1: %% The command below calls the default manuscript style,
2: %% which will produce a double-spaced document on one column.
3: %% Examples of commands for other substyles follow. Use
4: %% whichever is most appropriate for your purposes.
5: %\documentclass{aastex}
6: %% preprint produces a one-column, single-spaced document:
7: \documentclass[12pt, preprint]{aastex}
8: %% preprint2 produces a double-column, single-spaced document:
9: % \documentclass[preprint2]{aastex}
10: %% for information.
11: \newcommand{\hho}{H$_2$O$\,$}
12: \newcommand{\etal}{{\it et al.}}
13: \newcommand{\kms}{km s$^{-1}\,$}
14: \newcommand{\vlsr}{V$_{\rm LSR}$}
15: \newcommand{\msun}{\mbox{M$_{\sun}$}}
16: \newcommand{\lsun}{\mbox{L$_{\sun}$}}
17: \newcommand{\bm}{beam$^{-1}$}
18:
19: %% You can insert a short comment on the title page using the command below.
20:
21: %\slugcomment{Not to appear in Nonlearned J., 45.}
22:
23: %% If you wish, you may supply running head information, although
24: %% this information may be modified by the editorial offices.
25: %% The left head contains a list of authors,
26: %% usually a maximum of three (otherwise use et al.). The right
27: %% head is a modified title of up to roughly 44 characters. Running heads
28: %% will not print in the manuscript style.
29:
30:
31: \shorttitle{\hho megamasers in NGC\,5793}
32: \shortauthors{Hagiwara et al.}
33: %% This is the end of the preamble. Indicate the beginning of the
34: %% paper itself with \begin{document}.
35:
36: \begin{document}
37:
38: %% LaTeX will automatically break titles if they run longer than
39: %% one line. However, you may use \\ to force a line break if
40: %% you desire.
41:
42: \title{VLBI STUDY OF WATER MASER EMISSION IN THE SEYFERT 2 GALAXY NGC\,5793. I: IMAGING BLUESHIFTED EMISSION AND THE PARSEC-SCALE JET}
43:
44: %% Use \author, \affil, and the \and command to format
45: %% author and affiliation information.
46: %% Note that \email has replaced the old \authoremail command
47: %% from AASTeX v4.0. You can use \email to mark an email address
48: %% anywhere in the paper, not just in the front matter.
49: %% As in the title, you can use \\ to force line breaks.
50:
51: \author{\sc Yoshiaki Hagiwara}
52: \affil{Max-Planck-Institut f{\"u}r Radioastronomie, Auf dem H{\"u}gel 69, D-53121, Bonn, Germany}
53: \email{hagi@mpifr-bonn.mpg.de}
54: \author{\sc Philip J. Diamond}
55: \affil{Jodrell Bank Observatory, University of Manchester, Macclesfield Cheshire SK 11 9DL, UK}
56:
57: \and
58:
59: \author{\sc Naomasa Nakai and Ryohei Kawabe}
60: \affil{Nobeyama Radio Observatory, Minamimaki, Minamisaku, Nagano, 384-1305, Japan}
61:
62: %% Notice that each of these authors has alternate affiliations, which
63: %% are identified by the \altaffilmark after each name. Specify alternate
64: %% affiliation information with \altaffiltext, with one command per each
65: %% affiliation.
66:
67:
68:
69: %% Mark off your abstract in the ``abstract'' environment. In the manuscript
70: %% style, abstract will output a Received/Accepted line after the
71: %% title and affiliation information. No date will appear since the author
72: %% does not have this information. The dates will be filled in by the
73: %% editorial office after submission.
74: %
75: \begin{abstract}
76: We present the first result of VLBI observations of the blueshifted
77: water maser emission from the type 2 Seyfert galaxy NGC\,5793, which
78: we combine with new and previous VLBI observations of continuum
79: emission at 1.7, 5.0, 8.4, 15, and 22 GHz. Maser emission was detected
80: earlier in single-dish observations and found to have both red- and
81: blueshifted features relative to the systemic velocity. We could image
82: only the blueshifted emission, which is located 3.6 pc southwest of
83: the 22 GHz continuum peak. The blueshifted emission was found to
84: originate in two clusters that are separated by 0.7 milliarcsecond
85: (0.16 pc). No compact continuum emission was found within 3.6~pc of the
86: maser spot. A compact continuum source showing a marginally inverted
87: spectrum between 1.7 and 5.0 GHz was found 4.2 pc southwest of the
88: maser position. The spectral turnover might be due to synchrotron
89: self-absorption caused by a shock in the jet owing to collision with
90: dense gas, or it might be due to free-free absorption in an ionized
91: screen possibly the inner part of a disk, foreground to the jet.
92:
93: The water maser may be part of a maser disk. If so, it would be
94: rotating in the opposite sense to the highly inclined galactic disk
95: observed in CO emission. We estimate a binding mass within 1 pc of the
96: presumed nucleus to be on the order of 10$^7$ $M_{\sun}$.
97: Alternatively, the maser emission could result from the
98: amplification of a radio jet by foreground circumnuclear molecular
99: gas. In this case, the high blueshift of the maser emission might mean
100: that the masing region is moving outward away from the molecular gas
101: surrounding an active nucleus.
102: \end{abstract}
103: %
104: %% Keywords should appear after the \end{abstract} command. The uncommented
105: %% example has been keyed in ApJ style. See the instructions to authors
106: %% for the journal to which you are submitting your paper to determine
107: %% what keyword punctuation is appropriate.
108:
109: \keywords{water masers: extragalactic, jet---AGN: NGC\,5793}
110:
111: %% From the front matter, we move on to the body of the paper.
112: %% In the first two sections, notice the use of the natbib \citep
113: %% and \citet commands to identify citations. The citations are
114: %% tied to the reference list via symbolic KEYs. The KEY corresponds
115: %% to the KEY in the \bibitem in the reference list below. We have
116: %% chosen the first three characters of the first author's name plus
117: %% the last two numeral of the year of publication as our KEY for
118: %% each reference.
119:
120:
121: \section{INTRODUCTION}
122: %
123: %
124: %
125: VLBI observations of water megamasers on (sub-)pc scales in active
126: galaxies have proven to be a basic tool for investigating the central
127: regions of active galactic nuclei (AGN). The most impressive result
128: from this kind of research was the discovery of a thin, nearly edge-on
129: rotating disk around a super massive black hole at the center of
130: NGC\,4258 \cite{miyo95, linc95}. The VLBI observations revealed the
131: size and the shape of the slightly-warped maser disk with a central
132: radio jet outflowing perpendicularly to the disk plane
133: \cite{herr98}. VLBI imaging of the water megamasers in NGC\,1068,
134: NGC\,4945 and the Circinus galaxy showed the possible existence of
135: super massive objects at their nuclei similar to NGC\,4258
136: \cite{linc96, linc96b, linc97, linc01}. According to the most recent
137: result, in the Seyfert 2 galaxy IC\,2560, the barely resolved maser
138: spots which constitute the systemic emission show a linear velocity
139: gradient along the north-south elongation and are accompanied by
140: continuum emission, probably suggesting a super massive black hole
141: inside a rotating disk \cite{naka98, ishi00}.
142: %
143: %
144: \indent In the Seyfert 2 galaxy NGC\,1068, in addition to the nuclear
145: disk-maser, non-disk maser emission was found some 30 pc away along
146: the jet from the radio nucleus, that might be the result of
147: interaction between the jet and the interstellar gas \cite{gall96}. In
148: the elliptical galaxy NGC\,1052, the interpretation of non-disk masers
149: was introduced to explain the water masers seen lying along the jet
150: \cite{clau98}.
151: %
152: %
153: \indent Water megamasers seem to fall into two categories: nuclear
154: masers residing in a (sub-)parsec scale disk around an active nucleus
155: and non-nuclear masers which are seen significantly further from the
156: nucleus. The latter can also be classified into two types, that is,
157: jet masers associated with radio jets and outflow masers -- a wind
158: component discovered in Circinus \cite{linc01}. The study of non-disk
159: maser emission is equally important for the investigation of the
160: overall structure surrounding an active nucleus. In order to
161: understand the megamaser phenomenon, and to deduce the general
162: characteristics of the central parsecs of AGN, such as black hole
163: masses and accretion rates, it is important to investigate as many
164: megamasers as possible with VLBI imaging at milliarcsecond angular
165: resolutions. So far, more than 20 galaxies are reported to contain
166: megamasers (e.g., Braatz et al. 1997, Falcke et al. 2000). Most of
167: them, however, show weak intensities, and some are thus not observable
168: with VLBI \cite{mora99}. \\
169: %
170: %
171: \indent NGC\,5793 is an edge-on disk galaxy, at a distance of 46 Mpc,
172: with a compact nucleus seen in radio continuum emission
173: \cite{gard92}. Baan et al. (1998) classified the nucleus as being a
174: Seyfert 2 from its optical emission. Hagiwara et al. (1997), using
175: the Nobeyama 45-m telescope, first detected the water maser emission
176: with highly Doppler-shifted satellite features, displaced by 245 \kms
177: on either side of the systemic velocity ($V_{\rm sys, \, LSR}$ = 3442
178: \kms; Palumbo et al. 1983) of the galaxy. The total isotropic
179: luminosity of the maser was estimated to be 106 \lsun . Fig.~1 shows
180: a new spectrum of blueshifted maser features measured with the Parkes
181: 64-m telescope. There may be a suggestion of features near the
182: systemic velocity, which needs confirming. The highly redshifted
183: features at \vlsr = 3677 \kms, which were initially detected with the
184: 45-m remain undetected since February 1996. From 1996 we began VLBI
185: observations using the NRAO\footnote{~The National Radio Astronomy
186: Observatory is a facility of the National Science Foundation operated
187: under cooperative agreement by Associated Universities, Inc.} VLBA and
188: VLA, investigating the spatial distribution of the water maser
189: emission and the relationship between the water megamasers and the
190: nuclear radio activity of NGC\,5793. \\
191: %
192: %
193: \indent In this paper we describe the VLBA observations in Section 2,
194: present new images of the water vapour maser line and continuum
195: emission in the central parsecs of NGC\,5793 in Section 3, and discuss
196: the nature of the water maser in Section 4. We adopt a distance of 46
197: Mpc to NGC\,5793, corresponding to a scale of 1 milliarcsecond~(mas)
198: $\approx$ 0.23 pc.
199: %
200: %
201: %
202: \section{VLBA OBSERVATIONS AND DATA ANALYSIS}
203: %
204: \subsection{\it 22 GHz Observations}
205: %
206: We observed the 6$_{16} - 5_{23}$ maser line at 22 GHz (rest
207: frequency: 22.23508 GHz) toward NGC\,5793 for 8 hours on 1998 May 10
208: using the VLBA and the phased VLA. The use of the VLA as the eleventh
209: array element was critical to the observation because the peak flux
210: density of the maser emission is known to be very weak ($\sim$ 0.1
211: Jy). The data were recorded in left circular polarization with 2~bit
212: sampling in four 8 MHz IF bands of 128 spectral channels each
213: providing a velocity coverage of 108 \kms. The four IF bands were
214: centered on \vlsr = 3190, 3442, 3551, and 3667 \kms. The observed
215: velocity ranges are displayed in a single-dish spectrum in
216: Fig.~\ref{fig1}. The channel spacing in each band was 0.83 \kms
217: before averaging. We observed in a phase-referencing mode, using a
218: bright calibration source 1507$-$168 about 2 degrees away from
219: NGC\,5793. The total cycle time was two minutes; 1-minute scans on
220: 1507$-$168 and NGC\,5793 alternated. The total amount of time spent
221: integrating on the target source was about 150 minutes. 4C\,39.25 and
222: 3C\,345 were used for both bandpass and amplitude calibration. All the
223: data were correlated in NRAO Socorro and calibrated and reduced with
224: AIPS. The fringe-fit solutions (phase, rate, and delay) obtained from
225: the calibrator were applied to the IFs containing both the line plus
226: continuum emission. Five spectral channels were averaged together to
227: 4.2 \kms to obtain reasonable sensitivities. The positions of the
228: maser emission on the sky were determined by 2-D Gaussian fitting to
229: the maps. To make a continuum map, we averaged all spectral channels
230: in a line-emission-free IF, centered on 3667 \kms. \\
231: %
232: %
233: \indent The continuum and line maps were both made from differently
234: weighted data to get reasonable sensitivities, resulting in a
235: synthesized beam size of 2.8 $\times$ 1.8 mas for the continuum map,
236: and 1.8 $\times$ 1.0 mas for the line map. The rms noise levels are
237: 1.3 mJy \bm\ for the continuum map produced from a single bandwidth of
238: 8~MHz and 4.3 mJy \bm\ for the continuum-subtracted line map for each
239: averaged channel. The continuum emission was subtracted only in the
240: region close to C1(C).
241: %
242: %
243: \subsection{\it Observations at 1.7, 8.4, and 15 GHz}
244: %
245: The 1.7, 8.4, and 15~GHz continuum observations were made on 1999
246: October 10, using 8 elements of the VLBA. The VLBA stations
247: Brewster~(BR) and Saint Croix~(SC) did not take part in the
248: observations. The four 8 MHz IF bands with left circular polarization
249: were recorded using 2 bit sampling across a total bandwidth of 32
250: MHz. We used the frequency agility of the VLBA to switch rapidly among
251: the three frequencies to obtain better ($u$, $v$) coverages with
252: optimal observing time. The observing frequencies were switched
253: typically every 7 --10 minutes between 8.4 and 15~GHz. The 1.7 GHz
254: observation was inserted among the other two frequencies about every
255: 40 minutes. The 8.4 and 15~GHz observations were made using the
256: phase-referencing method, with the same calibrator as at 22~GHz. At
257: 8.4 GHz a total cycle time of about 4 minutes was used for the
258: calibrator and the galaxy; 3-minute scans on the calibrator and
259: 1-minute scans for the target source were employed. At 15~GHz,
260: 1-minute scans on the calibrator and the target source were
261: alternated. The total integration time spent on the target source at
262: each frequency ranged from 60 to 120 minutes. The correlation of all
263: the data was performed at Socorro. The data were calibrated using
264: standard AIPS software. The amplitude calibration was made with
265: observations of 4C\,39.25 and 3C\,345. The centroid position of the
266: 8.4 GHz continuum source was determined from the phase-referenced
267: map. Finally, self-calibration was used on the data at 1.7 and 8.4 GHz
268: to improve the sensitivity of images with DIFMAP. \\
269: %
270: %
271: \indent The maps were made from uniformly weighted data and the
272: resultant synthesized beam sizes are 23 $\times$ 3.7 mas at 1.7 GHz,
273: 7.0 $\times$ 2.7 mas at 8.4 GHz, and 2.9 $\times$ 0.6 mas at 15~GHz.
274: The rms noise levels are 0.86 mJy \bm\ for the 1.7 GHz map, 0.35 mJy
275: \bm\ for the 8.4 GHz map, and 0.82 mJy \bm\ for the 15~GHz map at
276: total bandwidths of 32~MHz.
277: %
278: \subsection{\it Analysis of Positional Errors}
279: %
280: The positional errors ($\Delta$\,$\theta$) in the VLBI images, dominated by statistical noise, were estimated from the synthesized beam size ($\theta _{b}$) divided by the signal-to-noise ratios (SNR). When we estimate relative positional errors between line and continuum maps, they are approximately given by the the equation
281: \begin{displaymath}
282: %
283: \Delta \,\theta_{mc} \simeq \sqrt{ \left( \frac{\theta _{bm}}{2SNR_{m}} \right)^{2} + \left( \frac{\theta _{bc}}{2SNR_{c}} \right)^{2}} ,
284: %
285: %
286: \end{displaymath}
287: where $m$ and $c$ represent maser and continuum emission images. Thus, the relative positional error between one of the maser lines at the \vlsr = 3194 \kms feature and the 22 GHz continuum peak C1(C) is $\theta _{mc}$ $\sim$ 0.22 mas.\\
288: \indent
289: The common reference positions (0, 0) (continuum peak C1(C)) were determined at various frequencies by phase-referencing to the compact source, 1507$-$168. This same calibrator was used for all the observing epochs and there seemed to be neither structural nor positional changes between the two epochs, May 1998 and October 1999. The alignment of the reference positions between the 22 and 8.4 GHz continuum maps was determined from the phase-referenced maps. Assuming that the position error of the common reference source and any structural changes of the reference source between epochs are negligible, the relative positional error between 22 GHz ($k$) and 8.4 GHz ($x$) is estimated as the following,
290: %
291: \begin{displaymath}
292: %
293: \Delta \,\theta_{mcx} \simeq \sqrt{ \left( \frac{\theta _{bm}}{2SNR_{m}} \right)^{2} + \left( \frac{\theta _{bcx}}{2SNR_{cx}} \right)^{2} + \left( \frac{\theta _{brk}}{2SNR_{rk}} \right)^{2} + \left( \frac{\theta _{brx}}{2SNR_{rx}} \right)^{2} } ,
294: %
295: %
296: \end{displaymath}
297: %
298: where $c$ and $r$ represent continuum images of a target source and a phase-reference source. In our analysis, the third and fourth term in the formula are negligible as SNR of reference calibrator images are better by on the order of 10 -- 100~than those of masers or 22 and 8.4 GHz continuum images of NGC\,5793. Consequently, the relative uncertainty of the maser position with respect to the reference point at 8.4 GHz in Fig.~\ref{fig3} is $\sqrt{ \left(\theta _{bm} / 2SNR_{m} \right)^{2} + \left(\theta _{bcx} / 2SNR_{cx} \right)^{2} }$ = $\sqrt{ \left( 2/ (2 \cdot 11) \right)^{2} + \left( 7/ (2 \cdot 22) \right)^{2} }$ $\sim$ 0.2 mas. (The inserted values were adopted from the phase-referenced maps without self-calibration.)
299: %
300: %
301: %
302: \section{RESULTS}
303: %
304: \subsection{\it Continuum Emission}
305: %
306: %
307: Fig.~2 and Fig.~3 show radio continuum images at 22 and 8.4 GHz, from
308: which we can identify at least four continuum components C1(C), C2(E),
309: C2(C) and C2(W). The naming of the components follows the convention
310: used in Hagiwara et al. (2000). The 22 GHz peak is coincident with
311: C1(C) in the 8.4 GHz map to an accuracy of 0.25 mas, as estimated in
312: the previous section. The basic morphological structure is quite
313: similar to those at 1.7 and 5.0 GHz given in Fig.~2 and Fig.~3 of the
314: paper by Hagiwara et al. (2000), but at both 15 and 22 GHz only C1(C)
315: is detected. Table~\ref{tbl-1} lists the Gaussian-fitted peak flux
316: densities of the major compact continuum sources identified at the
317: five frequencies, and Table~\ref{tbl-2} lists the spectral indices of
318: each component. The flux density and spectral indices were obtained
319: by convolving with the same beam size corresponding to that of the 1.7
320: GHz map in 1996 to avoid resolution effects. The poor ($u$, $v$)
321: coverage for the 1.7 GHz observing session in Oct. 1999 resulted in
322: relatively inaccurate component flux densities in comparison with
323: those in Nov. 1996. Fig.~\ref{fig4} shows spectra from four major
324: components imaged in Fig~\ref{fig3}. C1(C) shows steep spectral
325: indices $\alpha < - 0.70$ (using the convention that S$_{\nu}$
326: $\propto$ $\nu^{\alpha}$, where S$_{\nu}$ is the flux density at
327: frequency $\nu$) from 1.7 GHz to 15~GHz but a flat index $\alpha \,
328: \simeq\, -0.02$ between 15~GHz and 22~GHz. C2(C) and C2(W) have
329: similar spectra: $\alpha = -0.88$ and $-0.99$ between 5.0 and 8.4 GHz,
330: and $\alpha = -0.46$ and $-0.72$ between 1.7 and 5.0 GHz. The compact
331: component C2(E) that lies within the jet extension and shows an
332: inverted spectrum $\alpha \simeq +0.13$ between 1.7 GHz and 5.0
333: GHz. The continuum peak of C2(E) was not clearly seen at 1.7 GHz in
334: both epochs, but was seen in a 1.4 GHz map obtained in Dec. 1996
335: \cite{pihl00}.
336:
337:
338:
339:
340: %
341: %
342: %
343: \subsection{\it Water Maser Emission}
344: %
345: We detected maser features that are blueshifted with respect to the
346: systemic velocity of the galaxy with the VLBA. The maser features are
347: seen in the averaged channels ranging from \vlsr = 3190 \kms -- 3210
348: \kms. Inset into Fig.~\ref{fig5} is a blueshifted spectrum that shows
349: the total flux density obtained from our VLBA observations. The
350: spectrum appears to be a scaled-down version of the the single-dish
351: spectrum, which suggests that about 50 \% of the peak flux density is
352: detected in VLBA observations. This might be explained by the
353: intrinsic intensity variability of the water emission or by a loss of
354: coherence during phase referencing \cite{cari99} or, of course, by a
355: combination of both. Given the total maser intensity (0.78 Jy \kms)
356: estimated from the VLBA spectrum shown in the inset in
357: Fig.~\ref{fig5}, the isotropic luminosity of the blueshifted emission
358: is 38 \lsun, which is comparable to that of the redshifted emission
359: (22 \lsun) in NGC\,4258 \cite{naka95}. Fig.~\ref{fig5} shows the
360: spatial distribution of the water maser spots, where the position of
361: each maser feature was determined by Gaussian fitting for each
362: emission peak in the deconvolved images that was detected above
363: 4$\sigma$ level. Our analysis showed that the detected maser emission
364: is split into two clusters with a separation of $\sim$ 0.7 mas, or 0.16
365: pc, but there is no discernible velocity structure. Each maser clump,
366: assuming a nominal size of 0.2 mas, yields a lower limit to the
367: brightness temperature, T$_{\rm b}$ $\sim$ 10$^9$ K. The centroid
368: position of the brightest maser feature centered at \vlsr = 3194 \kms\
369: is marked with a cross in Fig.~\ref{fig2} and Fig.~\ref{fig3} in which
370: the continuum maps are superposed. No continuum emission was detected
371: at the maser position at any of our observed frequencies. The maser
372: spot is located 15.5 mas or 3.6 pc from the unresolved radio continuum
373: peak C1(C) at 22 GHz, which is coincident with a reference point (0, 0) in the 8.4 GHz map in Fig.~\ref{fig3}. From VLA-A observations with
374: a beam size of 130 $\times$ 80 mas (30 $\times$ 18 pc) in 1997
375: January, this blueshifted maser emission remained unresolved, and was
376: coincident with the unresolved continuum peak C1(C), supporting our
377: VLBA results in this paper. The peak of C2(E) at 8.4 GHz is about 18
378: mas, some 4 pc away from the maser spot, suggesting that the maser
379: clusters are located nearly at the midpoint of the line joining C1(C)
380: and C2(E). \\
381: %
382: %
383: \indent We found no velocity features of the water maser corresponding
384: to the velocities of the HI and OH absorptions, observed with the VLA
385: by Gardner et al. (1986).
386: %
387: %
388: %
389: %authors \cite[]{hag86}; we briefly summarize below.
390: %Results of this task are shown in table~\ref{tbl-1}.
391: % $M_{\sun}$, 90\degr\
392: %
393: %
394: %% Included in this acknowledgments section are examples of the
395: %% AASTeX hypertext markup commands. Use \url without the optional [HREF]
396: %% argument when you want to print the url directly in the text. Otherwise,
397: %% use either \url or \anchor, with the HREF as the first argument and the
398: %% text to be printed in the second.
399: %
400: %
401: \section{DISCUSSION}
402: %
403: %
404: \subsection{\it Location of the Nucleus}
405: %
406: %
407: %
408: One might ask where the nucleus of NGC\,5793 lies? In order to
409: understand the kinematics of the nuclear region in the galaxy we need
410: to address this question. Component C1(C) shows a flat-spectrum at
411: higher frequencies, at which it is not resolved, suggesting that it
412: might be the 'true' nucleus. On the other hand, the `true' nucleus may
413: lie in the continuum emission gap between C1(C) and C2(E) where the
414: maser spots are located. This would mean that the radio continuum
415: nucleus could be highly obscured by a screen of dense ionized gas. In
416: some megamasers like NGC\,1068 \cite{linc96}, the Circinus galaxy
417: \cite{linc01}, and possibly NGC\,4945 \cite{linc97}, the existence of
418: maser disks has been confirmed but continuum emission has not been
419: detected in the vicinity of the water masers. \\
420: %
421: %
422: \indent The measured surface brightness of C2(E) is at least 3.2
423: $\times$ 10$^8$ K and could in principle be as high as the $\sim$
424: 10$^{10}$ K required for synchrotron self-absorption \cite{kell69}. It
425: is conceivable that component C2(E) was produced by a strong shock
426: with the circumnuclear gas. A jet ejected from the nucleus could be
427: bent and deflected near C2(E) and extend further out to component
428: C2(W). Alternatively, component C2(E) could be free-free absorbed by
429: dense ionized foreground gas. The spectral index of +0.13 between 1.7
430: and 5.0 GHz is not steep enough for free-free absorption, but as it is
431: a lower limit, it is not inconsistent with such an hypothesis. If
432: C2(E) is a free-free absorbed component, it could be the inner part of
433: the disk/torus. However, at this stage we $conservatively$ consider
434: that the spectral turnover of C2(E) is due to synchrotron
435: self-absorption caused by the collision of a jet from the central
436: engine with a dense gas in the circumnuclear region. \\
437: %
438: %
439: \indent There might be a change in flux density during the two
440: observing sessions between Oct. 1999 at 1.7 GHz and Oct. 1997 at 5.0
441: GHz. If so, the determination of the spectral index of C2(E) is
442: unreliable. However, the flux density of C2(E) for 1.7 GHz in
443: Nov. 1996 is quite similar with that in Oct. 1999, hence it is less
444: likely that there was significant flux density variation from 1996 to
445: 1999 as for C2(E). In any case, there is no straightforward
446: interpretation for the location of the 'true' core. For this reason it
447: is impossible to decide at present whether the water emission in
448: NGC\,5793 is a nuclear or non-nuclear maser.
449: %
450: %
451: \subsection{\it Origins of the Water Maser}
452: %
453: %
454: Interpretation of the geometry of the detected maser emission in
455: NGC\,5793 is not straightforward, because the blueshifted features are
456: located between continuum peaks and the velocity fields do not show
457: any systematic trends. However, we discuss two possibilities to explain
458: the nature of the maser in the galaxy:
459: \begin{itemize}
460: \item[1] The water maser in NGC\,5793 results from the amplification
461: of a radio jet by the circumnuclear molecular gas foreground to the
462: jet.
463: \item[2] The maser lies in a in a compact molecular disk rotating
464: around the galactic nucleus, like that in NGC\,4258.
465: \end{itemize}
466: %
467: %
468: \indent Support for the validity of the first model comes from the
469: presence of molecular (OH) gas with a parsec-scale distribution seen
470: in absorption against the continuum structure with the VLBA
471: \cite{hagi00}. The brightness of a putative radio jet component may
472: not be sufficient to be detected by VLBI in NGC\,5793 at 22~GHz. The
473: brightness temperature of any undetected 22 GHz continuum source near
474: the maser spots is less than $\sim$ $10^7$ K, assuming that the peak
475: flux density is 6.5 mJy \bm\ (5$\sigma$ noise level). The beam
476: averaged brightness temperature of the unresolved maser spot is 1.3
477: $\times$ 10$^8$ K, so that the maser gain would be a factor of $>$
478: 10. Usually, this situation is seen in OH megamasers functioning as
479: low-gain amplifiers against background continuum sources. Non-nuclear
480: and non-disk masers have been seen in NGC\,1052 \cite{clau98} and
481: NGC\,1068 \cite{gall96}. The non-nuclear water maser known to exist in
482: NGC\,1068, located at 30 pc from the nucleaus along the jet, has not
483: been detected with VLBI, suggesting that the maser spot is resolved or
484: not intense enough for VLBI detection. In NGC\,1052 two maser spots
485: are seen lying 0.07 pc from the presumed core along a radio jet; they
486: show no evidence for disk structure \cite{clau98}. Neither of the
487: above cases seems to explain the maser in NGC\,5793 because it
488: exhibits different characteristics. If the water maser of NGC\,5793 is
489: emitted from the nuclear region, the large blueshift ($\sim$ 245 \kms)
490: of the maser emission implies that it traces molecular outflows moving
491: at high speed relative to the molecular disk surrounding the
492: nucleus. A model of a non-disk maser as described here also has
493: problems in explaining the single-dish spectrum of the emission.\\
494: %
495: %
496: \indent We now consider the the second possibility, namely that the
497: masers lie in a molecular disk rotating around the nucleus of the
498: galaxy as proposed by Hagiwara et al. (1997). Unfortunately, during
499: the period of observations the systemic and redshifted features became
500: weaker, and we could not measure the distribution and the velocity
501: structure of the water masers. If we assume the presence of rotating
502: disk, the blueshifted features could arise from the edge of the
503: disk. The emission has been detectable and does not show any velocity
504: drifts since its discovery in January 1996. In addition, the
505: peak flux density is less variable than other features. Such properties
506: are similar to the characteristics of the blue- and
507: redshifted masers in NGC\,4258 \cite{linc95b}. By contrast, the peak
508: flux densities of features around the systemic velocity have been
509: particularly variable. Greenhill et al. (1995a) suggested the
510: possibility that the high-velocity features in NGC\,4258 are visible
511: for substantially longer times than the systemic ones because the
512: emission might be self-amplified in the long gain paths along the edge
513: of the disk. Considering these facts, the high-velocity blueshifted
514: features may trace only a part of the disk, possibly near the tangent
515: points.
516: %
517: %
518: \indent With such a supposition in mind, we propose a model
519: (Fig.~\ref{fig6}) for the nuclear region of NGC\,5793. The location of
520: the nucleus is presumed to lie on the midpoint of the axis connecting
521: C1(C) to C2(E). This axis may mark the location of a two-sided jet.
522: The blueshifted maser features are offset from this axis by $\sim$ 4.5
523: mas, or 1.0 pc. The southern side of the rotating disk, where the
524: maser clumps lie, is approaching, while the northern side is
525: receding. The disk orientation is nearly perpendicular to the jet. In
526: this model, the inferred free-free absorbed source C2(E) lies within
527: the inner radius of the disk. We estimate that the total mass enclosed
528: within a radius of 1 pc is $\sim$ 1 $\times$ 10$^7$ $M_{\sun}$,
529: assuming a rotation velocity of 240 \kms and a disk inclination of
530: 73$^{\circ}$ \cite{roth94}. The mass density is $\sim 2 \times 10^6 \,
531: M_{\sun} \, {\rm pc}^{-3} $, a value that is the smallest among the
532: megamasers observed with VLBI \cite{mora99}. The enclosed mass of
533: $\sim$ 10$^7$ $M_{\sun}$ is lower approximately by a factor of 10 than
534: that estimated from the OH absorption in the central region of $\sim$
535: 10 pc \cite{hagi00}, because we adopted a smaller radius. The sense of
536: the disk rotation is opposite to that of the galactic disk, which has
537: a radius of $\sim$ 1 kpc and is observed in CO(J = 1 -- 0) emission,
538: but is the same as that of the OH absorbing gas disk, implying the
539: existence of an independent kinematical system within the central few
540: parsecs of the circumnuclear region. \\
541: %
542: \indent From recent single-dish measurements made in early 2000,
543: systemic features lying at \vlsr = 3430 \kms to 3520 \kms were found
544: to be flaring and variable in flux density (Henkel, private
545: communication), suggesting that the features might arise against the
546: continuum source at the nucleus along the line of sight, and that the
547: masing cloud is amplifying the background continuum source.
548: %
549: %
550: %
551:
552: %
553: %
554:
555:
556:
557: %
558: \section{SUMMARY}
559: %
560: We have observed the water megamasers and continuum emission in the
561: nuclear region of the Seyfert 2 galaxy NGC\,5793 with the
562: VLBA. Because of the faintness of the maser intensities, only the
563: blueshifted emission could be imaged. This emission, which is located
564: 3.6 pc southwest of the radio continuum source, is observed to consist
565: of two clusters with a separation of 0.7 mas (0.16 pc). No compact
566: continuum emission was detected near to the maser spot. The detected
567: water maser could be a part of a maser disk as in the case of
568: NGC\,4258. We propose a model in which the masers lie on the southern
569: edge (approaching side) of a rotating molecular gas disk whose
570: orientation is nearly perpendicular to the jet. Assuming the location
571: of the nucleus at a midpoint between C1 and C2(E) one can estimate a
572: binding mass on the order of 10$^7$ $M_{\sun}$. Alternatively, the
573: masers could result from the amplification of a radio jet by
574: foreground circumnuclear gas. The highly blueshifted velocity of the
575: maser features might mean that the masing cloud traces outflowing
576: gas surrounding an active nucleus. \\
577: %
578: %
579: \indent The compact continuum source, C2(E) appears to have an
580: inverted spectrum between 1.7 and 5.0 GHz, assuming no intrinsic
581: variability, and the spectral turnover might be due to synchrotron
582: self-absorption caused by a shock in the jet owing to collision with
583: the dense gas, or a free-free absorbing screen in a part of an ionized
584: torus/disk. \\
585: %
586: %
587: \indent Further VLBI observations of the water masers and continuum
588: emission are crucial to clarify the dynamical structures of the maser
589: emission in relation to the active nucleus. The systemic and
590: redshifted velocity features flared in early 2000 and additional VLBA
591: observations, following up our present result, are already underway.
592: %
593: %
594: %
595: %
596: \acknowledgments
597: %
598: We are grateful to Drs. R. Porcas, E. Ros, C. Henkel, and A. Roy for
599: their helpful comments and suggestions. We also appreciate J. Conway
600: and Y. Pihlstr{\"o}m for providing us the 1.4 GHz image. YH
601: appreciates the efforts of P. Hall, E. Troup, and other staff members
602: at Parkes for their assistance in the observations and data
603: analysis. Part of the VLBA observations appearing in this paper were conducted
604: while PJD was an NRAO staff member. This research has made use of NASA's Astrophysics Data System Abstract Service.
605:
606:
607:
608:
609:
610: %% The reference list follows the main body and any appendices.
611: %% Use LaTeX's thebibliography environment to mark up your reference list.
612: %% Note \begin{thebibliography} is followed by an empty set of
613: %% curly braces. If you forget this, LaTeX will generate the error
614: %% "Perhaps a missing \item?".
615: %%
616: %% thebibliography produces citations in the text using \bibitem-\cite
617: %% cross-referencing. Each reference is preceded by a
618: %% \bibitem command that defines in curly braces the KEY that corresponds
619: %% to the KEY in the \cite commands (see the first section above).
620: %% Make sure that you provide a unique KEY for every \bibitem or else the
621: %% paper will not LaTeX. The square brackets should contain
622: %% the citation text that LaTeX will insert in
623: %% place of the \cite commands.
624:
625: %% We have used macros to produce journal name abbreviations.
626: %% AASTeX provides a number of these for the more frequently-cited journals.
627: %% See the Author Guide for a list of them.
628:
629: %% Note that the style of the \bibitem labels (in []) is slightly
630: %% different from previous examples. The natbib system solves a host
631: %% of citation expression problems, but it is necessary to clearly
632: %% delimit the year from the author name used in the citation.
633: %% See the natbib documentation for more details and options.
634: \clearpage
635: \begin{thebibliography}{}
636: %
637: \bibitem[Baan et al. 1998]{baan98} Baan, W. A., Salzer, J. J., \& LeWinter, R. D. 1998, \apj, {509}, 633
638: %
639: \bibitem[Braatz et al. 1997]{braa97} Braatz, J. A., Wilson, A. S., \& Henkel, C. 1997, \apjs, {110}, 321
640: %
641: \bibitem[Carilli and Holdaway 1999]{cari99} Carilli, C. L., \& Holdaway, M. A. 1999, Radio Science, 34, 817
642: %
643: \bibitem[Claussen et al. 1998]{clau98}
644: Claussen, M. J., Diamond, P. J., Braatz, J. A., Wilson, A. S., \& Henkel, C. 1998, \apjl, 500, L129.
645: %
646: \bibitem[Falcke et al. 2000]{falc00}
647: Falcke, H., Henkel, Chr., Peck, A. B., Hagiwara, Y., Prieto, M. A., \& Gallimore, J. F. 2000, \aap, {358}, L17
648: %
649: %\bibitem[Gallimore et al. 1996a]{gall96}
650: %Gallimore, J. F., Baum, S. A., O'Dea, C. P., et al. 1996a, \apj, {462}, 740
651: %
652: \bibitem[Gallimore et al. 1996]{gall96}
653: Gallimore, J. F., Baum, S. A., \& O'Dea, C. P. 1996, \apj, {464}, 198
654: %
655: \bibitem[Gardner and Whiteoak 1986]{gard86}
656: Gardner, F. F., \& Whiteoak, J. B. 1986, MNRAS, {221}, 537
657: \bibitem[Gardner et al. 1992]{gard92}
658: Gardner, F. F., Whiteoak, J. B., Norris, R. P., \& Diamond, P. J. 1992, \mnras, {258}, 296
659: %NGC 4258
660: \bibitem[Greenhill et al. 1995a]{linc95}
661: Greenhill, L. J., Jiang, D. R., Moran, J. M., Reid, M. J., Lo, K.-Y., \& Claussen, M. J. 1995a \apj, {440}, 619
662: %NGC 4258 acceleration paper
663: \bibitem[Greenhill et al. 1995b]{linc95b}
664: Greenhill, L. J., Henkel, C., Becker, R., Wilson, T. L., \& Wouterloot, J. G. A. 1995b, \aap, {304}, 21
665: % NGC1068 blueshift-maser
666: \bibitem[Greenhill et al. 1996]{linc96}
667: Greenhill, L. J., Gwinn, C. R., Antonucci, R., \& Barvanis, R. 1996, \apjl, {472}, L21
668: % NGC1068 redshift-maser
669: \bibitem[Greenhill \& Gwinn 1997]{linc96b}
670: Greenhill, L. J., \& Gwinn, C. R. 1997, \apss, {248}, 261
671: % NGC4945 maser disk
672: \bibitem[Greenhill et al. 1997]{linc97}
673: Greenhill, L. J., Moran, J. M., \& Herrnstein, J. R. 1997, \apjl, {481}, L23
674: %
675: \bibitem[Greenhill 2000]{linc01}
676: Greenhill, L. J. 2000, in the Proceedings of the 5th EVN Symposium. eds. Conway, J. E., Polatidis, A. G., Booth, R. S. \& Pihlstr{\"o}m, Y. M. (Onsala Space Observatory, Sweden), p101
677: %
678: \bibitem[Hagiwara et al. 1997]{hagi97}
679: Hagiwara, Y., Kohno, K., Nakai, N., \& Kawabe, R. 1997, \pasj, {49}, 171
680: %
681: \bibitem[Hagiwara 1998]{hagi98} Hagiwara, Y. 1998, Ph.D Thesis, The Graduate University for Advanced Studies
682: \bibitem[Hagiwara et al. 2000]{hagi00}
683: Hagiwara, Y., Diamond, P. J., Nakai, N., \& Kawabe, R. 2000, \aap, 360, 49.
684: \bibitem[Herrnstein et al. 1998]{herr98}
685: Herrnstein, J. R., Greenhill, L. J., Moran, J. M., Diamond, P. J., Inoue, M., Nakai, N., \& Miyoshi, M. 1998, \apjl, 497, 69
686: %
687: \bibitem[Ishihara et al. 2001]{ishi00}
688: Ishihara, Y., Nakai, N., Iyomoto, N., Makishima, K., Diamond, P., \& Hall, P. 2001, \pasj, 53, 215
689: %
690: \bibitem[Kellermann \& Pauliny-Toth 1969]{kell69}
691: Kellermann, K. I., \& Pauliny-Toth, I. I. K. 1969, \apjl, 155, 71
692: %
693: \bibitem[Miyoshi et al. 1995]{miyo95}
694: Miyoshi, M., Moran, J., Herrnstein, J., Greenhill, L., Nakai, N., Diamond, P., \& Inoue, M. 1995, Nature, 373, 127
695: %
696: \bibitem[Moran et al. 1999]{mora99}
697: Moran, J. M., Greenhill, L. J., \& Herrnstein, J. R. 1999, J. Astrophys. Astron., 20, 165
698: %
699: \bibitem[Nakai et al. 1995]{naka95}
700: Nakai, N., Inoue, M., Miyazawa, K., Miyoshi, M., \& Hall, P. 1995, \pasj, 47, 771
701: %
702: \bibitem[Nakai et al. 1998]{naka98}
703: Nakai, N., Inoue, M., Hagiwara, Y., Miyoshi, M., \& Diamond, P. J. 1998, in Radio Emission from Galactic and Extragalactic Compact Sources., ed. J. A. Zensus, G.B. Taylor, \& J. M. Wrobel (San Francisco: ASP), 237
704: %
705: \bibitem[Palumbo et al. 1983]{palu83}
706: Palumbo, G. G. C., Tanzella-Nitti, G., \& Vettolani, G. 1983, in Catalogue of radial velocities of galaxies., (New York: Gordon and Breach Science Publishers)
707: %
708: \bibitem[Pihlstr{\"o}m et al. 2000]{pihl00}
709: Pihlstr{\"o}m, Y. M., Conway, J. E., Booth, R. S., Diamond, P. J., \& Koribalski, B. S. 2000, \aap, 357, 7
710: %
711: \bibitem[Roth 1994]{roth94}
712: Roth, J. 1994, \aj, 108, 862
713: %
714: \end{thebibliography}
715:
716: %% Tables should be submitted one per page, so put a \clearpage before
717: %% each one.
718:
719: %% Two options are available to the author for producing tables: the
720: %% deluxetable environment provided by the AASTeX package or the LaTeX
721: %% table environment. Use of deluxetable is preferred.
722: %%
723:
724: %% Three table samples follow, two marked up in the deluxetable environment,
725: %% one marked up as a LaTeX table.
726:
727:
728: %% In this first example, note that the \tabletypesize{}
729: %% command has been used to reduce the font size of the table.
730: %% Note also that the \label command needs to be placed
731: %% inside the \tablecaption.
732:
733: \clearpage
734: %
735: %
736: \begin{figure}
737: \plotone{f1.eps} \figcaption[f1.eps]{Spectrum of the \hho maser
738: towards the center of NGC\,5793, observed with the Parkes radio
739: telescope of the CSIRO. The spectrum is averaged over three days from
740: 18-20 August 1997. Accuracy of the flux density is about 10 \%. A
741: downward arrow indicates the adopted systemic velocity of the galaxy,
742: \vlsr = 3442 \kms. Solid bars indicate the velocity ranges covered by
743: four IFs which we observed with the VLBA in May 1998. \label{fig1}}
744: \end{figure}
745: %
746: %
747: \begin{figure}
748: \plotone{f2.eps} \figcaption[f2.eps]{Radio continuum image at 22 GHz
749: with the VLBA. The levels of contours are -3, 3, 4, 5, 6, 7$\sigma$ (1
750: $\sigma$ = 1.3 mJy \bm). The peak flux density is 9.2 mJy \bm. The
751: synthesized beam of 2.8 $\times$ 1.8 mas at P.A. = 23$\,^{\circ}$ is
752: plotted in the left-hand corner. The cross marks the Gaussian-fitted
753: center of a maser feature at \vlsr = 3194 \kms. The relative alignment of the
754: maser and continuum is accurate to 0.22 mas.
755: \label{fig2}}
756: \end{figure}
757: %
758: %
759: \begin{figure}
760: \plotone{f3.eps} \figcaption[f3.eps]{8.4 GHz radio continuum image of
761: the nuclear region made with the VLBA. The contour levels are $-2.7,
762: 2.7, 3.7, 5.2, 7.2, 10, 14, 19, 27, 37, 52, 72, ~{\rm and}~100\,\%$ of
763: the peak flux density of 51.1 mJy \bm. The synthesized beam of 7.0
764: $\times$ 2.7 mas at P.A. = -- 4.7$\,^{\circ}$ is shown in the
765: left-hand corner. The position of the maser feature at \vlsr = 3194
766: \kms is marked by a cross. The center position (0, 0) in the map
767: coincides with that of the 22 GHz continuum peak in Fig.~2. The
768: relative accuracy of the 8.4 GHz continuum and 22 GHz maser positions
769: is $\sim$ 0.2 mas.\label{fig3}}
770: \end{figure}
771: %
772: %
773: \begin{figure}
774: \plotone{f4.eps} \figcaption[f4.eps]{Continuum spectra of the four
775: compact components C1(C), C2(E), C2(C), and C2(W) at 1.7, 5.0, 8.4, 15
776: and 22 GHz. The flux density of C2(E) at 1.7 GHz is an upper
777: limit. \label{fig4}}
778: \end{figure}
779: %
780: %
781: \begin{figure}
782: \plotone{f5.eps} \figcaption[f5.eps]{Distributions of water maser
783: emission. The position of each component was estimated by 2-D
784: Gaussian-model fitting, and error bars indicate total positional
785: uncertainties of 1$\sigma$. The point (0, 0) is referred to position
786: of the crosses in Fig. 2 and Fig. 3. The maser spots are labelled by
787: their velocities in \kms. The inset shows the VLBA spectrum of the
788: blueshifted maser emission. Five spectral channels were averaged
789: to 4.2 \kms (channel spacing before averaging is 0.84 \kms). Accuracy
790: of the flux density scale is typically 5 \%. \label{fig5}}
791: \end{figure}
792: %
793: %
794: \begin{figure}
795: \plotone{f6.eps} \figcaption[f6.eps]{Schematic view of the nuclear
796: region of NGC 5793. The position of the nucleus is assumed to lie at
797: the midpoint on the line joining C1 and C2(E). The blueshifted maser
798: clusters are located on the approaching side of a postulated edge-on
799: rotating disk. The P.A. of the disk lying nearly perpendicular to the
800: jet is well aligned with the kilo-parsec-scale disk observed in CO
801: emission, and the pc-scale disk imaged in OH absorption
802: \cite{hagi00}. An arrow denotes the sense of the rotation of the maser
803: disk. The sense is the same as that of the OH disk, but reversed when
804: compared to that of the CO disk.
805: \label{fig6}}
806: \end{figure}
807: %\input{tbl-1.tex}
808: %\input{tbl-2.tex}
809: %
810: \clearpage
811: %----------------------table-1-----------------------
812: %
813:
814: %
815: \begin{deluxetable}{crrrrrr}
816: \tabletypesize{\scriptsize}
817: \tablecolumns{7}
818: \tablewidth{0pc}
819: \tablecaption{\sc Flux Densities of Major Components \label{tbl-1}}
820: %\begin{tabular}{crrrrrr}
821:
822: \tablehead
823: {
824: \colhead{} & \multicolumn{4}{c}{Component Flux Densities} \\
825: \colhead{} & \multicolumn{4}{c}{(mJy \bm)} \\
826: \colhead{Frequency}& \cline{1-4}
827: \colhead{Band} & \colhead{C1(C)} & \colhead{C2(E)} & \colhead{C2(C)} &\colhead{C2(
828: W)} & \colhead{Epoch} & \colhead{Reference} \\
829: \multicolumn{1}{c}{(GHz)} &\colhead{}
830: }
831: \startdata
832: 1.4\nodata &304&16.3&15.6&23.1& Dec. 96&\colhead{a}\\
833: 1.7\nodata &308&$<$16.1\tablenotemark{d}&18.0 & 28.3&Nov. 96&\colhead{b} \\
834: &274 &$<$ 15.2\tablenotemark{d} &13.9 &14.7 &Oct. 99&\colhead{c} \\
835: 5.0\nodata &139& 18.5&10.6&12.2& Oct. 97&\colhead{b}\\
836: 8.4\nodata & 64.5 &9.42 &6.73 &7.31 &Oct. 99&\colhead{c} \\
837: 15\nodata & 16.4 &\nodata & \nodata& \nodata&Oct. 99&\colhead{c} \\
838: 22\nodata & 16.3 &\nodata&\nodata&\nodata &May 98&\colhead{c} \\
839: \enddata
840: %% Any table notes must follow the \end{tabular} command.
841: \tablenotetext{a}{Y. Pihlstr{\"o}m, private communication}
842: \tablenotetext{b}{Hagiwara et al. 2000}
843: \tablenotetext{c}{This paper}
844: \tablenotetext{d}{An upper limit value}
845:
846:
847: \tablecomments{Flux uncertainties assume 5 persent calibration uncertainty added. All the flux densities are estimated with the synthesized beam at 1.7 GHz (12.4 $\times$ 4.4 mas).}
848: \end{deluxetable}
849: %
850: %----------------table 2-----------------------
851: %
852: %\clearpage
853: %
854: \begin{deluxetable}{crrrrrr}
855: \tabletypesize{\scriptsize}
856: \tablecolumns{5}
857: \tablewidth{0pc}
858: \tablecaption{\sc Spectral Indices of Major Components \label{tbl-2}}
859: %\begin{tabular}{crrrrrr}
860:
861: \tablehead
862: {
863: \colhead{} & \multicolumn{4}{c}{Component Spectra, $\alpha$} \\
864: \colhead{} & \multicolumn{4}{c}{(S$_{\nu}$ $\propto$ $\nu^{\alpha}$)} \\
865: \colhead{Frequency Range}& \cline{1-4}
866: \colhead{(GHz)} & \colhead{C1(C)} & \colhead{C2(E)} & \colhead{C2(C)} &\colhead{C2(W)}
867: }
868: \startdata
869: \colhead{1.7\tablenotemark{a} $-$ 5.0}\nodata&$-$0.70&0.13&$-$0.46 & $-$0.72\\
870: \colhead{5.0 $-$ 8.4}\nodata&$-$1.5&$-$1.3&$-$0.88&$-$0.99\\
871: \colhead{8.4 $-$ 15}\nodata&$-$2.3&\nodata &\nodata&\nodata \\
872: \colhead{15 $-$ 22}\nodata&$-$0.017& \nodata&\nodata&\nodata
873: \enddata
874: %% Any table notes must follow the \end{tabular} command.
875:
876: \tablenotetext{a}{Data from Nov. 1996}
877:
878:
879: \end{deluxetable}
880:
881:
882: \end{document}
883:
884: %%
885: %% End of file `subm.tex'.
886: