0704.1681/ms.tex
1: % mn2esample.tex
2: %
3: % v2.1 released 22nd May 2002 (G. Hutton)
4: %
5: % The mnsample.tex file has been amended to highlight
6: % the proper use of LaTeX2e code with the class file
7: % and using natbib cross-referencing. These changes
8: % do not reflect the original paper by A. V. Raveendran.
9: %
10: % Previous versions of this sample document were
11: % compatible with the LaTeX 2.09 style file mn.sty
12: % v1.2 released 5th September 1994 (M. Reed)
13: % v1.1 released 18th July 1994
14: % v1.0 released 28th January 1994
15: 
16: \documentclass[usegraphicx,useAMS,usenatbib]{mn2e}
17: 
18: % If your system does not have the AMS fonts version 2.0 installed, then
19: % remove the useAMS option.
20: %
21: % useAMS allows you to obtain upright Greek characters.
22: % e.g. \umu, \upi etc.  See the section on "Upright Greek characters" in
23: % this guide for further information.
24: %
25: % If you are using AMS 2.0 fonts, bold math letters/symbols are available
26: % at a larger range of sizes for NFSS release 1 and 2 (using \boldmath or
27: % preferably \bmath).
28: %
29: % The usenatbib command allows the use of Patrick Daly's natbib.sty for
30: % cross-referencing.
31: %
32: % If you wish to typeset the paper in Times font (if you do not have the
33: % PostScript Type 1 Computer Modern fonts you will need to do this to get
34: % smoother fonts in a PDF file) then uncomment the next line
35: % \usepackage{Times}
36: 
37: %%%%% AUTHORS - PLACE YOUR OWN MACROS HERE %%%%%
38: \def\plotfiddle#1#2#3#4#5#6#7{\centering \leavevmode
39:     \vbox to#2{\rule{0pt}{#2}}
40:     \special{psfile=#1 voffset=#7 hoffset=#6 vscale=#5 hscale=#4 angle=#3}}
41: \newcommand{\Mdot}{\dot M}
42: \newcommand{\Mdotd}{\dot M_d}
43: \newcommand{\Mdotp}{\dot M_p}
44: \newcommand{\Mdotstar}{\dot M_*}
45: \newcommand{\Mstar}{M_*}
46: \newcommand{\Mdisk}{M_{\rm disc}}
47: \newcommand{\Msun}{M_{\odot}}
48: \newcommand{\Rin}{R_{\rm in}}
49: \newcommand{\Rhole}{R_{\rm hole}}
50: \newcommand{\Rinner}{R_{\rm inner}}
51: \newcommand{\Msunperyr}{M_{\odot}\,{\rm yr}^{-1}}
52: \newcommand{\Lsun}{L_{\odot}}
53: \newcommand{\pcc}{{\rm cm}^{-3}}
54: \newcommand{\percc}{\rm \,cm^{-3}}
55: \newcommand{\psqcm}{{\rm cm}^{-2}}
56: \newcommand{\persqcm}{\rm \,cm^{-2}}
57: \newcommand{\gpersqcm}{\rm \,g\,cm^{-2}}
58: \newcommand{\gpercc}{\rm \,g\,cm^{-3}}
59: \newcommand{\ps}{{\rm s}^{-1}}
60: \newcommand{\pyr}{{\rm yr}^{-1}}
61: \newcommand{\ccps}{{\rm cm}^{3} {\rm s}^{-1}}
62: \newcommand{\kmps}{{\rm km}\,{\rm s}^{-1}}
63: \newcommand{\kms}{{\rm km}\,{\rm s}^{-1}}
64: \newcommand{\erg}{{\rm erg}}
65: \newcommand{\kb}{k_{\rm B}}
66: \def\micron{\hbox{$\mu$m}}
67: 
68: 
69: 
70: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
71: 
72: \title[Demographics of Transition Objects]{Demographics of Transition Objects}
73: \author[J. R. Najita, S. E. Strom, and J. Muzerolle]{Joan R. Najita$^{1}$, Stephen E. Strom$^{1}$ and James Muzerolle$^{2}$ \\ 
74: $^{1}$National Optical Astronomy Observatory, 950 N. Cherry Ave.,
75: Tucson, AZ 85719\\
76: $^{2}$Steward Observatory, 933 N. Cherry Ave., Tucson, AZ, 85721
77: }
78: 
79: \begin{document}
80: 
81: \date{Submitted 2006 November.}
82: 
83: \pagerange{\pageref{firstpage}--\pageref{lastpage}} \pubyear{2002}
84: 
85: \maketitle
86: 
87: \label{firstpage}
88: 
89: \begin{abstract}
90: The unusual properties of transition objects (young stars with
91: an optically thin inner disc surrounded by an optically thick outer
92: disc) suggest that significant disc evolution has occured in 
93: these systems.
94: We explore the nature of these systems by examining their demographics,
95: specifically their stellar accretion rates $\Mdotstar$ and
96: disc masses $\Mdisk$ compared to
97: those of accreting T Tauri stars of comparable age.
98: We find that transition objects in Taurus occupy a restricted region
99: of the $\Mdotstar$ vs.\ $\Mdisk$ plane.  Compared to non-transition
100: single stars in Taurus, they have stellar accretion rates that are
101: typically $\sim 10$ times lower at the same disc mass and
102: median disc masses $\sim 4$ times larger.
103: These properties are anticipated by several proposed 
104: planet formation theories and 
105: suggest that the formation of Jovian mass planets may play a significant 
106: role in explaining the origin of at least some transition objects. 
107: Considering transition objects as a distinct demographic group
108: among accreting T Tauri stars 
109: leads to a tighter relationship between
110: disc masses and stellar accretion rates, with a slope
111: between the two quantities that is close to the value of unity
112: expected in simple theories of disc accretion.
113: \end{abstract}
114: 
115: \begin{keywords}
116: (stars:) circumstellar matter ---
117: (stars:) planetary systems: formation ---
118: (stars:) planetary systems: protoplanetary disc ---
119: stars: pre-main sequence
120: \end{keywords}
121: 
122: \section{Introduction}
123: 
124: Transition objects are an interesting class of young stellar objects
125: whose properties indicate that significant disc evolution has
126: occurred.
127: Their spectral energy distributions (SEDs) imply the presence of
128: an optically thin inner region (defined by an ``opacity hole'' of
129: radius $\Rhole \sim 1-20$\,AU) that is surrounded by an optically
130: thick outer disc (beyond $\Rhole$).
131: Such an SED indicates that the circumstellar disc has evolved
132: significantly from the radially continuous optically thick disc
133: that is believed to surround all stars at birth.
134: 
135: Transition objects were first identified from the analysis of
136: the SEDs of large samples of solar-like pre-main sequence
137: stars in nearby star-forming regions (e.g. Strom et al.\ 1989;
138: Skrutskie et al.\ 1990; Gauvin and Strom 1992; Marsh \& Mahoney 1992;
139: Wolk \& Walter 1996).
140: They were selected to have SEDs whose long wavelength excesses
141: ($\lambda \ga 10\micron$) were equal to or exceeded that
142: expected from a
143: geometrically flat optically thick disc, and whose shorter
144: wavelength excesses were below those of an optically
145: thick disc that extends to within a few stellar radii.
146: Thus the definition included both objects with {\it no
147: dust} within some disc radius as well as objects with
148: {\it optically thin} inner dust discs.
149: 
150: Transitional SEDs are expected to arise via multiple pathways.
151: For example, a transitional SED could arise as grains in
152: the inner disc grow into larger bodies, thereby
153: decreasing the continuum optical depth of the inner disc
154: (Strom et al.\ 1989; Dullemond \& Dominik 2005).
155: Alternatively, as suggested by Skrutskie et al.\ (1990),
156: a transitional SED may arise through the creation of a
157: large gap, the result of the dynamical isolation
158: of the inner and outer discs by a sufficiently massive giant planet
159: (see also Marsh \& Mahoney 1992;
160: Calvet et al.\ 2002; Rice et al.\ 2003; D'Alessio et al.\ 2005;
161: Calvet et al.\ 2005; Quillen et al.\ 2004).
162: More recently, disc photoevaporation has been suggested as an
163: explanation for the properties of some transition objects
164: (Clarke, Gendrin \& Sotomayor 2001; 
165: Alexander, Clarke \& Pringle 2006; McCabe et al.\ 2006).
166: Because all of the above scenarios can in principle produce
167: similar SEDs, additional diagnostic tools beyond the SED may
168: be needed to determine the physical state of a transition
169: object.
170: 
171: To take steps in this direction, we therefore examine in this
172: contribution the demographics of transition objects.  Specifically,
173: we compare the stellar accretion rates and disc masses
174: of transition objects with those of accreting T Tauri stars of
175: comparable age.
176: Our methods and results are described in \S 2 and \S 3 respectively.
177: The implications of the results are discussed in \S 4 and
178: summarized in \S 5.
179: 
180: 
181: 
182: 
183: 
184: \section{Methods}
185: 
186: \subsection{Utility of a Demographic Approach}
187: In order to explore the nature of transition objects, we
188: take a demographic approach, examining the properties of
189: an ensemble of transition objects compared to the properties of
190: a parent population of classical T Tauri stars.  Specifically, we
191: compare two well-studied properties of these systems, their stellar
192: accretion rates and their disc masses with those of accreting
193: T Tauri stars of comparable age.
194: A demographic approach has the advantage of being robust against 
195: {\it systematic\/}
196: errors, since systematic errors (e.g., of the kind described by
197: White \& Hillenbrand 2004 and Hartmann et al.\ 2006)
198: can shift demographic trends up or down in $\Mdotstar$ or
199: $\Mdisk$ but should not obscure them.
200: Since the use of a large sample of T Tauri stars and transition objects
201: also allows us to ``average over'' random measurement uncertainties
202: associated with the individual measurements,
203: such an approach may prove useful in identifying demographic trends
204: and, thereby, provide clues to the physical state of transition objects.
205: As a result of significant
206: recent efforts to measure stellar accretion rates and disc masses
207: for young stars, measurements of both quantities are now available
208: for a relatively large number of sources in the Taurus star forming region.
209: 
210: \subsection{Stellar Accretion Rates}
211: 
212: We adopted stellar accretion rates from the literature, all of which
213: were determined using some measure of the UV/optical accretion
214: excess emission.  Stellar accretion rates thus derived are fairly
215: robust.  They have been used, for example, to demonstrate 
216: correlations between $\Mdotstar$ and stellar mass
217: (e.g., Muzerolle et al.\ 2003; Natta et al.\ 2004) as 
218: well as $\Mdotstar$ and age 
219: (e.g., Sicilia-Aguilar et al.\ 2005).
220: 
221: Stellar accretion rates (Table 1) were adopted from the following references,
222: in order of preference: Gullbring et al.\ (1998, hereafter, G98);
223: Hartmann et al.\ (1998, hereafter, H98); White \& Ghez (2001, hereafter, WG01);
224: Calvet et al.\ (2004); Gullbring et al.\ (2000);
225: Hartigan, Edwards \& Ghandour (1995, hereafter, HEG95).  Values from WG01 were used
226: for all binaries when available, since their observations were specifically
227: tailored for resolving most binary systems. 
228: Systematic differences between some of these determinations arise from several
229: factors, including different extinction estimates, bolometric corrections,
230: and pre-main sequence tracks adopted to estimate stellar masses.
231: We have thus attempted to place all the values on a consistent scale
232: by calculating any systematic offsets for objects with measurements
233: in common with H98 (which includes all values from G98).
234: 
235: We show two such comparisons with measurements from WG01
236: and HEG95 in Figure~1. One can see in
237: both cases a similar trend with a relatively small dispersion
238: of $\sim 0.2$ dex, which is, perhaps surprisingly, 
239: smaller than the typical 
240: $\sim 0.5$ dex measurement uncertainties discussed by G98
241: and the typical veiling variability of a factor of 2 or less
242: (e.g., Hartigan et al.\ 1991).  However, there are significant
243: systematic offsets.  The WG01 values are lower on average by
244: a factor of 2.  Since they use the same methodology as H98,
245: most of the offset is likely a result of their adoption of
246: a different set of pre-main sequence tracks (Baraffe et al.\ 1998; 
247: Palla \& Stahler 1999)
248: which give systematically higher stellar masses than the D'Antona \&
249: Mazzitelli (1998) tracks used by H98.  Meanwhile,
250: the HEG95 values are higher on average by a factor of 8.
251: This discrepancy, discussed in detail in G98, results from
252: differences in extinction estimates, bolometric corrections, 
253: and the assumed accretion geometry.  Based on these comparisons,
254: we have scaled the values from WG01 up by a factor of 2
255: and the values from HEG95 down by a factor of 0.12 to place them
256: on the same scale as H98.
257: 
258: Two objects, CoKu Tau/4 and UX Tau A, lack published UV measurements.
259: We have thus estimated their stellar accretion rates using magnetospheric
260: accretion models of H$\alpha$ profiles 
261: (e.g., Muzerolle, Calvet \& Hartmann 2001
262: and references therein).  The value for CoKu Tau/4 is an upper limit
263: based on the line equivalent width since no resolved profiles exist
264: in the literature.  
265: Based on the scatter in the Figure 1, we estimate that the 
266: resulting (random) uncertainty in the stellar accretion rates adopted
267: here is likely to be less than a factor of $\sim 3$ or so, much less
268: than the spread of $\sim 100$ in the stellar
269: accretion rate at a given disc mass that is 
270: found in Taurus (\S 3). 
271: 
272: \setcounter{figure}{0}
273: \begin{figure}
274: %\epsscale{0.7}
275: \includegraphics[width=84mm]{fig1.eps}
276: \caption{Comparison of stellar accretion rates (in $\Msunperyr$) 
277: from Hartmann et al.\ (1998)
278: with those from White \& Ghez (2001) (top panel, including only single stars)
279: and Hartigan et al.\ (1995) (bottom panel).  Note the systematic offsets
280: in both cases, while the general trend with relatively small dispersion
281: is maintained.
282: }
283: \end{figure}
284: 
285: \subsection{Disk Masses}
286: 
287: We adopted the disc masses (Table 1) estimated for Taurus sources
288: based on submillimeter continuum measurements
289: (Andrews \& Williams 2005).
290: The data set has the advantage of being large and homogeneous.
291: While Andrews \& Williams (2005) estimated disc masses from SED fits where
292: possible, the disc masses for most of the sample were determined from an
293: empirical conversion between submillimeter fluxes and disc mass
294: based on the SED fits.  The random error in the disc masses is
295: expected to be $\la 0.5$\,dex, primarily the result of the
296: uncertainty in the spectral index at submillimeter wavelengths.
297: Uncertainties in this range are much less than the two orders of
298: magnitude spread in disc mass at a given stellar accretion rate
299: that is found in Taurus (\S 3).
300: 
301: Systematic errors are also possible.  For example,
302: Hartmann et al.\ (2006) suggest that disc masses are
303: likely to have been systematically underestimated because
304: maximum grain sizes 3 -- 10 times larger or smaller than 1 mm result in
305: grain opacities significantly lower than is typically assumed.
306: We can estimate the extent to which disc masses may be
307: underestimated by looking at the upper end of the disc mass
308: distribution reported by Andrews \& Williams.
309: Correction factors larger than 0.5 dex cannot be tolerated at
310: the upper mass end without driving the disc masses over the
311: theoretical gravitational instability limit. 
312: In any event, a demographic approach is robust against systematic errors 
313: in disk masses arising from under- or over-estimates of grain opacity,  
314: since systematic errors will shift demographic trends up or down in
315: $\Mdisk$ but will not obscure them.
316: 
317: \subsection{SED Classification}
318: 
319: In Table 1, we provide a rudimentary classification of
320: the observed SEDs of the Taurus sources based on
321: ground-based photometry and mid-IR spectra from Spitzer/IRS
322: (Furlan et al.\ 2006).  Objects were classified
323: as transition objects (T) if they show weak or no infrared excess
324: shortward of $10\,\mu$m and a significant excess at longer wavelengths.
325: The initial selection and classification was made by visual
326: inspection of the SED fits of Furlan et al.\ (2006).  This resulted
327: in our selecting objects whose excess emission fell by $>0.2$ dex 
328: below that of the ``Taurus Median'' obtained from the analysis 
329: of all of the class II sources examined by D'Alessio et al.\ (1999).
330: Systems with no significant infrared excess out to 
331: $\sim 30\,\mu$m, indicating no substantial disc material, were classified
332: as ``weak'' objects (W).  All other systems exhibit strong
333: excess emission indicative
334: of optically thick discs extending all the way in to the dust sublimation
335: radius, either ``classical'' discs (C) or ``flat-spectrum'' (F) sources with
336: elevated accretion rates and possible remnant envelopes.
337: 
338: Note that in several cases, the transition 
339: object classification may be influenced by the presence of unresolved
340: companions with possible infrared excesses of their own.  Also, for stars
341: with lower masses, even ``classical'' discs can often exhibit lower
342: excess emission at shorter wavelengths because the stellar irradiation flux,
343: which is typically the dominant source of dust heating, is generally smaller.
344: The likelihood of incorrectly classifying an object as 
345: ``transition'' as opposed to ``classical'' is thus greater among
346: low mass, later spectral type stars owing to the decreased contrast
347: between disc and photosphere.   
348: We therefore limit our analysis to stars with spectral types earlier 
349: than M3. 
350: We have indicated the remaining ambiguous cases in Table 1 (C?, T?, C/T).
351: Table 2 captures the decrements from the Taurus median for all 
352: sources that have a transition-like SED (i.e., sources
353: with a T, T?, or C/T SED classification in Table 1).  
354: 
355: 
356: Our classification scheme is broader than that adopted in some recent
357: studies in which only systems with {\it no} apparent excess anywhere
358: shortward of a given wavelength are considered to be transition objects
359: (Muzerolle et al.\ 2006; Sicilia-Aguilar et al.\ 2006).  
360: We accept a broader range of excesses as transition objects, including
361: those with weak excesses shortward of $10\micron$, in order to
362: avoid excluding systems in possibly interesting phases of planet formation.
363: For example,
364: systems that are dissipating their discs via photoevaporation
365: are expected to have essentially
366: no dust within $\Rhole$ and hence no excess above photospheric
367: levels at short wavelengths.  In contrast,
368: a significant continuum excess might be expected from discs that
369: have formed a Jovian mass planet and are replenishing,
370: via accretion streams, an inner disc with gas and {\it dust} from an outer disc.
371: In order to capture objects that are possibly in these states of evolution,
372: we selected objects that show a significant decrement ($\ge 0.2$\,dex)
373: relative to the Taurus median SED.
374: 
375: \setcounter{figure}{1}
376: \begin{figure}
377: %\epsscale{1.0}
378: %\figurenum{2a}
379: \includegraphics[width=84mm]{fig2a.eps}
380: \caption{SEDs for all stars earlier than M3 that we identify as transition 
381: objects.  Diamonds: optical and IRAS photometry from Kenyon \& Hartmann 
382: (1995); near-infrared photometry from 2MASS; IRAC photometry from 
383: Hartmann et al. (2005).  Black solid lines: IRS spectra from Furlan et al. (2006).  Blue solid lines: Kurucz model photospheres of the appropriate spectral type,
384: normalized to the dereddened J-band flux.  Hashed region: the upper and lower 
385: quartiles of the Taurus CTTS median SED from D'Alessio et al. (1999).
386: }
387: \end{figure}
388: 
389: \setcounter{figure}{1}
390: \begin{figure}
391: %\epsscale{1.0}
392: %\figurenum{2b}
393: \includegraphics[width=84mm]{fig2b.eps}
394: \caption{{\it continued} Additional SEDs for objects identified as transition 
395: objects.}
396: \end{figure}
397: 
398: 
399: \section{Results}
400: 
401: Figure 3 plots the recalibrated stellar accretion rates from \S 2.2 against
402: the disc masses described in \S 2.3.
403: The top panel plots all sources for which detections are available
404: for both axes.
405: The subset of sources that are
406: known binaries, with orbital separations $<2.5\arcsec$ (Table 1),
407: are shown as gray-brown filled points
408: (White \& Ghez 2001; Prato et al.\ 2002;
409: Mathieu, Mart\'in \& Magazzu 1996; and references therein).
410: The submillimeter observations do not resolve the binaries, so the 
411: reported fluxes may may contain contributions from circumstellar 
412: disks as well as a circumbinary disk. 
413: The presence of an orbiting stellar companion may have a significant
414: impact on the disc properties, e.g., by truncating a disc and thereby
415: both reducing the disc mass and altering the SED
416: (Jensen, Mathieu \& Fuller 1996).
417: Indeed, the median disc mass for the binary population is
418: $\log \Mdisk = -2.8$, lower than the median disc mass
419: of $\log \Mdisk = -2.3$ for the entire sample shown.
420: Since a binary companion
421: introduces additional complexity and may obscure important trends
422: that result from other physical processes,
423: we exclude these systems from our analysis.
424: The remaining filled black points indicate sources that have no known
425: binary companion within $2.5\arcsec$.
426: 
427: In the lower panel, the non-binary systems are further divided
428: into those that have a transition-like SED (crosses), as described
429: in \S 2.4, and those that do not (filled black points).
430: The sources with transition-like SEDs are:
431: GO Tau, LkCa 15, DN Tau, DM Tau, GM Aur,
432: V836 Tau, UX Tau A, and CX Tau
433: in order of decreasing disc mass.
434: The SEDs of GO Tau and DN Tau (Furlan et al.\ 2006)
435: are similar to those of the well studied transition objects 
436: GM Aur and DM Tau in that they also lack strong infrared excesses at 
437: short wavelengths but have optically thick outer discs.  
438: This is also the case for
439: V836 Tau, which was originally classified as a transition object
440: by Strom et al.\ (1989; see also Padgett et al.\ 2006).
441: UX Tau A is noted as a transition object by Furlan et al.\ (2006).
442: LkCa 15 has an SED that indicates an inner disc ``gap'' of a few AU
443: (Bergin et al.\ 2004).  
444: 
445: The concentration of crosses toward the lower right region of the plot
446: suggests that transition objects inhabit a restricted region of the  
447: $\Mdotstar$ vs. $\Mdisk$ plane.  
448: The existence of identifiable demographic groups demonstrates 
449: that the random errors in stellar accretion rates and disc masses
450: (\S 3.2 and 3.2) are apparently insufficient to obscure significant
451: population trends.
452: We can estimate statistically the likelihood that the transition
453: objects represent a distinct population in terms of their disc masses 
454: and accretion rates.  Figure 4 compares the distributions of stellar
455: accretion rates (top) and disc masses (bottom) for transition objects
456: (dashed histograms) and non-transition single stars (solid
457: histograms).  For sparse data of this kind, the two-sided KS test
458: is a good way to estimate the likelihood that the two distributions
459: are drawn from the same parent distribution.
460: With this approach, we find that there is a less than 3\% probability that the
461: stellar accretion rates of the transition and non-transition objects
462: are drawn from the same parent distribution.  
463: For the disc masses, there is a less than 12\% probability 
464: that the disc masses of the transition and non-transition objects are
465: drawn from the same parent distribution.
466: 
467: A more useful estimate would jointly compare the distributions in
468: both $\Mdotstar$ and $\Mdisk$.
469: A two-dimensional two-sided KS test that accounts for the distribution
470: of the two samples in both quantities estimates that there is a 
471: less than 0.8\% probability that the transition and non-transition
472: objects are drawn from the same parent distribution.
473: Thus, even with the limited data available at present, it appears  
474: that the transition objects represent a distinct population in terms of
475: their stellar accretion rates and disc masses.
476: 
477: For the stellar accretion rates, the median of the distribution 
478: is $\log \Mdotstar = -8.5$ for the transition objects and
479: $\log \Mdotstar = -7.6$ for the non-transition objects.
480: For the disc masses, the median of the distribution
481: is $\log \Mdisk = -1.6$ for the transition objects and
482: $\log \Mdisk = -2.2$ for the non-transition objects.
483: Thus, compared to non-transition single stars, transition objects are
484: biased to lower stellar accretion rates and larger disc masses. 
485: We discuss the possible origin of these biases in \S 4.
486: 
487: The likelihood that transition objects represent a 
488: population distinct from other classical T Tauri stars 
489: has possible implications for our understanding  
490: of the large dispersion in the $\Mdotstar$ vs.\ $\Mdisk$ plane. 
491: In simple disc models the disc accretion rate
492: $\Mdotd \propto \nu\Sigma$ locally, where
493: $\Sigma$ is the disc mass surface density and
494: $\nu =\alpha c_s H$ is the
495: viscosity parametrized in terms of the sound speed $c_s$ and the
496: disc scale height $H$.
497: If all discs have a similar sizes (e.g., outer disc radii) and 
498: radial mass distributions
499: (e.g., in a steady accretion disc $\Sigma \propto r^{-1}$;
500: Hartmann et al.\ 1998),
501: $\Sigma$ would be proportional to $\Mdisk$.
502: If, in addition, the disc accretion rate equals the stellar
503: accretion rate, we would expect $\Mdotstar \propto \Mdisk$.
504: In contrast, Figure 3 (top) shows considerable scatter with no
505: apparent trend between the two quantities.
506: The apparent lack of the expected correlation has been noted
507: previously (e.g., Dullemond, Natta, \& Testi 2006).
508: 
509: 
510: If the transition objects represent a distinct population
511: whose stellar accretion rates have been modified downward from
512: their initial values (see \S 4), this downward evolution can account
513: for some of the dispersion in the
514: $\Mdotstar$ vs.\ $\Mdisk$ plane.
515: To illustrate this, we show least absolute deviation fits to the
516: distributions in Figure 3.
517: In comparison to a minimized chi-square error criterion,
518: least absolute deviation fits are less sensitive to statistical
519: outliers.
520: In the top panel of Figure 3, a least absolute deviation fit to the
521: $\Mdotstar$ vs.\ $\Mdisk$ values for all sources for which there are
522: detections in both axes gives a slope of 0.22, relatively far from the linear
523: relation that would be expected in simple disc models.
524: 
525: If we remove the binaries from the sample (bottom panel) and fit
526: separately the slopes of the the non-transition single stars
527: (solid line) and the transition objects (dashed line),
528: both populations
529: have a slope of 0.94, close to the value of unity that would
530: be expected if mass accretion rate is proportional to disc mass.
531: The dashed line is offset downward from the solid line by approximately
532: an order of magnitude, similar to the difference in the median
533: accretion rates between the transition objects and the non-transition
534: single stars.
535: Because of the small number of sources that make up the transition
536: object sample, the slope for that population is not well-determined.
537: 
538: \setcounter{figure}{2}
539: \begin{figure}
540: %\epsscale{0.5}
541: %\figurenum{3}
542: \includegraphics[scale=0.5]{fig3.eps}
543: \caption{Stellar accretion rates and outer disc masses for
544: sources in Taurus-Auriga.  Top panel: systems possessing known a
545: binary companion within $2.5\arcsec$ (gray-brown points) have lower
546: median disc masses compared to systems without such binary
547: companions (black points).  A least absolute deviation fit to
548: all of the points shown gives a slope of 0.22 (solid line).
549: Bottom panel: among the non-binary systems, those with transition-like
550: SEDs (crosses) are segregated from those with non-transition
551: SEDs (black points), with the transition objects having lower stellar
552: accretion rates for their disc masses.
553: Separately fitting the single, non-transition stars (solid line)
554: and the transition objects (dashed line) gives a slope of 0.94
555: for each population.
556: The upper limit to the accretion rate determined for
557: CoKu Tau/4 is included in the fit for the transition objects. 
558: }
559: \end{figure}
560: 
561: \section{Discussion} 
562: 
563: We have found that transition objects in Taurus occupy a restricted region
564: of the $\Mdotstar$ vs.\ $\Mdisk$ plane.  Compared to non-transition
565: single stars in Taurus, they have stellar accretion rates that are
566: typically $\sim 10$ times lower at the same disc mass.
567: In addition, the median disc mass for the transition objects is
568: $\sim 4$ times larger than the median disc mass for the
569: non-transition single stars.  The results place useful constraints
570: on the possible explanations for the origin of the transitional SEDs
571: in these systems.  They also shed some light on the nature of the 
572: relation between disc masses and disc accretion rates.
573: For example, the observed properties allow us to sort among the
574: following proposed explanations for transitional SEDs.  
575: 
576: (1) {\it Grain growth and the formation of rocky planetary cores in the
577: inner disc} (e.g., Strom et al.\ 1989; Dullemond \& Dominik 2005).
578: If small grains are consumed in this process, the inner disc will
579: become optically thin while remaining gas rich.  As a result, the
580: accretion rate through the disc and on to the star will continue
581: unabated.  
582: 
583: (2) {\it The dynamical clearing of a large gap by a Jovian mass planet}
584: (e.g., Skrutskie et al\ 1990; Marsh \& Mahoney 1992; 
585: Calvet et al.\ 2002; Rice et al.\ 2003; D'Alessio et al.\ 2005;
586: Calvet et al.\ 2005; Quillen et al.\ 2004).  
587: When a giant planet forms in a disc with a mass sufficient
588: to open a gap, gap crossing streams, from the outer disc 
589: (beyond $\Rhole$) to the planet, and from the planet to the 
590: inner disc (within $\Rinner$), are expected to allow 
591: continued accretion on to both planet and star, the latter through
592: the replenishment of an inner disc (Lubow, Seibert \& Artymowicz 1999; 
593: see also Kley 1999; 
594: Bryden et al.\ 1999; 
595: D'Angelo, Henning, \& Kley 2002;  
596: Bate et al.\ 2003; Lubow \& D'Angelo 2006).  
597: By the time the planet is massive enough to open a gap, it is also
598: able to divert a significant fraction of the mass accreting from
599: the outer disc on to itself.  Thus the stellar accretion rate is 
600: predicted to be reduced by a factor $\sim 0.1$ 
601: (Lubow \& D'Angelo 2006)
602: to $\sim 0.05$ (Varni\`ere et al.\ 2006)
603: compared to the mass accretion rate through the outer disc.
604: 
605: (3) {\it The isolation of the outer disc by a supra-Jovian mass planet,
606: followed by the viscous draining of the inner disc.}  As the
607: planet grows in mass via accretion from the gap-crossing streams,
608: accretion past the planet effectively ceases
609: (Lubow et al.\ 1999), isolating the inner and outer disc.
610: When the remaining inner disc material accretes on to the star,
611: the system is left with a large inner hole,
612: devoid of both gas and dust, and no
613: further stellar accretion.
614: 
615: (4) {\it Disk photoevaporation.}  Irradiation by energetic
616: (UV and X-ray) photons from the star heat
617: the disc surface layers to high temperatures $\sim 10^4$\,K.
618: Beyond a radius of $\sim$7--10\,AU where the thermal velocity of the
619: surface layers exceeds the escape speed ($\sim 10\,\kms$),
620: a photoevaporative flow (characterized by a mass loss rate of
621: $\sim 4\times 10^{-10}\,\Msunperyr$; Hollenbach, Yorke \& Johnstone 2000)
622: is expected to develop.
623: As discs viscously spread and accrete, eventually the column density
624: of the outer disc drops sufficiently that the disc accretion rate
625: becomes comparable to the disc photoevaporation rate.
626: At this point, the region beyond the photoevaporation radius
627: is losing material via photoevaporation more rapidly
628: than it can resupply the inner disc through viscous
629: accretion.  The inner disc is thereby decoupled from the outer
630: disc.  Material in the inner disc accretes on to the star,
631: leaving behind a large inner hole,
632: devoid of gas and dust (the ``UV Switch Model''; Clarke et al.\
633: 2001), producing a transitional SED.
634: 
635: 
636: \setcounter{figure}{3}
637: \begin{figure}
638: %\epsscale{.60}
639: %\figurenum{4}
640: \includegraphics[width=65mm]{fig4.eps}
641: \caption{Comparison of the distributions of stellar accretion rates 
642: (top) and disc masses (bottom) for transition objects (dashed line) 
643: and single, non-transition stars (solid line) in the Taurus star
644: forming region.  Compared to single, non-transition stars,
645: transition objects are biased to lower stellar accretion rates
646: and larger disc masses.
647: }
648: \end{figure}
649: 
650: 
651: 
652: When compared with these scenarios, the factor of $\sim 10$ decrement
653: in the observed stellar accretion rates of transition objects
654: compared to those of non-transition single stars at a given disc mass
655: is remarkably similar to predictions for the Jovian mass planet formation 
656: scenario described above.
657: The higher average disc masses for the transition objects,
658: compared to those of non-transition single stars may also be consistent
659: with this scenario since
660: more massive discs are expected to be conducive to the
661: formation of giant planets, either in the core accretion scenario
662: (e.g., Lissauer \& Stevenson 2006) or
663: if gravitational instabilities play a role (e.g., Boss 2005;
664: Durisen et al.\ 2005).
665: These similarities are suggestive that the formation of Jovian mass
666: planets plays a role in explaining the origin of at least 
667: some transitional SEDs in Taurus.   
668: 
669: Indeed, for the Taurus transition objects with higher disc masses
670: ($\ga 0.01\Msun$),  
671: it is relatively straightforward to rule out 
672: the UV Switch model as an explanation for the transitional SEDs.
673: The high disc masses of these systems imply large 
674: gas column densities at the gravitational radius, suggesting that
675: photoevaporation can hardly be on the verge of decoupling the
676: inner and outer discs. 
677: The stellar accretion rates of these sources
678: ($10^{-9} - 10^{-8}\Msunperyr$) are in excess of
679: the nominal disc photoevaporation rate of $4\times 10^{-10}\Msunperyr$,
680: further arguing against photoevaporation as the origin of the
681: transitional SED. 
682: Since disk photoevaporation rates are poorly known, sources 
683: with stellar accretion rates somewhat larger than the nominal 
684: $4\times 10^{-10}\Msunperyr$
685: could be on the verge of clearing their inner disks.  However, 
686: the inner disk is expected to clear rapidly once the 
687: stellar accretion rate drops down to the disk photoevaporation 
688: rate (Clarke et al.\ 2001); as a result, we would expect that 
689: few of the sources with stellar accretion rates of 
690: $\sim 10^{-9} \Msunperyr$ are passing through this 
691: brief evolutionary phase.  The high disk masses of these 
692: transition objects ($\ga 0.01\Msun$) support this interpretation.
693: Similarly, large accretion rates ($10^{-9} - 10^{-8}\Msunperyr$) 
694: are not expected 
695: if these systems have formed a supra-Jovian mass planet.
696: In contrast, significant accretion rates {\it are} expected when discs form
697: planetesimals.  
698: However, unlike the presence of a Jovian mass planet, this mechanism
699: does not provide a simple explanation for the low accretion
700: rates of these objects for their disc masses. 
701: 
702: The presence of an orbiting Jovian mass planet has indeed been the
703: favored explanation for the SEDs of the Taurus transition objects
704: GM Aur and DM Tau (Marsh \& Mahoney 1992; Rice et al.\ 2003;
705: Bergin et al.\ 2004; Calvet et al.\ 2005).
706: If such a planet were present, with a mass large enough to reduce
707: the accretion rate on to the star,
708: we would infer artificially small values of $\alpha$  
709: for the outer disc if we assumed that the accretion rate through the 
710: outer disc equalled the observed stellar accretion rate.
711: Indeed, a low value of the viscosity parameter $\alpha \sim 0.001$
712: is inferred for DM Tau under this assumption (Calvet et al.\ 2005).
713: Instead, if a Jovian mass planet is present,
714: most of the mass accreting through the outer disc may accrete
715: on to the planet rather than the star.
716: If the outer disc accretion
717: rate is then $\sim 10$ times the stellar accretion rate
718: (Lubow \& D'Angelo 2006), the inferred value of $\alpha$ characterizing
719: the outer disc of DM Tau would be $\sim 10$ times larger,
720: i.e., more similar to the value of $\alpha \simeq 0.01$ that is
721: believed to be typical of T Tauri discs (Hartmann et al.\ 1998).
722: 
723: 
724: 
725: Notably, at least one of the transition objects in Taurus
726: (CoKu Tau/4) has a disc mass
727: ($\la 10^{-3}\,\Msun$) and
728: a stellar accretion rate
729: ($\la 4\times 10^{-10}\Msunperyr$) that are both low enough
730: that its transitional SEDs may result from disc photoevaporation
731: rather than giant planet formation. 
732: CX Tau is a possibly similar case, although its estimated disk mass 
733: and stellar accretion rate are somewhat larger than the above limits.
734: While photoevaporation is not expected to erode away the inner regions of 
735: high mass discs ($\ga 0.01\Msun$) until they are 6--15 Myr old
736: (Clarke et al.\ 2001; Alexander et al.\ 2006), discs with sufficiently low
737: initial masses can photoevaporate away on much shorter timescales, 
738: comparable to the age of Taurus. 
739: For example, scaling the standard accretion disc model of
740: Hartmann et al.\ (1998) down in mass
741: shows that a disc with $\alpha=10^{-2}$ and $M_d=10^{-3}\Msun$
742: would spread beyond 100\,AU and evolve to
743: $\Mdotstar < 3\times 10^{-10}\Msunperyr$
744: on a time scale of $\sim 1$ Myr, a mass accretion rate low enough
745: that photoevaporation would be able to create an inner hole.
746: 
747: 
748: If we interpret CX Tau and CoKu Tau/4 as systems whose SEDs are
749: the result of photoevaporation, the remaining number of transition
750: objects is roughly consistent with the interpretation that
751: they are systems undergoing giant planet formation.
752: The fraction of such transition objects is 7 out of a total of
753: 28 non-binary T Tauri stars or 25\%.  This is larger than the
754: 5-10\% of nearby FGK stars currently known to harbor extrasolar
755: giant planets.
756: If the current detection statistics are extrapolated to 20 AU,
757: the expected fraction of planet-bearing systems rises to $\sim 12$\%
758: (Marcy et al.\ 2005).  
759: This is not totally inconsistent with the 25\% fraction of
760: transition objects identified here, given both the small sample of 
761: objects studied and the possible role of 
762: planetary orbital migration in accounting
763: for any difference in the incidence rates of accreting transition
764: objects and extrasolar giant planets.
765: 
766: Perhaps surprisingly, there is a dearth of transition objects
767: intermixed with the population of non-transition, single stars.
768: Systems that are forming planetesimals and protoplanets (i.e., 
769: objects too low in mass to open a gap) would be expected to
770: occupy this region of the 
771: diagram.  
772: That is, they would appear as systems with optically thin inner
773: discs but otherwise unaltered stellar accretion rates.
774: Perhaps the lack of such systems indicates that
775: planetesimal and protoplanet formation is inefficient in
776: clearing the dust from the inner regions of accreting discs
777: (cf.\ Dullemond \& Dominik 2005).
778: This might be the case if the accretion of small dust grains from 
779: larger disk radii is effective in ``filling in'' the opacity 
780: holes created by grain growth at smaller radii. 
781: Alternatively, the phase of planetesimal and protoplanet formation
782: may be very short-lived. 
783: 
784: A tempting counter-hypothesis is that systems that have experienced
785: significant grain growth and settling (and are forming planetesimals
786: and protoplanets) are intermixed with the transition objects we have
787: identified as candidate giant-planet forming systems.  This might
788: explain the diversity of transition object SEDs (Figure 2) that 
789: range from those with large ``dips'' (e.g., GM Aur) to 
790: ``anemic'' SEDs (e.g., GO Tau) that might be more characteristic 
791: of systems that have undergone grain settling and growth.  
792: However, as already noted, it is difficult to understand why systems
793: that have experienced significant grain growth 
794: would show lower accretion rates for their disk masses.  By reducing
795: the small grain population in the surface layers of the disk, 
796: grain growth is expected to lead to larger ionization
797: fractions and therefore larger surface disk columns that are
798: susceptible to the magnetorotational instability (e.g., Sano et
799: al.\ 2000).  We would then naively expect that such disks would
800: have {\it larger} accretion rates for their disk masses, rather
801: than the reduced accretion rates that are observed.  Whether or 
802: not this is in fact the case, 
803: we may be able to determine observationally whether transition 
804: objects produced via grain growth are intermixed with systems that 
805: have formed giant planets,  
806: by measuring the radial distribution of gas in the disk. 
807: The radial distribution of gas would be continuous for the grain 
808: growth scenario, whereas it would show a radial gap if a 
809: Jovian mass planet has formed.   
810: 
811: Finally, we note that if we interpret the transition objects
812: as systems in which the stellar accretion rate has been reduced
813: from its original value (either by the presence of a giant 
814: planet or by photoevaporation) so that the accretion rate on to 
815: the star no longer traces the outer disc accretion rate, 
816: this implies a reduced scatter in the relation between
817: disc mass and accretion rate. 
818: If we consider only the single, non-transition stars as the
819: systems in which  
820: the stellar accretion rate might more closely trace the
821: {\it disc\/} accretion rate (Figure 3, bottom),
822: the range in $\Mdotstar$ at a given $\Mdisk$ is reduced by a factor
823: of $\sim 3$ compared to the range when the transition objects are
824: included.
825: 
826: By considering only single, non-transition objects, 
827: the relation between disc mass and stellar accretion rate has 
828: steeper a slope compared to that in Figure 3 (top), quite close to
829: the expected value of unity.   
830: There are many potential explanations for the remaining scatter.
831: Disks may have substantial dead zones that do 
832: not participate in accretion (Gammie 1996).   
833: Disk grain opacities may evolve (D'Alessio et al.\ 2006;
834: Natta et al.\ 2006), or the dust content 
835: that is measured by the submillimeter continuum may not reliably 
836: trace the gas (Takeuchi \& Lin 2005).  
837: Alternatively, 
838: the viscosity parameter $\alpha$ may not be a constant from
839: star to star, or  
840: discs may possess a range of initial angular momenta 
841: (Hartmann et al.\ 1998; Dullemond et al.\ 2006).
842: 
843: 
844: One potential problem with the interpretation that the transition
845: objects with higher disc masses have formed Jovian mass companions
846: is that their SEDs do not appear to show the distinctive
847: ``gap'' structure one expects if a Jovian mass planet is present.
848: Several of these sources (GM Aur, DM Tau) have been fit with little
849: to no dust in the inner disc (e.g., Calvet et al.\ 2005), in marked
850: contrast to the highly optically thick inner disc (within
851: $\Rinner < \Rhole$) that would be expected for transition objects
852: with T Tauri-like accretion rates of $10^{-9}-10^{-8}\Msunperyr$.
853: This conundrum suggests the importance of developing a deeper
854: understanding of the processes that affect the vertical distribution 
855: of grains in the outer disk, the transport of grains from the outer 
856: to the inner disc, as well as the transformation of grain properties
857: in the inner disc (e.g., grain growth, destruction in spiral shocks)
858: in order to use SEDs as a finely honed tool for distinguishing among
859: the possibly physical origins of a transition SED.
860: 
861: Rice et al.\ (2006) have suggested that the filtering of
862: grains at gap edges can help to explain the lack of dust opacity in
863: the inner disc (within $\Rinner$).  As they describe, gas-grain 
864: coupling may drive grains
865: larger than 1--10$\micron$ away from the gap edge allowing only a
866: small fraction of
867: the solid mass to accrete past $\Rhole$.  Such a mechanism will
868: reduce the amount of solids reaching both the inner disc and
869: the planet in this stage of evolution.
870: As an additional mechanism, the high grain
871: densities that arise as material from the outer disc is
872: concentrated into narrow accretion streams may also give rise
873: to rapid grain growth and a further reduction in the disc opacity.
874: Similarly, the possible presence of a dead zone (Gammie 1996)
875: in the outer disc, combined with grain growth and settling out of
876: the surface regions of the disk (i.e., the region undergoing accretion), 
877: may also reduce the mass of small grains reaching $\Rhole$.
878: A dead zone has been invoked, similarly, by Ciesla (2006) in the 
879: context of {\it non}-accreting disks, where a transition-like SED 
880: may be produced in disks that have experienced significant grain 
881: growth but have not formed giant planets.  
882: 
883: An additional potential problem with the interpretation that the 
884: transition objects with higher disc masses have formed Jovian mass 
885: companions is that the mass doubling time for the putative 
886: accreting planet is quite short.  The median stellar accretion 
887: rate for these systems is $\sim 3 \times 10^{-9}\Msunperyr$, which
888: for a planetary accretion rate 10 times larger implies a mass 
889: doubling time of 0.03 Myr, an interval short compared
890: to the $\sim 1$ Myr age of Taurus.  
891: Such a mass accretion timescale might be consistent with the 
892: observed fraction of transition objects if sources in Taurus are 
893: approximately coeval, giant planet formation is common, and it takes 
894: close to 1 Myr to form a planet massive enough to open a gap.
895: Alternatively, 
896: other processes for forming transition disks may be at work. 
897: If so, they must provide a natural explanation for the 
898: apparent tendency for transition objects to have large disk 
899: masses and low stellar accretion rates.   
900: 
901: Given these uncertainties, it would be useful to identify
902: additional ways of confirming the presence of a giant planet beyond
903: the modeling of SEDs.
904: Perhaps one of the most convincing approaches would be to
905: search for direct evidence of the presence of such a planet.
906: If the stellar accretion rate in these systems is $\sim 0.1$
907: of the outer disc accretion rate, the planet then accretes
908: at a high rate, $\sim 9$ times the stellar accretion rate,
909: making the planet easier to detect.
910: Among the transition objects in Taurus,
911: GM Aur is the most promising system for such a search because
912: of the potentially large separation between the planet and star
913: ($\sim$ 20\,AU or $0.13\arcsec$).  It also has
914: a large accretion rate. 
915: The stellar accretion rate of $10^{-8}\Msunperyr$ implies an
916: accretion rate of $\sim 9 \times 10^{-8}\Msunperyr$ on to the planet.
917: Extrapolating from the calculations of 
918: Hubickyj, Bodenheimer \& Lissauer (2005) and 
919: Papaloizou \& Nelson (2005), the accretion luminosity 
920: of the planet is expected to be substantial, $\sim 0.02\Lsun$.
921: The resulting modest contrast ratio between the accreting planet 
922: and the central star may greatly facilitate the detection of such 
923: an object. 
924: 
925: 
926: 
927: \section{Summary}
928: 
929: Using the approach introduced here, a study of the demographics of 
930: accreting T Tauri stars, we find that 
931: transition objects inhabit restricted regions of
932: the $\Mdotstar$ vs. $\Mdisk$ plane.  
933: Compared to non-transition 
934: single stars in Taurus, they have stellar accretion rates that are 
935: typically $\sim 10$ times lower at the same disc mass. 
936: In addition, the median disc mass for the transition objects is
937: $\sim 4$ times larger than the median disc mass for the 
938: non-transition single stars.  
939: The decrement in the stellar accretion rates and the higher
940: average disc masses are suggestive that the formation of Jovian
941: mass planets plays a role in explaining the origin of 
942: some transitional SEDs in Taurus. 
943: 
944: For the transition objects with higher disc masses ($\ga 0.01\Msun$),
945: it seems plausible that a Jovian mass planet has created a large gap
946: and suppressed the accretion rate on to the star 
947: (Lubow \& D'Angelo 2006; Varni\`ere et al.\ 2006).   
948: The high accretion rate $\sim 9\Mdotstar$ inferred for the
949: Jovian mass planets in these systems would make an 
950: accreting giant planetary companion easier to detect using
951: either direct imaging or through infrared interferometry.
952: The remaining transition objects
953: have the low disc masses $\la 0.001\Msun$ expected 
954: for discs that have cleared inner holes through photoevaporation.
955: In contrast, the paucity of sources
956: with stellar accretion rates and disc masses similar 
957: to those of non-transition, single T Tauri stars 
958: leads us to question the formation of planetesimals or low mass
959: protoplanets as a likely pathway to a transitional SED. 
960: Planetesimal or low mass protoplanet formation may be unable to
961: completely remove small dust grains from the disc;
962: alternatively, this phase of evolution may be very short-lived.
963: 
964: Accreting transition objects that arise as a consequence 
965: of the formation of either planetesimals or a Jovian mass planet
966: present a significant theoretical challenge: how to sustain an
967: optically thin inner region while accreting material from a
968: dusty outer disc.  Understanding
969: the physical mechanisms that can sustain an optically thin
970: inner region in an accreting transition disc
971: represents a critical step toward understanding
972: the evolutionary state of these objects.
973: 
974: Finally, considering transition objects as a distinct
975: demographic group among accreting T Tauri stars, whose
976: stellar accretion rates have been altered as a
977: consequence of either giant planet formation or disc
978: photoevaporation, leads to a tighter relation between
979: disc masses and stellar accretion rates, with a slope
980: between the two quantities that is close to the value of
981: unity expected in theories of disc accretion.
982: These results, while suggestive, are based on a small sample
983: of objects.  Future observations aimed at determining disc masses,
984: stellar accretion rates, and SEDs for large samples of T Tauri stars
985: will be critical in improving our understanding
986: of the demographics and nature of transition objects.
987: 
988: 
989: 
990: 
991: 
992: 
993: 
994: \section*{Acknowledgments}
995: We are grateful to Lee Hartmann, Nuria  
996: Calvet, Doug Lin, and Geoff Bryden for stimulating and
997: insightful discussions on this topic.  
998: We also thank the anonymous referee for thoughtful comments 
999: that improved the manuscript.  
1000: 
1001: 
1002: 
1003: \begin{thebibliography}{}
1004: 
1005: \bibitem[\protect\citeauthoryear{}{}]{} 
1006: Alexander R. D., Clarke C. J., Pringle J. E. 2006, MNRAS, 369, 229
1007: 
1008: \bibitem[\protect\citeauthoryear{}{}]{} 
1009: Andrews S. M., Williams J. P. 2005, ApJ, 631, 1134
1010: 
1011: \bibitem[\protect\citeauthoryear{}{}]{} 
1012: Baraffe I., Chabrier G., Allard F., Hauschildt P. H. 1998, A\&A, 377, 
1013: 403
1014: 
1015: \bibitem[\protect\citeauthoryear{}{}]{} 
1016: Bate M. R., Lubow S. H., Ogilvie G. I., Miller K. A.
1017: 2003, MNRAS, 341, 213
1018: 
1019: \bibitem[\protect\citeauthoryear{}{}]{} 
1020: Bergin E., et al. 2004, ApJ, 614, L133
1021: 
1022: \bibitem[\protect\citeauthoryear{}{}]{} 
1023: Boss A. P. 2005, ApJ, 629, 535
1024: 
1025: \bibitem[\protect\citeauthoryear{}{}]{} 
1026: Bryden G., Chen X., Lin D. N. C., Nelson R. P., Papaloizou J. C. B.
1027: 1999, ApJ, 514, 344
1028: 
1029: \bibitem[\protect\citeauthoryear{}{}]{} 
1030: Calvet N., D'Alessio P., Hartmann L., Wilner D., Walsh A., 
1031: Sitko M. 2002, ApJ, 568, 1008 
1032: 
1033: \bibitem[\protect\citeauthoryear{}{}]{} 
1034: Calvet N., Muzerolle J., Brice\~no C., Hern\'andez J., Hartmann L., 
1035: Saucedo J. L., Gordon K. D. 2004, AJ, 128, 1294 (C04) 
1036: 
1037: \bibitem[\protect\citeauthoryear{}{}]{} 
1038: Calvet N., et al.\ 2005, ApJ, 630, L185
1039: 
1040: \bibitem[\protect\citeauthoryear{}{}]{} 
1041: Ciesla, F. J. 2006, ApJ, in press (astro-ph/0611811)
1042: 
1043: \bibitem[\protect\citeauthoryear{}{}]{} 
1044: Clarke C. J., Gendrin A., Sotomayor M. 2001, MNRAS, 328, 485
1045: 
1046: \bibitem[\protect\citeauthoryear{}{}]{} 
1047: D'Alessio P., Calvet N., Hartmann L., Lizano S., Cant\'o J. 
1048: 1999, ApJ, 527, 893
1049: 
1050: \bibitem[\protect\citeauthoryear{}{}]{} 
1051: D'Alessio P. et al.\ 2005, ApJ, 621, 461
1052: 
1053: \bibitem[\protect\citeauthoryear{}{}]{} 
1054: D'Alessio P., Calvet N., Hartmann L., Franco-Hern\'andez R.,
1055: Serv\'in H. 2006, ApJ, 638, 314
1056: 
1057: \bibitem[\protect\citeauthoryear{}{}]{} 
1058: D'Angelo G., Henning T., Kley W. 2003, ApJ, 599, 548
1059: 
1060: \bibitem[\protect\citeauthoryear{}{}]{} 
1061: D'Antona F., Mazzitelli I. 1998, 
1062: in R. Rebolo, E. L. Martin, M. R. Zapatero Osorio, eds,
1063: ASP Conf. Ser. 134, 
1064: Astron. Soc. Pac., San Francisco, p. 442
1065: 
1066: \bibitem[\protect\citeauthoryear{}{}]{} 
1067: Dullemond C. P., Dominik C. 2005, A\&A, 434, 971
1068: 
1069: \bibitem[\protect\citeauthoryear{}{}]{} 
1070: Dullemond C. P., Natta A., Testi L. 2006, ApJ, 645, L69
1071: 
1072: \bibitem[\protect\citeauthoryear{}{}]{} 
1073: Durisen R. H., Cai K., Mej\'ia A. C., Pickett M. K. 2005,
1074: Icarus, 173, 417 
1075: 
1076: \bibitem[\protect\citeauthoryear{}{}]{} 
1077: Furlan E., et al.\ 2006, ApJS, 165, 568 
1078: 
1079: \bibitem[\protect\citeauthoryear{}{}]{} 
1080: Gammie C. F. 1996, ApJ, 457, 355
1081: 
1082: \bibitem[\protect\citeauthoryear{}{}]{} 
1083: Gauvin L. S., Strom K. M. 1992, ApJ, 385, 217
1084: 
1085: \bibitem[\protect\citeauthoryear{}{}]{} 
1086: Gullbring E., Hartmann L., Brice\~no C., Calvet N.
1087: 1998, ApJ, 492, 323 (G98)
1088: 
1089: \bibitem[\protect\citeauthoryear{}{}]{} 
1090: Gullbring E., Calvet N., Muzerolle J., Hartmann L. 2000, ApJ, 544, 927 (G00)
1091: 
1092: \bibitem[\protect\citeauthoryear{}{}]{} 
1093: Hartigan P., Edwards S., Ghandour L. 1995, ApJ, 452, 736 (HEG95)
1094: 
1095: \bibitem[\protect\citeauthoryear{}{}]{} 
1096: Hartmann L., Calvet N., Gullbring E., D'Alessio P. 1998,
1097: ApJ, 495, 385 (H98)
1098: 
1099: \bibitem[\protect\citeauthoryear{}{}]{} 
1100: Hartmann L., Megeath S. T., Allen L., Luhman K., Calvet N., 
1101: D'Alessio P., Franco-Hernandez R., Fazio G. 2005, ApJ, 
1102: 629, 881
1103: 
1104: \bibitem[\protect\citeauthoryear{}{}]{} 
1105: Hartmann L., D'Alessio P., Calvet N., Muzerolle J. 2006, ApJ, 648, 484
1106: 
1107: \bibitem[\protect\citeauthoryear{}{}]{} 
1108: Hollenbach D. J., Yorke H. W., Johnstone D. 2000, in
1109: Mannings, V., Boss, A. P., Russell, S. S., eds,  
1110: Protostars and Planets IV, University of Arizona, Tucson, p. 401
1111: 
1112: \bibitem[\protect\citeauthoryear{}{}]{} 
1113: Hubickyj O., Bodenheimer P., Lissauer J. J. 2005, Icarus,
1114: 179, 415
1115: 
1116: \bibitem[\protect\citeauthoryear{}{}]{} 
1117: Jensen E. L. N., Mathieu R. D., Fuller G. A. 1996,
1118: ApJ, 458, 312
1119: 
1120: \bibitem[\protect\citeauthoryear{}{}]{} 
1121: Kenyon S. J., Hartmann L. 1995, ApJS, 101, 117
1122: 
1123: \bibitem[\protect\citeauthoryear{}{}]{} 
1124: Kley W. 1999, MNRAS, 303, 696
1125: 
1126: \bibitem[\protect\citeauthoryear{}{}]{} 
1127: Lissauer J. J., Stevenson D. J. 2006, in 
1128: B. Reipurth, ed, Protostars and Planets V, University of Arizona Press, 
1129: Tucson, in press
1130: 
1131: \bibitem[\protect\citeauthoryear{}{}]{} 
1132: Lubow S. H., Seibert M., Artymowicz P. 1999, ApJ, 526, 1001
1133: 
1134: \bibitem[\protect\citeauthoryear{}{}]{} 
1135: Lubow S. H., D'Angelo G. 2006, ApJ, 641, 526
1136: 
1137: \bibitem[\protect\citeauthoryear{}{}]{} 
1138: Marcy G., Butler R. P., Fischer D., Vogt S., Wright J. T.,
1139: Tinney C. G., Jones H. R. A. 2005, Prog. Theor. Phys. Supp.,
1140: 158, 2005 
1141: 
1142: \bibitem[\protect\citeauthoryear{}{}]{} 
1143: Marsh K. A., Mahoney M. J. 1992, ApJ, 395, L115
1144: 
1145: \bibitem[\protect\citeauthoryear{}{}]{} 
1146: Mathieu R. E., Mart\'in E. L., Magazzu A. 1996, BAAS, 188, 60.05
1147: 
1148: \bibitem[\protect\citeauthoryear{}{}]{} 
1149: McCabe C., Ghez A. M., Prato L., Duch\^ene G., Fisher R. S.,
1150: Telesco C. 2006, ApJ, 636, 932
1151: 
1152: \bibitem[\protect\citeauthoryear{}{}]{} 
1153: Muzerolle J., Calvet N., Hartmann L. 2001, ApJ, 550, 944
1154: 
1155: \bibitem[\protect\citeauthoryear{}{}]{} 
1156: Muzerolle J., Hillenbrand L., Calvet N., Brice\~no C., 
1157: Hartmann L. 2003, ApJ, 592, 266
1158: 
1159: \bibitem[\protect\citeauthoryear{}{}]{} 
1160: Muzerolle J., et al.\ 2006, ApJ, 643, 1003
1161: 
1162: \bibitem[\protect\citeauthoryear{}{}]{} 
1163: Natta A., Testi L., Muzerolle J., Randich S., Comer\'on F.,
1164: Persi P. 2004, A\&A, 424, 603
1165: 
1166: \bibitem[\protect\citeauthoryear{}{}]{} 
1167: Natta A., Testi L., Calvet N., Henning T., Waters R.,
1168: Wilner D. 2006, in 
1169: B. Reipurth, ed, Protostars and Planets V, University of Arizona Press, 
1170: Tucson, in press
1171: 
1172: \bibitem[\protect\citeauthoryear{}{}]{} 
1173: Padgett D. L., et al.\ 2006, ApJ, 645, 1283
1174: 
1175: \bibitem[\protect\citeauthoryear{}{}]{} 
1176: Palla F., Stahler S. 1999, ApJ, 525, 772 
1177: 
1178: \bibitem[\protect\citeauthoryear{}{}]{} 
1179: Papaloizou J. C. B., Nelson R. P. 2005, A\&A, 433, 247
1180: 
1181: \bibitem[\protect\citeauthoryear{}{}]{} 
1182: Prato L., Simon M., Mazeh T., McLean I. S., Norman D., 
1183: Zucker S. 2002, ApJ, 569, 863
1184: 
1185: \bibitem[\protect\citeauthoryear{}{}]{} 
1186: Quillen A. C., Blackman E. G., Frank A., Varni\`ere P.
1187: 2004, ApJ, 612, L137
1188: 
1189: \bibitem[\protect\citeauthoryear{}{}]{} 
1190: Rice W. K. M., Wood K., Armitage P. J., Whitney B. A., 
1191: Bjorkman J. E. 2003, MNRAS, 342, 79
1192: 
1193: \bibitem[\protect\citeauthoryear{}{}]{} 
1194: Rice W. K. M., Armitage P. J., Wood K., Lodato G.
1195: 2006, astro-ph/0609808
1196: 
1197: \bibitem[\protect\citeauthoryear{}{}]{} 
1198: Sano, T., Miyama, S. M., Umebayashi, T., Nakano, T. 
1199: 2000, ApJ, 543, 486 
1200: 
1201: \bibitem[\protect\citeauthoryear{}{}]{} 
1202: Sicilia-Aguilar A., et al.\ 2006, 638, 897
1203: 
1204: \bibitem[\protect\citeauthoryear{}{}]{} 
1205: Skrutskie M. F., Dutkevitch D., Strom S. E., Edwards S.,
1206: Strom K. M., Shure M. A.  1990, AJ, 99, 1187
1207: 
1208: \bibitem[\protect\citeauthoryear{}{}]{} 
1209: Strom K. M., Strom S. E., Edwards S., Cabrit S., 
1210: Skrutskie M. F. 1989, AJ, 97, 1451
1211: 
1212: \bibitem[\protect\citeauthoryear{}{}]{} 
1213: Takeuchi T., Lin D. N. C. 2005, ApJ, 623, 482
1214: 
1215: \bibitem[\protect\citeauthoryear{}{}]{} 
1216: Varni\`ere P., Blackman E. G., Frank A., Quillen A. C. 2006,
1217: ApJ, 640, 1110
1218: 
1219: \bibitem[\protect\citeauthoryear{}{}]{} 
1220: White R. J., Ghez A. M. 2001, ApJ, 556, 265 (WG01)
1221: 
1222: \bibitem[\protect\citeauthoryear{}{}]{} 
1223: White R. J., Hillenbrand L. A. 2004, ApJ, 616, 998
1224: 
1225: %\bibitem[\protect\citeauthoryear{}{}]{} 
1226: %White R. J., Hillenbrand L. A. 2005, ApJ, 621, L65 (WH05)
1227: 
1228: \bibitem[\protect\citeauthoryear{}{}]{} 
1229: Wolk S. J., Walter F. M. 1996, AJ, 111, 2066
1230: 
1231: \end{thebibliography}
1232: 
1233: 
1234: 
1235: 
1236: \begin{table*}
1237:  \centering
1238:  \begin{minipage}{140mm}
1239:   \caption{Taurus-Auriga Stellar Accretion Rates and Disk Masses.}
1240: %  \begin{tabular}{@{}llllrlr@{}}
1241:   \begin{tabular}{@{}llllrlr}
1242:   \hline
1243: Name & 
1244: Spectral Type & 
1245: Binarity$^{\rm a}$ &
1246: SED$^{\rm b}$ &
1247: $\log \Mdotstar^{\rm c}$ &
1248: $\Mdotstar$ Ref. &
1249: $\log M_d$$^{\rm d}$ \\
1250:  \hline
1251: AA Tau   & K7 &  S  &  C  &  -8.48 &  G98 & -1.89\\
1252: Anon 1   & M0 &  S  &  W  &  $<$-8.56 & WG01 & $<$-3.40\\
1253: BP Tau   & K7 &  S  &  C?  &  -7.54 &  G98 & -1.74\\
1254: CI Tau   & K7 &  S  &  C  &  -7.19 & H98 & -1.55\\
1255: CoKu Tau$/$4 & M1.5 & X &  T  &  $<$-10 & $^{\rm e}$ & -3.34\\
1256: CW Tau   & K3 &  S  &  C  &  -7.61 & WG01 & -2.62\\
1257: CX Tau   & M2.5 &  S  &  T  &  -8.97 & H98 &  -2.92\\
1258: CY Tau   & M1 &  S  &  C  &  -8.12 & G98 &  -2.20\\
1259: DD Tau A & M1 &  B &  C  &  -8.72 & WG01 & -3.13\\
1260: DE Tau   & M2 &  S  &  C  &  -7.59 & G98 &  -2.28\\
1261: DF Tau A & M0.5 &  B  &  C  &  -7.62 & WG01 & -3.40\\
1262: DG Tau   & K7-M0 &  S  &  F  &  -6.30 & G00 & -1.62\\
1263: DH Tau   & M1 &  S  &  C  &  -8.30 & H98 & -2.48\\
1264: DK Tau   & K7 &  B  &  C  &  -7.42 & G98 & -2.30\\
1265: DL Tau   & K7 &  S  &  C  &  -6.79 & WG01 & -1.05\\
1266: DM Tau   & M1 &  S  &  T  &  -7.95 & H98 & -1.62\\
1267: DN Tau   & M0 &  S  &  C/T  &  -8.46 & G98 & -1.54\\
1268: DO Tau   & M0 &  S  &  C/F  &  -6.85 & G98 & -2.15\\
1269: DP Tau   & M0 &  S  &  C/F  &  -7.88 & H98 & $<$-3.30\\
1270: DQ Tau   & M0 &  B  &  C  &  -9.22 & G98 & -1.72\\
1271: DR Tau   & K7 &  S  &  F  &  -6.50 & G00 & -1.72\\
1272: DS Tau   & K5 &  S  &  C  &  -7.89 & G98 & -2.22\\
1273: FM Tau   & M0 &  S  &  C  &  -8.45 & H98 & -2.82\\
1274: FO Tau A & M2 &  B  &  T?  &  -7.90 & WG01 & -3.19\\
1275: FQ Tau   & M2 &  B &  T? &  -6.45 & H98  & -2.87\\
1276: FS Tau   & M1 &  B &  C/F &  -9.12 & WG01 & -2.64\\
1277: FV Tau A & K5 &  B &  C  &  -7.32 & WG01 & -2.96\\
1278: FX Tau   & M1 &  B &  C  &  -8.65 & H98 & -3.05\\
1279: FY Tau   & K7 &  S  &  X  &  -7.41 & H98 & -3.17\\
1280: FZ Tau   & M0 &  S  &  C  &  -7.32 & WG01 & -2.70\\
1281: GG Tau Aa & K7 &  B  &  C  &  -7.52 & WG01 & -0.64\\
1282: GH Tau A & M2 &  B &  C  &  -8.52 & WG01 & -3.13\\
1283: GK Tau   & K7 &  B &  C  &  -8.19 & G98 & -2.80\\
1284: GM Aur   & K3 &  S  &  T  &  -8.02 & G98 & -1.60\\
1285: GO Tau   & M0 &  S  &  C/T  &  -7.93 & H98 & -1.15\\
1286: Haro 6$-$37 & K6 & B  &  C  &  -7.00 & H98 & -1.97\\
1287: HBC 376  & K7 &  S  &  X  &  $<$-8.92 & WG01 & $<$-3.54\\
1288: HBC 388  & K1 &  S  &  W  &  $<$-8.03 & WG01 & $<$-3.49\\
1289: HO Tau   & M0.5 &  S  &  C  &  -8.86 & H98 & -2.68\\
1290: Hubble 4 & K7 &  S  &  W  &  $<$-7.78 & WG01 & $<$-3.36\\
1291: IP Tau   & M0 &  S  &  C  &  -9.10 & G98 & -2.55\\
1292: IQ Tau   & M0.5 &  S  &  C  &  -7.55 & H98 & -1.66\\
1293: IS Tau A & K7 &  B &  C  &  -7.72 & WG01 & -2.84\\
1294: L1551$-$51 & K7 &  S  &  W  &  $<$-9.12 & WG01 & $<$-3.19\\
1295: L1551$-$55 & K7 &  S  &  X  &  $<$-9.32 & WG01 & $<$-3.56\\
1296: LkCa 14  & M0 &  S  &  X  &  $<$-8.47 & WG01 & $<$-3.35\\
1297: LkCa 15  & K5 &  S  &  C/T  &  -8.87 & H98 & -1.32\\
1298: LkCa 19  & K0 &  S  &  X  &  $<$-9.62 & WG01 & $<$-3.30\\
1299: LkCa 4   & K7 &  S  &  W  &  $<$-8.35 & WG01 & $<$-3.70\\
1300: LkCa 5   & M2 &  S  &  W  &  $<$-9.62 & WG01 & $<$-3.72\\
1301: RW Aur A & K3 &  B  &  C  &  -7.12 & WG01 & -2.44\\
1302: RY Tau   & G1 &  S  &  C  &  -7.11 & C04 & -1.74\\
1303: SU Aur   & G1 & X  &  C  &  -8.26 &  C04 & -3.05\\
1304: T Tau A  & G6 & B &  F  &  -7.12 & WG01 & -2.09\\
1305: UX Tau A & K2 &  S  &  T  &  -9.00 & $^{\rm f}$ & -2.29\\
1306: UY Aur   & K7 &  B &  C  &  -7.18 & G98 & -2.74\\
1307: UZ Tau E & M1 &  B &  C  &  -6.48 & VBJ93  & -1.72\\
1308: V410 Tau A & K3 &  B  &  W  &  $<$-8.42 & WG01 & $<$-3.44\\
1309: \hline
1310: \end{tabular}
1311: \end{minipage}
1312: \end{table*}
1313: 
1314: \begin{table*}
1315:  \centering
1316:  \begin{minipage}{140mm}
1317:   \contcaption{Taurus-Auriga Stellar Accretion Rates and Disk Masses.}
1318: %  \begin{tabular}{@{}llllrlr@{}}
1319:   \begin{tabular}{@{}llllrlr}
1320:   \hline
1321: Name & 
1322: Spectral Type & 
1323: Binarity$^{\rm a}$ &
1324: SED$^{\rm b}$ &
1325: $\log \Mdotstar^{\rm c}$ &
1326: $\Mdotstar$ Ref. &
1327: $\log M_d$$^{\rm d}$ \\
1328:  \hline
1329: V773 Tau & K3 &  B  &  C/T  &  $<$-9.62  & WG01 & -3.33\\
1330: V807 Tau A & K7 &  B &  X  &  -8.02 & WG01 & -3.01\\
1331: V819 Tau & K7 &  S  &  W$^{\rm g}$  &  $<$-8.48 & WG01 & $<$-3.35\\
1332: V827 Tau & K7 &  S  &  W  &  $<$-8.15 & WG01 & $<$-3.50\\
1333: V836 Tau & K7 &  S  &  T?  & -8.98 & HEG95 & -2.00\\
1334: V955 Tau & K7 &  B &  C  &  -8.12 & WG01 & -3.30 \\
1335: \hline
1336: \end{tabular}
1337: \medskip \\
1338: $^{\rm a}$ Binarity where S$=$single; 
1339: B$=$binary with orbital separation $< 2.5\arcsec$; \\
1340: $^{\rm b}$ SED class where C$=$classical T Tauri star;
1341: F$=$extremely active accretor or flat spectrum source;
1342: T$=$transition object, i.e., no IR excess shortward of
1343: $10\micron$ and a significant excess at longer wavelengths;
1344: W$=$optically thin or no IR excess shortward of $30\micron$;
1345: X$=$unknown.
1346: Entries with a C$?$, T$?$ or c$/$T indicate uncertainty in the
1347: true level of the short-wavelength IR excess either because of
1348: variability (e.g., BP Tau, V836 Tau) or companions (e.g., V773 Tau). 
1349: These designations are also used in cases of ambiguity between
1350: a transition phenomenon and naturally lower near-IR excess emission around
1351: later-type stars where the irradiation flux impinging on the disc is smaller
1352: (e.g., FO Tau, FQ Tau, GO Tau). \\
1353: $^{\rm c}$ Stellar accretion rate in units of $\Msunperyr$. \\
1354: $^{\rm d}$ Disk masses in units of $\Msun$ from
1355: Andrews \& Williams (2005). \\
1356: $^{\rm e}$ $\Mdotstar$ upper limit estimated assuming no accretion
1357: component is measurable at H$\alpha$. \\
1358: $^{\rm f}$ $\Mdotstar$ estimated from an accretion model of
1359: the H$\alpha$ profile shown in Alencar \& Basri (2000). \\
1360: $^{\rm g}$ Furlan et al.\ (2006) note that a 2MASS companion
1361: with unknown mid-infrared properties, which also lay in the long-low slit,
1362: possibly accounts for the longer wavelength excess in the measured SED.
1363: \end{minipage}
1364: \end{table*}
1365: 
1366: 
1367: \begin{table*}
1368:  \centering
1369:  \begin{minipage}{140mm}
1370:   \caption{SED Decrements of Transition Objects.}
1371:   \begin{tabular}{@{}lcll@{}}
1372:   \hline
1373: Name &
1374: SED &
1375: $\Delta_{3.5}$ &
1376: $\Delta_{5.0}$ \\
1377:     &
1378:     &
1379: dex &
1380: dex \\
1381:  \hline
1382: CoKu Tau$/$4 & T & 0.4  & 0.7 \\
1383: CX Tau   & T   &   0.4  & 0.3 \\
1384: DM Tau   & T   &   0.2  & $>$0.3 \\
1385: DN Tau   & C/T &   0.2  & 0.1 \\
1386: GM Aur   & T &   0.5  & 0.4 \\
1387: FO Tau A & T?  &   0.25  & 0.15 \\
1388: FQ Tau   & T?  &   0.4  & 0.3 \\
1389: GO Tau   & C/T &   0.3  & 0.25 \\
1390: LkCa 15  & C/T &   0.2  & 0.2 \\
1391: UX Tau A & T   &   0.15  & 0.25 \\
1392: V773 Tau & C/T &   0.35  & 0.25 \\
1393: V836 Tau & T?  &   0.35  & 0.05 \\
1394: \hline
1395: \end{tabular}
1396: \end{minipage}
1397: \end{table*}
1398: 
1399: 
1400: 
1401: 
1402: \end{document}
1403: