1: \documentclass[english,a4paper,12pt]{article}
2: \usepackage{graphicx}
3: \textwidth=15cm \textheight=22cm
4: \newcommand{\rytau}{RY~Tau\ }
5: \newcommand{\cii}{[C~{\sc II}]\ }
6: \newcommand{\ciii}{C~{\sc III}]\ }
7: \newcommand{\alii}{Al~{\sc II}]\ }
8: \newcommand{\siii}{Si~{\sc III}]\ }
9: \newcommand{\civ}{C~{\sc IV}\ }
10: \newcommand{\feii}{Fe~{\sc II}\ }
11: \newcommand{\oxigen}{[O~{\sc II}]\ }
12: \newcommand{\mgii}{Mg~{\sc II}\ }
13: \newcommand{\mgi}{Mg~{\sc I}\ }
14:
15:
16: \begin{document}
17:
18: \begin{titlepage}
19:
20: \begin{center}
21: \textbf{\large Evidence for stellar driven outflows\\
22: from the Classical T Tauri Star RY Tau }
23: \end{center}
24: \vskip 1cm
25: \begin{center}
26: \textbf{Ana I. G\'omez de Castro$^{1}$ and Eva Verdugo$^{2}$}
27: \end{center}
28: \begin{tabular}{ll}
29: $^{1}$ & Instituto de Astronom\'{\i}a y Geodesia (CSIC-UCM)\\
30: & Facultad de CC. Matem\'aticas, Univ. Complutense \\
31: & Madrid, E28040 Spain\\
32: $^{2}$ & European Space Astronomy Centre (ESAC) \\
33: & Research \& Scientific Support Dept.- ESA \\
34: & P.O.Box 50727, Madrid, E28080 Spain \\ \\
35: \end{tabular}
36: \begin{center}
37: {\bf ABSTRACT} \vskip 0.5cm
38: \end{center}
39: RY~Tau is a rapidly rotating Classical T Tauri star observed close
40: to edge-on. The combination of new HST/STIS observations obtained
41: in 2001 with HST/GHRS Archive data from 1993 has allowed us to
42: get, for the first time, information on the thermal structure and
43: the velocity law of the wind. The repeated observations of the
44: \siii\ and \ciii\ lines show a lack of changes with time in the
45: blue side of the profile (dominated by the wind contribution).
46: Very high temperature plasma ($\log T_e = 4.8$) is detected at
47: densities of $9.5 \leq \log n_e {\rm (cm^{-3})}\leq 10.2$
48: associated with the wind. The emitting volumes are $\sim
49: (0.35R_{\odot})^3$ suggesting a stellar origin. The wind
50: kinematics derived from the profiles (\siii, \ciii and \oxigen)
51: does not satisfy the theoretical predictions of MHD centrifugally
52: driven disk winds. The profiles' asymmetry, large velocity
53: dispersions and small variability as well as the small emitting
54: volumes are best explained if the wind is produced by the
55: contribution of several outflows from atmospheric open field
56: structures as those observed in the Sun.
57:
58:
59: \noindent
60: {\bf Keywords:} stars: pre-main sequence --- stars:
61: winds,outflows
62: --- stars : individual(RY Tau)
63:
64:
65:
66: \end{titlepage}
67:
68: \section{Introduction}
69:
70: Winds from T Tauri Stars (TTSs) were assumed to be cold,
71: in spite of the observational evidence of our own Sun, because
72: they were first detected as bipolar molecular outflows (Loren et al 1981)
73: and optical jets with low ionization fractions.
74: The detection of \ciii and \siii semiforbidden line emission
75: from \rytau and RU~Lup wind (G\'omez de Castro
76: \& Verdugo, 2001; hereafter GdCV) provided the first evidence
77: that electron temperatures as
78: high as $\log Te(K) \simeq 4.5$ could be achieved at the base of
79: the TTSs outflows; this high
80: temperature was derived from ultraviolet (UV) observations and
81: could not be inferred from the classical optical/infrared/radio tracers.
82: Recently, Dupree et al (2005) have claimed that O~{\sc VI}
83: absorption has been detected in TW~Hya and T~Tau.
84: These new data represent a challenge to the current theories
85: for outflow generation in TTSs.
86: In this letter, we provide observational
87: evidence of hot winds being driven from the stellar atmosphere
88: of RY~Tau, a rapidly rotating TTS ($v\sin i = 51.6$ km/s, Hartmann
89: \& Stauffer, 1989). RY Tau has been classified as an UX~Ori
90: star: a pre-main sequence star which shows
91: aperiodic eclipse-like minima caused by
92: variable obscuration produced by circumstellar dust associated with the disk
93: (Grinin, 1992). This suggests that the disk is seen close to edge-on.
94: Also, the high projected rotation velocity of
95: the stellar atmosphere suggests that the star inclination is
96: close to 90$^o$.
97:
98: \section{Data}
99:
100: Some few spectral tracers have been selected for the subsequent
101: study, namely, the \civ, \ciii, \cii, \siii and \oxigen
102: lines observed in the UV spectrum of RY~Tau
103: (see Table~1 for more details on the observations and the
104: line fluxes, and
105: Fig.~1 for the profiles). All data are public and can
106: be withdrawn from the HST Archive (they were obtained either
107: by us or by other HST users). Data have been processed with the
108: Routine Science Data Pipeline (RSDP). The major source of
109: inaccuracy in the calibration is the centering
110: of the target in the aperture that can account for as much as
111: 20~km~s$^{-1}$ in the wavelength zeropoint accuracy of the GHRS data
112: (see GdCV) and $<$0.1-0.2 pixels in the STIS data ({\it e.g.}
113: velocity zeropoint accuracies better than
114: 0.6~km~s$^{-1}$ and 2.8~km~s$^{-1}$ for the E230M and G230M
115: spectra, respectively). WAVECALs were obtained during the observations
116: and the peakup procedure (ACQ/PEAK) was used for the acquisition
117: to center the source accurately, to about a 5\% of the slit width
118: (see www.stsci.edu/hst/stis/ for more details).
119: This accuracy was cross-checked for the
120: E230M spectra using as reference the narrow absorption features observed in
121: the Fe~{\sc ii}, Mg~{\sc ii} and Mg~{\sc i} resonance lines
122: produced by the interstellar medium.
123: The STIS spectra
124: were co-added after this check to improve the SNR of the data since
125: no variations are detected. The profiles have been
126: classified in Fig~1 by observing date. The left panel
127: displays the very asymmetric \ciii and \siii profiles
128: that led GdCV to conclude that these lines are
129: produced at the base of the wind. The \civ line observed $\sim 2$~hours
130: before is also displayed; the profile is
131: very similar to the \ciii and \siii lines.
132: Bluewards shifted
133: emission is also detected in the \oxigen lines (see central panel)
134: observed in Feb. 2001. The \cii , \mgii
135: and \mgi lines are obtained in the same spectra. A minimum is detected
136: at a similar velocity in the \mgii lines;
137: these lines have large optical depths in dense plasmas with $T\simeq 10^4$K.
138: A narrow absorption component, probably of interstellar origin,
139: is detected at the stellar rest velocity.
140: There also seems to be an absorption component in the
141: \mgi line at the same velocity. The last observations of the
142: \ciii and \siii lines are displayed in the right panel of Fig.1.
143: The profiles
144: changed significantly from 1993 to 2001, although the bluewards
145: shifted part of the profile remains stable.
146:
147:
148: \section{Interpretation and Analysis}
149:
150: The detection of emission from high velocity gas
151: in \civ , \siii , \ciii , \oxigen , [O~{\sc I}] (Hamann 1994) indicates
152: that {\it there is a broad range of temperatures in the RY Tau
153: wind at similar velocities.} The wind properties seem to
154: be rather stable; {\it the emission peaks at similar
155: bluewards shifted velocities in observations obtained over 10 years},
156: compare Fig.~1 with Hamann (1994) that measures a blueshifted
157: [O~{\sc I}] emission peaking at -79 km/s.
158:
159: Line ratios can be used to constrain the wind temperature
160: and density taking into account that the regions
161: traced by each spectral tracer only overlap partially.
162: Also occultation effects may be relevant (see Sect.4).
163: Line emissivities have been calculated for
164: a collisional plasma over a broad range of
165: densities ($n_e = 10^7 - 10^{13}$~cm$^{-3}$) and
166: temperatures ($\log T_e (K) = 4.0 -5.0$) using the
167: atomic parameters in the CHIANTI data base (Dere et al 1997)
168: and assuming solar abundances and ionization equilibrium.
169:
170: \noindent
171: {\it The high temperature region:} from the
172: GHRS observations we derive: \civ/\siii~$=0.6 \pm 0.7$
173: and \siii/\ciii ~=$1.1 \pm 0.6$. Then from
174: Fig.~2 we derive: $\log (T_e) \simeq 4.8$ and
175: $ 9.5 < \log (n_e) <10.2$.
176: {\it Neither the line ratios nor the UV line fluxes
177: vary by more than a factor of 2 in the fifteen observations
178: available from
179: 1979 to 2001 (GdCV)}. Notice that the high resolution data
180: indicate that the variable contribution from
181: 0-velocity and redshifted material to the \siii and \ciii line
182: fluxes does not significantly change their line ratios
183: \footnote{
184: $\left[\frac {Si III]}{C III]}\right]_{GHRS} = 1.1 \pm 0.6$, $\left[\frac {Si III]}{C III]}\right]_{STIS} = 1.7 \pm 0.5$,
185: $\left[\frac {Si III]_{STIS}-Si III]_{GHRS}}
186: {C III]_{STIS}-C III]_{GHRS}}\right] = 4 \pm 3$
187: Thus the {\it average} physical conditions of this
188: hot plasma in the wind are similar to those
189: derived from the redshifted material.}.
190:
191: \noindent
192: {\it The low temperature region:}
193: The \cii emission feature consists of five spectral lines
194: (see Fig.~1 and Table~1). The individual features are not
195: resolved in RY~Tau; a broad blend is observed instead.
196: Some ratios among the individual lines in the feature
197: are {\it density sensitive} in the range
198: $n_e = 10^8-10^{11}$~cm$^{-3}$.
199: An accurate determination of the electron density in the
200: \cii emission region requires us to derive simultaneously the
201: ratios among the multiplet lines and the underlying
202: \cii profile. A least squares scheme has been used to solve
203: this inverse problem. The optimal fit gives
204: a high lower limit of $n_e > 10 ^{10}$~cm$^{-3}$
205: for the \cii emitting region, similar
206: to that determined from the \ciii line.
207: Note that the \cii profile is too
208: noisy to narrow this range (the contribution from
209: the Fe~II(uv3) line at vacuum wavelength 2328.11\AA\ is
210: negligible as shown by the weakness of nearby features
211: corresponding to the same multiplet).
212: The fit also indicates that the underlying \cii profile
213: is similar to the \ciii (STIS) profile; {\it e.g.}
214: a blueshifted wind profile plus an excess from
215: 0-velocity to the redwards shifted edge.
216: From that fit, we also obtain
217: $
218: \frac {F(C~{II]}_{2326.117})}{F(C~{ III]}_{1908.7})}
219: = 1.3 \pm 0.1
220: $.
221: This ratio is {\it temperature sensitive} and
222: from that we derive that $4.4 < \log (T_e) < 4.5$.
223:
224:
225:
226:
227: \section{Implications for wind models}
228:
229: TTSs outflows are characterized by their collimation
230: at large scales, {\it i.e.}, it is expected that whatever
231: the ejection mechanism is, the wind terminal
232: speed corresponds to a longitudinal motion along
233: the collimation axis. Current models make
234: use of magnetic fields to guarantee the large scale collimation.
235: In this section we use the models of G\'omez de Castro \&
236: Ferro-Font\'an 2005 (hereafter GdCFF) to compare to data.
237: Disk wind models by other authors have a similar kinematics
238: though the scales are different (see Uzdensky 2004 for a review).
239:
240: In GdCFF models, the wind is launched from the disk corona
241: by a combination of thermal pressure and centrifugal force.
242: This results in a radial expansion, away from
243: the rotation axis, at the base and produces an increasing
244: broadening of the profiles in this acceleration region.
245: After this, the restoring toroidal component of the
246: field narrows down the beam and the velocity field is
247: progressively dominated by the motion along the rotation
248: axis until it reaches the asymptotic regime, {\it i.e.}
249: flow along the axis at the terminal velocity and basically no
250: radial expansion. The evolution of the radial expansion,
251: $V_r$, with the height above the disk is plotted in the top
252: panel of Fig~3 for the warm disk winds model.
253: Line emission is produced at different locations in the flow
254: depending on densities and temperatures thus, the line profiles
255: depend on the region of the wind sampled
256: by each particular spectral indicator (as well as on
257: the inclination).
258:
259:
260: In order to compare these theoretical
261: predictions with the observations, we have defined two observables
262: that are measured only from the bluewards shifted emission of the relevant
263: spectral tracers to avoid the pollution by the contribution
264: of the accretion flow. Notice that line
265: emission from high temperature infalling plasma is produced at scales
266: $<<R_*$ pumped by the X-ray radiation released at the accretion
267: shocks. Thus occultation by the stellar disk causes the
268: contribution to the line flux from accretion to be redwards
269: shifted (see Beristain et al (2001) for more details).
270: These observables are the velocity
271: dispersion, $dV$, and the line centroid,$V_c$. The line
272: profiles may be understood as
273: velocity-flux plots (or $V-f$ plots). The centroid of the
274: bluewards shifted region of the profile is defined as,
275: $
276: V_c = \frac {\Sigma _{i=1}^{i=N}f_i (V_0+i\Delta V)}{\Sigma _{i=1}^{i=N}f_i}
277: $
278: and the dispersion, $dV$, as,
279: $
280: dV = \sqrt{\frac{\Sigma _{i=1}^{i=N}f_i ((V_0+i\Delta V)^2 - V_c^2)}
281: {\Sigma _{i=1}^{i=N}f_i}}
282: $
283: for a uniformly sampled profile with step $\Delta V$ and $N$ points.
284: $V_0$ represents the velocity at the blue edge of the profile
285: and $N = ABS(V_0/\Delta V)+1$. These observables have been
286: calculated for three spectral tracers that sample very different
287: plasma densities: \siii, \ciii and \oxigen.
288: The
289: critical electron densities for the collisional quenching of
290: the \ciii and \oxigen lines are
291: $1.5 \times 10^{10}$~cm$^{-3}$ and $ 0.5
292: \times 10^7$~cm$^{-3}$ respectively. For comparison,
293: the critical density of the \siii line is
294: $2.0 \times 10^{12}$~cm$^{-3}$ for a fiducial temperature
295: of $2.5\times 10^4$K.
296: $V_c$ and $dV$ are plotted in the bottom panel of Fig 3 for the
297: GdCFF theoretical predictions and various inclinations;
298: notice that the \oxigen line is predicted to be formed in the
299: rarified expanding region while the \ciii and \siii lines are
300: predicted to be formed in the hot disk corona where the
301: wind is accelerated.
302:
303:
304: $V_c$ and $dV$ have been computed for the \oxigen , \ciii and \siii
305: profiles observed in RY~Tau. To take into account the effect of the noise
306: on these observables, we have computed these values both on the
307: observed profiles and on filtered profiles obtained by applying
308: a white noise filter to the data. As shown in Fig.~3, the major effect
309: of the noise is to increase the velocity dispersion and shift
310: the centroid some few km~s$^{-1}$ to the blue; however the
311: overall trend remains. Two important results are
312: derived from the figure:
313: \begin{enumerate}
314: \item The line centroid shifts to higher velocity
315: as the density traced by the spectral line
316: becomes lower as expected (see critical densities),
317: \item The velocity dispersion is high,
318: $> 40$km~s$^{-1}$, and similar for all the tracers.
319: In addition, the profiles are very asymmetric (see Fig.~1).
320: Rather constant velocity dispersions and asymmetric
321: profiles are expected for very small inclinations,
322: while for large inclinations the profiles should be
323: symmetric and the dispersion increase steadily with
324: the line centroid. However, all evidence points to
325: RY~Tau being observed cloes to edge-on (see Sect.~1).
326: \end{enumerate}
327: Thus RY~Tau observations cannot be
328: reconciled with models
329: based on the centrifugal launching of the TTSs jets
330: from the accretion disk.
331:
332:
333: Another important piece of evidence comes from the
334: emitting volumes. The volume of plasma producing the
335: observed \siii and \ciii flux can be derived from the
336: line emissivity\footnote{The
337: emissivities are: j(\civ ) = $1.44 \times
338: 10^{-2}$erg~cm$^{-3}$~s$^{-1}$ for $\log (T_e) =4.8$ and $\log (n_e) =11$;
339: j(\siii ) = $1.49 \times
340: 10^{-2}$erg~cm$^{-3}$~s$^{-1}$ for $\log (T_e) =4.8$ and $\log (n_e) =11$;
341: and j(\ciii ) = $0.97 \times
342: 10^{-3}$erg~cm$^{-3}$~s$^{-1}$ for $\log (T_e) =4.8$ and $\log (n_e) =11$.}.
343: The emitting volume producing the \siii emission is
344: $2.07\times 10^{31}$cm$^3$ ({\it i.e.} (0.24R$_{\odot}$)$^3$)
345: and similar to the derived for the \ciii and \civ lines. Thus,
346: line emission is produced in either a small structure
347: or in dense clumps in the outflow. Notice that even
348: adopting a filling factor of 0.1\%--1\% for the clumps, as that
349: derived from observations of optical jets (Liseau et al 1996), the volume
350: of the region producing the observed \ciii and \siii profiles
351: just increases to about $ 1.1-2.4$~R$_{\odot}$.
352: Thus, it is highly likely that mass ejection occurs from
353: the stellar surface.
354:
355:
356: As RY~Tau is a fast rotator the
357: centrifugal lever arm from the stellar surface is an efficient
358: source of thrust (see {\it e.g.} Sakurai 1987).
359: Thus, the wind could be
360: ejected from the stellar surface from
361: open field structures like those observed in the
362: Solar corona. The interaction between the wind and
363: the circumstellar material (and a probable slow wind)
364: could produce shocked knots like the Corotating
365: Interaction Regions (CIRs) observed in the interplanetary
366: medium and in young stars like AB~Dor (G\'omez de
367: Castro 2002) that could contribute to the UX~Ori
368: phenomenon observed in RY~Tau.
369:
370: This suggestive possibility would indicate that
371: the observed profiles represent just an average
372: of many small scale outflows and although these
373: would not be stationary on a one-by-one basis, the
374: average of all of them could produce a rather
375: stable profile (provided that the spectral resolution
376: is not good enough to resolve the individual components).
377: If the magnetic field configuration is not axisymmetric
378: (as observed in cool main sequence stars) this wind would
379: naturally lead to the formation of nonaxisymmetric outflows.
380: Tracking the variability of the
381: \civ and \ciii profiles is crucial for testing
382: this possibility.
383:
384: \noindent {\bf Acknowledgments:} We thank an anonymous referee for
385: helping us to sharpen the definition of the observables in Fig.~3.
386:
387: \begin{thebibliography}{}
388:
389: \bibitem{}
390: Ardila, D. R., Basri, G., Walter, F. M., Valenti, J. A.,
391: Johns-Krull, C. M., 2002, ApJ, 566, 1100
392: \bibitem{}
393: Beristain, G. Edwards, S., Kwan, J., 2001, ApJ, 551, 1037
394: \bibitem{}
395: Dere, K. P., Landi, E., Mason, H. E., Monsignori Fossi, B. C., Young, P. R.,
396: 1997, A\&AS, 125, 149
397: \bibitem{}
398: Dupre\'e, A. K., Brickhouse, N. S., Smith, Graeme H., Strader, Jay, 2005, ApJ, 626, L59
399: \bibitem{}
400: G\'omez de Castro, A.I., Verdugo, E., 2001, ApJ, 548, 976
401: \bibitem{}
402: G\'omez de Castro, A.I., 2002, MNRAS, 332, 409
403: \bibitem{}
404: G\'omez de Castro, A.I., Ferro-Font\'an, C., 2005, MNRAS, 365, 569
405: \bibitem{}
406: Grinin, V.P., 1992, A\&AT, 3, 17
407: \bibitem{}
408: Hamman, F., 1994, ApJS, 93, 485
409: \bibitem{}
410: Hartmann, L., Stauffer, J.R., 1989, AJ, 97, 873
411: \bibitem{}
412: Liseau, R., Huldtgren, M., Fridlund, C.V.M., Cameron, M., A\& A, 306, 255
413: \bibitem{}
414: Sakurai, T., 1985, A\& A, 152, 121
415: \bibitem{}
416: Uzdensky, D., 2004, Ap\&SS, 292, 573
417: \end{thebibliography}
418:
419: \clearpage
420:
421:
422: \begin{figure}
423: \includegraphics[width=12cm]{f1.eps}
424: \caption{ UV lines observed in RY~Tau
425: and used in this work; RSDP processed data (see Table~1 for the
426: binning) are plotted with a thin line and the 3-pixels average
427: profile with a thick line. The rest wavelength of the lines and
428: the velocity of the unresolved jet at $\simeq -80$ km/s (from GdCV
429: and Hamann 1994) are marked with dashed lines. {\it Left panel}:
430: Observations obtained in Dec. 31st, 1993 with the GHRS. The \civ
431: line is blended with a very weak H$_2$ feature (see Ardila et al
432: 2002). {\it Central panel}: Profiles from STIS observations
433: obtained in Feb. 20-21, 2001. The 5 components of the \cii
434: multiplet are marked. {\it Right panel}: \ciii and \siii lines
435: observed in March 27th, 2001. Both lines show an excess of flux in
436: the red wing compared with the 1993 observations; this excess is
437: shaded in the figure.}
438: \end{figure}
439:
440:
441: \clearpage
442: \begin{figure}
443: \includegraphics[width=12cm]{f2.eps}
444: \caption{ \civ/ \siii versus \siii/ \ciii flux
445: ratios (in erg/s/cm$^{-2}$) for collisional plasmas with electron
446: densities between $10^9-10^{13}$cm$^{-3}$, temperatures
447: $\log(T_e)$=4-5 and solar abundances. The value (with the error bars) derived from
448: high resolution observations is marked with thick line;
449: the 13 low resolution IUE observations are plotted with thin lines.
450: Isothermal and isochoric lines are plotted with
451: continuous and dashed lines respectively.
452: }
453: \end{figure}
454:
455:
456: \clearpage
457:
458:
459: \begin{figure}
460: \includegraphics[width=12cm]{f3.eps}
461: \caption{{\it Top:} variation of the wind expansion
462: velocity with respect to the jet axis, $V_r$, with the height above
463: the disk, $z$, from GdCFF model. The location of the \siii, \ciii and
464: \oxigen emission regions is sketched (see GdCFF for the detailed
465: calculations). {\it Bottom:} Comparison between the theoretical
466: predictions from the GdCFF model and RY Tau observations.
467: $V_c$ and $dV$ values measured from
468: the \siii, \ciii and \oxigen profiles observed in RY Tau are marked with
469: big squares, triangles and circles respectively.
470: $V_c$ and
471: $dV$ are plotted from the \siii, \ciii and \oxigen profiles derived
472: from the theory (small symbols) for various inclinations. High $dV$ values
473: are obtained from the observed profile while the slightly
474: smaller $dV$ values are measured over filtered profiles (see text). }
475: \end{figure}
476:
477:
478: \clearpage
479:
480: \setlength{\oddsidemargin}{-2cm}
481: \begin{table}
482: \begin{scriptsize}
483: \caption{HST/STIS observations.}
484: \begin{tabular}{lllllcccl}
485: \hline
486: Inst. & Grating & Res. &Scale$^{(a)}$& Slit & Spec. Range & Obs. Date & Target$^{(b)}$ & Flux $^{(c)}$ \\
487: & & & & & & & Features & \\
488: & & &\AA /elem & & \AA &(dd-mm-yyyy)& &$10^{-14}$ (erg/s/cm$^2$) \\
489: \hline
490: GHRS$^{(d)}$ & G160M & 20000 &0.069& LSA$^{(e)}$& 1532 - 1567 & 31-12-1993& C~{\sc IV} & F(C~{\sc IV})$=8.2\pm 7.1^{(g)}$\\
491: & G200M & 20000 & 0.078 &LSA&1880 - 1920 & 31-12-1993& Si~{\sc III}], C~{\sc III}] & F(Si~{\sc III}])$=13\pm 4$\\
492: & & & & & & & F(C~{\sc III}])$=12\pm 3$ \\
493: STIS$^{(g)}$ & E230M & 30000 &1/60,000&0".2x0".2& 2303 - 3111 & 19-02-2001 & [O~{\sc II}], [C~{\sc II}],Mg~{\sc II}, Mg~{\sc I} & F([O~{\sc II}])$=5.5\pm 1.4 $ \\
494: & E230M & 30000 &1/60,000& 0".2x0".2&2303 - 3111 & 20-02-2001 &[O~{\sc II}], [C~{\sc II}], Mg~{\sc II}, Mg~{\sc I} &
495: F([C~{\sc II}]$_{2326.1}$)$=19\pm 2$ \\
496: & E230M & 30000 &1/60,000& 0".2x0".2&2303 - 3111 & 20-02-2001 &[O~{\sc II}], [C~{\sc II}], Mg~{\sc II}, Mg~{\sc I} &\\
497: & G230M & 10500 &0.09 & 52"x0".2&1839 - 1929 & 27-03-2001 &Si~{\sc III}], C~{\sc III}] & F(Si~{\sc III}])$=26\pm 3$\\
498: & & & & & & & & F(C~{\sc III}])$=15\pm 3$ \\
499: \hline
500: \end{tabular}
501: \end{scriptsize}
502: \begin{scriptsize}
503: \begin{tabular}{ll}
504: $^{(a)}$ & The element is a diode for GHRS spectra and a pixel for STIS spectra.\\
505: $^{(b)}$ & The vacuum wavelengths of the target features are:
506: C~{\sc IV}(1548\AA , 1550\AA), Si~{\sc III}](1892\AA), C~{\sc III}](1908\AA), [O~{\sc II}] (2471\AA), \\
507: & [C~{\sc II}](2324.2\AA,2325.4\AA,2326.1\AA,2327.6\AA,2328.8\AA),
508: Mg~{\sc II} (2796\AA,2803\AA),Mg~{\sc I} (2853\AA) \\
509:
510: $^{(c)}$ &The Interstellar reddening law has been used since the extinction
511: law towards T Tauri systems is not known.\\
512: & Parameters: A$_V$=0.55mag; A$_{CIV}$/A$_V$ = 2.60;
513: A$_{SiIII}$/A$_V$ = 2.66; A$_{CIII}$/A$_V$ = 2.66; A$_{CII}$/A$_V$ = 2.60;
514: A$_{OII}$/A$_V$ = 2.44.\\
515: & The flux of the [C~{\sc II}](2326.1\AA) has been derived from the fitting of the \cii multiplet (see text).\\
516: $^{(d)}$ & Goddard High Resolution Spectrograph removed in February 1997\\
517: $^{(e)}$ & Large Science Aperture (2'')\\
518: $^{(f)}$ & The H$_2$ (3-8 R(3)) line contribution to the C~{\sc IV} flux is
519: negligible (see Ardila et al 2002).\\
520: $^{(g)}$ & Space Telescope Imaging Spectrograph stopped science operations in August 2004\\
521: \end{tabular}
522: \end{scriptsize}
523: \end{table}
524:
525:
526:
527:
528:
529: \end{document}
530: