0803.3481/ms.tex
1: \documentclass[numberedappendix]{emulateapj}
2: \begin{document}
3: 
4: \slugcomment{Accepted for publication in ApJ: March 20, 2008}
5: 
6: \title{A Resolved Molecular Gas Disk around the Nearby A Star 49 Ceti}
7: 
8: \author{A. M. Hughes \altaffilmark{1}, 
9: D. J. Wilner \altaffilmark{1}, 
10: I. Kamp \altaffilmark{2}, 
11: M. R. Hogerheijde \altaffilmark{3}}
12: 
13: \email{mhughes@cfa.harvard.edu}
14: 
15: \altaffiltext{1}{Harvard-Smithsonian Center for Astrophysics,
16:   60 Garden Street, Cambridge, MA 02138}
17: \altaffiltext{2}{Kapteyn Astronomical Institute, University of Groningen, 
18:   9700 AV Groningen, The Netherlands}
19: \altaffiltext{3}{Leiden Observatory, Leiden University, P.O. Box 9513, 2300 RA,
20:   Leiden, The Netherlands}
21: 
22: 
23: % document history
24: % 13aug07 amh draft
25: % 04sep07 ik model section revised
26: % 28oct07 amh revision
27: % 29oct07 ik edits
28: % 02nov07 ik CO chem section added
29: % 04nov07 amh edits
30: % 09nov07 ik new figures, paragraph on gas temp
31: % 12nov07 amh edits
32: % 13nov07 djw tweaks
33: % 13nov07 amh edits
34: % 27nov07 amh edits in response to Sean & Michiel's comments
35: % 03/07jan08 amh edits incorporating ik comments
36: % 08jan08 submitted to ApJ
37: % 03mar08 revisions
38: 
39: \bibliographystyle{apj}
40: 
41: \begin{abstract}
42: 
43: The A star 49 Ceti, at a distance of 61 pc, is unusual in retaining a 
44: substantial quantity of molecular gas while exhibiting dust properties similar 
45: to those of a debris disk.  We present resolved observations of the disk 
46: around 49 Ceti from the Submillimeter Array in the J=2-1 rotational transition 
47: of CO with a resolution of 1.0$\times$1.2 arcsec.  The observed emission 
48: reveals an extended rotating structure viewed approximately edge-on and clear 
49: of detectable CO emission out to a distance of $\sim 90$~AU from the star.  No 
50: 1.3 millimeter continuum emission is detected at a $3\sigma$ sensitivity of 
51: 2.1 mJy/beam.  Models of disk structure and chemistry indicate that the inner 
52: disk is devoid of molecular gas, while the outer gas disk between 40 and 200 AU 
53: from the star is dominated by photochemistry from stellar and interstellar 
54: radiation.  We determine parameters for a model that reproduces the basic 
55: features of the spatially resolved CO J=2-1 emission, the spectral energy 
56: distribution, and the unresolved CO J=3-2 spectrum. We investigate 
57: variations in disk chemistry and observable properties for a range of 
58: structural parameters. 49 Ceti appears to be a rare example of a system in a 
59: late stage of transition between a gas-rich protoplanetary disk and a tenuous, 
60: virtually gas-free debris disk.
61: 
62: \keywords{astrochemistry --- circumstellar matter --- planetary systems: 
63: protoplanetary disks --- stars:individual (49 Ceti)}
64: 
65: \end{abstract}
66: 
67: \section{Introduction}
68: 
69: A key to understanding the formation of planetary systems lies in 
70: characterizing the transitional phase between the gas-rich primordial 
71: disks found around young T Tauri stars and the tenuous, virtually 
72: gas-free debris disks around their main-sequence counterparts.  Unfortunately, 
73: disks in this transitional phase are rare and difficult to identify.
74: Dust disks around young stars are commonly identified through the 
75: ``Vega-excess'' phenomenon (first observed using the Infrared 
76: Astronomical Satellite by \citealt{aum84}; see review by \citealt{zuc01}), 
77: in which an infared excess over the stellar photosphere is attributed to 
78: reprocessing of optical and ultraviolet 
79: starlight by thermally emitting circumstellar dust grains.  
80: 49 Ceti was first identified in this way by \citet{sad86}.
81: The quantity $\tau = L_{IR}/L_{bol}$  is often used as an indicator of the 
82: ``optical depth'' of the dust disk, as it provides a rough estimate of the 
83: quantity of optical/ultraviolet light intercepted and reemitted by the dust.  
84: \citet{jur93} correlated the IRAS Point Source Catalog with the Yale Bright 
85: Star Catalog \citep{hof91} and identified three A stars with 
86: $\tau > 10^{-3}$, indicative of tenuous, optically thin circumstellar dust.  
87: Two were the stars $\beta$ Pic and HR4796, which are now known to host 
88: debris disks.  The third was 49 Ceti, which unlike the other two defies 
89: classification as a debris disk because it retains a substantial quantity of 
90: molecular gas, first observed in the CO J=2-1 line \citep{zuc95} and 
91: later confirmed in J=3-2 \citep{den05}.  At a distance of only 61 pc 
92: (Hipparcos), it is one of the closest known gas-rich circumstellar 
93: disks, farther only than TW Hydrae \citep[51pc;][]{mam05}.  Its outward 
94: similarity to a debris disk, combined with the substantial quantity of 
95: molecular gas still present in the system, suggest that the disk may be
96: in an unusual transitional evolutionary phase.
97: 
98: All three high-$\tau$ A stars are young: HR 4796A has an age of $8 \pm 2$ Myr \citep{sta95}
99: and $\beta$ Pic has been placed at $\sim 20$ Myr by \citet{thi01a}, consistent 
100: with the age determination of $20 \pm 10$ Myr by \citet{bar99}.  
101: The age of 49 Ceti is uncertain due to its isolation; unlike $\beta$ Pic or
102: HR 4796A there are no known associated low-mass stars to provide a corroborating
103: age estimate.  \citet{jur98} demonstrate that on an HR diagram, all three 
104: stars exhibit a low luminosity for their color, which is likely attributable 
105: to their young ages ($\sim 10$~Myr).  Using the evolutionary tracks of 
106: \citet{sie00}, \citet{thi01a} assign an age of 7.8~Myr to 49 Ceti based on 
107: its position on the HR diagram. 
108: 
109: Few conclusive measurements have been made of the dust properties in the 
110: 49 Ceti system. HST/NICMOS coronographic observations of 49 Ceti failed to 
111: detect any scattered light in the near infrared at $r>1\farcs6$ \citep{wei99}. 
112: Recent subarcsecond-scale imaging at mid-infrared wavelengths with Keck 
113: \citep{wah07} revealed dust emission at 12.5 and 17.9 $\mu$m, extended along 
114: a NW-SE axis and apparently inclined at an angle of $60^\circ$. Simple 
115: models of the dust emission suggest a radial size segregation of 
116: dust grains, with a population of very small grains ($a \sim 0.1\mu$m) 
117: confined between 30 and 60 AU from the star, and a population of larger 
118: grains ($a \sim 15 \mu$m) from 60 to 900 AU from the star. 
119: However, the outer radius of this latter component is uncertain due to its
120: dependence on the millimeter flux, which is not well determined.  There are two 
121: contradictory single dish measurements of the millimeter dust emission, both 
122: with modest signal-to-noise.  \citet{boc94} report a IRAM 1.2~mm flux of 
123: $12.7 \pm 2.3$ mJy, while \citet{son04} report a JCMT/SCUBA 850 $\mu$m flux of 
124: $8.2 \pm 1.9$ mJy.  These measurements are mutually inconsistent for either
125: a thermal spectrum ($F_\lambda \propto \lambda^{-2}$) or a typical optically
126: thin circumstellar disk spectrum ($F_\lambda \propto \lambda^{-3}$) in this
127: wavelength regime.  
128: 
129: If we accept the lower value of the 850 $\mu$m flux and make standard 
130: assumptions about the dust opacity \citep[e.g.][]{bec91}, then the total mass 
131: of the 49 Ceti dust disk is 0.1 M$_\earth$.  If we compare this to other
132: nearby dusty disks at potentially similar stages of evolution, we find that
133: 49 Ceti, with an 850 $\mu$m flux of 8.2 mJy at a distance of 61 pc, has a 
134: dust mass ($\propto F_{850\mu m} d^2$) approximately 80\% that of $\beta$ 
135: Pic \citep[104.3 mJy, 19.3 pc;][]{hol98} but only 0.3\% that of 
136: the typical Herbig Ae star HD 169142 \citep[554 mJy, 145 pc;][]{syl96}.
137: Thus the 49 Ceti disk appears to have a tenuous dust disk more akin to that
138: of the debris disk around $\beta$ Pic than a gas-rich protoplanetary disk.
139: 
140: Studies of the distribution of gas in the 49 Ceti system have been similarly 
141: inconclusive, particularly since it is not obvious that a substantial 
142: reservoir of molecular gas should persist in the strong UV field of an A star 
143: at this apparently advanced stage. Attempts to detect pure rotational 
144: transitions of the H$_2$ molecule have resulted in contradictory reports, 
145: with \citet{thi01n} reporting a marginal detection using SWS/ISO, which 
146: \citet{che06} did not confirm with Spitzer/IRS observations; nor did 
147: \citet{car07} detect H$_2$ emission with VLT/CRIRES observations.  Models of 
148: the double-peaked JCMT CO J=3-2 line profile observed by \citet{den05} 
149: indicated that the gas is likely distributed in either a very compact disk 
150: with $\sim 16^\circ$ inclination or a more inclined ring of radius $\sim 50$ AU 
151: and inclination $\sim 35^\circ$.  The latter was deemed more consistent with 
152: the dust distribution seen in the mid-infrared, although it fails to reproduce 
153: the high-velocity wings that may be present in the CO J=3-2 line profile. 
154: 
155: In order to obtain spatially resolved information on the distribution of  
156: material in the system, we observed 49 Ceti with the Submillimeter 
157: Array\footnote{The Submillimeter Array is a joint project between the 
158: Smithsonian Astrophysical Observatory and the Academia Sinica Institute of 
159: Astronomy and Astrophysics and is funded by the Smithsonian Institution and the 
160: Academia Sinica.}
161: in the J=2-1 transition of CO and associated continuum.  We detect a rotating 
162: structure of much greater extent than predicted from the single-dish 
163: measurements, with a large central region devoid of molecular gas emission.  
164: We also model the disk emission using COSTAR 
165: \citep{kam00,kam01}, a code that combines thin hydrostatic equilibrium 
166: models of disks with a rich chemistry network and a detailed heating and 
167: cooling balance to determine gas properties.  The models provide some insight 
168: into basic properties of the disk, including the region of photodissociation 
169: of CO in the inner disk and the spatial extent of the emission.  
170: 
171: The observations are described in \S\ref{sec:obs}, and results presented
172: in \S\ref{sec:res}.  In \S\ref{sec:mod} we discuss the process undertaken
173: to model the data, including the basic model structure, the initial conditions
174: for the chemistry, and the initial model adopted from the dust emission analysis
175: of \citet{wah07}, as well as adjustments to that fiducial model necessitated
176: by the new observations.  The parameter space is explored in 
177: \S\ref{sec:grid}, including an investigation of the varying influence of 
178: chemistry across the model grid, and \S\ref{sec:SED} discusses the dust
179: properties in the context of the spectral energy distribution predicted from
180: the gas model.  The best-fit model is discussed in \S\ref{sec:best}, including 
181: an a posteriori comparison of the model prediction with the observed CO J=3-2 
182: spectrum; inadequacies of the model are also noted.  The results are discussed
183: in the broader context of disk evolution in \S\ref{sec:dis}, and a summary is 
184: presented in \S\ref{sec:con}.
185: 
186: 
187: \section{Observations}
188: \label{sec:obs}
189: 
190: We observed 49 Ceti with the SMA at 230 GHz during an 11-hour track on
191: the night of October 13, 2006.	Atmospheric phase was extremely stable,
192: with typical phase changes of $<15^\circ$ between calibrator scans (every 25
193: minutes).  Seven antennas were used in the ``extended'' configuration, with 
194: projected baselines between 15 and 130 meters.
195: The primary flux calibrator was Uranus, and the passband calibrators were
196: the quasars 3C454.3 and J0530+135. Gain calibration was carried out using
197: the quasar J0132-169, located just 1.3$^\circ$ from 49 Ceti; the flux derived
198: for this quasar was 0.93 Jy.  The nearby quasar J0006-063 was also included
199: to test the quality of the phase transfer from J0132-169.
200: 
201: Two 2-GHz sidebands separated by 10 GHz were used, yielding a continuum
202: sensitivity of 0.7 mJy (1$\sigma$).  Spectral resolution in the line was
203: 0.26 km/s, subsequently binned to 2.1 km/s, with rms sensitivity 0.030 Jy in a 
204: single 2.1 km/s channel.  The LSR velocities were converted to heliocentric 
205: using an offset of -9.14 km/s.  The synthesized naturally weighted beam in the 
206: CO J=2-1 line was 1\farcs0$\times$1\farcs2, at a position angle of 
207: -78.6$^\circ$.
208: Imaging was carried out using the MIRIAD software package.
209: 
210: \begin{figure}
211: \centering
212: \includegraphics[scale=0.6,angle=90]{f1.eps}
213: \caption{A renzogram of SMA observations of 49 Ceti in the CO J=2-1 line.  
214: The beam size is 1\farcs0$\times$1\farcs2, and the position angle is 
215: $-79^\circ$.  Contours are -3, 3, and 5 $\times$ 37 mJy/beam (the rms noise).  
216: The position of 49 Ceti is marked with a star symbol, while the green line 
217: indicates the position angle derived by \citet{wah07} from mid-IR imaging.
218: The contour colors indicate heliocentric line-of-sight velocity; 
219: the four distinct velocities shown are 9.0, 11.1, 13.2, and 15.3 km/s, in the
220: %the four distinct velocities shown are -0.1, 2.0, 4.1, and 6.2 km/s, in the
221: order of bluest to reddest channel.  No emission was detected outside this
222: velocity range.
223: }
224: \label{fig:map}
225: \end{figure}
226: 
227: \section{Results and Analysis}
228: \label{sec:res}
229: 
230: Figure \ref{fig:map} shows the observed line emission from the region around
231: 49 Ceti.  Four velocity channels are shown, with the velocity indicated by 
232: the color of the contour lines.  The observations are centered on the J2000
233: coordinates of 49 Ceti; the star symbol indicates the position corrected
234: for the proper motion measured by {\em Hipparcos}.  The maximum signal-to-noise 
235: ratio in the line is 8.  The CO J=2-1 emission appears to be in an extended 
236: rotating structure of $>2$" radius, apparently viewed close to edge-on.  
237: The symmetric distribution of the emission in the four velocity channels 
238: implies a heliocentric velocity near 12.2 km/s, consistent with previous 
239: determinations of the systemic velocity \citep[10.5 and 9.9 km/s for the disk 
240: and the star, respectively; see][and references therein]{den05}. 
241: No emission is detected outside the range of velocities shown.
242: The wide separation of the emission peaks, combined with a lack
243: of compact, high-velocity emission, suggests that the central regions are
244: clear of CO J=2-1 emission out to $\sim$90~AU radius ($\sim$1\farcs5),
245: twice that of the larger ring predicted from the CO J=3-2 single dish data.
246: %In addition, while the position angle of the inner velocity channels is
247: %consistent with that seen in the mid-infrared by \citet{wah07}, the
248: %outer velocity channels are misaligned from this axis, perhaps indicative of
249: %either a warp or a more complex geometry, such as an incomplete or ``broken''
250: %ring.  
251: Table \ref{tab} lists the observed parameters of the system.
252: 
253: \begin{table*}
254: \caption{Observational parameters for 49 Ceti}
255: \begin{center}
256: \begin{tabular}{lcccc}
257: \hline
258: \multicolumn{1}{c}{Parameter} & $^{12}$CO(3-2)$^a$ & $^{12}$CO(2-1) & $^{13}$CO(2-1) & continuum \\
259: \hline
260: \hline
261: Rest frequency (GHz) & 345.796 & 230.538 & 220.399 & 230.5 (USB$^b$) \\
262: Channel width & 0.27 km s$^{-1}$ & 2.1 km s$^{-1}$ & 8.4 km s$^{-1}$ & 2$\times$2 GHz \\
263: Beam size (FWHM) & 14" & 1\farcs0$\times$1\farcs2 & 1\farcs0$\times$1\farcs2 & 1\farcs0$\times$1\farcs2 \\
264: ~~~PA & -- & -78.6$^\circ$ & -78.6$^\circ$ & -78.6$^\circ$ \\
265: rms noise (Jy beam$^{-1}$) & 0.22 & 0.030 & 0.017 & $7.0\times 10^{-4}$ \\
266: Dust flux (mJy) & -- & -- & -- & $<2.1$ \\
267: Peak brightness temperature (K) & 0.076$\pm$0.008 & 3.5$\pm$0.5 & $<0.8$ & -- \\
268: Integrated intensity (Jy km s$^{-1}$) & 9.5$\pm$1.9 & 2.0$\pm$0.3 & $<0.5$ & --
269: \\
270: \hline
271: \end{tabular}
272: \end{center}
273: \tablenotetext{a}{~\citet{den05}}
274: \tablenotetext{b}{~Upper sideband frequency; lower sideband is centered at 
275: 220.5 GHz.  Both sidebands have 2 GHz width.}
276: \label{tab}
277: \end{table*}
278: 
279: Assuming optically thin lines and LTE, the total mass in CO probed by the
280: J=2-1 transition is given by
281: \begin{eqnarray}
282: M=\frac{4 \pi}{h \nu_{21}} \frac{F_{21} m d^2}{A_{21} x_{2}}
283: \end{eqnarray}
284: where the subscript 21 refers to the CO(2-1) transition, $F$ is the
285: integrated flux in the line, $d$ is the distance to the source (61 pc;
286: Hipparcos), $m$ is the mass of the CO molecule, $\nu$ is the rest frequency
287: of the transition, $h$ is Planck's constant, and
288: $x_{2} \equiv \frac{N_2}{N_{tot}}$ where $N_2$ is the population in the
289: J=2 rotational level while $N_{tot}$ is the total CO population.
290: The CO mass calculated using this method is $2.2 \times 10^{-4}$ M$_\earth$.
291: Using the canonical CO/H$_2$ ratio of 10$^{-4}$ this yields a molecular
292: hydrogen mass of $2.2$ M$_\earth$, consistent with the value
293: of $6.3\times 10^{-3}$ M$_{Jup}$ = $2.0$ M$_\earth$ calculated by \citet{zuc95}.
294: 
295: No continuum emission was detected at this combination of resolution and
296: sensitivity.  This indicates one of two things: either the continuum flux is
297: concentrated at the center of the disk but the total flux is too low to be
298: detected, or the total flux may be larger but spread over many beams, so that 
299: the brightness within each beam is below our detection threshold.  These 
300: observations were sensitive enough to detect the higher continuum flux 
301: reported by \citet{boc94} if it were concentrated within a few synthesized 
302: beams.  However, an extrapolation of the \citet{son04} value for a typical 
303: circumstellar dust spectrum predicts a lower flux by a factor of 6, which is 
304: just below the detection threshold.  The lack of an SMA continuum detection 
305: at 230\,GHz is therefore inconclusive: if the \citet{son04} value is correct, 
306: we would not expect to detect even centrally concentrated emission, and so we 
307: cannot constrain the spatial extent of dust emission through the nondetection 
308: at 230\,GHz.
309: 
310: \section{Disk Modeling}
311: \label{sec:mod}
312: 
313: In order to gain insight into the physical processes at work in the 49 Ceti
314: system, we carried out modeling of the disk with COSTAR \citep{kam00,kam01},
315: a code which solves the chemical equilibrium simultaneously with a detailed 
316: heating and cooling balance to determine gas properties of circumstellar disks.
317: In the following, the salient features of these models are summarized. 
318: The chemistry is modeled using a network of 48 different species covering the 
319: elements H, He, C, O, S, Mg, Si, and Fe. The elemental abundances and key 
320: parameters of these models, including the stellar mass, radius, effective 
321: temperature, surface gravity, and ultraviolet flux, are summarized in 
322: Table~\ref{models_par}. The 48 species are connected through 281 reactions, 
323: including cosmic ray chemistry, photochemistry and the chemistry of excited 
324: H$_2$. We compute equilibrium chemistry using a modified Newton-Raphson 
325: algorithm. The solution then only depends on the element abundances and not 
326: on initial conditions.
327: 
328: We use the results of dust modeling by \citet{wah07} and assume large 
329: $30~\mu$m black body grains with radiative efficiencies of $Q_\lambda = 2\pi 
330: a/\lambda$ for $\lambda > 2\pi a$ and $Q_\lambda = 1$ otherwise. These 
331: grains are efficient absorbers and inefficient emitters, thus achieving dust 
332: radiative equilibrium temperatures of \begin{equation}
333: T_{\rm dust} = 324 \left(\frac{L_*}{L_\odot}\right)^{0.2} (a_{\rm \mu m})^{-0.2} (r_{\rm AU})^{-0.4}~~~{\rm K}\,\,\,.
334: \end{equation}
335: Here, $L_*$ and $L_\odot$ are the stellar and solar luminosity 
336: respectively, $a_{\rm \mu m}$ is the grain size in micron and $r_{\rm AU}$ the 
337: distance from the star in astronomical units.  The gas temperature is derived 
338: from a detailed energy balance including the most relevant heating and cooling 
339: processes \citep{kam01}.  
340: 
341: \begin{table}[htdp]
342: \caption{Element abundances and parameters used in the disk models}
343: \begin{center}
344: \begin{tabular}{lc}
345: \hline
346: Parameter$^a$      & Value \\
347: \hline
348: \hline
349: $A_{\rm He}$   &  $7.5 \times 10^{-2}$ \\ 
350: $A_{\rm C}$    &  $1.3 \times 10^{-4}$ \\ 
351: $A_{\rm O}$    &  $2.9 \times 10^{-4}$ \\ 
352: $A_{\rm Mg}$   &  $4.2 \times 10^{-6}$ \\
353: $A_{\rm Si}$   &  $8.0 \times 10^{-6}$ \\
354: $A_{\rm S}$    &  $1.9 \times 10^{-6}$ \\
355: $A_{\rm Fe}$   &  $4.3 \times 10^{-6}$ \\
356: T$_{\rm eff}$  &  10\,000~K            \\
357: $\log g$       & 4.5                   \\
358: R$_\ast$       & 1.7~R$_\odot$         \\
359: M$_\ast$       & 2.3~M$_\odot$         \\
360: $\sigma_{\rm UV}$ & $4.68\,10^{-24}~{\rm cm}^{-2}~{\rm H-atom}^{-1}$\\
361: \hline
362: \end{tabular}
363: \tablenotetext{a}{Gas-phase abundances ($A$) are relative to hydrogen.}
364: \end{center}
365: \label{models_par}
366: \end{table}
367: 
368: The radiation field consists of both stellar and interstellar components. 
369: The stellar properties are determined by a Kurucz model fit to photometric
370: points collected from the literature \citep{wah07,syl96,boc94,son04}; 
371: using T$_{\rm eff}$=10000~K and $\log g$ = 4.5,
372: consistent with the values quoted by \citet{che06}, the derived stellar 
373: luminosity is $L_*=26.1 L_\sun$ and the radius is 1.7 R$_\sun$. 
374: The spectral energy distribution and Kurucz model are plotted in Figure
375: \ref{fig:SED}, including dereddening according to extinction derived
376: by \citet{syl96} and using a \citet{car89} extinction law.  The solid
377: line in the figure denotes the fit to the photometry of a Kurucz stellar
378: atmosphere model at the Hipparcos distance of 61 pc.  The dashed line shows
379: the spectral energy distribution of the best-fit model of the outer disk as 
380: described in \S\ref{sec:best}. 
381: The interstellar radiation field in the ultraviolet is assumed to be 
382: $1.2 \times 10^7$ cm$^{-2}$ s$^{-1}$ \citep{hab68}.  
383: 
384: A basic model of the dust disk was constructed according to the 
385: Bayesian analysis of mid-infrared emission carried out by \citet{wah07}.  
386: Their model consists of an inner disk extending from 30 to 60 AU, composed 
387: primarily of small grains ($a \sim 0.1$ $\mu$m) with a surface density of 
388: $5\times10^{-8}$ g/cm$^2$, and an outer disk extending from 60 to 900 AU 
389: composed of larger grains ($a \sim 15$ $\mu$m) with a surface density of 
390: $3 \times 10^{-6}$ g/cm$^2$.  They derive a surface density distribution for 
391: the outer disk that is constant with radius, yielding a total disk mass of 
392: 0.35 M$_\earth$.  From the mid-IR images, they also determine a position angle 
393: of $125^\circ \pm 10^\circ$ (indicated in Figure \ref{fig:map}) and an 
394: inclination of $60^\circ \pm 15^\circ$.  We use this model as a starting
395: point for the disk structure, since it reflects the best available information
396: on the dust density distribution.  However, since the molecular gas emission
397: provides better constraints on some aspects of disk structure, including the 
398: vertical density distribution and the surface density structure of the outer 
399: disk, we introduce refinements to this initial model where justified, as 
400: described in \S\ref{sec:rad} and \S\ref{sec:grid} below.  For the large grain
401: population, our model uses 30~$\mu$m grains instead of 15~$\mu$m grains, 
402: although the grain size used in this simple model is highly degenerate with 
403: other disk properties, as discussed in \S\ref{sec:SED}.
404: 
405: To predict gas properties from this dust model, we make two 
406: primary assumptions: (1) gas and dust are well-mixed, (2) the gas:dust mass
407: ratio is constant.  We initially assume a constant scale height $H$=2\,AU,
408: since there is no information on disk scale height from the dust model of 
409: \citet{wah07}; we also begin by retaining the inner and outer radii and 
410: radially constant surface density structure from the \citet{wah07} model, 
411: although these assumptions are modified in \S\ref{sec:rad} below.
412: Throughout the modeling process, we use the canonical gas:dust mass ratio of 
413: 100 and  assume that the disk is embedded in interstellar material of density 
414: 10~cm$^{-3}$ to avoid model densities dropping to unrealistically low values 
415: near the boundaries of the numerical grid.
416: 
417: To compare our models with the SMA data, we use the radiative transfer 
418: code RATRAN \citep{hog00} to generate a sky-projected image of the CO J=2-1 
419: emission predicted for the physical model.  We then use the MIRIAD task {\em 
420: uvmodel} to sample the image with the combination of spatial frequencies 
421: and visibility weights appropriate for our SMA data.  We allow the inclination 
422: and position angle of the system to vary in order to best match the data.
423: 
424: \begin{center}
425: \begin{figure}[htp]
426: \centering
427: \includegraphics[scale=0.6,viewport=110 30 400 350]{f2.eps}
428: \caption{
429: Spectral energy distribution (de-reddened according to extinction derived 
430: by \citealt{syl96} and \citealt{car89} extinction law) for 49 Ceti using 
431: available optical, infrared, and submillimeter photometry.  The solid line
432: denotes a Kurucz stellar atmosphere model fitted to the photometry using
433: the Hipparcos distance of 61 pc. 
434: The dot-dashed line shows the SED for the best-fit model of the outer disk 
435: see text of \S\ref{sec:SED} for details.
436: }
437: \label{fig:SED}
438: \end{figure}
439: \end{center}
440: 
441: \subsubsection{Inner Disk}
442: 
443: In the inner disk, inside 60 AU, composed primarily of small grains, 
444: the stellar radiation field raises the dust temperature to 1000-2000 K and 
445: dissociates most of the molecular gas.  In this region, the dominant form of 
446: carbon is C$^+$, and even hydrogen is predominantly atomic.  We therefore 
447: ignore the inner disk component in subsequent modeling and focus on 
448: reproducing the observed CO emission with only the outer disk component.  
449: 
450: This lack of molecular gas in the inner disk is consistent with the 
451: non-detection of warm H$_2$ by \citet{che06} and \citet{car07}, and with the 
452: lack of high-velocity CO emission in Figure \ref{fig:map}.
453: The lack of CO emission more than 4.3 km/s from the stellar velocity is 
454: consistent with an absence of CO within a radius of $\sim90$ AU, 
455: for gas in Keplerian rotation around a star of 2.3~$M_{\sun}$.
456: %The agreement between the velocity structure of the data (Figure \ref{fig:map})
457: %and the model (Figure \ref{fig:model}) demonstrate that the velocity structure
458: %we observe in the 49 Ceti system is consistent with Keplerian rotation around a
459: %star of 2.3 $M_\sun$.  The lack of emission more than 4.3 km/s from the 
460: %velocity center therefore indicates a depletion of material within a distance 
461: %of $\sim 90$~AU from the star.
462: 
463: \subsubsection{Outer Disk}
464: \label{sec:rad}
465: 
466: There are three primary features of the observed CO emission from the outer 
467: disk that we attempted to reproduce with this modeling effort: (1) the 
468: separation of the emission peaks in the outer channels ($\sim3$"), (2) the 
469: spatial extent of the CO emission in all channels, and (3) the strength of the 
470: emission. Reproducing these features of the observed CO emission requires 
471: several modifications to the best-fit \citet{wah07} model of the outer 
472: dust disk, including adjustments to the inner and outer radii and a departure
473: from the constant surface density prescription. 
474: 
475: At first glance, the inner radius of 60 AU derived by \citet{wah07} might 
476: seem consistent with the lack of emission within 90 AU derived from the 
477: missing high-velocity wings in our data; however, there is a large region at 
478: the inner edge of the outer disk subject to photodissociation by stellar 
479: radiation which therefore contributes little to the CO emission.  In order to 
480: reproduce the separation of the emission peaks, material is required 
481: interior to this 60 AU radius.  We therefore take the uncertainties in the 
482: \citet{wah07} dust distribution into account and allow the inner disk radius to 
483: vary.  However, moving the inner radius closer than $\sim 40$ AU to the star 
484: results in high-velocity emission that we do not observe in the data, while 
485: still producing emission peaks wider than observed.  We therefore set the
486: disk inner radius at 40 AU, and then adjust the gas densities to further 
487: reduce the separation of the emission peaks.
488: 
489: Increasing the total gas mass leads to an elongated morphology with an aspect 
490: ratio larger than the observations, as the optical depth rises throughout the 
491: disk.  To meet the three criteria of (1) enough gas-phase CO near the inner 
492: disk edge to reproduce the observed peak separation, (2) low enough optical 
493: depth in the outer parts of the disk to keep the emission from becoming more
494: elongated than the data (through photodissociation by interstellar UV 
495: photons), and (3) maintaining an inner radius large enough to avoid generating 
496: high-velocity emission that is not present in the data, we must ``pile up'' 
497: material at the inner disk edge to enhance shielding and concentrate 
498: emission.  We therefore modify the initial assumption of constant surface 
499: density as derived from the infrared analysis, instead adopting an 
500: $r^{-\epsilon}$ density profile.  We simultaneously relax the constant scale
501: height assumption, introducing a scale height $H$ that increases linearly 
502: with radius $r$, with proportionality constant $h=H/r$. The full 2-D density 
503: structure then becomes $n(r,z) = r^{-\epsilon} \exp{(-z^2/2 H^2)}$, where the 
504: exponent $\epsilon$ and scale height constant $h$ are varied to obtain the 
505: best fit to the CO data.  
506: 
507: The power-law surface density profile results in a much better match between 
508: the model and the observed emission peak separation.  It also curbs the 
509: elongation of the emission to some extent, as the vertical column density of 
510: the outer disk drops and the material far from the star becomes subject to 
511: dissociation by interstellar radiation. However, even steep power law indices 
512: for the surface density profile do not result in a completely photoevaporated 
513: outer disk and consequently produce emission that is much more elongated than 
514: observed.  In a next step, we therefore reduce the outer radius from 900 to 
515: 200~AU.  While this is at the lower end of the range allowed by \citet{wah07}, 
516: their derived outer radius was based largely on the uncertain millimeter flux 
517: measurement, and the gas geometry is likely a better probe of the disk extent.
518: 
519: \subsection{Grid of Disk Models}
520: \label{sec:grid}
521: 
522: After these initial studies of the outer disk, it became clear that several 
523: model parameters were ill-constrained by previously existing data.
524: Specifically, the disk mass is constrained only by the weakly-detected and 
525: contradictory millimeter flux measurements; similarly, the density power law 
526: index $\epsilon$ is ill-determined by the infrared observations, which are 
527: primarily sensitive to inner disk emission. The scale height $h$ is also 
528: completely unconstrained by the continuum or single-dish measurements, 
529: neither of which is sensitive to disk structure in the vertical direction.  
530: The disk geometry (PA and inclination) quoted by \citet{wah07} is also 
531: subject to large uncertainties, due to the irregular shape of the emission 
532: observed in the infrared.  We therefore attempt to better constrain these 
533: disk parameters by using our resolved CO gas line observations. Gas lines 
534: are generally more sensitive than dust emission to temperature and density 
535: gradients, and can thus provide means to break model degeneracies. We 
536: ran grids of models for the three structural parameters (disk mass, 
537: density index, scale height) and two geometrical parameters (PA, 
538: inclination), finding the best-fit values by calculating and minimizing 
539: a $\chi^2$ value comparing the model to the observed emission from the disk.  
540: Due to the computational intensity of the calculations necessary to 
541: determine the chemistry and radiative transfer solutions for each model, 
542: we ran only a sparsely sampled grid of models.  In order to ensure that the
543: final model reflects all available observational constraints, we centered the 
544: grid on the fiducial model of \S\ref{sec:mod} and adjusted the parameters only 
545: as necessary to better reproduce the new CO(2-1) observations, moving from 
546: coarse to fine grids to ensure adequate exploration of the parameter space.  
547: We use the modeling primarily as a demonstration that the basic features of 
548: the observed CO emission can be reproduced by a simple azimuthally symmetric 
549: model of disk structure; the ``best-fit'' model should therefore be viewed 
550: as representative of an initial understanding of the features of the 
551: system rather than as a conclusive determination of the disk structural 
552: parameters.
553: 
554: \subsubsection{CO Chemistry Across the Model Grid}
555: 
556: The CO chemistry is dominated by photodissociation in a number of UV bands
557: and thus the abundance of CO in each model is mostly dependent on the radial
558: and vertical column densities being able to shield the stellar and interstellar
559: UV radiation respectively. In the following we briefly discuss some basic 
560: characteristics of the model grid.
561: 
562: The surface density in the models is independent of the scale height and hence
563: the radial mass distribution in each model can be written as $M(R) \propto R^{-
564: \epsilon +3}$, where $M(R)$ denotes the mass inside a radius $R$. So, as we
565: increase the density power law exponent $\epsilon$, the inner region of the 
566: outer disk harbors a larger fraction of the total mass. The densities in this 
567: region of the disk become higher and hence it is easier to obtain the critical 
568: column densities necessary for UV shielding in the radial direction. On the 
569: other hand, a shallower gradient for the density distribution 
570: translates into higher densities in the outer parts of the disk, thus 
571: enhancing the vertical shielding in the outer disk compared to models with 
572: high $\epsilon$.  None of our models is optically thick in the dust continuum, 
573: so the UV shielding is mainly H$_2$ shielding of the CO bands due to their 
574: overlap in wavelengths; CO self-shielding also plays a role.
575: 
576: With this basic picture, we can understand the CO chemistry displayed in Fig.~
577: \ref{COoverview1} as a function of disk mass (right column) and density 
578: gradient $\epsilon$ (center column).
579: As the total disk mass is increased, CO first starts to build up in the radial 
580: direction.  It can still be dissociated by the vertically impinging 
581: interstellar UV radiation field in the outer regions of the disk (150-200~AU) 
582: until the disk reaches a mass of $\sim17$~M$_{\earth}$, at which point it 
583: becomes opaque in the CO bands even in the vertical direction. A shallow 
584: density gradient always leads to smaller radial column densities at the same 
585: reference radius, thus pushing the C$^+$/C/CO transition further out in radial 
586: distance. In our best-fit model of 13~M$_{\earth}$, a change in $\epsilon$ 
587: from 2.5 to 1.1 changes the radius for the C$^+$/C/CO front from close to 
588: 40~AU to 190~AU. 
589: 
590: \begin{figure*}
591: \centering
592: \includegraphics[width=0.99\textwidth]{f3.eps}
593: \caption{
594: Two-dimensional CO abundances in a subset of disk models.  The center panel 
595: shows the best-fit model ($M=13$M$_\earth$, $\epsilon = 2.5$, $h=0.02$), while 
596: the rows of models above and below show the effects of incrementing and 
597: decrementing, respectively, each of the three structural parameters that were 
598: allowed to vary during the fitting process: $h$ {\em (left column)}, $\epsilon$ 
599: {\em (center column)}, and M$_{disk}$ {\em (right column)}.  The values of the
600: parameters shown are $h= 0.01$, $0.03$; $\epsilon=1.1$, $3.5$; and $M=9$, $17$ 
601: M$_\earth$.  
602: }
603: \label{COoverview1}
604: \end{figure*}
605: 
606: The scale height $h$ of the models affects only the vertical density structure 
607: in the models. However, since density and chemistry are closely intertwined, it 
608: can strongly impact the overall radial and vertical structure of the CO 
609: chemistry.  From a comparison of the center panel with the bottom left panel in
610: Figure \ref{COoverview1}, we see that a factor 2 lower scale height with 
611: respect to the best fit model ($h=0.02$), enhances the CO abundance in the disk 
612: significantly, leading to radial and vertical column densities that are more 
613: than a factor 10 higher with respect to the best fit model. The total CO mass 
614: increases by a factor of 10 as well, with the integrated emission undergoing
615: a corresponding dramatic increase.  
616: 
617: Table~\ref{grid_CO_tab} displays some key results from a subset of grid models
618: such as characteristic radial and vertical CO column densities, CO masses and
619: total CO J=2-1 line emission.  For all models in the table, the inner radius 
620: is fixed at 40~AU and the outer radius at 200~AU.  
621: 
622: \begin{table*}[htdp]
623: \caption{Derived quantities from a subset of the 49 Ceti disk models}
624: \begin{center}
625: \begin{tabular}{rccrrrr}
626: \hline
627: $M_{\rm disk}^a$ & $\epsilon$ & $h$ & $N({\rm CO})_{\rm radial}^b$ & $N({\rm
628: CO})_{\rm vertical}^{\rm 100 AU,c}$  & $M_{\rm CO}^d$ & $I_{\rm CO}$(J=2-1)$^e$ \\
629:  (M$_{\earth})$  &                      &               &  
630: ($10^{18}$\,cm$^{-2}$)     &  
631: ($10^ {15}$\,cm$^{-2}$)
632:  & ($10^{-4}$\,M$_{\earth}$)    & (Jy km s$^{-1}$) \\
633: \hline
634: \hline
635: 13 & 2.5 & 0.020 &   2.76 &  4.23  &  9.66 &  2.6 \\
636:  9 & 2.5 & 0.020 &   0.32 &  1.82  &  2.46 &  1.2 \\    
637: 17 & 2.5 & 0.020 &   13.5 &  9.06  &  37.2 &  6.9 \\
638: 13 & 3.5 & 0.020 &   15.1 &  4.47  &  98.0 & 11.7 \\
639: 13 & 1.1 & 0.020 &   0.13 &  0.91  &  3.74 &  2.3 \\
640: 13 & 2.5 & 0.010 &   42.8 &  78.4  &  96.6 & 14.5 \\
641: 13 & 2.5 & 0.030 &   0.12 &  2.20  &  2.97 &  1.5 \\
642: \hline
643: \end{tabular}
644: \tablenotetext{a}{Total disk gas mass}
645: \tablenotetext{b}{Total radial CO column density through the midplane}
646: \tablenotetext{c}{CO vertical column density at 100 AU}
647: \tablenotetext{d}{Total CO mass in the disk}
648: \tablenotetext{e}{Integrated CO(J=2-1) line emission}
649: \end{center}
650: \label{grid_CO_tab}
651: \end{table*}
652: 
653: 
654: \subsubsection{From Chemistry to Observables}
655: 
656: The predicted CO J=2-1 emission for the models in Figure~\ref{COoverview1} 
657: is displayed in Figure~\ref{fig:model}; a comparison of these figures
658: illustrates the ways in which differences in chemical structure are manifested
659: in the observable properties of the CO emission.  The CO emission is sampled
660: with the same spatial frequencies and visibility weights as the SMA data and 
661: displayed in renzogram form with the same velocity structure as in 
662: Figure~\ref{fig:map}.  In order to emphasize the relative structural 
663: differences between models, the contour levels are 15\% of the peak flux for 
664: each model, with the absolute flux indicated by the thickness of the 
665: contours, and also printed explicitly at the top of each panel.  
666: 
667: \begin{figure*}
668: \centering
669: \includegraphics[angle=90,totalheight=0.66\textheight]{f4.eps}
670: \caption{
671: CO J=2-1 emission predicted for the subset of models shown in Figure 
672: \ref{COoverview1}, sampled with the same spatial frequencies and visibility 
673: weights as the SMA data in Figure~\ref{fig:map}.  The center panel shows the 
674: best-fit model, while the rows of models above and below show the effects of 
675: incrementing and decrementing, respectively, each of the three structural 
676: parameters that we allowed to vary during the fitting process: $h$ {\em (left 
677: column)}, $\epsilon$ {\em (center column)}, and M$_{disk}$ 
678: {\em (right column)}.  The contour levels are displayed in the upper left 
679: corner of each panel; they are set at 3 and 5 $\times$ 15\% of the peak flux 
680: for each model. The thickness of the contours is proportional to the absolute 
681: flux: thicker contours indicate that the source is brighter than the data, 
682: while thinner contours indicate that it is fainter than the data.  The contour
683: levels in the center panel are identical to those in Figure~\ref{fig:map}.  
684: Table \ref{tab:best} gives the full list of parameters for the best-fit model.  
685: }
686: \label{fig:model}
687: \end{figure*}
688: 
689: The decreased shielding in the inner disk caused by reducing the density 
690: gradient $\epsilon$ is visible as a lengthening of the emission in the central
691: channels and a widening of the emission peaks in the outer channels in the 
692: low-epsilon model (bottom center panel).  Increasing $\epsilon$ (top center 
693: panel) leads to enhanced shielding at the disk inner edge, causing much higher 
694: CO fluxes in the outer part of the disk and extremely high contrast between the 
695: inner and outer velocity channels. 
696: 
697: The primary observable consequence of adjusting the mass (right panels, top 
698: and bottom) is that the increased or decreased shielding from extra gas leads 
699: to a corresponding increase or decrease in the total CO flux; changes to the 
700: shape of the emission are minimal, and the primary difference between models 
701: of different mass over the mass range under consideration is simply in the 
702: relative brightness of the emission.  
703: 
704: Differences in the scale height of the disk similarly manifest as differences
705: in the flux scale; however, decreasing the scale height (bottom left panel) 
706: also causes greater shielding at the inner disk edge, leading to greater 
707: elongation of the outer velocity channels and causing the inner velocity 
708: channels to draw together and overlap as the CO flux rises throughout the 
709: inner areas of the disk.  An increase in scale height (top left panel) leads 
710: to a greater area in the front and back of the disk, projected along 
711: our line of sight, which increases the flux in the central channels and leads 
712: to a lower contrast between the inner and outer channels of the disk.
713: 
714: 
715: \subsection{Spectral Energy Distribution}
716: \label{sec:SED}
717: 
718: After converging initially on a model that was able to reproduce the observed
719: CO J=2-1 emission, we used that model to predict the spectral energy 
720: distribution.  This serves as an {\em a posteriori} test of the consistency 
721: between the gas and dust properties in the models and the available observables.
722: 
723: We integrate over the disk volume to obtain the flux as a function of 
724: wavelength 
725: \begin{equation}
726: F_\lambda = (\pi a^2 / d^2) \int \int 2 \pi r \, B_\lambda(T_{dust}(r,z)) n_{dust}(r,z) \, Q(\lambda) \, dz \, dr\,\,\,, 
727: \end{equation}
728: where $d$ is the distance to the source and $n_{dust}$ is the number density of 
729: dust grains in cm$^{-3}$.  We assume throughout a grain density of 2.5 g/cm$^3$.
730: 
731: While the predicted shape of the spectral energy distribution matches the
732: observations well, the absolute fluxes are too high by a factor of $\sim 5$.  
733: Adjusting the temperature of the dust grains alters the shape of the SED curve, 
734: causing it to deviate from the observed shape; we were therefore required to 
735: increase the gas:dust ratio from 100 to 500 in order to reproduce the observed 
736: photometry.  This unusually high ratio is likely an artifact of the simple 
737: assumptions of the model, since little information is available about the dust
738: distribution in this system (and none at all from our data).  For example, the 
739: mass of the system is likely not all in 30 $\mu$m grains, and a significant 
740: fraction of the mass may be in larger grains that contribute little to the 
741: infrared emission.  Another possibility is that the overall gas:dust ratio is
742: consistent with the standard value, but that gas and dust are not well-mixed:
743: for example, much of the excess emission may arise from the inner edge of the
744: disk, which will be directly illuminated and heated by the stellar radiation.
745: Resolved observations of the dust continuum emission would test this hypothesis
746: by placing constraints on the spatial distribution of the emitting region.
747: Including effects such as this would significantly complicate the model
748: presented here, as the H$_2$ formation rate would be affected by varying the
749: abundance of the dust on which it forms.  In general, the dust size and 
750: gas:dust mass ratio are strongly related by the total dust surface required 
751: to maintain the observed quantity of molecular gas; these are in turn 
752: dependent on the stellar properties determining dust grain temperatures.  None
753: of these dust-dependent quantities are well constrained by available data.  
754: Given the observations available and the extremely simplified dust model, 
755: which not only neglects the size distribution but also the possibility of a 
756: mixture of compositions and opacities, we use the simplest assumption of an 
757: altered gas:dust ratio in order to conduct a consistency check of the 
758: temperature and density structure of the gas model.
759: 
760: Decreasing the total dust mass in the model to match the SED reduces the grain 
761: surface area for H$_2$ formation. Thus molecular hydrogen begins to form at 
762: larger radii and greater depth, compared to the initial model with the 
763: canonical gas:dust ratio of 100.  As a consequence of less effective UV 
764: shielding, the total CO mass decreases. Hence the total mass of the best-fit 
765: model has to be increased slightly to compensate for the lower molecular gas 
766: fraction. As a secondary effect, the overall gas temperature of the 
767: dust-depleted model also decreases due to the diminished photoelectric 
768: heating in the disk.  The corresponding SED predicted for these parameters 
769: is indicated by the dashed line in Figure \ref{fig:SED}.  The mid-infrared 
770: flux points are underestimated by this model because we do not include the 
771: inner disk component of \citet{wah07}; as our data provide no constraints 
772: on the properties of the inner disk, we ignore this component and 
773: concentrate on the fit to the outer disk.  The flux predicted by the model 
774: SED is consistent with our own continuum upper limit reported in 
775: Table~\ref{tab}.
776: 
777: \subsection{Best-Fit Disk Model}
778: \label{sec:best}
779: 
780: The center panel of Figure~\ref{fig:model} shows the best-fit model from the 
781: grid, with the minor modifications introduced by reproducing simultaneously the 
782: spectral energy distribution.  The structural and geometric parameters for this 
783: model are listed in Table \ref{tab:best}.  The errors given in the table are
784: the approximate 1-$\sigma$ uncertainty range interpolated from the $\chi^2$ 
785: grid.
786: 
787: \begin{table}
788: \caption{Parameters for Best-Fit Disk Model}
789: \begin{center}
790: \begin{tabular}{lc}
791: \hline
792: $h$ & $0.020^{+0.015}_{-0.005}$ \\
793: $\epsilon$ & $2.5^{+0.5}_{-1.0}$ \\
794: $M_{gas}$ & $13 \pm 3$ M$_\earth$ \\
795: $M_{dust}$ & $0.02 \pm 0.01$ M$_\earth$ \\
796: $i$ & $90^\circ \pm 5^\circ$ \\
797: PA & $-70^\circ \pm 10^\circ$ \\
798: $R_{in}$ & 40 AU$^a$ \\
799: $R_{out}$ & 200 AU$^a$ \\
800: \hline
801: \end{tabular}
802: \end{center}
803: \tablenotetext{a}{For a description of the constraints on the inner and
804: outer radii, see \S\ref{sec:rad}}
805: \label{tab:best}
806: \end{table}
807: 
808: This model reproduces the basic features of the CO J=2-1 emission well,
809: including the strength of the emission, the separation of the emission peaks,
810: and the spatial extent of the emission.  There are still several important
811: differences between the model and the data, however, including (1) an inability
812: to reproduce the changes in position angle with radius evident in the data
813: (the ``wings'' of emission extending to the southeast and northwest of the
814: position angle axis), and (2) the separation of the innermost, low-velocity 
815: channels.  Both of these may be indicative of departures from azimuthal 
816: symmetry in the disk structure, the former possibly indicating a warp in the
817: disk and the latter apparently pointing to a deficit of emission along the 
818: minor axis of the disk.  In none of our models were we able to reproduce the 
819: wide separation between the inner channels; while the signal-to-noise ratio in 
820: these channels is low, the observed CO morphology is difficult to reproduce in
821: detail with a simple, azimuthally symmetric disk model.  The CO emission for
822: this best-fit model is optically thin in both the J=2-1 and J=3-2 transitions,
823: even for the edge-on disk orientation, and therefore traces the full column 
824: density of disk material.
825: 
826: The densities in the disk are too low for efficient gas-dust coupling and 
827: thus the gas finds its own equilibrium temperature determined mainly by 
828: photoelectric heating and line cooling. The most important cooling lines from 
829: surface to midplane are [C\,{\sc ii}], [O\,{\sc i}], and H$_2$. CO abundances 
830: are only high in a region between 45 and 70~AU (Fig.~\ref{COoverview1}). 
831: Outside that region, CO cooling is less important for the energy balance. 
832: Fig.~\ref{gastemp} summarizes the most important heating and cooling processes 
833: and also shows the resulting gas temperature structure. The disk surface stays 
834: well below 100~K due to efficient fine structure line cooling. The molecular 
835: cooling is however less efficient in competing with the photoelectric heating 
836: from the large silicate grains \citep{kam01}, leading to temperatures of a few 
837: 100~K in the disk interior.
838: 
839: \begin{figure}
840: \centering
841: \includegraphics[scale=0.9]{f5.eps}
842: \caption{
843: Two-dimensional gas temperatures in the best fit disk model ($M_{\rm disk}
844: =13$~M$_\earth$,
845: $\epsilon= 2.5$, $h=0.02$. Shown are the most important heating (top panel) 
846: and cooling (middle panel) processes as well as the gas temperature (bottom 
847: panel). The dust temperature, which depends only on radius, is overlaid in
848: white 
849: contour lines (steps of 10~K).
850: }
851: \label{gastemp}
852: \end{figure}
853: 
854: In order to test the robustness of the best-fit model to the gas properties,
855: we used this model to predict the CO J=3-2 spectrum.  It compares
856: favorably with the spectrum observed by \citet{den05}, reproduced in  
857: Figure~\ref{fig:spec}.  The heavy solid line shows the J=3-2 spectrum predicted 
858: from the best-fit disk model, while the light solid line shows the 
859: observed JCMT spectrum.  
860: %The x-axis shows heliocentric velocity in km/s, while 
861: %the y-axis gives the JCMT main beam brightness temperature, assuming a 
862: %main beam efficiency of 0.62.  
863: Although the observed spectrum is noisy and likely 
864: subject to an absolute calibration uncertainty, the overall agreement is 
865: within $\sim 30$\%, which is very good given that the CO J=3-2 spectrum was 
866: not used {\em a priori} to determine these model parameters.
867: 
868: \begin{figure}
869: \centering
870: \includegraphics[scale=0.7]{f6.eps}
871: \caption{
872: CO J=3-2 spectrum predicted for the model that provides the best fit to the 
873: resolved J=2-1 emission (heavy solid line), compared with the \citet{den05} 
874: JCMT CO J=3-2 spectrum (light solid line).  The x-axis shows heliocentric
875: velocity while the y-axis gives the JCMT main beam brightness temperature.  
876: }
877: \label{fig:spec}
878: \end{figure}
879: 
880: \section{Discussion}
881: \label{sec:dis}
882: 
883: The processes determining the amount and distribution of gas and dust in
884: transition disks like that around 49 Ceti are the same processes that shape 
885: the features of emergent planetary systems around these young stars. Resolved 
886: observations of individual disks in this phase are desirable to address such 
887: basic questions as when in the lifetime of a star its disk disperses, whether 
888: the gas clears before the dust, and whether the disk clears from the center or 
889: in a radially invariant manner.  
890: 
891: In the 49 Ceti system, the infrared dust properties appear similar to those of
892: a debris disk \citep{wah07}.  Yet observations presented here indicate that 
893: a substantial quantity of molecular gas persists in the outer disk, between
894: radii of 40 and 200 AU, where photochemistry from stellar and interstellar 
895: radiation dominates.  The lack of molecular gas emission interior to this 
896: radius as indicated by our observations, combined with the lack of dust emission
897: within a radius of 30 AU inferred by \citet{wah07}, implies that the 49 Ceti
898: system appears to be clearing its gas and dust from the center out.  
899: The mechanism responsible for this central clearing is not 
900: indicated; in general, the best-developed theories to explain this transitional
901: morphology are (1) central clearing through the influence of a massive planet 
902: and (2) photoevaporation by radiation from the central star.  
903: 
904: The clearing of gaps and inner holes has long been predicted as a consequence
905: of the formation of massive planets in circumstellar disks 
906: \citep[e.g.][]{lin86,bry99}.  In the case of 49 Ceti, the formation of a 
907: Jupiter-mass planet would be required at a distance of $\sim 40$ AU from the 
908: star, roughly the inner radius of the observed hole in the gas distribution.  
909: Such a scenario could also help to explain the size segregation of dust grains 
910: observed by \citet{wah07}; a predicted consequence of inner disk clearing 
911: by gravitational influence of a massive planet is a filtration of dust grains
912: by size, with only those below a certain threshold (typically 1-10 $\mu$m) 
913: accreted across the gap along with a reduced amount of gas \citep{ric06}.  
914: However, this scenario ultimately requires the accretion of substantial 
915: amounts of gas into the inner disk, and searches for molecular gas in the 
916: inner disk of 49 Ceti \citep{che06,car07} have not detected such a population. 
917: Another indication that an inner hole is likely caused by a massive planet
918: in formation would be non-axisymmetric features resulting from its
919: gravitational influence, such as spiral waves. While the CO emission from
920: 49 Ceti does not appear asymmetric within the limits of the SMA observations,
921: more sensitive spatially resolved observations could address this hypothesis.
922: 
923: The absence of gas in the inner disk is, however, consistent with a 
924: photoevaporation scenario: as the photoevaporative wind produced by stellar
925: radiation becomes comparable to the accretion rate in the disk, material 
926: within the gravitational radius $R_g = GM_\star / c_s^2$ will quickly drain 
927: onto the star, leaving an evacuated inner hole free of gas and dust 
928: \citep[e.g.][]{hol94,ale06}.  The gravitational radius for 49 Ceti is roughly
929: 20 AU, which is comparable to the inferred inner radius of 40 AU
930: for the outer disk.  The larger outer radius may in fact be consistent with
931: the later stages of photoevaporation, after the inner disk has become optically
932: thin to ultraviolet radiation and the inner disk radius slowly increases under 
933: the influence of the photoevaporative wind \citep{ale06}.  
934: \citet{ale07} propose a method of discriminating between inner holes caused
935: by photoevaporation and those caused by the formation of a giant planet, 
936: involving a simple comparison between two observables: the disk mass and the 
937: accretion rate.  As there is no measured accretion rate for 49 Ceti, we cannot
938: apply the criteria presented by these authors; however, we note that the low
939: disk mass does indeed fall within the parameter space consistent with 
940: a photoevaporative scenario.  Further observations are necessary to determine
941: the origin of the inner hole; in particular, stringent limits on the accretion
942: rate could suggest a photoevaporative mechanism. 
943: 
944: There are few disks which appear to be in a similar evolutionary stage to that
945: of 49 Ceti; a rare example is the disk around the A star HD 141569.  Like 49 
946: Ceti, it hosts a disk composed primarily of sub$\mu$m-size grains with infrared 
947: properties approaching those of a debris disk \citep{wah07,mar02}, while still 
948: retaining a substantial quantity of molecular gas with central region clear of 
949: gas emission, in this case out to a radius of $\sim11$ AU \citep{got06,bri07}.  
950: It exhibits a transitional SED \citep{mer04}, and observations of the 
951: rovibrational CO spectrum reveal gas with disparate rotational and vibrational 
952: temperatures \citep[200 K and 5000 K respectively; ][]{bri07}, indicative of UV 
953: fluorescence on the outer edges of an inner disk region cleared of gas and 
954: dust.  An analysis of the chemistry and gas properties of the system similar 
955: to the one presented here for 49 Ceti was conducted by \citet{jon06}. While 
956: the presence and extent of the inner hole are clearly indicated, the physical 
957: origin of this clearing is less obvious.  The Br$\gamma$ profile is indicative 
958: of substantial accretion, and \citet{bri07} deem a photoevaporative clearing 
959: mechanism unlikely due to the large column density outside the cleared region 
960: and the lack of evidence for a photoevaporative wind in the FUV \citep{mar05}.  
961: However, \citet{mer04} place a much lower limit of $10^{-11}$ M$_\sun$/yr on the
962: accretion rate, based on the assumed gas:dust ratio of 100 and the low optical
963: depth of the inner disk, which would be much more consistent with a 
964: photoevaporation scenario.  \citet{got06} note that the rough coincidence of
965: the inner rim of the disk with the gravitational radius suggests that 
966: photoevaporation in concert with viscous accretion is a likely cause for the 
967: inner disk clearing. 
968: 
969: Whatever the origin of their morphology, the observed gas and dust properties
970: indicate that the disks surrounding both 49 Ceti and HD 141569 appear to be in 
971: a transitional state of evolution during which the dust properties are 
972: beginning to appear more like those of a debris disk, while the gas is in the 
973: process of being cleared from the disk from the center out.  
974: 
975: \section{Conclusions}
976: \label{sec:con}
977: 
978: The SMA CO J=2-1 observations presented here provide the first spatially 
979: resolved observations of molecular gas in the 49 Ceti system.  The data reveal 
980: a surprisingly extended and complex molecular gas distribution in rotation 
981: about the central star, viewed approximately edge on and clear of molecular gas 
982: emission in the central region of the disk.  Modeling the disk structure and 
983: chemistry in this system indicates that the inner disk is entirely devoid of 
984: molecular gas due to irradiation by the central star, while a ring of molecular 
985: gas persists between 40 and 200 AU, subject to photodissociation 
986: at the inner edge by stellar radiation.  The disk model presented here 
987: reproduces well the observed properties of the system, including the resolved 
988: CO J=2-1 emission, the CO J=3-2 spectrum, and the spectral energy distribution.
989: With dust properties similar to those of a debris disk and a substantial 
990: reservoir of gas maintained in the outer disk, 49 Ceti appears to be a rare 
991: example of a system in a late stage of transition between a gas-rich 
992: protoplanetary disk and a tenuous, gas-free debris disk.
993: 
994: 
995: \acknowledgements
996: The authors would like to thank Bill Dent for providing the JCMT CO J=3-2
997: spectrum.  Partial support for this work was provided by NASA Origins
998: of Solar Systems Program Grant NAG5-11777.  A. M. H. acknowledges support
999: from a National Science Foundation Graduate Research Fellowship.
1000: 
1001: 
1002: \begin{thebibliography}{40}
1003: 
1004: \bibitem[{{Alexander} \& {Armitage}(2007)}]{ale07}
1005: {Alexander}, R.~D. \& {Armitage}, P.~J. 2007, \mnras, 375, 500
1006: 
1007: \bibitem[{{Alexander} {et~al.}(2006){Alexander}, {Clarke}, \&
1008:   {Pringle}}]{ale06}
1009: {Alexander}, R.~D., {Clarke}, C.~J., \& {Pringle}, J.~E. 2006, \mnras, 369, 229
1010: 
1011: \bibitem[{{Aumann} {et~al.}(1984){Aumann}, {Beichman}, {Gillett}, {de Jong},
1012:   {Houck}, {Low}, {Neugebauer}, {Walker}, \& {Wesselius}}]{aum84}
1013: {Aumann}, H.~H., et~al. 1984, \apjl, 278, L23
1014: 
1015: \bibitem[{{Barrado y Navascu{\'e}s} {et~al.}(1999){Barrado y Navascu{\'e}s},
1016:   {Stauffer}, {Song}, \& {Caillault}}]{bar99}
1017: {Barrado y Navascu{\'e}s}, D., {Stauffer}, J.~R., {Song}, I., \& {Caillault},
1018:   J.-P. 1999, \apjl, 520, L123
1019: 
1020: \bibitem[{{Beckwith} \& {Sargent}(1991)}]{bec91}
1021: {Beckwith}, S.~V.~W. \& {Sargent}, A.~I. 1991, \apj, 381, 250
1022: 
1023: \bibitem[{{Bockel{\'e}e-Morvan} {et~al.}(1994){Bockel{\'e}e-Morvan},
1024:   {Andr{\'e}}, {Colom}, {Colas}, {Crovisier}, {Despois}, \& {Jorda}}]{boc94}
1025: {Bockel{\'e}e-Morvan}, D., {Andr{\'e}}, P., {Colom}, P., {Colas}, F.,
1026:   {Crovisier}, J., {Despois}, D., \& {Jorda}, L. 1994, in Circumstellar Dust
1027:   Disks and Planet Formation, ed. R.~{Ferlet} \& A.~{Vidal-Madjar}, 341--+
1028: 
1029: \bibitem[{{Brittain} {et~al.}(2007){Brittain}, {Simon}, {Najita}, \&
1030:   {Rettig}}]{bri07}
1031: {Brittain}, S.~D., {Simon}, T., {Najita}, J.~R., \& {Rettig}, T.~W. 2007, \apj,
1032:   659, 685
1033: 
1034: \bibitem[{{Bryden} {et~al.}(1999){Bryden}, {Chen}, {Lin}, {Nelson}, \&
1035:   {Papaloizou}}]{bry99}
1036: {Bryden}, G., {Chen}, X., {Lin}, D.~N.~C., {Nelson}, R.~P., \& {Papaloizou},
1037:   J.~C.~B. 1999, \apj, 514, 344
1038: 
1039: \bibitem[{{Cardelli} {et~al.}(1989){Cardelli}, {Clayton}, \& {Mathis}}]{car89}
1040: {Cardelli}, J.~A., {Clayton}, G.~C., \& {Mathis}, J.~S. 1989, \apj, 345, 245
1041: 
1042: \bibitem[{{Carmona} {et~al.}(2007){Carmona}, {van den Ancker}, {Henning},
1043:   {Goto}, {Fedele}, \& {Stecklum}}]{car07}
1044: {Carmona}, A., {van den Ancker}, M.~E., {Henning}, T., {Goto}, M., {Fedele},
1045:   D., \& {Stecklum}, B. 2007, ArXiv e-prints, 710
1046: 
1047: \bibitem[{{Chen} {et~al.}(2006){Chen}, {Sargent}, {Bohac}, {Kim},
1048:   {Leibensperger}, {Jura}, {Najita}, {Forrest}, {Watson}, {Sloan}, \&
1049:   {Keller}}]{che06}
1050: {Chen}, C.~H., et~al. 2006, \apjs, 166, 351
1051: 
1052: \bibitem[{{Dent} {et~al.}(2005){Dent}, {Greaves}, \& {Coulson}}]{den05}
1053: {Dent}, W.~R.~F., {Greaves}, J.~S., \& {Coulson}, I.~M. 2005, \mnras, 359, 663
1054: 
1055: \bibitem[{{Goto} {et~al.}(2006){Goto}, {Usuda}, {Dullemond}, {Henning}, {Linz},
1056:   {Stecklum}, \& {Suto}}]{got06}
1057: {Goto}, M., {Usuda}, T., {Dullemond}, C.~P., {Henning}, T., {Linz}, H.,
1058:   {Stecklum}, B., \& {Suto}, H. 2006, \apj, 652, 758
1059: 
1060: \bibitem[{{Habing}(1968)}]{hab68}
1061: {Habing}, H.~J. 1968, \bain, 19, 421
1062: 
1063: \bibitem[{{Hoffleit} \& {Jaschek}(1991)}]{hof91}
1064: {Hoffleit}, D. \& {Jaschek}, C.~. 1991, {The Bright star catalogue} (New Haven,
1065:   Conn.: Yale University Observatory, |c1991, 5th rev.ed., edited by Hoffleit,
1066:   Dorrit; Jaschek, Carlos |v(coll.))
1067: 
1068: \bibitem[{{Hogerheijde} \& {van der Tak}(2000)}]{hog00}
1069: {Hogerheijde}, M.~R. \& {van der Tak}, F.~F.~S. 2000, \aap, 362, 697
1070: 
1071: \bibitem[{{Holland} {et~al.}(1998){Holland}, {Greaves}, {Zuckerman}, {Webb},
1072:   {McCarthy}, {Coulson}, {Walther}, {Dent}, {Gear}, \& {Robson}}]{hol98}
1073: {Holland}, W.~S., et~al. 1998, \nat, 392, 788
1074: 
1075: \bibitem[{{Hollenbach} {et~al.}(1994){Hollenbach}, {Johnstone}, {Lizano}, \&
1076:   {Shu}}]{hol94}
1077: {Hollenbach}, D., {Johnstone}, D., {Lizano}, S., \& {Shu}, F. 1994, \apj, 428,
1078:   654
1079: 
1080: \bibitem[{{Jonkheid} {et~al.}(2006){Jonkheid}, {Kamp}, {Augereau}, \& {van
1081:   Dishoeck}}]{jon06}
1082: {Jonkheid}, B., {Kamp}, I., {Augereau}, J.-C., \& {van Dishoeck}, E.~F. 2006,
1083:   \aap, 453, 163
1084: 
1085: \bibitem[{{Jura} {et~al.}(1998){Jura}, {Malkan}, {White}, {Telesco}, {Pina}, \&
1086:   {Fisher}}]{jur98}
1087: {Jura}, M., {Malkan}, M., {White}, R., {Telesco}, C., {Pina}, R., \& {Fisher},
1088:   R.~S. 1998, \apj, 505, 897
1089: 
1090: \bibitem[{{Jura} {et~al.}(1993){Jura}, {Zuckerman}, {Becklin}, \&
1091:   {Smith}}]{jur93}
1092: {Jura}, M., {Zuckerman}, B., {Becklin}, E.~E., \& {Smith}, R.~C. 1993, \apjl,
1093:   418, L37+
1094: 
1095: \bibitem[{{Kamp} \& {Bertoldi}(2000)}]{kam00}
1096: {Kamp}, I. \& {Bertoldi}, F. 2000, \aap, 353, 276
1097: 
1098: \bibitem[{{Kamp} \& {van Zadelhoff}(2001)}]{kam01}
1099: {Kamp}, I. \& {van Zadelhoff}, G.-J. 2001, \aap, 373, 641
1100: 
1101: \bibitem[{{Lin} \& {Papaloizou}(1986)}]{lin86}
1102: {Lin}, D.~N.~C. \& {Papaloizou}, J. 1986, \apj, 309, 846
1103: 
1104: \bibitem[{{Mamajek}(2005)}]{mam05}
1105: {Mamajek}, E.~E. 2005, \apj, 634, 1385
1106: 
1107: \bibitem[{{Marsh} {et~al.}(2002){Marsh}, {Silverstone}, {Becklin}, {Koerner},
1108:   {Werner}, {Weinberger}, \& {Ressler}}]{mar02}
1109: {Marsh}, K.~A., {Silverstone}, M.~D., {Becklin}, E.~E., {Koerner}, D.~W.,
1110:   {Werner}, M.~W., {Weinberger}, A.~J., \& {Ressler}, M.~E. 2002, \apj, 573,
1111:   425
1112: 
1113: \bibitem[{{Martin-Za{\"i}di} {et~al.}(2005){Martin-Za{\"i}di}, {Deleuil},
1114:   {Simon}, {Bouret}, {Roberge}, {Feldman}, {Lecavelier Des Etangs}, \&
1115:   {Vidal-Madjar}}]{mar05}
1116: {Martin-Za{\"i}di}, C., {Deleuil}, M., {Simon}, T., {Bouret}, J.-C., {Roberge},
1117:   A., {Feldman}, P.~D., {Lecavelier Des Etangs}, A., \& {Vidal-Madjar}, A.
1118:   2005, \aap, 440, 921
1119: 
1120: \bibitem[{{Mer{\'{\i}}n} {et~al.}(2004){Mer{\'{\i}}n}, {Montesinos}, {Eiroa},
1121:   {Solano}, {Mora}, {D'Alessio}, {Calvet}, {Oudmaijer}, {de Winter}, {Davies},
1122:   {Harris}, {Cameron}, {Deeg}, {Ferlet}, {Garz{\'o}n}, {Grady}, {Horne},
1123:   {Miranda}, {Palacios}, {Penny}, {Quirrenbach}, {Rauer}, {Schneider}, \&
1124:   {Wesselius}}]{mer04}
1125: {Mer{\'{\i}}n}, B., et~al. 2004, \aap, 419, 301
1126: 
1127: \bibitem[{{Rice} {et~al.}(2006){Rice}, {Armitage}, {Wood}, \& {Lodato}}]{ric06}
1128: {Rice}, W.~K.~M., {Armitage}, P.~J., {Wood}, K., \& {Lodato}, G. 2006, \mnras,
1129:   373, 1619
1130: 
1131: \bibitem[{{Sadakane} \& {Nishida}(1986)}]{sad86}
1132: {Sadakane}, K. \& {Nishida}, M. 1986, \pasp, 98, 685
1133: 
1134: \bibitem[{{Siess} {et~al.}(2000){Siess}, {Dufour}, \& {Forestini}}]{sie00}
1135: {Siess}, L., {Dufour}, E., \& {Forestini}, M. 2000, \aap, 358, 593
1136: 
1137: \bibitem[{{Song} {et~al.}(2004){Song}, {Sandell}, \& {Friberg}}]{son04}
1138: {Song}, I., {Sandell}, G., \& {Friberg}, P. 2004, in ASP Conf. Ser. 324: Debris
1139:   Disks and the Formation of Planets, ed. L.~{Caroff}, L.~J. {Moon},
1140:   D.~{Backman}, \& E.~{Praton}, 250--+
1141: 
1142: \bibitem[{{Stauffer} {et~al.}(1995){Stauffer}, {Hartmann}, \& {Barrado y
1143:   Navascues}}]{sta95}
1144: {Stauffer}, J.~R., {Hartmann}, L.~W., \& {Barrado y Navascues}, D. 1995, \apj,
1145:   454, 910
1146: 
1147: \bibitem[{{Sylvester} {et~al.}(1996){Sylvester}, {Skinner}, {Barlow}, \&
1148:   {Mannings}}]{syl96}
1149: {Sylvester}, R.~J., {Skinner}, C.~J., {Barlow}, M.~J., \& {Mannings}, V. 1996,
1150:   \mnras, 279, 915
1151: 
1152: \bibitem[{{Thi} {et~al.}(2001{\natexlab{a}}){Thi}, {Blake}, {van Dishoeck},
1153:   {van Zadelhoff}, {Horn}, {Becklin}, {Mannings}, {Sargent}, {van den Ancker},
1154:   \& {Natta}}]{thi01n}
1155: {Thi}, W.~F., et~al. 2001{\natexlab{a}}, \nat, 409, 60
1156: 
1157: \bibitem[{{Thi} {et~al.}(2001{\natexlab{b}}){Thi}, {van Dishoeck}, {Blake},
1158:   {van Zadelhoff}, {Horn}, {Becklin}, {Mannings}, {Sargent}, {van den Ancker},
1159:   {Natta}, \& {Kessler}}]{thi01a}
1160: {Thi}, W.~F., et~al. 2001{\natexlab{b}}, \apj, 561, 1074
1161: 
1162: \bibitem[{{Wahhaj} {et~al.}(2007){Wahhaj}, {Koerner}, \& {Sargent}}]{wah07}
1163: {Wahhaj}, Z., {Koerner}, D.~W., \& {Sargent}, A.~I. 2007, ArXiv Astrophysics
1164:   e-prints
1165: 
1166: \bibitem[{{Weinberger} {et~al.}(1999){Weinberger}, {Becklin}, {Schneider},
1167:   {Smith}, {Lowrance}, {Silverstone}, {Zuckerman}, \& {Terrile}}]{wei99}
1168: {Weinberger}, A.~J., {Becklin}, E.~E., {Schneider}, G., {Smith}, B.~A.,
1169:   {Lowrance}, P.~J., {Silverstone}, M.~D., {Zuckerman}, B., \& {Terrile}, R.~J.
1170:   1999, \apjl, 525, L53
1171: 
1172: \bibitem[{{Zuckerman}(2001)}]{zuc01}
1173: {Zuckerman}, B. 2001, \araa, 39, 549
1174: 
1175: \bibitem[{{Zuckerman} {et~al.}(1995){Zuckerman}, {Forveille}, \&
1176:   {Kastner}}]{zuc95}
1177: {Zuckerman}, B., {Forveille}, T., \& {Kastner}, J.~H. 1995, \nat, 373, 494
1178: 
1179: \end{thebibliography}
1180: 
1181: 
1182: \end{document}
1183: 
1184: