1: \documentclass[12pt,preprint]{aastex}
2:
3: %NEW COMMANDS
4: \newcommand{\phunit}{photons cm$^{-2}$ sr$^{-1}$ s$^{-1}$ \AA$^{-1}$}
5: \newcommand{\ergunit}{ergs cm$^{-2}$ sr$^{-1}$ s$^{-1}$}
6: \newcommand{\lya}{Ly~$\alpha$}
7: \newcommand{\lyb}{Ly~$\beta$}
8: \newcommand{\fdegree}{\,\hbox{$.\!\!^\circ$}}
9: \newcommand{\fuse}{{\it FUSE}}
10: \newcommand{\thc}{$\theta^{1}$ Ori C}
11:
12: %Title info
13: \shorttitle{Intense Diffuse Far-UV Emission from the Orion Nebula}
14: \shortauthors{Murthy et al.}
15:
16: \begin{document}
17:
18: \title{Intense Diffuse Far-UV Emission from the Orion Nebula}
19: \author{Jayant Murthy}
20: \affil{Indian Institute of Astrophysics, Koramangala, Bangalore 560 034}
21: \email{jmurthy@yahoo.com}
22:
23: \author{David J. Sahnow and R. C. Henry}
24: \affil{The Johns Hopkins University, Baltimore, MD 21218}
25: \email{sahnow@pha.jhu.edu}
26: \email{henry@jhu.edu}
27:
28: \begin{abstract}
29:
30: We present spectra of the diffuse FUV (900 -- 1200 \AA) emission from a region near the Orion
31: Nebula: the first high resolution spectra of the diffuse background radiation. These observations
32: were made using serendipitous \fuse\ observations and were only possible because of the strength of
33: the diffuse emission ($\approx 3 \times\ 10^{5}$ \phunit) and the sensitivity of the \fuse\ instrument.
34: Preliminary modeling suggests that the light is
35: scattered starlight from the Trapezium stars, primarily \thc. However, a comparison of the spectra
36: with nearby stars shows significant differences in the presence and strength of the absorption lines,
37: particularly in the \lyb\ line where there is
38: much less interstellar absorption in our diffuse spectrum. We believe that we are indeed observing scattered
39: light from the Trapezium stars but through a line of sight with much less matter than the direct line
40: to the stars.
41:
42: \end{abstract}
43:
44: \keywords{ISM: general, ISM: individual M42, ultraviolet: ISM}
45:
46: \section{Introduction}
47:
48: The Orion Nebula (M42) is one of the best studied of all astronomical objects with emission in
49: different wavelengths reflecting different aspects of its physics and morphology. An excellent
50: description of the observations and the resulting model of the region is given by \citet{odell01}.
51: Intense ultraviolet emission from the Orion Nebula has been observed by rocket flights
52: (\citet{boh82}, \citet{ona84}) and from the International Ultraviolet Explorer \citep{mat81} and is
53: presumed to be due to the forward scattering of the light of the brightest of the
54: Trapezium stars --- \thc\ (HD 37022) --- by dust close to and in front of the star. Note that the
55: scattering is not actually from the Orion Nebula, which is defined as a region of
56: ionized gas in front of the Orion Molecular Cloud (OMC-1), but is rather from dust in the same
57: direction. In this
58: paper we present the first observations of diffuse emission from Orion in the far ultraviolet (FUV)
59: using serendipitous observations made by the Far Ultraviolet Spectroscopic Explorer (\fuse).
60:
61: \section{Observations and Data Analysis}
62:
63: \fuse\ was
64: launched on June 24, 1999 into a low Earth orbit (LEO) by a Delta II rocket and has been observing
65: astronomical targets since then. The spacecraft and mission have been described by \citet{M00}
66: and \citet{S00}. The instrument
67: consists of four coaligned optical channels, two of which are coated with silicon carbide (SiC)
68: and two with lithium fluoride (LiF) over aluminum, providing coverage over the spectral range from 905 --
69: 1187 \AA. Diffuse sources, such as we report on here, will be visible in all 3 apertures (LWRS,
70: MDRS, and HIRS) but useful spectra can only be obtained from the LWRS ($30\arcsec\ \times\ 30\arcsec$)
71: and, for sufficiently bright sources, MDRS ($4\arcsec\ \times\ 20\arcsec$) apertures.
72:
73: Two of the observations in the S405/505 program, an operational program used as part of the
74: channel realignment of the \fuse\ spectrographs, were of a region of empty space near the star
75: HD 36981 (Fig. \ref{dss_fig}, Table \ref{obs_log}). A check of the attitude has shown that the actual pointing
76: was within a few arcseconds of the nominal position.
77: We have downloaded the data from these two observations
78: and processed them using the standard CALFUSE pipeline (v2.4, Dixon, Kruk, \& Murphy 2002) with
79: the modifications described by \citet{mur04}. One of these modifications was to use only the
80: data from the ``NIGHT'' part of the orbit in order to exclude all airglow emission, apart from
81: the Lyman lines of atmospheric hydrogen. Although there
82: may still be residual amounts of the \ion{O}{1} lines around 1040 \AA\ and the \ion{N}{1} lines at
83: 1134 \AA\ \citep{F01}, these will not contribute to the intense continuum emission observed in Orion.
84: Other modifications to the standard pipeline were that we added together all the different exposures
85: which form a single observation and replaced the background subtraction procedure
86: with one where we estimated the background from the region on either side of the aperture. Finally, in
87: order to increase the signal-to-noise ratio, we binned together the spectra by a factor of 32 in
88: wavelength, to reduce the effective resolution to about 700. We have used this procedure to reduce the
89: data from most of the S405/S505 observations and the results are discussed by \citet{mur04}. This
90: is the brightest of the diffuse fields observed in that program.
91:
92: The resultant spectra are shown in Fig. \ref{spectra} in which we have coadded the data from the two
93: different observations. There are three different spectra shown in the Figure, all from the LWRS
94: aperture: the coadded 1B and 2A SiC spectra (900 -- 1000 \AA); the 1A LiF spectrum (980 -- 1100 \AA);
95: and the coadded 1B and 2A LiF spectra (1080 -- 1200 \AA). The same data are shown at higher resolution
96: in Figs. \ref{lines} and \ref{long_spec}.
97: For clarity, we have not shown the
98: spectra from the other \fuse\ segments which are entirely consistent with those plotted here. Because
99: the diffuse emission is so bright, we were also able to extract spectra from the MDRS aperture which
100: was pointed about 3\arcmin\ away from the LWRS. Despite the slightly different pointing,
101: the MDRS spectra were essentially identical with the LWRS spectra.
102:
103: \section{Results and Discussion}
104:
105: The \fuse\ spectrographs are the first instruments with sufficient sensitivity to
106: perform absorption line spectroscopy of the diffuse ISM, albeit only in bright
107: areas such as Orion. The diffuse emission is due to scattered light from the nearby stars and a
108: comparison of the spectra can provide important clues about the source of the photons and the
109: geometry of the dust. Although our first assumption was that we were observing scattering of
110: the light of HD 36981 (B5V, V = 8), only about 2\arcmin\ away, a comparison of the spectra near
111: \lyb\ shows that
112: HD 36981 cannot be the major contributor to the scattered radiation because the broad intrinsic
113: \lyb\ absorption in the star is not reflected in our observed spectrum (Fig. \ref{lines}).
114:
115: Despite its much greater distance (12\arcmin) from our observed locations,
116: \thc\ (HD 37022; O5V; V = 5.1) ---
117: the brightest of the Trapezium stars --- may contribute as much or more energy as
118: HD 36981, depending on the relative geometry. Unfortunately HD 37022 is much too bright to observe with
119: \fuse\ but was observed by Copernicus. We downloaded the spectrum from the MAST
120: archives and applied the calibration of \citet{sno77} and this is plotted in
121: Fig. \ref{spectra}. The shape matches the diffuse spectrum well until about 1150 \AA\ when interstellar
122: \lya\ absorption eats into the stellar spectrum. It was difficult to establish
123: the absolute calibration of the Copernicus spectrograph and instead we
124: downloaded an International Ultraviolet Explorer observation (SWP 5085) from MAST to compare
125: the stellar flux with the diffuse spectrum, at least at wavelengths above 1150 \AA\
126: (Fig. \ref{long_spec}).
127:
128: Despite the overall similarity in the shape of the spectra, a detailed comparison
129: (Figs. \ref{lines} and \ref{long_spec}) shows many differences. As with HD36981,
130: the \lyb\ absorption line is much broader in the spectrum of \thc\ than in our diffuse spectrum but,
131: in this case, is due to interstellar absorption in the several interstellar clouds
132: in front of the Trapezium \citep{pr01}. Many of the other lines in the diffuse spectrum
133: have been identified as
134: being due to molecular hydrogen in both absorption and emission (France - personal communication,
135: \citet{M03}) which, surprisingly, do not
136: appear with the same strength in either of the stellar spectra, even though the stars are approximately
137: co-located with the scattering dust.
138:
139: While a definitive interpretation must await a more detailed
140: analysis of the scattering geometry, it seems likely that we are observing the scattered light from
141: \thc\ through a much lower column density line of sight than the direct line to the star.
142: We have begun the process of modeling the entire spectrum including the absorption lines in order
143: to determine the scattering properties of the dust grains near Orion.
144:
145: \section{Conclusions}
146:
147: We present the first high resolution spectra of diffuse scattering in the FUV using serendipitous
148: \fuse\ observations of a blank patch of sky near HD 36981 and just off the Orion Nebula. We
149: propose that the emission is due to scattering of the light from \thc\ (12\arcmin\ away) rather than
150: the much closer HD 36981 (2\arcmin\ away), based on a comparison of the spectra. The HD 36981 spectrum
151: has broad self-absorbed Lyman absorption lines which are not seen in the diffuse spectrum. Although
152: there are also broad Lyman absorption lines in the spectrum of \thc, these are due to interstellar
153: absorption in foreground gas and suggest that the path traversed by the scattered photons includes
154: much less gas than the direct path to the star.
155:
156: We believe that such high resolution observations of the diffuse scattering provide an important new tool
157: for the study of interstellar dust scattering not just in understanding the spectral properties of the
158: interstellar dust but also in providing clues to the origin of the diffuse emission. We have
159: proposed new \fuse\ observations of several locations in and around M42 and hope to integrate these
160: observations with archival {\it IUE} data in the region \citep[for example, ][]{mat81}. Through these
161: data and making use of the exceptionally well understood geometry around Orion, we should be able
162: to, for the first time, pin down the optical properties of the interstellar dust grains from the
163: FUV through the visible.
164:
165: \acknowledgments
166: We thank the FUSE team for their support and quick response to the many questions we
167: raised while working on this problem. In particular, Bill Blair, Van Dixon, Alex Fullerton
168: and B-G Andersson provided clarification on many points. Steve McCandliss and Kevin France
169: pointed out the
170: importance of H$_2$ absorption and emission in the diffuse spectrum. The data presented in this paper
171: was obtained from the Multimission Archive at the Space Telescope Science Institute (MAST).
172: STScI is operated by the Association of Universities for Research in Astronomy, Inc.,
173: under NASA contract NAS5-26555. Support for MAST for non-HST data is provided by the
174: NASA Office of Space Science via grant NAG5-7584 and by other grants and contracts. This
175: research has made use of NASA's Astrophysics Data System and the SIMBAD database, operated at
176: CDS, Strasbourg, France.
177:
178: \begin{thebibliography}{}
179:
180: \bibitem[Bohlin et al.(1982)]{boh82}
181: Bohlin, R. C., Hill, J. K., Stecher, T. P., \& Witt 1982, ApJ, 255, 87
182:
183: \bibitem[Dixon et al.(2002)]{D02}
184: Dixon, W. V. D., Kruk, J. \& Murphy The CalFUSE Pipeline Reference Guide V1.3
185: (http://fuse.pha.jhu.edu)
186:
187: \bibitem[Feldman et al.(2001)]{F01}
188: Feldman, P. D., Sahnow, D. J., Kruk, J. W., Murphy, E. M., \& Moos, H. W. 2001, JGR, 106, 8119
189:
190: \bibitem[Mathis et al.(1981)]{mat81}
191: Mathis, J. S., Perinotto, M., Patriarchi, P., \& Shiffer, F. H 1981, ApJ, 249, 99
192:
193: \bibitem[McCandliss(2003)]{M03}
194: McCandliss, S. R. 2003, PASP, 115, 651
195:
196: \bibitem[Moos et al.(2000)]{M00}
197: Moos, H. W. et al. 2000, ApJL, 538, 1
198:
199: \bibitem[Murthy \& Sahnow(2004)]{mur04}
200: Murthy, J., \& Sahnow, D. J. 2004 ApJ, 615, 315
201:
202: \bibitem[O'Dell(2001)]{odell01}
203: O'Dell, C. R. 2001, ARA\&A, 39, 99
204:
205: \bibitem[Price et al.(2001)]{pr01}
206: Price, R. J., Crawford, I. A., Barlow, M. J., \& Howarth, I. D. 2001, MNRAS 328, 555
207:
208: \bibitem[Sahnow et al.(2000)]{S00}
209: Sahnow, D. J. et al. 2000, ApJ, 538, L7
210:
211: \bibitem[Schlegel et al.(1998)]{S98}
212: Schlegel, D. J., Finkbeiner, D. P., \& Davis, M., 1998, ApJ, 500, 525
213:
214: \bibitem[Snow \& Jenkins(1977)]{sno77}
215: Snow, T.P. \& Jenkins, E. B. 1977, ApJSS, 33, 269
216:
217: \bibitem[Tanaka et al.(1984)]{ona84}
218: Tanaka, W., Onaka, T., Sawamura, M., Watanabe, T., Kodaira, K., \& Nishi, K. 1984, ApJ, 280, 213
219:
220: \end{thebibliography}
221:
222: \clearpage
223:
224: \begin{figure}
225: \plotone{f1.eps}
226: \caption{An interstellar dust map \citep{S98} is shown with the Orion Nebula (M42) in the center of the image.
227: Overplotted are the hot stars in Orion with the size of the symbol proportional to the TD-1 magnitude
228: of the star. The two S40546 observations are very near the star HD36981 (red circle
229: at the NE edge of M42). The LWRS
230: ($30\arcsec \times 30\arcsec$) aperture is centered 105\arcsec\ to the NE of the star and the MDRS
231: aperture 100\arcsec\ to the SW. The observations are about 12\arcmin\ away from \thc\
232: which provides most of the power for the surrounding Orion Nebula.}
233: \label{dss_fig}
234: \end{figure}
235:
236: \begin{figure}
237: \plotone{f2.eps}
238: \caption{The spectrum for our diffuse observation is shown. Only the spectra from three of the
239: detector segments are shown for
240: clarity; the extracted spectra from the other segments are consistent with these. Also omitted for clarity are the error bars which
241: are on the order of 10000 \phunit. Note the excellent agreement between segments at the edges, validating
242: our empirical background subtraction procedure. Superimposed on the plot is a scaled
243: spectrum of $\theta^{1}$ Ori (red line) taken from the Copernicus archive. The spectral shape is
244: similar to the diffuse spectrum until the onset of interstellar \lya\ absorption near 1200 \AA.
245: }
246: \label{spectra}
247: \end{figure}
248:
249: \begin{figure}
250: \plotone{f3.eps}
251: \caption{
252: Plotted are (from top to bottom) the \fuse\ spectrum of HD 36981, the Copernicus spectrum of \thc\
253: and the \fuse\ observation reported here. The \lyb\ line is much broader in both stellar spectra
254: than in the diffuse observation; in the case of HD 36981 due to self-absorption in the star and
255: due to interstellar absorption in the case of \thc. The geocoronal \lyb\ emission
256: line is seen in the middle of the interstellar \lyb\ (1026 \AA)
257: absorption line in our diffuse \fuse\ observation. Note that, surprisingly, the diffuse spectrum is
258: rich in absorption lines which are not seen in either of the stellar spectra. France
259: (personal communication) has suggested that these are largely due to H$_2$ absorption but also with
260: contributions from flurescent H$_2$ emission; however, a full
261: analysis is beyond the scope of this work.
262: }
263: \label{lines}
264: \end{figure}
265:
266: \begin{figure}
267: \plotone{f4.eps}
268: \caption{
269: Plotted is our diffuse spectrum at long wavelengths with a calibrated Copernicus spectrum
270: of \thc\ and the diffuse observation are plotted. Because of an uncertain
271: calibration of the Copernicus data, we have also plotted an IUE observation of \thc\
272: which is well calibrated (red line).
273: The units are, respectively, photons cm$^{-2}$ sr$^{-1}$ s$^{-1}$ for the stellar spectra on the left hand
274: side and \phunit\ for the diffuse spectrum, on the right hand side.
275: }
276: \label{long_spec}
277: \end{figure}
278:
279: \clearpage
280:
281: \begin{deluxetable}{rcccc}
282: \tablecaption{Observational Parameters}
283: \tablewidth{0pt}
284: \tablehead{
285: \colhead{Observation } & \colhead{Aperture} & \colhead{Night Exposure Time (s)} & \colhead{L} & \colhead{B}}
286: \startdata
287: S4054601 & LWRS & 10565 & 208.81 & -19.31\\
288: S4054601 & MDRS & 10565 & 208.81 & -19.31\\
289: S4054602 & LWRS & 5696 & 208.82 & -19.36\\
290: S4054602 & MDRS & 5696 & 208.81 & -19.36\\
291: \enddata
292: \label{obs_log}
293: \end{deluxetable}
294:
295: \end{document}
296: