1: %
2: %\documentclass[12pt,manuscript]{aastex}
3: \documentclass[12pt,preprint]{aastex}
4:
5: \shorttitle{IRC+10216 CS(14--13)}
6: \shortauthors{Young et al.}
7:
8: %\received{2003 October 1}
9: \begin{document}
10:
11: \slugcomment{Accepted by ApJ Letters: March 8, 2004}
12:
13: \title{Submillimeter Array Observations of
14: CS J=14--13 Emission from the Evolved Star IRC+10216}
15:
16: \author{K.H. Young\altaffilmark{1},
17: T.R. Hunter\altaffilmark{1},
18: D.J. Wilner\altaffilmark{1},
19: M.A. Gurwell\altaffilmark{1}, \\
20: J.W. Barrett\altaffilmark{1},
21: R. Blundell\altaffilmark{1},
22: R. Christensen\altaffilmark{2},
23: D. Fong\altaffilmark{2},
24: N. Hirano\altaffilmark{3},
25: P.T.P. Ho\altaffilmark{1,3},
26: S.Y. Liu\altaffilmark{3},
27: K.Y. Lo\altaffilmark{3,5},
28: R. Martin\altaffilmark{3},
29: S. Matsushita\altaffilmark{3},
30: J.M. Moran\altaffilmark{1},
31: N. Ohashi\altaffilmark{3},
32: D.C. Papa\altaffilmark{1},
33: N. Patel\altaffilmark{1},
34: F. Patt\altaffilmark{3},
35: A. Peck\altaffilmark{2},
36: C. Qi\altaffilmark{1},
37: M. Saito\altaffilmark{1,4},
38: A. Schinckel\altaffilmark{2},
39: H. Shinnaga\altaffilmark{2},
40: T.K. Sridharan\altaffilmark{1},
41: S. Takakuwa\altaffilmark{2},
42: C.E. Tong\altaffilmark{1},
43: D.V. Trung\altaffilmark{3}
44: }
45: \email{dwilner@cfa.harvard.edu}
46: \altaffiltext{1}{Harvard-Smithsonian Center for Astrophysics,
47: 60 Garden Street, Cambridge, MA 02138, USA}
48: \altaffiltext{2}{Harvard-Smithsonian Center for Astrophysics,
49: Submillimeter Array, 645 N. A'ohoku Place, Hilo, HI 96721, USA}
50: \altaffiltext{3}{Academia Sinica Institute of Astronomy \&
51: Astrophysics, P.O. Box 23-141, Taipei, Taiwan, 106, R.O.C.}
52: \altaffiltext{4}{National Astronomical Observatory of Japan,
53: 2-21-1 Osawa, Mitaka, Tokyo 181-8588, Japan}
54: \altaffiltext{5}{National Radio Astronomy Observatory,
55: 520 Edgemont Road, Charlottesville, VA 22903, USA}
56:
57:
58: \begin{abstract}
59: We present imaging observations of the evolved star IRC+10216
60: in the CS J=14--13 line at 685.4 GHz and associated submillimeter
61: continuum at $\sim 2''$ resolution made with the partially constructed
62: Submillimeter Array.
63: The CS J=14--13 line emission from the stellar envelope is
64: well resolved both spatially and spectrally.
65: The strong central concentration of the line emission provides
66: direct evidence that CS is a parent molecule that forms close to
67: the stellar photosphere, in accord with previous images of the
68: lower excitation CS J=2--1 line and inferences from unresolved
69: observations of vibrationally excited transitions.
70: The continuum emission is dominated by a compact, unresolved
71: component, consistent with the photospheric emission,
72: that accounts for $\sim20\%$ of the broadband 450~$\mu$m flux.
73: These are the first interferometer imaging observations
74: made in the semi-transparent 450~$\mu$m atmospheric window.
75: \end{abstract}
76:
77: \keywords{circumstellar matter --- stars: individual (IRC+10216)
78: --- astrochemistry --- techniques: interferometric}
79:
80: \section{Introduction}
81: The nearby carbon star IRC+10216 (CW Leo, IRAS 09452+1330) is surrounded
82: by a dusty, expanding envelope with relatively simple geometry and kinematics
83: that has been the subject of many studies of astrochemical processes.
84: A rich chemistry develops in the envelope as gas is lost from the
85: star at a high rate ($\sim3\times10^{-5}$~M$_{\odot}$~yr$^{-1}$,
86: Crosas \& Menten 1997) and is modified by thermodynamic equilibrium and
87: non-equilibrium reactions, photochemical reactions, ion-molecule reactions,
88: and the condensation of dust grains.
89: The range of physical conditions in the envelope make it especially
90: well suited to observations of rotational emission lines from molecules,
91: and numerous species have been identified in spectral scans at millimeter
92: and submillimeter wavelengths
93: (e.g. Cernicharo, Guelin \& Kahane 2000; Groesbeck, Phillips \& Blake 1994,
94: Avery et al. 1992).
95:
96: The close proximity ($\sim150$~pc) and copious mass loss of IRC+10216 allows
97: studies of the circumstellar envelope on a wide range of size scales.
98: On the largest scales, an extensive envelope of dust
99: (Mauron \& Huggins 1999) and gas (Fong, Meixner \& Shah 2003) has been
100: found to extend to $\sim0.15$~pc ($\sim3\farcm3$) in radius.
101: On much smaller scales,
102: interferometric imaging at millimeter wavelengths has revealed the spatial
103: distribution of molecular species within the central arcminute
104: (e.g. Bieging \& Tafalla 1993,
105: Guelin, Lucas \& Cernicharo 1993,
106: Lucas \& Guelin 1999).
107: Parent molecules are found to be centrally peaked (e.g. SiS, HCN), while
108: daughter molecules are found with shell distributions
109: (e.g. CN, C$_2$H, HNC). High angular resolution is especially important for
110: imaging lines of high excitation that emerge from the innermost regions of
111: the envelope, where the temperatures and densities are highest.
112:
113: Since the high excitation rotational lines of many common molecules
114: lie at short submillimeter wavelengths,
115: the Submillimeter Array\footnote{
116: The Submillimeter Array is a joint
117: project between the Smithsonian Astrophysical Observatory and the
118: Academia Sinica Institute of Astronomy and Astrophysics, and is
119: funded by the Smithsonian Institution and the Academia Sinica.}
120: (SMA)
121: opens new possibilities for probing physical conditions and chemistry
122: by enabling imaging observations with arcsecond resolution.
123: In this {\em Letter}, we present observations of IRC+10216 in CS J=14--13
124: emission made with the partially constructed SMA (Ho, Moran \& Lo 2004).
125: These are the first interferometer images from observations in the
126: atmospheric window centered near 450~$\mu$m, where $\sim$30\% or better
127: transmission occurs at high, dry sites like the summit of Mauna Kea a
128: significant fraction of the time. They build on the pioneering efforts of
129: Carlstrom et al. (1994),
130: who obtained fringes on the nearby single baseline between the
131: Caltech~Submillimeter~Observatory and the James~Clerk~Maxwell~Telescope.
132: The results provide constraints on sulfur chemistry in the
133: inner wind of the IRC+10216 prototype carbon-rich circumstellar envelope.
134:
135: \section{Observations}
136:
137: We observed IRC+10216 on 10 December 2002 with the
138: SMA when three of the 6~meter diameter antennas were equipped
139: with receivers for the ``600--700~GHz'' frequency band.
140: The weather was very good, with 225~GHz atmospheric opacity of
141: 0.035 measured at the nearby Caltech Submillimeter Observatory
142: through the night, or $\sim0.7$ at the SMA observing frequency.
143: Table 1 summarizes the observational parameters.
144: We adopt the CS J=14--13 line rest frequency of 685.435923 GHz,
145: recently calculated from newly determined
146: spectroscopic constants (Gottlieb, Myers \& Thaddeus 2003).
147: The SMA observations provided three independent baselines ranging
148: from 14 to 25 meters in length, resulting in $\sim2''$ resolution.
149:
150: A major difficulty for interferometric imaging at these high
151: frequencies is the lack of sufficiently strong sources with
152: known structure for the derivation of complex gains to track
153: instrumental phase drifts and atmospheric fluctuations. In particular,
154: nearly all of the point-like quasars
155: are too weak for the SMA to detect at shorter wavelengths
156: in a few minutes of integration time.
157: Compact thermal sources such as planetary
158: moons or asteroids provide one alternative. For these observations,
159: we took advantage of the close proximity of Jupiter to IRC+10216
160: in the sky ($<7$~degrees).
161: While Jupiter was far too large to be a suitable calibrator
162: (diameter $40''$),
163: the Galilean moons were only slightly resolved on these baselines,
164: and high signal-to-noise observations of Callisto (diameter $1\farcs4$)
165: could be efficiently interleaved with those of IRC+10216.
166: Jupiter was located well outside the small SMA field of view
167: during the observations of Callisto, and the data show no evidence
168: for any contamination from the planetary emission.
169: IRC+10216 was tracked over the hour angle range $-4$h to $+4$h.
170: The predicted flux
171: of Callisto varied by less than $10$\% due to resolution effects.
172:
173: The partially complete digital correlator was configured with four overlapping
174: ``chunks'', each of 104 MHz bandwidth and 128 channels, with the
175: CS J=14--13 line centered in one of them. A calibration of the bandpass
176: shape was obtained from observations of Mars. The system temperatures
177: ranged from 1200 to 2200 K (DSB).
178: %, with the higher value at lowest elevation.
179: The absolute flux scale was set assuming a full disk flux of 55~Jy
180: for Callisto, based on a brightness temperature of 120 K determined
181: by comparison with Titan, whose submillimeter spectrum is better known
182: (Mark Gurwell, private communication). The overall flux scale is unlikely
183: to be more accurate than 30\%.
184: %All calibration was performed with MIR, and suite of routines in IDL
185: %based on the OVRO MMA software (Scoville et al. 1993), and subsequent
186: %imaging was done with MIRIAD (Sault, Teuben \& Wright 1995).
187: Images were made with a variety of weighting schemes that provided
188: slightly different angular resolutions, sensitivities, and
189: synthesized beam sidelobe patterns.
190: %All images were cleaned in the
191: %to a cutoff equal to twice the r.m.s. noise level.
192:
193: \section{Results and Discussion}
194:
195: \subsection{Spectrum at Image Center}
196: Figure~\ref{fig:cs_spectrum} shows the spectrum from the center of
197: the image cube for the full USB correlator band, averaging groups of
198: 8 channels, corresponding to a velocity width of 2.85 km~s$^{-1}$.
199: The CS J=14--13 line is the prominent feature centered at $-26$~km~s$^{-1}$
200: and an approximate width at zero power of $\sim30$~km~s$^{-1}$,
201: values which are consistent with the systemic velocity of the source
202: and the expansion velocity of the envelope, respectively.
203: The line shape is difficult to interpret because spatial filtering
204: by the interferometer results in missing flux that varies with
205: velocity across the profile (see \S\ref{discussion:line}).
206: The peak brightness temperature of the line at the
207: full angular resolution of the data is well over 100~K.
208: An additional feature appears in the spectrum near $-100$~km~s$^{-1}$
209: with a double peaked line shape and an approximate width consistent with
210: the circumstellar expansion.
211: A search of the JPL spectral line catalog suggests a tentative identification
212: of this feature with the C$_3$H$_2$ $8(4,4)-7(3,5)$ line at 685.6125565~GHz.
213: This species was previously detected in transitions at lower frequencies
214: (Kawaguchi et al. 1995, Cernicharo et al. 2000).
215: No alternative candidates line were found in the catalog of
216: Cernicharo et al. (2000) that contains
217: 1050 molecular species (Cernicharo, private communication).
218: If the assignment is correct, then the line might be thought
219: to emerge from the external shell where various
220: carbon chain radicals are found (Guelin et al. 1993).
221: This feature appears to be spatially compact, though,
222: which perhaps renders the identification problematic.
223: %The irregular
224: %double peaked line profile might be the result of optically thin emission
225: %from the expanding envelope.
226:
227: \subsection{Continuum Emission}
228: Figure~\ref{fig:cont} shows the continuum image obtained from the
229: channels free of strong line emission. Visibilities from both sidebands,
230: separated by 10~GHz, were combined in a multifrequency synthesis, resulting
231: in an effective frequency of 680 GHz. The continuum peak position is
232: consistent with previous determinations, in particular the 95~GHz observation
233: from the IRAM PdBI (Guelin et al. 1993), which is marked by
234: the cross. The peak continuum flux is 3.8~Jy, with an uncertainty
235: dominated by systematic effects.
236:
237: Previous bolometer observations of IRC+10216 at 450~$\mu$m indicate
238: that the SMA recovers approximately 20\% of the broadband flux at
239: this wavelength. Sandell (1994) noted that IRC+10216 has long period
240: variations in the submillimeter and found $19\pm3$~Jy in an $18''$
241: aperture near the 635 day maximum. Jenness et al. (2002) analyzed
242: several years of SCUBA observations of IRC+10216 at 850~$\mu$m and
243: 450~$\mu$m and found a period and maximum flux consistent with the
244: earlier single pixel measurements. The SMA observations were made
245: approximately a month before the predicted maximum, when $\sim95\%$ of
246: the peak flux would be expected. Remarkably, the CS J=14--13 line emission
247: {\em alone} accounts for $1\%$ of the flux in the $\sim68$~GHz wide
248: filter used for the 450~$\mu$m bolometer measurements. Given the
249: many strong spectral lines excited in the inner envelope at these
250: high frequencies, it is clear that spectral line emission
251: makes an important contribution to the broadband ``continuum'' flux.
252:
253: A large fraction of the continuum emission from the compact component
254: detected by the SMA at 680~GHz likely comes from the stellar photosphere.
255: Lucas \& Guelin (1999) report a ``point source'' component in IRAM PdBI
256: observations of $65\pm1$~mJy at 89~GHz
257: and $486\pm7$~mJy at 242~GHz, identified as photospheric emission.
258: Assuming an optically thick spectrum, $S_\nu\propto\nu^2$,
259: an extrapolation to 680~GHz gives 3.8~Jy, consistent with the measured
260: value. Thermal emission from dust in the inner envelope cannot contribute
261: substantial additional flux in the small synthesized beam of the SMA at
262: this high frequency.
263:
264: \subsection{CS J=14--13 Emission}
265: \label{discussion:line}
266: Figure~\ref{fig:cs_channels} shows images for a series of velocity intervals
267: that span the CS J=14--13 line. Below each of the images is a plot of
268: visibility amplitude vs. $(u,v)$ distance for the corresponding velocity
269: interval to give an idea of spatial extent, which is not easy
270: to ascertain from the images.
271: The overall structure in the visibility amplitudes is consistent
272: with simple expectations for a spherically expanding envelope.
273: At the extreme velocities, the amplitude is approximately constant,
274: as expected for
275: the unresolved ``caps'' of blueshifted and redshifted emission, while at
276: intermediate velocities, the fall-off of visibility amplitude with
277: $(u,v)$ distance demonstrates that the emission is spatially extended and
278: resolved. If the gas is expanding radially with approximately spherical
279: symmetry, then the narrow velocity interval around the central velocity
280: corresponds to a cross section through the envelope in the plane of the sky.
281:
282: The central peak of CS J=14--13 emission distribution provides additional
283: direct evidence that CS is ``parent'' molecule, in accord with the
284: detection of ro-vibrational transitions in absorption in the infrared
285: (Keady \& Ridgeway 1993), the detection of vibrationally excited CS emission
286: (Turner 1987, Lucas \& Guelin 1999, Highberger et al. 2000),
287: and interferometric imaging of CS J=2--1 emission (Lucas et al. 1995).
288: The CS molecules must originate close to the stellar photosphere, and
289: they are lost from the expanding envelope through reactions that
290: build more complex species, perhaps aided by shocks as suggested by
291: Willacy \& Cherchneff (1998).
292: The J=14-13 emission distribution shows no indication of the
293: extended $\sim15''$ radius ring visible in the J=2-1 line,
294: but emission on that size scale, if present, could not be detected
295: on account of the small SMA field of view.
296:
297: The reaction network for sulfur bearing carbon chains has been explored
298: by Millar, Flores \& Markwick (2001), who concluded that an initial
299: CS abundance of $4\times10^{-6}$ is needed to match observations of
300: C$_3$S and C$_5$S. Models where CS is not a parent species, in which the
301: CS fractional abundance rises from $<10^{-11}$ at radius $10^{16}$~cm,
302: or $4\farcs$5 at 150~pc, to a peak more than four orders of magnitude larger
303: at radius $10^{17}$~cm, could not match observations of the longer carbon
304: chains. The SMA observations provide a robust {\em lower limit} to the
305: CS abundance within a radius of $\sim34$~R$_*$ ($2.25\times10^{15}$~cm)
306: that supports the parent species scenario.
307: Under the assumption of a thermalized level population, the
308: total CS column density is given by
309: \begin{equation}
310: N(CS) = \frac{3 k f}{8 \pi^3 B \mu^2}
311: \frac{\exp{(hBJ_l(J_l+1)/kT)}}{J_l+1}
312: \frac{T + hB/3k}{1- \exp{(-h\nu/kT)}}
313: \int \tau dv,
314: \end{equation}
315: where the molecular constants are $\mu=1.96$ Debye, $B=24.584$ GHz, and
316: the filling factor $f$ may be estimated as (e.g. Scoville et al. 1986)
317: \begin{equation}
318: f = \frac{T_R^*/\eta_c}{(h\nu/k)/(\exp{(h\nu/kT)-1})(1 - \exp{-\tau}) }.
319: \end{equation}
320: The lower limit to the line flux in a $\sim2''$ beam is
321: $\sim 5000$~Jy~km~s$^{-1}$, or $\sim 1250$~K~km~s$^{-1}$
322: in brightness units.
323: Assuming small optical depth, $\tau<<1$, a kinetic temperature equal to the
324: excitation of the upper level of the observed transition (246.8~K), which
325: minimizes the column density required to explain the observed line flux,
326: and a spherical volume with molecular hydrogen density equal to the
327: critical density of the transition of $3.2\times10^8$~cm$^{-3}$, results in
328: a lower limit to the CS fractional abundance of $\sim3.4\times10^{-9}$.
329: The abundance close to the photosphere is likely substantially higher
330: than this estimate; Highberger et al. (2000) derive $3-7\times10^{-5}$
331: at a radius $\sim14$~R$_*$ ($9\times10^{14}$~cm) from their multi-transition
332: single dish observations. Though the uncertainties are large, models
333: such as those investigated by Millar et al. (2001) where the CS abundance
334: rises from $<10^{-11}$ at radius $10^{16}$~cm are clearly in conflict with
335: the high resolution images of the CS J=14--13 emission.
336:
337: The SMA observations show directly that the CS molecules are formed close
338: to the stellar photosphere. Detailed radiative transfer models that
339: account for the observed spatial distribution are needed for an accurate
340: determination of the CS abundance in the inner envelope. The effect of
341: pulsational shocks could be important in establishing the CS abundance as
342: this species is injected into the wind, and
343: careful modeling may provide insight and constraints on the shock chemistry.
344:
345: %\section{Conclusions}
346: Imaging the CS J=14--13 line in the inner envelope of IRC+10216 offers
347: a first demonstration of the efficacy of the SMA
348: %for making observations
349: in the semi-transparent 450~$\mu$m atmospheric window.
350: %The line emission is clearly resolved and shows a velocity and spatial
351: %structure consistent
352: %with an outflow that originates near the stellar photosphere.
353: Calibration at these high frequencies should become considerably easier
354: as more receivers for these frequencies are deployed at the SMA,
355: the correlator bandwidth is increased to 2~GHz, and simultaneous
356: operation of a lower frequency band enables phase transfer.
357: Many more evolved stars, as well as other objects detected but
358: with presently unresolved emission at short submillimeter wavelengths,
359: are now accessible to imaging at the arcsecond scale.
360:
361: \acknowledgements
362: We thank all of members of the SMA Team that made these observations
363: possible.
364: We thank John Bieging and the referee Jose Cernicharo for very helpful
365: comments.
366: We extend special thanks to those of Hawaiian ancestry on whose sacred
367: mountain we are privileged to be guests.
368:
369: \clearpage
370: \begin{references}
371: \reference{}Avery, L.W. et al. 1992, \apjs, 83, 363
372:
373: \reference{}Bieging, J.H., Tafalla, M. 1993, \aj, 105, 576
374:
375: \reference{}Carlstrom, J.E., Hills, R.E., Lay, O.P., Force, B., Hall, C.G.,
376: Phillips, T.G., Schinckel, A.E. 1994,
377: in {\em Astronomy with Millimeter and Submillimeter Wave Interferometry},
378: IAU Colloquium 140, eds. M. Ishiguro and J. Welch,
379: ASP Conference Series, Vol. 59, p.35
380:
381: \reference{}Cernicharo, J., Guelin, M., Kahane, C. 2000, \aaps, 142, 181
382:
383: \reference{}Crosas, M., Menten, K.M. 1997, \apj, 483, 913
384:
385: \reference{}Gottlieb, C.A., Myers, P.C., Thaddeus, P. 2003, \apj, 588, 655
386:
387: \reference{}Fong, D., Meixner, M., Shah, R.Y. 2003, \apj, 582, L39
388:
389: \reference{}Groesbeck T.D., Phillips T.G., Blake, Geoffrey A., 1994,
390: \apjs, 94, 187
391:
392: \reference{}Guelin, M., Lucas, R., Cernicharo, J. 1993, \aap, 280, L19
393:
394: \reference{}Highberger, J.L., Apponi, A.J., Bieging, J.H., Ziurys, L.M.,
395: Mangum, J.G. 2000, \apj, 544, 881
396:
397: \reference{}Ho, P.T.P, Moran, J.M., Lo, K.Y. 2004, \apj, submitted
398:
399: \reference{}Kawaguchi, K., Kasai, Y., Ishikawa, S.-I., Kaifu, N. 1995,
400: \pasj, 47, 853
401:
402: \reference{}Keady, J.J. \& Ridgeway, S.T. 1993, \apj, 406, 199
403:
404: \reference{}Lucas, R., Guelin, M. 1999, IAU Symp. 191, Asymptotic Giant
405: Branch Stars, eds. T LeBetre, A Lebre, C Waelkens (San Francisco: ASP),
406: p. 305
407:
408: \reference{}Lucas, R., Guelin, M., Kahane, C., Audinos, P, Cernicharo, J.
409: 1995, Ap\&SS, 224, 293
410:
411: \reference{}Mauron, N. \& Huggins, P.J. 1999, \astap, 349, 203
412:
413: \reference{}Millar, T.J., Flores, J.R., Markwick, A.J. 2001, \mnras, 327,
414: 1173
415:
416: %\reference{}Sault, R.J., Teuben, P.J., Wright, M.C.H. 1995,
417: % in {\em Astronomical Data Analysis Software and Systems IV},
418: % eds. R.A. Shaw, H.E. Payne, and J.J.E. Hayes,
419: % ASP Conference Series, Vol. 77, p. 433
420:
421: \reference{}Scoville, N.Z., Sargent, A.I., Sanders, D.B., Claussen, M.J.,
422: Masson, C.R., Lo, K.Y., Phillips, T. G. 1986, \apj, 303, 416
423:
424: %\reference{}Scoville, N.Z., Carlstrom, J.E., Chandler, C.J., Phillips, J.A.,
425: % Scott, S.L., Tilanus, R.P.J., Wang, Z. 1993, \pasp, 105, 1482
426:
427: \reference{}Turner, B.E. 1987, \aap, 182, L15
428:
429: \reference{}Willacy, K., Cherchneff, I. 1998, \aap, 330, 676
430:
431: \end{references}
432:
433: %\clearpage
434:
435: \begin{deluxetable}{lc}
436: \tablecolumns{3}
437: \tablewidth{0pt}
438: \tablecaption{IRC+10216 Observational Parameters}
439: \tablehead{
440: \colhead{Parameter} & \colhead{Value} }
441: \startdata
442: Observations: & 2002 Dec 10 (3 antennas)\\
443: Min/Max baseline: & 14 to 25~meters \\
444: Pointing center (J2000): &
445: $\alpha=9^{h}47^{m}57\fs39$, $\delta=13^{h}16^{m}43\fs90$ \\
446: Calibrator (flux): & Callisto (56 Jy) \\
447: Primary beam HPBW: & $17''$ \\
448: Synthesized beam HPBW:
449: & $2\farcs3\times1\farcs5$ P.A. $28^{\circ} $ \\
450: K/Jy: & 0.76 \\
451: %r.m.s. (continuum image): & 0.3 mJy/beam \\
452: Spectral Line Correlator: & $4\times128$ channels \\
453: ~~~usable bandwidth: & 328 MHz \\
454: ~~~species/transition: & CS J=14--13 \\
455: ~~~frequency: & 685.435923 GHz\tablenotemark{a} \\
456: ~~~channel spacing: & 0.36 km~s$^{-1}$ \\
457: r.m.s. (line images): & 8 Jy~beam$^{-1}$ \\
458: \enddata
459: \tablenotetext{a}{Gottlieb, Myers \& Thaddeus 2003}
460:
461: \label{tab:obs}
462: \end{deluxetable}
463:
464: \clearpage
465:
466: \begin{figure}
467: \epsscale{0.53}
468: \rotatebox{-90}{
469: \plotone{f1.eps}}
470: \caption{
471: Spectrum of IRC+10216 across the full bandwidth,
472: showing the four overlapping 104 MHz correlator ``chunks'',
473: binned to 2.8 km~s$^{-1}$ resolution,
474: in a $4\farcs6\times1\farcs6$ p.a. $-68^{\circ}$ synthesized beam.
475: Strong CS J=14--13 line emission is visible, as well as an interloping
476: line at low velocities, tentatively identified as C$_3$H$_2$.
477: }
478: \label{fig:cs_spectrum}
479: \end{figure}
480:
481: \clearpage
482:
483: \begin{figure}
484: \epsscale{0.9}
485: \rotatebox{-90}{
486: \plotone{f2.eps}}
487: \caption{
488: IRC+10216 continuum emission at 685 GHz.
489: The contour levels are $\pm2,4,6,...\times0.3$~Jy~beam$^{-1}$.
490: Negative contours are dotted.
491: The ellipse in the lower left corner shows the
492: $2\farcs3\times1\farcs5$ p.a. $28^{\circ}$ synthesized beam.
493: The dashed circle indicates the $17''$ FWHM primary beam size.
494: }
495: \label{fig:cont}
496: \end{figure}
497:
498: \clearpage
499:
500: \begin{figure}
501: \epsscale{0.66}
502: \rotatebox{-90}{
503: \plotone{f3.eps}}
504: \caption{
505: {\em upper:} Velocity channel maps (width 5.7 km~s$^{-1}$)
506: of CS J=14--13 line emission from IRC+10216.
507: The contour levels are $\pm1,3,6,9,...\times15$~Jy~beam$^{-1}$.
508: Negative contours are dotted.
509: The ellipse in the lower left corner of the leftmost panel shows
510: the $2\farcs3\times1\farcs5$ p.a. $28^{\circ}$ synthesized beam.
511: The dashed circle indicates the $17''$ FWHM primary beam size.
512: {\em lower:}
513: Visibility amplitude of the CS J=14--13 emission vs. baseline length,
514: for the same velocity channels, annularly averaged in 10~k$\lambda$ bins.
515: The error bars represent $\pm1$ standard deviation for each bin.
516: The falloff of amplitude with baseline length shows that the emission
517: is spatially resolved at the central velocities, as expected for a
518: symmetrically expanding source.
519: }
520: \label{fig:cs_channels}
521: \end{figure}
522:
523: \end{document}
524: