1: \documentclass[12pt,preprint]{aastex}
2:
3: %\documentclass{emulateapj}
4: %\usepackage{rotating}
5:
6: % Revision History:
7: % 2001-Oct-02 Valenti Reformatted paragraphs. Migrated towards AASTeX
8: % and ApJ conventions. Switched bibliography to
9: % \cite and \bibitem formulation.
10:
11: % New commands.
12: \newcommand{\Lya}{\hbox{Ly$\alpha$}}
13: \newcommand{\Lyb}{\hbox{Ly$\beta$}}
14: \newcommand{\Jup}{\hbox{$J^\prime$}}
15: \newcommand{\Jlo}{\hbox{$J^{\prime\prime}$}}
16: \newcommand{\Vup}{\hbox{$v^\prime$}}
17: \newcommand{\Vlo}{\hbox{$v^{\prime\prime}$}}
18: \newcommand{\Elo}{\hbox{$E^{\prime\prime}$}}
19: \newcommand{\kms}{\hbox{km s$^{-1}$}}
20: \newcommand{\erg}{\hbox{erg cm$^{-2}$ s$^{-1}$}}
21: \newcommand{\erga}{\hbox{erg cm$^{-2}$ s$^{-1}$ \AA$^{-1}$}}
22: \newcommand{\mdotyr}{\hbox{$M_\odot$ yr$^{-1}$}}
23: \newcommand{\IUE}{\textit{IUE}}
24: \newcommand{\Chandra}{\textit{Chandra}}
25: \newcommand{\HST}{\textit{HST}}
26: \newcommand{\STIS}{STIS} % supposed to be roman typeface -JAV
27: \newcommand{\FUSE}{\textit{FUSE}}
28: \newcommand{\IDL}{IDL} % assume roman typeface for now -JAV
29:
30: % Temporarily suppress figures insertion when EPS files not available.
31: % \renewcommand{\plotone}[1]{}
32: % \renewcommand{\plottwo}[2]{}
33:
34: % AASTeX macros.
35: %\slugcomment{To be submitted to ApJ Letters}
36: \shorttitle{[Ne II] Emission from TW Hya}
37: \shortauthors{Herczeg et al.}
38:
39: % Begin document.
40: \begin{document}
41:
42: % Title.
43: \title{High-resolution Spectroscopy of [Ne II] Emission from TW Hya}
44:
45: % Authors.
46: \author{Gregory J. Herczeg\altaffilmark{1}, Joan
47: Najita\altaffilmark{2}, Lynne A. Hillenbrand\altaffilmark{1}, Ilaria Pascucci\altaffilmark{2}}
48:
49:
50: \altaffiltext{1}{Caltech, MC105-24, 1200 E. California Blvd., Pasadena, CA 91125}
51: \altaffiltext{2}{National Optical Astronomy Observatory, Tucson, AZ 85719}
52: \altaffiltext{3}{Steward Observatory, The University of Arizona, Tucson, AZ 85721}
53:
54:
55: % Abstract
56: \begin{abstract}
57: We present high-resolution echelle spectra of [\ion{Ne}{2}]
58: $12.81\mu$m emission from the classical T Tauri star (CTTS) TW Hya
59: obtained with MICHELLE on Gemini North. The line is centered at the
60: stellar radial velocity and
61: has an intrinsic FWHM of $21\pm4$ \kms.
62: %, and is unresolved at the angular resolution of $0\farcs75$.
63: The line width is broader than other narrow emission lines typically
64: associated with the disk around TW Hya. If formed in a disk, the line broadening could result from
65: turbulence in a warm disk atmosphere, Keplerian rotation at an average
66: distance of 0.1 AU
67: from the star, or a photoevaporative flow from the optically-thin region
68: of the disk.
69: We place upper limits on the [\ion{Ne}{2}] emission flux from the CTTSs DP Tau and BP Tau.
70: \end{abstract}
71:
72: % Official ApJ keywords in alphabetical order.
73: % See http://www.journals.uchicago.edu/ApJ/information.html
74: \keywords{
75: stars: pre-main sequence --- stars: planetary systems:
76: protoplanetary disks matter --- infrared: stars: TW Hya, DP Tau, BP Tau}
77:
78: % Citation commands:
79: %
80: % \citet{Nam98} for text citations, e.g. "Name (1998) says..."
81: % \citep{Nam98} for parenthetical citations, e.g. "is true (Name 1998)."
82: % \citep{Nam98,Alt01} to cite multiple refs, e.g. "(Name 1998; Alt 2001)"
83: % \citep{Nam98,Nam01} automatically merges, e.g. "(Name 1998; 2001)"
84: % \citet*{Nam98} or \citep* for the first citation of 3 author papers.
85: % \citet*{Nam98} or \citep* for the first citation of 3 author papers.
86: % \citep[e.g.,][]{Nam98} for leading text, e.g. "(e.g., Name 1998)"
87: % \citep[, for example]{Nam98} for trailing text.
88: % \citep[e.g.,][, for example]{Nam98} for leading and trailing text.
89: % \citealt{Nam98} is \citet{Nam98} without parentheses.
90: % \citealp{Nam98} is \citep{Nam98} without parentheses.
91: % \citeauthor{Nam98} only lists authors, e.g. "Name"
92: % \citeyear{Nam98} only lists the year, e.g. "1998"
93: % \citeyearpar{Nam98} only lists the year in parentheses, e.g. "(1998)"
94: % \citetext{\citealp{Nam98}; see also \citealp{Alt01}} embeds text, e.g.
95: % "(Name 1998; see also Alt 2001)"
96: %
97: % Look up "natbib" on the web for a more detailed description.
98:
99: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
100:
101: \section{INTRODUCTION}
102: X-ray and UV emission from classical T Tauri stars (CTTSs) can affect the chemical structure of their protoplanetary disks by ionizing the disk surface, dissociating molecules, and heating the gas and grains \citep[see reviews by][]{Naj07,Dul07,Ber07}.
103: Direct evidence of irradiated disks is seen in far-UV H$_2$ emission, which requires excitation by \ion{H}{1} Ly$\alpha$ irradiation and can be attributed to emission from the inner disk of some CTTSs \citep{Her06}. Other observational evidence for FUV- and X-ray-irradiated disks around CTTSs has been more circumstantial.
104: The underabundance of HCN relative to CN in disks around several CTTSs
105: may be attributed to strong Ly$\alpha$ emission \citep{Ber03}.
106: \citet{Bar03} suggest that X-ray irradiation warms the disk surface
107: enough to produce emission in the H$_2$ 1-0 S(1) line. The excess FUV
108: continuum emission may result from the electronic cascade and
109: dissociation following excitation of H$_2$ by energetic electrons
110: \citep{Her04,Ber04}. EUV irradiation of the disk can produce a photoevaporative wind, which could be visible in the low-velocity component of some optical forbidden lines \citep[see review by][]{Dul07}.
111:
112: Recently, \citet{Pas07} and \citet{Lah07} may have found a new probe of gas at the disk surface by detecting
113: emission in the [\ion{Ne}{2}] $12.81\mu$m fine-structure line in
114: low-resolution {\it Spitzer} IRS spectra of CTTSs. [\ion{Ne}{2}] was
115: detected in four of six targets in the \citet{Pas07} sample and in 15 of 76 targets in the \citet{Lah07} sample. Relative to the $\sim 1$ Myr old CTTSs in Taurus, the
116: six stars observed by \citet{Pas07} have faint mid-IR continuum
117: emission, are X-ray bright, and have mass accretion rates an order of
118: magnitude lower than the average CTTS of similar spectral type. The
119: \citet{Lah07} sample and detections are comprised mostly of T Tauri stars with an IR
120: excess. The only class I source in their sample was also detected in
121: [\ion{Ne}{2}],
122: and several Herbig Ae/Be stars were undetected in [\ion{Ne}{2}].
123: The sources both with and without [\ion{Ne}{2}] detections from \citet{Lah07}
124: show a range of X-ray luminosities.
125: %The det However, only 5 of the 15 detections have
126: % X-ray detections, and several strong X-ray emitters were undetected in [\ion{Ne}{2}].
127: % The strongest [\ion{Ne}{2}] flux was from T Cha, which has a
128: % transition disk.
129: \citet{Lah07} also detected [\ion{Ne}{3}] emission from one source (Sz102),
130: and four of the six targets with H$_2$ 0--0 S(2) emission also show [\ion{Ne}{2}] emission.
131:
132: \citet{Pas07} and \citet{Lah07} both argue that the irradiated disk surface is the most likely
133: origin for this emission. Two ionization paths can ionize Ne and thereby
134: produce the [\ion{Ne}{2}] emission. \citet{Gla07} propose that the
135: line is formed because of Ne ionization by K-shell absorption of
136: stellar X-rays at energies $>0.9$ keV. The observed [\ion{Ne}{2}]
137: fluxes and the [\ion{Ne}{3}]/[\ion{Ne}{2}] flux ratio and lower limits
138: are consistent with predictions from this X-ray ionization model.
139: Alternatively, EUV photons shortward of the \ion{Ne}{1} ionization edge
140: at $575$ \AA\ can produce [\ion{Ne}{2}] emission if the EUV emission
141: is able to penetrate through neutral gas in the accretion column and
142: wind.
143: %\citet{Lah07} also discuss excitation of [\ion{Ne}{2}] where winds shock the disk surface, and argue based on line ratios that any such shock must be high-velocity ($\geq 70$ \kms).
144: If the disk origin is confirmed, then the [\ion{Ne}{2}] 12.81$\mu$m line would be a direct probe of the irradiation of the protoplanetary disk by the central star.
145:
146: In this paper, we present high-resolution echelle spectra of the 12.81$\mu$m
147: region of the CTTSs TW Hya, BP Tau, and DP Tau obtained with MICHELLE
148: on Gemini North. TW Hya is a nearby 10 Myr old CTTS with a disk that
149: is believed to be optically-thin within 4 AU from the star based on
150: models of its infrared spectral energy distribution \citep{Cal02} and
151: an inner hole and bright ring imaged at 7mm with the VLA \citep{Wil07}. BP Tau and DP Tau are 1 Myr old CTTSs with optically-thick inner disks \citep{Ken95,Muz03}. All three stars are X-ray bright (Table 1).
152: We find strong
153: [\ion{Ne}{2}] emission from TW Hya and place upper limits on the
154: [\ion{Ne}{2}] flux from DP Tau and BP Tau.
155: %The [\ion{Ne}{2}] emission from TW Hya
156: %has an intrinsic FWHM of 22.5 \kms\ and is centered at the radial velocity of the star.
157: %, and
158: %is not spatially extended beyond $0\farcs75$ (40 AU at the distance of TW Hya).
159: %Najita (??) reports that {\it Spitzer} previously obtained a flux of ??? from TW Hya.
160: We discuss the origin of the
161: emission in the context of profiles of other emission lines from TW
162: Hya and suggest that the line is formed in a disk, although formation
163: in a heated photosphere cannot be ruled out.
164: % at a spectral resolution similar to our MICHELLE spectra.
165: If the [\ion{Ne}{2}] emission is formed in a disk, the line width could result from Keplerian
166: rotation at a distance of 0.1 AU from the star, tran-sonic turbulence,
167: or detecting the front and back of a photoevaporative flow.
168:
169: \section{OBSERVATIONS AND DATA REDUCTION}
170: We used the echelle mode of MICHELLE on Gemini North to observe the
171: CTTSs TW Hya, DP Tau, and BP Tau on 9 March 2007. MICHELLE has a
172: 320x240 Si:As array with a plate scale of $0\farcs201$ \citep{Gla97}. When used
173: with the echelle
174: spectrograph MICHELLE has a spectral resolution of $\sim 30,000$.
175: We centered the targets on the slit using images obtained with the
176: Si-5 filter, which has an effective wavelength of $11.7$ $\mu$m. We used a 2-pixel wide slit and the Si-6 filter to obtain spectra centered at
177: 12.815$\mu$m and covering $\sim0.049$ $\mu$m. We nodded
178: $10^{\prime\prime}$ off the slit at 35s intervals. Our total
179: on-source integration was 40.4 min for TW Hya, 20 min for BP Tau and
180: 22.5 min for DP Tau. These three targets were observed at median
181: airmasses of 1.8, 1.3, and 1.9, respectively. Observations of
182: $\alpha$ CMa and the asteroid Iris were used as telluric standards.
183: The seeing varied from $0\farcs6$ to $0\farcs85$, as measured in the cross-dispersion spatial profile from our unresolved telluric standards.
184: %Poor weather forced a premature end to the run.
185:
186: In each set of four nod positions, the counts in each column of the off-source images are scaled to and then subtracted from the counts in the same column of the on-source sky background. This scaling is necessary because the second derivative of the sky brightness versus time was non-zero.
187: We correct for fringing in the cross-dispersion direction by fitting a high-order polynomial to the residual background in each column.
188: We extract the spectrum from a 7-pixel window centered on the source,
189: and then subtract background sky spectra obtained from 20-pixel
190: windows on both sides of the source. Fringing in the spectral
191: direction is not detected or corrected for because of the low S/N in
192: the continuum (Table 1).
193: %However, any such fringing should not affect the narrow region around the [\ion{Ne}{2}] line.
194:
195: We estimate the telluric correction for each observation by
196: interpolating the standard spectra of Iris and $\alpha$ CMa at
197: intermediate airmasses. The atmospheric correction is applied to each
198: nod prior to combining the images. Several O$_3$, CO$_2$, and H$_2$O
199: lines are present in the wavelength range of our spectra.
200: %, including a strong CO$_2$ line located at -34 \kms\ from the [\ion{Ne}{2}] line from TW Hya at the time of our observation.
201: We calibrate wavelengths to an accuracy of $\sim 2$ \kms\ per pixel
202: using a 2nd-order fit to sky emission lines identified with ATRAN
203: \citep{Lor92}. Each pixel covers $\sim 3.6$ \kms. We estimate a
204: resolution of $3.2\pm0.5$ pixels ($11.5\pm1.8$ \kms) from the FWHM of
205: four weak, narrow sky emission lines.
206:
207: TW Hya was also observed on 12 April 2006 during re-commissioning of the East Arm Echelle
208: Spectrograph \citep{Lib95} on the 5m Hale Telescope at Palomar. The
209: instrument is fed by a bundle of fibers that together subtend
210: $1\farcs5$ and illuminate a $2048\times2048$ pixel CCD. The spectrum
211: covers most of the 4000-9000 \AA\ range at a resolution of
212: 15,000-45,000, depending on location on the chip. The relative wavelength solution in each order was obtained using a ThAr lamp. The absolute
213: wavelength solution is determined separately for each echelle order by matching photospheric lines to the radial velocity of TW Hya and is accurate to $\sim 3$ \kms.
214:
215: No sufficient flat field was obtained during the commissioning.
216: However, observations of the white dwarf Feige 34 and the A0 star
217: HD74198 obtained in the same night yield spectra with a count rate
218: that peaks near the center of each order and falls off smoothly to the
219: edge of the detector. The absence of flat-fielding has a negligible effect on the
220: line profile shapes and equivalent width measurements because they are
221: extracted from narrow spectral regions.
222: %The fiber-fed design of the echelle spectrograph makes sky-subtraction difficult.
223: We use observations of several other stars to identify the approximate flux in telluric emission lines, which are located at -20 \kms\ relative to the photospheric velocity of TW Hya.
224: The telluric and stellar components of the [\ion{O}{1}] lines at
225: 5577, 6300, and 6363 \AA\ can therefore be separated. The \ion{Na}{1} D emission lines are undetected, with a sensitivity
226: limited by strong telluric emission.
227: We calculate line fluxes from equivalent widths by estimating the
228: continuum flux level from non-simultaneous archival {\it HST}/STIS
229: optical spectra \citep{Her04,Rob05}.
230:
231: The heliocentric radial velocity of TW Hya is $12.2\pm0.5$ \kms\
232: (Weintraub et al. 2000; see also Kastner et al. 1997, de la Reza et
233: al. 1989) and of BP Tau is 15.8 \kms\ \citep{Her95}. The heliocentric
234: radial velocity of DP Tau has not been measured. We estimate a radial
235: velocity of $17\pm5$ \kms\ by averaging the radial velocities of the T Tauri stars GO Tau and V955 Tau \citep{Her95}, which are both about 10$^\prime$ from DP Tau.
236:
237:
238: \section{CHARACTERIZING THE [\ion{Ne}{2}] EMISSION}
239: Figure 1 shows our MICHELLE spectra of TW Hya, BP Tau, and DP Tau. The
240: continuum is detected from each target. We measure equivalent widths
241: for the [\ion{Ne}{2}] fine-structure line ($^2P_{3/2}$ -- $^2P_{1/2}$)
242: at $12.81355\pm0.00002$ $\mu$m \citep{Yam85} of $62\pm11$ \AA\ for TW Hya and $2\sigma$ upper limits of $<10$ and $<11$ \AA\ for non-detections from BP Tau and DP Tau, respectively. The upper limits are calculated by assuming the same line profile as that of TW Hya, and would be larger if the hypothetical lines from these objects were broader.
243: We convert equivalent widths to $(5.9\pm1.1)\times10^{-14}$, $<0.7\times10^{-14}$, and $<1.1\times10^{-14}$ erg cm$^{-2}$ s$^{-1}$ using $12.8\mu$m continuum flux measurements from {\it Spitzer} IRS spectra \citep{Uch04,Fur06}. These
244: fluxes assume that the mid-IR continuum and the [\ion{Ne}{2}] emission are not
245: spatially extended beyond the seeing.
246:
247: The [\ion{Ne}{2}] line profile from TW Hya is consistent with a single Gaussian profile
248: centered at $-2\pm3$ \kms\ in the stellocentric frame with a FWHM of $6.7\pm0.8$ pixels ($24\pm3$ \kms).
249: %, much broader than the $\sim2.7$ pixel spectral resolution.
250: Deconvolving the line width from the instrumental resolution yields an
251: intrinsic line FWHM of $21\pm4$ \kms. Two components (a broad
252: component at the stellocentric velocity
253: with a flux enhancement on the blue side of the line)
254: are visually apparent in the line profile. However, based on the S/N
255: detected in the continuum (Table 1) the profile shape is not a significant deviation from a single Gaussian profile.
256: The telluric CO$_2$ line did not divide out sufficiently for some of
257: the TW Hya observations and results in increased uncertainty on the blue side of the line profile. The
258: [\ion{Ne}{2}] emission is more compact than the $0\farcs75$ angular
259: resolution (40 AU at the distance of TW Hya), which was estimated from both the mid-IR continuum of TW Hya and the telluric
260: standard $\alpha$ CMa, which was observed shortly before TW Hya.
261:
262: Table 2 and Figure 2 compare the [\ion{Ne}{2}] emission to a sample of other emission lines from TW
263: Hya, excluding those significantly affected by wind absorption (e.g.,
264: \ion{H}{1} $\lambda6563$, \ion{O}{1}
265: $\lambda 1302$, \ion{C}{2} $\lambda1335$, \ion{He}{1}
266: $\lambda10830$, see Herczeg et al. 2002, Edwards et al. 2006).
267: %When comparing profiles, it is useful to keep in mind that uncertainties may be introduced to the optical lines by the lack of sky subtraction or flat fielding.
268: These lines show a range of emission profile shapes, and were observed at resolutions similar to that of our MICHELLE spectra.
269: The velocity and FWHM listed in Table 2 are measured from Gaussian fits to the emission lines.
270: The energy of the upper level ($E^\prime$) for these transitions
271: (0.60--23.1 eV) are all substantially larger than that of the
272: fine-structure [\ion{Ne}{2}] line (0.10 eV).
273: Most lines, including the \ion{Ca}{2} IR triplet and the
274: [\ion{O}{1}] $\lambda6300,6363$, show narrow emission centered at
275: the stellocentric velocity. No high-velocity ($>100$ \kms) or
276: low-velocity ($<20$ \kms) component of other optical forbidden lines
277: are detected \citep{Ham94,Har95}. Several lines show broad
278: redshifted emission profiles, presumably produced by the accretion
279: flow. Many of the near-UV \ion{Fe}{2} lines have both narrow and broad
280: components centered at the stellocentric velocity.
281: % Most lines that trace $<6,000$ K gas are narrower than the [\ion{Ne}{2}] emission, but many hotter lines are very broad.
282:
283:
284: \section{DISCUSSION}
285:
286: \subsection{Comparison To Other Emission Lines From TW Hya}
287: Emission lines from CTTSs may be formed at or near the disk surface,
288: in the accretion funnel flow and shock, in outflows, in the stellar
289: chromosphere, or in the accretion-heated part of the stellar photosphere.
290: We first rule out formation of the [\ion{Ne}{2}] emission in the accretion
291: flow and shock. Lines commonly attributed to accretion, including \ion{He}{1} $\lambda5876$ and the
292: \ion{C}{4} $\lambda1549$, \ion{N}{5} $\lambda1240$, and \ion{O}{6}
293: $\lambda1035$ doublets \citep{Ale02,Her02,Lam04,Joh07,Yan07}, have much
294: broader line widths than the [\ion{Ne}{2}] emission from TW Hya and
295: are redshifted from the stellocentric velocity. Second,
296: we rule out fast outflows because the [\ion{Ne}{2}] line is centered
297: at or near the stellocentric velocity and is limited to low velocities. Formation in a very slow
298: outflow may be possible if both the front and back sides of the flow
299: are visible. Third, we rule out a
300: chromospheric origin. Some lines from Table 2 are attributed to
301: chromospheric emission and seen from main-sequence dwarfs and
302: weak-line T Tauri stars \citep{Ayr05}, but these lines are
303: narrower than the [\ion{Ne}{2}] line.
304: Several other emission lines, including the \ion{C}{1} $\lambda1656$
305: and \ion{C}{2}]
306: $\lambda2326$ multiplets, \ion{Cl}{1} $\lambda1351.7$, and some NUV
307: \ion{Fe}{2} lines, have slightly larger
308: widths than the [\ion{Ne}{2}] emission and may form in a
309: heated stellar photosphere. However, these lines have larger energy upper
310: levels than the fine-structure
311: [\ion{Ne}{2}] line and are unlikely to trace the same gas.
312: More likely is the possibility that the [\ion{Ne}{2}] line forms
313: either in the inner disk or in a photoevaporative flow from the
314: optically thin region of the disk, as suggested by
315: the excitation energy (0.10 eV), width (21 \kms), and
316: velocity centroid ($-2\pm3$ \kms) of the line.
317:
318: %Some
319: %lines from Table 2 are centered at the stellocentric velocity and are
320: %also seen from main-sequence and weak-line T Tauri stars
321: %\citep{Ayr05}. These lines are attributed to the stellar
322: %chromosphere, but are much more narrow than the [\ion{Ne}{2}] line.
323: %\ion{Ca}{2} IR triplet emission is also seen weakly from WTTSs
324: %\citep{Bat93,Muz98}. However, polarization measurements and a weak
325: %correlation between \ion{Ca}{2} flux and optical veiling indicate that
326: %this narrow emission from TW Hya is produced by some accretion-related process
327: %\citep{Yan07,Ale02}.
328:
329: Collisional ionization equilibrium would require
330: temperatures of $>30,000$ K to produce [\ion{Ne}{2}] emission.
331: All identified lines from TW Hya that probe $>20,000$ K have a larger width
332: ($>40$ \kms) than the [\ion{Ne}{2}] line. We therefore infer that the
333: [\ion{Ne}{2}] emission is not likely collisionally ionized.
334: Instead, the emission is likely ionized
335: by X-ray or EUV-irradiation of gas at or near the disk surface.
336: Indeed, Glassgold et al.~(2007) speculated that X-ray irradiated
337: disks would produce strong [\ion{Ne}{2}] emission. The observed
338: width of the [\ion{Ne}{2}] line is larger than that shown by
339: Glassgold et al.~(in their Fig.~5). However, Glassgold et al.~(2007)
340: considered only the emission that would arise from a non-turbulent
341: X-ray irradiated disk that extends beyond 1\,AU.
342:
343: Irradiated disks are believed to produce the molecular line emission
344: detected from TW Hya in the FUV H$_2$, near-IR H$_2$,
345: and fundamental CO transitions
346: \citep{Wei00,Her02,Ret04,Sal07,Naj07}.
347: The emission arises from warm ($500-3000$ K) gas, and the
348: FUV H$_2$ lines in particular require irradiation by strong FUV
349: emission from the central star. The line widths are $>16$, $8$, and $14$ \kms\ narrower for H$_2$ 1-0
350: S(1), FUV H$_2$, and fundamental CO lines, respectively, than the
351: [\ion{Ne}{2}] line. The narrow width of the H$_2$ 1-0 S(1) suggests
352: formation at large radii.
353:
354: Interestingly, the widths of the
355: fundamental CO, FUV H$_2$, and [\ion{Ne}{2}] lines
356: increase with their approximate formation temperatures of
357: $\sim 500-1500$\,K (Salyk et al. 2007; Najita et al.\ 2003),
358: $\sim 2500$\,K (as observed for TW Hya; Herczeg et al.\ 2004),
359: and 4000--10,000\,K (based on theoretical modeling;
360: Glassgold et al.\ 2007; Hollenbach et al.\ 2000),
361: respectively.
362: This trend might indicate a common origin for the lines in a disk
363: with an outwardly decreasing temperature gradient (e.g., Carr et al.\ 2004).
364: If the [\ion{Ne}{2}] line width results from
365: Keplerian disk rotation, then the emission arises at an average
366: distance of $\sim 0.1$
367: AU of the star, assuming a $7\pm1^\circ$ inclination of the disk
368: around TW Hya \citep{Qi04}.
369: The trend might also indicate a common origin for the lines in a turbulent
370: disk atmosphere with a vertically increasing temperature gradient,- i.e.,
371: increasing (transonic) turbulent broadening higher in the disk
372: atmosphere. For example,
373: the line width of the [\ion{Ne}{2}] emission could be explained
374: entirely by turbulent broadening in 10,000 K gas.
375: With a combination of Keplerian rotation and turbulence, the
376: line could form at disk radii larger than $\sim 0.1$ AU and in gas
377: cooler than $10,000$ K.
378: Alternatively,
379: the width of the [\ion{Ne}{2}] line could also be explained
380: if the emission arises from a photoevaporative flow from the
381: optically thin region of the disk (at $<4$ AU).
382: We explore these possibilities using simple scaling arguments to
383: point out the challenges associated with these interpretations.
384:
385: \subsection{X-ray Excitation of the Inner Disk}
386:
387: %Simple scaling arguments point out some challenges for interpreting
388: %the line broadening as either Keplerian
389: %rotation or photoevaporation.
390: The [\ion{Ne}{2}] flux is converted to a mass from the Einstein
391: A-value of $8.59\times10^{-3}$ s$^{-1}$ \citep{Kra06}, a solar Ne/H abundance
392: ratio of $8.9\times10^{-5}$ \citep{And89}, a Ne ionization
393: fraction of $\zeta$, and an average atomic mass per atom of 1.4. In LTE the fraction of \ion{Ne}{2} in the
394: $^2P_{3/2}$ level is 0.14 and 0.28 at 1000 and 5000 K if the electron
395: density is larger than the critical density, $10^5$ cm$^{-3}$, as
396: assumed here. If the line is optically-thin and not attenuated by
397: dust within the disk, then the gas mass is
398: $0.5-1.5\times10^{-9}{\zeta^{-1}}$ $M_\odot$, where this range
399: includes uncertainties in the line flux and the population of the upper level. If we assume that the
400: [\ion{Ne}{2}] emission arises in the annulus of an empty cylinder with
401: a height $H$,
402: radius $R$, and width $dR<R$, then the total hydrogen column density of the layer
403: measured parallel to the disk is $N($H$)\sim 1.5\times
404: 10^{22}\frac{0.2 AU}{R}\frac{0.2 AU}{H} \zeta^{-1}$ cm$^{-2}$.
405:
406: Either the disk atmosphere or the photoevaporative flow scenario requires that the ionizing
407: photons penetrate through the emission region, which
408: constrains the neutral hydrogen column density, $N$(\ion{H}{1}), as
409: measured radially from the central star.
410: Based on the {\it Chandra} X-ray spectrum of TW Hya \citep{Kas02},
411: 23\% of photons with energies $>0.9$ keV penetrate to depths of
412: $N$(\ion{H}{1})=$10^{22}$ cm$^{-2}$, but only the hardest 3\% of those
413: photons can penetrate to $N$(\ion{H}{1})=$10^{23}$ cm$^{-2}$.
414: However, almost all EUV photons longward of 100 and 500 \AA\ are attenuated
415: for $N$(\ion{H}{1})$>10^{20}$ and $10^{18}$ cm$^{-2}$, respectively.
416:
417: If we set $R\lesssim0.2$ AU, then $N$(H)$\sim 10^{22} \zeta^{-1}$
418: cm$^{-2}$. Even if the ionization fraction is high, $N$(\ion{H}{1})
419: will be large enough to require the ionization of
420: Ne by X-rays and a disk height comparable to the disk radius.
421: Such a
422: geometrically thick disk may also be required to explain the strength
423: of the H$_2$ fluorescence in FUV spectra of TW Hya \citep{Her04}.
424: The scale height of gas in hydrostatic equilibrium at $R=0.2$ AU from
425: TW Hya can be estimated as $0.04 \left(\frac{T}{10^4 K}\right)^{-0.5}$
426: AU. \citet{Eis06} calculated that the dust temperature is $\sim1100$
427: K at the inner dust truncation radius ($\sim 0.06$ AU). However,
428: since only a small amount of micron-sized dust is present within the 4
429: AU dust clearing radius \citep{Cal02}, the gas temperature at
430: these radii may be significantly hotter than the dust temperature.
431: Models of optically-thick disks with viscosity parameter $\alpha=0.01$ suggest a gas
432: surface density of 25--90 g cm$^{-2}$ at 0.1 AU from TW Hya
433: \citep{Dal98}. The mass of the [\ion{Ne}{2}] emitting region would be
434: about $\frac{0.1}{\zeta}$\% of the total disk mass within 0.2 AU of
435: the star. The [\ion{Ne}{2}] emission in this scenario traces a thin
436: surface layer of the disk higher than one scale height above the disk
437: midplane. This interpretation of [\ion{Ne}{2}] emission is
438: challenging if $\zeta<0.1$ because
439: the larger total mass traced by the
440: [\ion{Ne}{2}] emission would require an even larger geometrical
441: disk height.
442:
443:
444:
445: This situation can be ameliorated if turbulent broadening contributes
446: to or dominates the [\ion{Ne}{2}] line width. If the disk atmosphere
447: is turbulent and the turbulent velocity is close to the sound speed,
448: then the FWHM of a line would be $\sim 21$ \kms\ at $10,000$ K,
449: consistent with the observed line width. At this temperature the
450: [\ion{Ne}{2}] emission could arise from a wide range of radii, perhaps
451: extending out to 10-20 AU. \citet{Gla07}
452: suggest that at 1--10 AU, the gas within $N(H)=10^{20}$ cm$^{-2}$ of the disk
453: surface will be heated to 4000 K. At this temperature the average
454: radii for formation of [\ion{Ne}{2}] emission would be $\sim 0.5$ AU
455: from the star.
456:
457: Approximately transonic turbulence in disk atmospheres is inferred
458: from the intrinsic line broadening of CO overtone emission from
459: accretion disks (Carr et al. 2004; Najita et al. 1996). The
460: dissipation of turbulence in disk atmospheres could also produce the
461: extra heating that is needed to explain the strength of CO fundamental
462: emission from T Tauri disks \citep{Gla04}. While these
463: observations refer to emission from cool, molecular disk gas that is
464: located deeper in the disk atmosphere, similar transonic
465: turbulence may characterize warmer, atomic gas higher up in the disk
466: atmosphere where the [\ion{Ne}{2}] emission may arise. Indeed,
467: simulations of weakly magnetized disks find that MHD turbulence
468: generates a warm ($\sim 10,000$ K) disk corona with a transonic velocity
469: dispersion above several disk scale heights \citep{Mil00}.
470:
471: \subsection{Photoevaporation of an EUV-irradiated Disk Surface}
472:
473: On the other hand, in the disk dissipation models of Alexander et al. (2006ab)
474: EUV irradation of the disk produces a $10,000$ K flow with a velocity of
475: $\sim 10$ \kms\ from the disk surface, near the gravitational radius
476: $r_g$ (see also Hollenbach \& Gorti 2007).
477: Analytical formulations
478: for $r_g$ assuming an abiabatic gas \citep{Lif03} suggest that
479: $r_g\approx 1.4\frac{M_*}{M_\odot}\frac{10^4 K}{T}$AU$\approx 1$ AU
480: for a CTTS with the mass of TW Hya. In the photoevaporation simulations by \citet{Fon04},
481: most of the mass loss occurs inside $\sim 6$ AU for a source like TW
482: Hya. While these expectations are for a radially continuous gas disk,
483: \citet{Cal02}
484: find that at 10 $\mu$m the disk is optically thin within 4 AU of TW
485: Hya.
486: Any additional frontal illumination of the inner edge of the optically-thick
487: disk at 4 AU by FUV and X-ray photons may provide an additional heating
488: source that could help puff up the disk or help launch the
489: photoevaporative flow.
490: If the [\ion{Ne}{2}] emission is formed in a photoevaporative flow
491: within 4 AU of the star, the apparent line broadening could result
492: from detecting [\ion{Ne}{2}] emission from both the front and back
493: sides of the disk.
494: %The best-fit [\ion{Ne}{2}] line velocity of
495: %$-2\pm3$ \kms\ is consistent with both no velocity shift or a small
496: %blueshift.
497: The profile
498: could be slightly blueshifted if some [\ion{Ne}{2}] emission occurs in a photoevaporative
499: flow beyond the dust clearing radius, where emission on the back
500: side of the disk would not be detectable.
501: \citet{Ale06a} predict
502: that the photoevaporative flow will have velocities $>10$ \kms\ at
503: distances greater than 4 AU
504: from the star, implying the possibility of a stronger blueshifted
505: asymmetry than is observed.
506:
507: %If the gas has $N$(\ion{H}{1})$\lesssim10^{18.5}$ cm$^{-2}$, then EUV
508: %photoionization model requires $\frac{H\times R}{\zeta}>4$ (AU)$^2$,
509: %which may be reasonable if the bulk of the [\ion{Ne}{2}] emission
510: %occurs near the dust disk rim at 4 AU.
511:
512: If we assume the photoevaporative flow is produced at $R\sim4$ AU,
513: then the height of the flow (including both front and back sides of
514: the disk) is
515: $H=\frac{1.5\times10^{20} {\rm cm}^{-2}}{\zeta N({\rm H})}$
516: AU. If $\zeta=1$ and $N$(H)$\sim10^{19}$ cm$^{-2}$, then $H=15$ AU.
517: The velocity of a photoevaporative flow is expected to be $v\sim
518: 10$ \kms\ at 4 AU above the disk \citep{Ale06a}. The crossing time for the evaporating gas across the
519: emission region on either side of the disk ($H/2$) is then 3.5 yr.
520: The estimated mass loss rate is
521: $\frac{M_{{\rm Ne II}}}{3.5 {\rm yr}}\sim 3\times 10^{-10} f$ $M_\odot$ yr$^{-1}$, where $f$ is the fraction
522: of [\ion{Ne}{2}] emission that arises in the photoevaporative flow.
523:
524:
525: Such a low-velocity ($\sim 10$ \kms) photoevaporative flow would be distinct from the observed
526: high-velocity ($-50$ to $-200$ \kms) wind
527: that dominates blueshifted absorption profiles of
528: atomic and singly-ionized lines from
529: TW Hya (Edwards et al. 2006;
530: Johns-Krull et al. 2007).
531: The estimated photoevaporation mass loss rate
532: of about one-fifth the stellar mass accretion rate is similar to the $\sim
533: 3\times10^{-10}$ $M_\odot$ yr$^{-1}$ estimate for a
534: radially-continuous disk that is irradiated with an
535: EUV photon flux of $\sim10^{41}$ phot s$^{-1}$ \citep{Hol94}. Several
536: parameters in this simple scaling argument, including $f$, are uncertain and together could
537: suppress our estimate for the mass loss rate. This scenario
538: constrains the $N$(H) in the photoevaporative flow to a narrow range.
539: A smaller $N$(H) would increase
540: the volume and decrease the $n_e$ below the critical density by
541: increasing $H$, where any flow is likely less collimated.
542: A larger $N$(H) decreases the height and crossing time, thereby
543: increasing the estimate of the mass loss rate.
544:
545: The photoevaporation scenario is contingent upon some EUV emission
546: reaching the disk surface. The photon flux at $<912$ \AA\ from TW Hya
547: of $\sim 10^{41}-10^{42}$ phot s$^{-1}$ is dominated by line emission
548: \citep{Ale05}. Since both the $\sim 10^5$ K FUV lines and $10^6$ K X-ray
549: lines are attributed to the accretion flow \citep{Joh07,Kas02}, this
550: estimated EUV emission likely applies to emission produced near the
551: accretion shock. Whether this EUV emission reaches the disk depends
552: on a small neutral hydrogen column density. A smaller amount of EUV emission is
553: expected to be produced by the transition region in magnetic
554: structures and may reach the disk, depending on the geometry of the
555: accretion flow.
556: Based on an analysis of FUV H$_2$ emission, \citet{Her04} suggested
557: that an absorbing column of $N$(\ion{H}{1})$\sim 5\times10^{18}$ cm$^{-2}$
558: \ion{H}{1}, perhaps in a stellar outflow, is present between the \ion{H}{1} Ly$\alpha$
559: emission from TW Hya and the warm H$_2$ in the disk. If $N$(\ion{H}{1})$\lesssim2\times10^{18}$
560: between the EUV and [\ion{Ne}{2}] emission resions, then
561: 5--15\% of the EUV photons could reach the disk and lead to
562: some photoevaporation. A larger intervening $N$(\ion{H}{1}) would
563: prevent EUV photoevaporation.
564:
565:
566: %EUV emission from the
567: %accretion flow
568: %can be attenuated by relatively small $N$(\ion{H}{1}).
569: %EUV emission
570: %is also produced by the transition region in magnetic structures,
571: %although such emission
572: %could also be attenuated by the accretion flow \citep{Gre07} and the
573: %wind. \citet{Her04} used the \ion{H}{1}-Ly$\alpha$ pumped H$_2$
574: %emission from TW Hya to estimate the \ion{H}{1} Ly$\alpha$ emission that
575: %irradiates the disk. This reconstructed \ion{H}{1} Ly$\alpha$ profile
576: %suggests that neutral gas with
577: %$N$(\ion{H}{1})$\leq is present between the Ly$\alpha$ and
578: %H$_2$ emission regions. The neutral gas appears to be blueshifted relative to
579: %the H$_2$ gas by $\sim 70$ \kms.
580:
581: \subsection{Conclusions}
582:
583: Our lower luminosity limits for
584: [\ion{Ne}{2}] emission from BP Tau and DP Tau are comparable to the
585: line luminosity from TW Hya, despite having mass accretion rates an
586: order of magnitude larger than that onto TW Hya. [\ion{Ne}{2}]
587: is detected from several CTTSs with low mass accretion rates both here and
588: in \citet{Pas07}.
589: Of the few stars in the \citet{Lah07} sample with measured mass
590: accretion rates, [\ion{Ne}{2}] emission is detected from several weak
591: accretors but not from several strong accretors. If weak accretors do
592: show stronger
593: [\ion{Ne}{2}] emission than strong accretors, this effect
594: may be explained because the intervening
595: $N$(\ion{H}{1}) in the accretion flow and wind should scale with the mass accretion and mass loss
596: rates. However, a similar
597: effect may be seen for X-ray-irradiation models if older CTTSs are
598: more X-ray luminous than younger CTTSs.
599:
600: The [\ion{O}{1}] emission from TW Hya may be produced by recombination
601: subsequent to photoionization. The [\ion{O}{1}] lines
602: are centered at the
603: stellocentric velocity, do not correlate with optical veiling
604: \citep{Ale02}, and are not seen from WTTSs.
605: %Since the
606: %[\ion{O}{1}] line from TW Hya is not likely to be thermally excited, it likely
607: %traces recombination and requires the presence of photons shortward of
608: %the O ionization potential at 912 \AA.
609: [\ion{O}{1}] emission is thought to arise
610: from the disk surface of some Herbig Ae/Be stars
611: \citep{Boh94,Ack05}. [\ion{O}{1}] emission produced subsequent to
612: OH dissociation by FUV emission from the central star will be
613: undetectable from disks around CTTSs \citep{Sto00}. However,
614: models of EUV or X-ray irradiated disks may explain some [\ion{O}{1}]
615: emission in the low-velocity components of CTTSs \citep{Fon04}.
616: That the [\ion{O}{1}] lines are
617: narrower than the [\ion{Ne}{2}] line suggests that, if the [\ion{O}{1}]
618: emission is
619: produced in a disk, it traces less turbulent gas or gas at larger disk
620: radius than the [\ion{Ne}{2}] emission.
621: %Broader (FWHM=41 \kms) lines in the
622: %semi-forbidden \ion{C}{2}]
623: %$\lambda2325$ multiplet are also centered at the stellocentric
624: %velocity and suggest the presence of low-density ($n_e<10^{10}$
625: %cm$^{-3}$) gas outside of both the accretion flow and stellar wind.
626:
627:
628: High-resolution spectroscopy of [\ion{Ne}{2}] emission from TW Hya
629: suggests formation
630: in a disk irradiated by X-ray or EUV photons, although formation in a
631: heated photosphere cannot be ruled out. The origin of the line
632: broadening and
633: the ionization path can be identified by high-resolution spectra of
634: additional sources. If the line broadening is dominated by Keplerian
635: rotation, the [\ion{Ne}{2}] line will appear much broader from stars
636: with disks viewed at higher inclinations. If the line broadening is
637: dominated by turbulence, the observed line width from disks viewed at
638: higher inclinations will only be modestly larger.
639: If formed in a photoevaporative wind, the
640: [\ion{Ne}{2}] emission from stars with optically-thick inner disks
641: will likely appear blueshifted, as only the front side of the disk
642: will be detectable. The description of [\ion{Ne}{2}] emission from TW
643: Hya should apply generally to stars
644: with small mass accretion rates. [\ion{Ne}{2}] emission from stars
645: with large mass accretion rates may also be produced in shocks
646: associated with outflows.
647:
648:
649: \section{Acknowledgements}
650: We thank the anonymous referee for helpful comments that improved the clarity of the discussion. This observing time was awarded by Caltech as part of the Keck/Gemini exchange program.
651: We thank Kevin Volk for observing support with MICHELLE during the run and for helping to identify the atmospheric lines with ATRAN. We thank Andrew Pickles for observing TW Hya during the re-commissioning of the East Arm Echelle spectrograph. We thank David Hollenbach for discussing the application of photoevaporation models to the [\ion{Ne}{2}] line.
652:
653:
654: \begin{thebibliography}{}
655: \bibitem[Acke et al.(2005)]{Ack05}
656: Acke, B., van den Ancker, M.E., \& Dullemond, C.P. 2005, A\&A, 436, 209
657:
658: \bibitem[Alencar \& Batalha(2002)]{Ale02}
659: Alencar, S.H.P. \& Batalha, C. 2002, \apj, 571, 378
660:
661: \bibitem[Alexander et al.(2005)]{Ale05}
662: Alexander, R.D, Clarke, C.J., \& Pringle, J.E. 2005, MNRAS, 358, 283
663:
664: \bibitem[Alexander et al.(2006a)]{Ale06a}
665: Alexander, R.D, Clarke, C.J., \& Pringle, J.E. 2006a, MNRAS, 369, 216
666:
667: \bibitem[Alexander et al.(2006b)]{Ale06b}
668: Alexander, R.D, Clarke, C.J., \& Pringle, J.E. 2006b, MNRAS, 369, 229
669:
670: \bibitem[Anders \& Gervesse(1989)]{And89}
671: Anders, E., \& Grevesse, N. 1989, GeCoA, 53, 197
672:
673: %\bibitem[Alexander et al.(2006)]{Ale06}
674: %Alexander, R.D., Clarke, C.J., \& Pringle, J.E. 2006, MNRAS, 369, 229
675:
676: \bibitem[Ayres(2005)]{Ayr05}
677: Ayres, T.R. 2005, Proc. of the 13th Cambridge Workshop on Cool Stars, Stellar Systems and the Sun, Edited by F. Favata et al. ESA SP-560, European Space Agency, 2005., p.419
678:
679: %\bibitem[Bally et al.(2000)]{Bal00}
680: %Bally, J., O'Dell, C.R., \& McCaughrean, M.J. 2000, AJ, 119, 2919
681:
682: \bibitem[Bary et al.(2003)Bary, Weintraub, \& Kastner]{Bar03}
683: Bary, J.S., Weintraub, D.A., \& Kastner, J.H. 2003, \apj, 586, 1136
684:
685: \bibitem[Batalha \& Basri(1993)]{Bat93}
686: Batalha, C.C., \& Basri, G. 1993, ApJ, 412, 363
687:
688: \bibitem[B\"ohm \& Catala(1994)]{Boh94}
689: B\"ohm, T., \& Catala, C. 1994, A\&A, 290, 167
690:
691: \bibitem[Bergin et al.(2003)]{Ber03}
692: Bergin, E., Calvet, N., D'Alessio, P., \& Herczeg, G.J. 2003, ApJ, 591, L159
693:
694: \bibitem[Bergin et al.(2004)]{Ber04}
695: Bergin, E., et al. 2004, ApJ, 614, L133
696:
697: \bibitem[Bergin et al.(2007)]{Ber07}
698: Bergin, E.A., Aikawa, Y., Blake, G.A., \& van Dishoeck, E.F. 2007, PPV
699:
700: \bibitem[Briggs et al.(2007)]{Brig07}
701: Briggs, K.R., et al. 2007, A\&A
702:
703: %\bibitem[Brittain et al.(2007)]{Bri07}
704: %Brittain, S., Simon, T., Najita, J., \& Rettig, T. 2007, ApJ, accepted
705:
706: \bibitem[Calvet \& Gullbring(1998)]{Cal98}
707: Calvet, N., \& Gullbring, E. 1998, ApJ, 509, 802
708:
709: \bibitem[Calvet et al.(2002)]{Cal02}
710: Calvet, N., D'Alessio, P., Hartmann, L., Wilner, D., Walsh, A., \& Sitko,
711: M. 2002, \apj, 568, 1008
712:
713:
714: %\bibitem[Carr, et al.(2001)]{Car01}
715: %Carr, J.S., Mathieu, R.D., \& Najita, J.R. 2001, ApJ, 551, 454
716:
717: %\bibitem[Carr et al.(2004)]{Car04}
718: %Carr, J.S., Tokunaga, A.T., \& Najita, J. 2004, ApJ, 603, 213
719:
720: %\bibitem[Chiang \& Goldreich(1997)]{Chi97}
721: %Chiang, E.I., \& Goldreich, P. 1997, ApJ, 490, 368
722:
723: %\bibitem[Clarke et al.(2001)]{Cla01}
724: %Clarke, C.J., Gendrin, A., \& Sotomayor, M. 2001, MNRAS, 328, 485
725:
726: \bibitem[de la Reza et al.(1989)]{del89}
727: de la Reza, R., Torres, C.A.O., Quast, G.R., Castilho, B.V., \& Vieira, G.L. 1989, ApJ, 343, L61
728:
729: \bibitem[D'Alessio et al.(1998)]{Dal98}
730: D'Alessio, P., Canto, J., Calvet, N., \& Lizano, S. 1998, ApJ, 500, 411
731:
732: %\bibitem[Dullemond et al.(2001)]{Dul01}
733: %Dullemond, C.P., Dominik, C., \& Natta, A. 2001, ApJ, 560, 957
734:
735: \bibitem[Dullemond et al.(2007)]{Dul07}
736: Dullemond, C.P., Hollenbach, D., Kamp, I., D'Alessio, P. 2007, proceedings of {\it P\&PV}, eds. B. Reipurth, D. Jewitt, and K. Keil, 951, 555
737:
738: %\bibitem[Dutrey et al.(2003)]{Dut03}
739: %Dutrey, A., Guilloteau, S., \& Simon, M. 2003, A\&A, 402, 1003
740:
741: \bibitem[Edwards et al.(2006)]{Edw06}
742: Edwards, S., Fischer, W., Hillenbrand, L., \& Kwan, J. 2006, ApJ, 646, 319
743:
744: \bibitem[Eisner et al.(2006)]{Eis06}
745: Eisner, J.A., Chiang, E.I., \& Hillenbrand, L.A. 2006, ApJ, 637, L133
746:
747: \bibitem[Font et al.(2004)]{Fon04}
748: Font, A.S., McCarthy, I.G., Johnstone, D., \& Ballantyne, D.R. 2004, ApJ, 607, 890
749:
750: %\bibitem[Fromang et al.(2002)]{Fro02}
751: %Fromang, S., Terquem, C., \& Balbus, S.A. 2002, MNRAS, 329, 18
752:
753: \bibitem[Furlan et al.(2006)]{Fur06}
754: Furlan, E., et al. 2006, ApJS, 165, 568
755:
756: %\bibitem[Gammie(1996)]{Gam96}
757: %Gammie, C.F. 1996, ApJ, 457, 355
758:
759: \bibitem[Glasse et al.(1997)]{Gla97}
760: Glasse, A.C., Atad-Ettedgui, E.I., \& Harris, J.W. 1997, SPIE, 2871, 1197
761:
762: \bibitem[Glassgold et al.(2004)]{Gla04}
763: Glassgold, A.E., Najita, J.R., \& Igea, J. 2004, ApJ, 615, 972
764:
765: \bibitem[Glassgold et al.(2007)]{Gla07}
766: Glassgold, A.E., Najita, J.R., \& Igea, J. 2007, ApJ, 656, 515
767:
768: %\bibitem[Gorti \& Hollenbach(2007)]{Gor07}
769: %Gorti, U., \& Hollenbach, D. 2007, in prep.
770:
771: \bibitem[Gregory et al.(2007)]{Gre07}
772: Gregory, S.G., Wood, K., \& Jardine, M. 2007, MNRAS, accepted. astro-ph/07042958
773:
774: \bibitem[Hamann(1994)]{Ham94}
775: Hamann, F. 1994, ApJSS, 93, 485
776:
777: \bibitem[Hartigan et al.(1995)]{Har95}
778: Hartigan, P., Edwards, S., \& Ghandour, L. 1995, ApJ, 452, 736
779:
780: \bibitem[Hartmann et al.(1998)]{Har98}
781: Hartmann, L., Calvet, N., Gullbring, E., \& D'Alessio, P. 1998, ApJ, 495, 385
782:
783: \bibitem[Herbig \& Bell(1995)]{Her95}
784: Herbig, G.H., \& Bell, K.R. 1995, yCat, 5073
785:
786: \bibitem[Herczeg et al.(2002)]{Her02}
787: Herczeg, G. J., Linsky, J. L., Valenti, J.A., Johns-Krull, C.M. 2002,
788: \apj, 572, 310 (Paper I)
789:
790: \bibitem[Herczeg et al.(2004)]{Her04}
791: Herczeg, G.J., et al. 2004, ApJ, 607, 369
792:
793: \bibitem[Herczeg et al.(2006)]{Her06}
794: Herczeg, G.J., et al. 2006, ApJS, 165, 256
795:
796: \bibitem[Hollenbach et al.(1994)]{Hol94}
797: Hollenbach, D., Johnstone, D., Lizano, S., \& Shu, F. 1994, ApJ, 428, 654
798:
799: \bibitem[Hollenbach et al.(2000)]{Hol00}
800: Hollenbach, D., Yorke, H.W., \& Johnstone, D. 2000, in Mannings V., Boss A. P., Russell S. S., eds, Protostars and Planets IV. Univ. Arizona Press,
801: Tucson, p. 401
802:
803: \bibitem[Hollenbach \& Gorti(2007)]{Hol07}
804: Hollenbach, D., \& Gorti, U. 2007, to be submitted
805:
806: %\bibitem[Hollenbach \& Tielens(1997)]{Hol97}
807: %Hollenbach, D.J., \& Tielens, A.G.G.M. 1997, ARA\&A, 35, 179
808:
809: %\bibitem[Isella \& Natta(2005)]{Isa05}
810: %Isella, A., \& Natta, A. 2005, A\&A, 438, 899
811:
812: \bibitem[Hughes et al.(2007)]{Wil07}
813: Hughes, A.M., Wilner, D.J., Calvet, N., D'Alessio, P., Claussen, M.J.,
814: Hogerheijde, M.R. 2007, ApJ, accepted. astro-ph/07042442
815:
816:
817: \bibitem[Johns-Krull \& Herczeg(2007)]{Joh07}
818: Johns-Krull, C.M., \& Herczeg, G.J.
819:
820: %\bibitem[Johns-Krull \& Valenti(2001)]{Joh01}
821: %Johns-Krull, C.M., \& Valenti, J.A. 2001, ApJ, 561, 1060
822:
823: %\bibitem[Johnstone et al.(1998)]{Joh98}
824: %Johnstone, D., Hollenbach, D., \& Bally, J. 1998, ApJ, 499, 758
825:
826: %\bibitem[Kamp \& Dullemond(2004)]{Kam04}
827: %Kamp, I., \& Dullemond, C.P. 2004, ApJ, 615, 991
828:
829: \bibitem[Kastner et al.(1997)]{Kas97}
830: Kastner, J.H., Zuckerman, B., Weintraub, D.A., \& Forveille, T. 1997, {\it Science}, 277, 67
831:
832: \bibitem[Kastner et al.(2002)]{Kas02}
833: Kastner, J. H. Huenemoerder, D. P., Schulz, N. S., Canizares, C.R., \&
834: Weintraub, D.A. 2002, \apj, 567, 434
835:
836: \bibitem[Kenyon \& Hartmann(1995)]{Ken95}
837: Kenyon, S.J., \& Hartmann, L. 1995, \apjs, 101, 117
838:
839: \bibitem[Kramida \& Nave(2006)]{Kra06}
840: Kramida, A.E., \& Nave, G. Eur. Phys. J. D, 39, 331
841:
842: %\bibitem[Kwan \& Tademaru(1995)]{Kwa95}
843: %Kwan, J. \& Tademaru, E. 1995, \apj, 454, 382
844:
845: \bibitem[Lahuis et al.(2007)]{Lah07}
846: Lahuis, F., va Dischoeck, E.F., Blake, G.A., Evans, N.J.,
847: Kessler-Silacci, J.E., \& Pontpiddian, K.M. 2007, ApJ, accepted. astro-ph/07042305
848:
849: \bibitem[Lamzin et al.(2004)]{Lam04}
850: Lamzin, S.A., Kravtsova, A.S., Romanova, M.M., \& Batalha, C. 2004, AstL, 30, L413
851:
852: %\bibitem[Lang \& Willson(1979)]{Lan79}
853: %Lang, K.R., \& Willson, R.F. 1979, \apj, 227, L163
854:
855: %\bibitem[Lecavelier des Etangs et al.(2001)]{Lec01}
856: %Lecavelier des Etangs, A., et al. 2001, Nature, 412, L706
857:
858: \bibitem[Libbrecht \& Peri(1995)]{Lib95}
859: Libbrecht, K.G., \& Peri, M.L. 1995, PASP, 107, L62
860:
861: \bibitem[Liffman(2003)]{Lif03}
862: Liffman, K. 2003, PASJ, 2003, 337
863:
864: \bibitem[Lord(1992)]{Lor92}
865: Lord, S.D. 1992, NASA Technical Memor. 103957
866:
867: \bibitem[Mamajek(2005)]{Mam05}
868: Mamajek, E.E. 2005, ApJ, 634, 1385
869:
870: %\bibitem[Matsuyama et al.(2003)]{Mat03}
871: %Matsuyama, I., Johnstone, D., \& Hartmann, L. 2003, ApJ, 582, 893
872:
873: \bibitem[Mazzotta et al.(1998)]{Maz98}
874: Mazzotta, P., Mazzitelli, G., Colafrancesco, S., \& Vittorio, N. 1998, A\&AS, 133, 403
875:
876: \bibitem[Miller \& Stone(2000)]{Mil00}
877: Miller, K.A., \& Stone, J.M. 2000, ApJ, 534, 398
878:
879: \bibitem[Muzerolle et al.(1998)]{Muz98}
880: Muzerolle, J., Hartmann, L., \& Calvet, N. 1998, AJ, 116, 455
881:
882: \bibitem[Muzerolle et al.(2000)]{Muz00}
883: Muzerolle, J., Calvet, N., Briceno, C., Hartmann, L., \& Hillenbrand, L. 2000, 535, L47
884:
885: \bibitem[Muzerolle et al.(2003)]{Muz03}
886: Muzerolle, J., Calvet, N., Hartmann, L., \& D'Alessio, P. 2003, ApJ, 597, L149
887:
888: %\bibitem[Najita et al.(2003)]{Naj03}
889: %Najita, J., Carr, J.S., \& Mathieu, R.D. 2003, ApJ, 589, 931
890:
891: \bibitem[Najita et al.(2007)]{Naj07}
892: Najita, J., Carr, J.S., Glassgold, A.E., \& Valenti, J.A. 2007, PPV
893:
894:
895: %\bibitem[O'Dell \& Wen(1994)]{Odell94}
896: %O'Dell, C.R., \& Wen, Z. 1994, ApJ, 436, 194
897:
898: %\bibitem[Pascucci et al.(2006)]{Pas06}
899: %Pascucci, I. et al. 2006, ApJ, 651, 1177
900:
901: \bibitem[Pascucci et al.(2007)]{Pas07}
902: Pascucci, I., et al. 2007, ApJ, accepted. astro-ph/0703616
903:
904: \bibitem[Qi et al.(2004)]{Qi04}
905: Qi, C., et al. 2004, ApJ, 616, L11
906:
907:
908: %\bibitem[Ratzka et al.(2007)]{Rat07}
909: %Ratzka, et al. 2007, submitted
910:
911: \bibitem[Rettig et al.(2004)]{Ret04}
912: Rettig, T.W., Haywood, J., Simon, T., Brittain, S.D., \& Gibb, E. 2004, ApJ, 616, L163
913:
914:
915: %\bibitem[Richter et al.(2002)]{Ric02}
916: %Richter, M.J., Jaffe, D.T., Blake, G.A., \& Lacy, J.H. 2002, ApJ, 572, L161
917:
918: %\bibitem[Richter et al.(2005)]{Ric05}
919: %Richter, M.J., Lacy, J.H., Greathouse, T.K., Jaffe, D.T., \& Blake, G.A. 2005, proc. of {\it High Resolution Infrared Spectroscopy in Astronomy}, eds. R. Siebenmorgen \& A.F.M. Moorwood.
920:
921: %\bibitem[Roberge et al.(2005)]{Rob05}
922: %Roberge, A., Weinberger, A.J., Redfield, S., \& Feldman, P.D. 2005, \apj, 626, L105
923:
924: \bibitem[Roberge et al.(2005)]{Rob05}
925: Roberge, A., Weinberger, A.J., \& Malumuth, E.M. 2005, \apj, 622, 1171
926:
927:
928: %\bibitem[Sako et al.(2005)]{Sak05}
929: %Sako, S., Yamashita, T., Kataza, H., Miyata, T., Okamoto, Y.K., Honda, M., Fujiypshi, T., \& Onaka, T. 2005, ApJ, 620, 347
930:
931: \bibitem[Salyk et al.(2007)]{Sal07}
932: Salyk, C., Blake, G.A., Boogert, A.C.A., \& Brown, J.M. 2007, ApJ, 655, L105
933:
934: %\bibitem[Sheret et al.(2003)Sheret, Ramsay Howat, \& Dent]{She03}
935: %Sheret, I., Ramsay Howat, S.K., \& Dent, W.R.F. 2003, accepted by \mnras
936:
937: %\bibitem[Smith et al.(2003)]{Smi03}
938: %Smith, N., Bally, J., \& Morse, J.A. 2003, ApJ, 587, L105
939:
940: %\bibitem[Strom et al.(1989)]{Str89}
941: %Strom, K.M., Strom, S.E., Edwards, S., Cabrit, S., \& Skrutskie, M.. 1989, AJ, 97, 145
942:
943: \bibitem[Stelzer \& Schmitt(2004)]{Stel04}
944: Stelzer, B., Schmitt, J.H.H.M. 2004, A\&A, 418, 687
945:
946: \bibitem[St\"orzer \& Hollenbach(2000)]{Sto00}
947: St\"orzer, H., \& Hollenbach, D. 2000, ApJ, 539, 751
948:
949: \bibitem[Uchida et al.(2004)]{Uch04}
950: Uchida, K.I., et al. 2004, ApJS, 154, 439
951: %TW Hya 12.8 micron
952:
953: \bibitem[Valenti et al.(1993)Valenti, Basri, \& Johns]{Val93}
954: Valenti, J.A., Basri, G., \& Johns, C.M. 1993, \apj, 106, 2024
955:
956: \bibitem[Webb et al.(1999)]{Web99}
957: Webb, R.A., Zuckerman, B., Platais, I., Patience, J., White, R.J., Schwartz, M.J., \& McCarthy, C. 1999, ApJ, 512, L63
958:
959: \bibitem[Weintraub et al.(2000)Weintraub, Kastner, \& Bary]{Wei00}
960: Weintraub, D. A., Kastner, J. H. \& Bary, J. S. 2000, \apj, 541, 767
961:
962:
963: %\bibitem[Wolk \& Walter(1996)]{Wol96}
964: %Wolk, S.J., \& Water, F.M. 1996, AJ, 111, 2066
965:
966: \bibitem[Yamada et al.(1985)]{Yam85}
967: Yamada, C., Kanamori, H., \& Hirota, E. 1985, JCP, 83, 552
968:
969: \bibitem[Yang et al.(2007)]{Yan07}
970: Yang, H., Johns-Krull, C.M., \& Valenti, J.A. 2007, AJ, 133, 73
971:
972: \end{thebibliography}
973:
974: %\begin{figure}
975: %\epsscale{1.}
976: %\plotone{spat.eps}
977: %\caption{The spatial distribution of [\ion{Ne}{2}] emission compared with 12.81%$\mu$m continuum emission from TW Hya and $\alpha$ CMa. The seeing $\sim 0\far%cs75$...}
978: %\end{figure}
979: %12.813794
980:
981: \clearpage
982:
983: \begin{table}
984: \label{tab:obsx.tab}
985: %\caption{Properties of Sources}
986: %\rotatebox{90}
987: {\scriptsize
988: \begin{tabular}{ccccccccccccc}
989: \multicolumn{13}{l}{Table 1: Source Properties}\\
990: \hline
991: Star & SpT & d & $v_r$ & $L_X$ &$M_*$ &$R_*$ & $\dot{M}$ & $F_{12.8}^a$ & $S/N^b$ & EW ([\ion{Ne}{2}]) & $L_{Ne II}$ & Refs\\
992: & & pc & \kms\ & $10^{-4} L_\odot$ & $M_\odot$ & $R_\odot$ & $10^{-9} M_\odot$ yr$^{-1}$ & & &\AA & $10^{-6} L_\odot$ & \\
993: \hline
994: TW Hya & K7 & 51 & 12.2 & 1.4$^c$ & 0.7 & 1.0 & $1.8,0.5^d$ & $0.95$ & 1.3 &$62\pm11$ & $4.8$ &1,2,3,4\\
995: BP Tau & K7 & 140 & 15.8 & 1.8$^e$ & 0.5 & 2.0 & $24$ & $0.7$ & 1.5 &$<10$ & $<4$ &5,6,7,8 \\
996: DP Tau & M0 & 140 & 17$^f$ & 0.8$^e$ & 0.5 & 1.4 & $9.0$ & $1.0$ & 1.8 &$<11$ & $<7$ &5,6,7,8\\
997: %0.68, M0.5, L=-0.523, log T=3.55
998: \hline
999: \multicolumn{13}{l}{$^a$Continuum flux level ($10^{-15}$ erg cm$^{-2}$ s$^{-1}$ \AA$^{-1}$) at 12.8$\mu$m from {\it Spitzer} IRS spectra}\\
1000: \multicolumn{13}{l}{$^b$S/N per pixel in the continuum in our MICHELLE observations}\\
1001: \multicolumn{13}{l}{$^c$0.45--2.25 eV luminosity from high-resolution {\it XMM} spectra \citep{Stel04}.}\\
1002: \multicolumn{13}{l}{$^d$1.8 from \citet{Ale02} and \citet{Her04}, adjusted for 51 pc distance. 0.5 from \citet{Muz00}}\\
1003: \multicolumn{13}{l}{$^e$0.3--10 eV from models based on low-resolution {\it XMM} spectra \citep{Brig07}.}\\
1004: \multicolumn{13}{l}{$^f$Average radial velocities of GO Tau and V955 Tau from \citet{Her95}.}\\
1005: \multicolumn{3}{l}{$1$\citet{Mam05}} &
1006: %\multicolumn{9}{l}{$2$\citet{Her04}}\\
1007: \multicolumn{3}{l}{$2$\citet{Uch04}} &
1008: \multicolumn{3}{l}{$3$\citet{Wei00}} &
1009: \multicolumn{3}{l}{$4$\citet{Web99}}\\
1010: \multicolumn{3}{l}{$5$\citet{Val93}} &
1011: \multicolumn{3}{l}{$6$\citet{Her95}} &
1012: \multicolumn{3}{l}{$7$\citet{Brig07}}&
1013: \multicolumn{3}{l}{$8$\citet{Har98}}\\
1014: \end{tabular}}
1015: \end{table}
1016:
1017: %V955: 15.7
1018: %GO Tau: 21.3
1019: %AA Tau: 16.1
1020: %Dn Tau: 16.1
1021:
1022:
1023: %0.07956 0.084, 0.094, 0.0703
1024: %0.001045 (24.5\pm3), 0.000986
1025: %12.813806 0\pm2, 12.813703
1026:
1027: \clearpage
1028:
1029: \begin{table}
1030: %\label{tab:obsx.tab}
1031: %\caption{Selected Emission Lines from TW Hya, in order of FWHM$^{a,b}$}
1032: {\scriptsize \begin{tabular}{cccccccccc}
1033: \multicolumn{10}{l}{Table 2: Selected Emission Lines from TW Hya, in order of FWHM$^{a,b}$}\\
1034: \hline
1035: & Line & $\lambda_{obs}$ & $E^\prime$ & R & $v$ & FWHM$^a$ & EW
1036: & Flux$^c$ & Formation$^d$\\
1037: & ID & \AA & eV &$\lambda/\Delta\lambda$ & \kms & \kms & \AA\ & &\\
1038: \hline
1039: & H$_2^e$ & 21218 & 0.60 & 49000 & 0 & $<6$ & 0.02 & 1 & D\\
1040: & \multicolumn{2}{c}{CO (fundamental)$^f$} & & 24000 & 0 & 8.3 & - & - & D\\
1041: & \ion{Si}{1} & 2882.551 & 5.1 & 30000 & 1 & 11 & 0.14 & 0.96 & P\\
1042: & [\ion{O}{1}]$^b$ & (5577.571) &2.0 & 36000 & (0) & (12) & (0.11) & $<1.4$ & --\\
1043: & [ \ion{O}{1}] & 6300.549 &2.0 & 34000 & -1 & 12 & 0.67 & 8.7 & D?\\
1044: & [\ion{O}{1}] & 6364.050 &2.0& 26000 & 0 & 13 & 0.23 & 3.5 & D?\\
1045: & [\ion{S}{2}]$^b$ & (6730.81) & 1.8 & 20500.& (0) & (12) & $<0.06$ & $<0.9$ & --\\
1046: %& [\ion{S}{2}]$^b$ & (6724.52) & 22.5 & 19000 & (0) & (12) & $<0.06$ & $<0.9$ & --\\
1047: %& [\ion{S}{2}]$^b$ & (6716.44) & 1.8 & 18000 & (0) & (12) & $<0.10$ & $<1.5$ & --\\
1048: & [\ion{N}{2}]$^b$ & (6583.46) & 1.9 & 21000 & (0) & (12) & $<0.07$ & $<1.0$ & --\\
1049: %& [\ion{Fe}{2}]$^b$ & (7452.54) & 2.0 & 18000 & (0) & (12) & $<0.05$ & $<1.0$ & --\\
1050: & [\ion{Fe}{2}]$^b$ & (8616.952) & 1.7 & 27500 & (0) & (12) & $<0.07$ & $<1.2$ & --\\
1051: & \ion{Na}{1}$^g$ & (5889.951) & 2.1 & 22000 & (0) & (12) &
1052: $<0.45$ & $<4.5$ & --\\
1053: & \ion{Na}{1}$^g$ & (5895.924) & 2.1 & 22000 & (0) & (12) &
1054: $<0.45$ & $<4.5$& --\\
1055: & H$_2^h$ & FUV$^h$ & 1--2$^i$ & 25000 & 0 & 14 & - & - & D\\
1056: & \ion{Fe}{2} & 2371.319 & 5.6 & 30000 & -1 & 15 & 0.17 & 0.76 & ?\\
1057: & \ion{Fe}{2} & 2665.569 & 8.0 & 30000 & 0 & 15 & 0.09 & 0.55 & ?\\
1058: & \ion{Ca}{2} & 8542.362 & 3.2 & 16000 & -2 & 15 & 1.2 & 21 & P+A?\\
1059: & \ion{Ca}{2} & 8662.464 & 3.1& 43000 & -1 & 16 & 0.84 & 17 & P+A?\\
1060: & \ion{Mg}{2} & 2791.722 & 8.9 & 30000 & 1 & 16 & 0.27 & 0.19 & P?\\
1061: & {\bf [\ion{Ne}{2}]} & {\bf 128140} & {\bf 0.10} & {\bf 31000} & {\bf -2} & {\bf 22.5} & {\bf 62} & {\bf 5.9} & D\\
1062: & \ion{Fe}{2} (1)$^j$ & 2626.565 & 4.8 & 30000 & 1 & 25 & 1.3 & 6.3 & P/H?\\
1063: & (2)$^j$ & 2626.557 & 4.8 & 30000 & 0 & 130 &2.1 & 10.3 & H?\\
1064: & \ion{Cl}{1} & 1351.715 & 9.3 & 30000 & 0 & 41 & 0.18 & 0.27 & H?\\
1065: & \ion{C}{1} & 1656.346 & 7.5 & 30000 & 1 & 41 & 1.5 &4.6 & H?\\
1066: &\ion{C}{2}] & 2326.231 & 5.3 & 30000 & 1 & 42 &3.8 & 19 & H?\\
1067: & \ion{He}{1}$^k$ & 5876.060 & 23.1 & 19000 & 10 & 50 & 0.81 & 12 & A\\
1068: & \ion{S}{1} & 1296.231 & 9.6 & 30000 & 0 & 52 & 0.95 &1.3 & A?\\
1069: & \ion{N}{1}] & 1411.966 & 12.4 & 30000 & -7 & 83 & 1.9 & 2.8 & A?\\
1070: & \ion{He}{2}$^l$ & 1640.601 & 48 & 30000 & 20 & 88 & 310 & 94 & A\\
1071: & \ion{O}{3}] & 1666.093 & 7.5 & 30000 & -23 & 113 & 0.8 &1.6 & W?\\
1072: & \ion{C}{3} & 2297.842 & 18.1 & 30000 & 22 & 127 & 0.8 & 3.2 & A\\
1073: & H$\gamma$ & 4341.214 & 13.1 & 32000 & 39 & 195 & 8.1 & 81 & A\\
1074: & H$\beta^m$ & 4861.713 & 12.8 & 32000 & 11 & 199 & 34 & 44 & A\\
1075: & H$\delta$ & 4102.464 & 13.2 & 32000 & 41 & 201 & 2.9 & 23 & A\\
1076: & \ion{C}{4}$^l$ & 1548.757 & 8.0 & 30000 & 110 & 300 & 93 & 186 & A\\
1077: & \ion{N}{5}$^l$ & 1239.068 & 10.0 & 30000 & 60 & 340 & 15 & 30 & A\\
1078: \hline
1079: \multicolumn{10}{l}{$^a$Intrinsic FWHM after deconvolution from the instrumental resolution.}\\
1080: \multicolumn{10}{l}{$^b$Assumed properties for undetected lines in parentheses.}\\
1081: \multicolumn{10}{l}{$^c$$10^{-14}$ erg cm$^{-2}$ s$^{-1}$}\\
1082: \multicolumn{10}{l}{$^d$W=Wind, P=photosphere, H=Heated Photosphere, A=Accretion funnel, D=Disk}\\
1083: \multicolumn{10}{l}{$^e$From Weintraub et al. (2000).}\\
1084: \multicolumn{10}{l}{$^f$Averaged CO FWHM from Salyk et al. (2007).}\\
1085: \multicolumn{10}{l}{$^g$Not detected here. \citet{Ale02} found narrow
1086: \ion{Na}{1} D lines at the stellocentric velocity.}\\
1087: \multicolumn{10}{l}{$^h$Coadded FUV H$_2$ profiles by coadding many of the strongest lines from Herczeg et al. (2002).}\\
1088: \multicolumn{10}{l}{$^i$Excited from many levels between 1--2 eV (Herczeg et al. 2002).}\\
1089: \multicolumn{10}{l}{$^j$NUV \ion{Fe}{2} lines from TW Hya have narrow
1090: and broad components.}\\
1091: \multicolumn{10}{l}{$^j$Properties are for bright, narrower
1092: component. Excess weak redshifted emission is also present.}\\
1093: \multicolumn{10}{l}{$^l$Line center and width calculate from a Gaussian fit to a non-Gaussian profile, see Johns-Krull \& Herczeg (2007).}\\
1094: \multicolumn{10}{l}{$^m$Corrected for optically-thin wind absorption}\\
1095: %\multicolumn{6}{l}{$^b$CO FWHM from \citet{Joh01}.}\\
1096: \end{tabular}}
1097: \end{table}
1098:
1099: \clearpage
1100:
1101: \begin{figure}
1102: \plotone{f1.eps}
1103: \caption{The MICHELLE spectra of TW Hya, BP Tau, and DP Tau, shifted to the appropriate stellar velocity. The [\ion{Ne}{2}] line is strong from TW Hya but not detected from BP Tau or DP Tau. The mid-IR continuum is detected from all three sources. The bottom panel shows the spectrum of the telluric standard $\alpha$ CMa shifted to the observed velocity of TW Hya.}
1104: \end{figure}
1105:
1106: \clearpage
1107:
1108: \begin{figure}
1109: \plotone{f2.eps}
1110: \caption{A comparison of the [\ion{Ne}{2}] line profile with many
1111: optical and UV lines from TW Hya. The [\ion{Ne}{2}] line is much
1112: more narrow than lines associated with the accretion funnel
1113: (including the \ion{He}{1} $\lambda5876$ line shown here), but is broader than
1114: lines formed in the disk (H$_2$ and possibly [\ion{O}{1}] emission
1115: shown here) and in the photosphere (\ion{Si}{1} emission shown here). Telluric [\ion{O}{1}] emission is identified by the Earth symbol.}
1116: \end{figure}
1117:
1118: \end{document}
1119:
1120:
1121: