~E., {Hong}, J., {Zhao}, P., {Laycock}, S., {van den Berg}, M.,
437: {Koenig}, X., {Schlegel}, E.~M., {Cohn}, H.~N., {Lugger}, P.~M., \& {Rogel},
438: A.~B. 2005, \apj, 635, 920
439:
440: \bibitem[G{\"u}del(2004)]{gudel04} G{\"u}del, M.\ 2004, \aapr,
441: 12, 71
442:
443: \bibitem[{{Hong} {et~al.}(2005){Hong}, {van den Berg}, {Schlegel}, {Grindlay},
444: {Koenig}, {Laycock}, \& {Zhao}}]{hong05}
445: {Hong}, J., {van den Berg}, M., {Schlegel}, E.~M., {Grindlay}, J.~E., {Koenig},
446: X., {Laycock}, S., \& {Zhao}, P. 2005, \apj, 635, 907
447:
448: \bibitem[{{Kulkarni} {et~al.}(1992){Kulkarni}, {Vogel}, {Wang}, \&
449: {Wood}}]{kulkarni92}
450: {Kulkarni}, S.~R., {Vogel}, S.~N., {Wang}, Z., \& {Wood}, D.~O.~S. 1992, \nat,
451: 360, 139
452:
453: \bibitem[{{Nakajima} {et~al.}(2000){Nakajima}, {Iwamuro}, {Maihara},
454: {Motohara}, {Terada}, {Goto}, {Iwai}, {Tanabe}, {Taguchi}, {Hata},
455: {Yanagisawa}, {Iye}, {Kashikawa}, \& {Tamura}}]{nakajima00}
456: {Nakajima}, T., {Iwamuro}, F., {Maihara}, T., {Motohara}, K., {Terada}, H.,
457: {Goto}, M., {Iwai}, J., {Tanabe}, H., {Taguchi}, T., {Hata}, R.,
458: {Yanagisawa}, K., {Iye}, M., {Kashikawa}, N., \& {Tamura}, M. 2000, \aj, 120,
459: 2488
460:
461: \bibitem[{{Onello} {et~al.}(1995){Onello}, {Depree}, {Phillips}, \&
462: {Goss}}]{onello95}
463: {Onello}, J.~S., {Depree}, C.~G., {Phillips}, J.~A., \& {Goss}, W.~M. 1995,
464: \apjl, 449, L127+
465:
466: \bibitem[{{Phillips} {et~al.}(1993){Phillips}, {Onello}, \&
467: {Kulkarni}}]{phillips93}
468: {Phillips}, J.~A., {Onello}, J.~S., \& {Kulkarni}, S.~R. 1993, \apjl, 415,
469: L143+
470:
471: \bibitem[{{Predehl} \& {Schmitt}(1995)}]{predehl95}
472: {Predehl}, P. \& {Schmitt}, J.~H.~M.~M. 1995, \aap, 293, 889
473:
474: \bibitem[{{Reich} {et~al.}(1985){Reich}, {F\"{u}rst}, {Altenhoff}, {Reich}, \&
475: {Junkes}}]{reich+85}
476: {Reich}, W., {F\"{u}rst}, E., {Altenhoff}, W.~J., {Reich}, P., \& {Junkes}, N.
477: 1985, \aap, 151, L10
478:
479: \bibitem[{{Rutledge} {et~al.}(2001){Rutledge}, {Bildsten}, {Brown}, {Pavlov},
480: \& {Zavlin}}]{rutledge01}
481: {Rutledge}, R.~E., {Bildsten}, L., {Brown}, E.~F., {Pavlov}, G.~G., \&
482: {Zavlin}, V.~E. 2001, \apj, 559, 1054
483:
484: \bibitem[{{van Dam} {et~al.}(2006){van Dam}, {Bouchez}, {Le Mignant},
485: {Johansson}, {Wizinowich}, {Campbell}, {Chin}, {Hartman}, {Lafon}, {Stomski},
486: \& {Summers}}]{lgs2}
487: {van Dam}, M.~A., {Bouchez}, A.~H., {Le Mignant}, D., {Johansson}, E.~M.,
488: {Wizinowich}, P.~L., {Campbell}, R.~D., {Chin}, J.~C.~Y., {Hartman}, S.~K.,
489: {Lafon}, R.~E., {Stomski}, Jr., P.~J., \& {Summers}, D.~M. 2006, \pasp, 118,
490: 310
491:
492: \bibitem[{{Wizinowich} {et~al.}(2006){Wizinowich}, {Le Mignant}, {Bouchez},
493: {Campbell}, {Chin}, {Contos}, {van Dam}, {Hartman}, {Johansson}, {Lafon},
494: {Lewis}, {Stomski}, {Summers}, {Brown}, {Danforth}, {Max}, \&
495: {Pennington}}]{lgs1}
496: {Wizinowich}, P.~L., {Le Mignant}, D., {Bouchez}, A.~H., {Campbell}, R.~D.,
497: {Chin}, J.~C.~Y., {Contos}, A.~R., {van Dam}, M.~A., {Hartman}, S.~K.,
498: {Johansson}, E.~M., {Lafon}, R.~E., {Lewis}, H., {Stomski}, P.~J., {Summers},
499: D.~M., {Brown}, C.~G., {Danforth}, P.~M., {Max}, C.~E., \& {Pennington},
500: D.~M. 2006, \pasp, 118, 297
501:
502: \end{thebibliography}
503:
504: \clearpage
505:
506: \begin{deluxetable}{lccccc}
507: \tablecaption{\label{tab:fit} X-ray Spectral Fits.}
508: \tablehead{
509: \colhead{Model} & \colhead{$N_H$} & \colhead{$\Gamma$/$k_BT$} &
510: \colhead{Flux} & \colhead{$\chi^2/\nu$} \\
511: & \colhead{($10^{22}$~cm$^{-2}$)} & \colhead{(keV)} &
512: \colhead{($10^{-14}$~erg~cm$^{-2}$~s$^{-1}$)} & \\
513: \colhead{(1)}& \colhead{(2)}& \colhead{(3)}& \colhead{(4)}& \colhead{(5)}\\
514: }
515: \startdata
516: Diffuse Emission \\
517: \hline
518: Power-law & $0.70^{+0.19}_{-0.13}$ & $4.4^{+0.9}_{-0.5}$ & $31^{+22}_{-11}$ & 18.2/13\\
519: Raymond-Smith Plasma & $1.04^{+0.08}_{-0.07}$ & $0.71^{+0.05}_{-0.07}$ & $33^{+9}_{-5}$ & 9.75/13\\
520: \hline
521: \ptsrc \\
522: \hline
523: Power-law & $1.5^{+1.1}_{-1.0}$ & 1.8$^{+1.2}_{-1.1}$ & 1.7$^{+5.6}_{-1.7}$ & ---\\
524: Power-law ($\Gamma=2.0$)& $1.6^{+0.5}_{-0.4}$ & (2.0) & 1.6$^{+0.5}_{-0.4}$ & ---\\
525: Power-law ($N_H=1\times10^{22}$)& (1.0) & $1.3^{+0.5}_{-0.4}$ & $2.0^{+1.0}_{-0.7}$ & ---\\
526: \enddata
527: \tablecomments{ All errors are 68\% confidence levels. Values in parentheses are held fixed.
528: (1)~--~Absorbed spectral model.
529: (2)~--~Best-fit column density.
530: (3)~--~Measured photon index for power-law models and $k_BT$ for the
531: Raymond-Smith plasma.
532: (4)~--~The unabsorbed flux in the 0.5--2.5\,keV band for the diffuse emission
533: and 2--10\,keV band for \ptsrc.
534: (5)~--~The value of $\chi^2$ for diffuse emission models and the number
535: of degrees of freedom, $\nu$. This column is not applicable to
536: \ptsrc\ since the spectral fitting was performed with Cash
537: statistics.
538: }
539: \end{deluxetable}
540:
541: \clearpage
542:
543: \begin{figure}
544: \plotone{f1.eps}
545: \epsscale{0.8}
546: \caption{H-band image of \gsev\ with contours of the adaptively
547: smoothed X-ray emission (black lines) and the \chandra\ 99\%
548: confidence (0\farcs49) error circle (blue circle). The X-ray
549: contours are logarithmically spaced between 10\% and 90\% of the peak
550: emission.}
551: \label{fig:combo}
552: \end{figure}
553:
554: \clearpage
555:
556: \begin{figure}
557: \plotone{f2.eps}
558: \caption{Color magnitude diagram using data from \citet{bessell88}.
559: Filled circles show the IR counterpart for $A_V = 5.5$ and $A_V =
560: 10.5$ at a distance of 4.5\,kpc .}
561: \label{fig:cmd}
562: \end{figure}
563:
564: \clearpage
565:
566: \begin{figure}
567: \plotone{f3.eps}
568: \caption{Diagram of the geometry of \gsev. The newly discovered X-ray
569: binary moves into the far side of the molecular material powering the
570: radio/NIR/optical nebula, whereas the Be-star creates a reflection
571: nebula on the near-side. See the text for details.}
572: \label{fig:diagram}
573: \end{figure}
574:
575: \end{document}
576: