astro-ph0304531/ms.tex
1: \documentclass[12pt,preprint]{aastex}
2: \shorttitle{Fluorescent H$_2$ Emission Near T Tau}
3: \begin{document}
4: 
5: \title
6: {THE SPATIAL DISTRIBUTION OF FLUORESCENT H$_2$ EMISSION NEAR T TAU}
7: 
8: 
9: \author{Jos\'e Saucedo \altaffilmark{1,2}, Nuria Calvet \altaffilmark{1}, 
10: Lee Hartmann 
11: \altaffilmark{1}, and John Raymond \altaffilmark{1}}
12: 
13: \altaffiltext{1}{Harvard-Smithsonian Center for Astrophysics, 60 Garden 
14: St.,
15: Cambridge, MA 02138, USA, \email{\scriptsize jsaucedo@cfa.harvard.edu, 
16: ncalvet@cfa.harvard.edu, hartmann@cfa.harvard.edu, 
17: raymond@cfa.harvard.edu}}
18: 
19: 
20: \altaffiltext{2}{Instituto de Astronom\'{\i}a,
21: UNAM, Ap. Postal 70-264, Cd. Universitaria, 04510 M\'exico D.F.,
22: M\'exico}
23: 
24: 
25: \begin{abstract}
26: 
27: New subarcsecond FUV observations of T Tau with HST/STIS show spatially 
28: resolved structures in the 2"$\times$2" area around the star. The 
29: structures show in multiline emission of fluorescent 
30: H$_2$ pumped by Lyman $\alpha$. One emission structure follows the 
31: cavity walls 
32: observed around T Tau N in scattered light in the optical. 
33: A temperature
34: of $\ge$ 1000K is required to have enough population in the
35: H$_2$ to produce the observed fluorescent lines; in the cool environment
36: of the T Tau system, shock heating is required to achieve
37: this temperature at distances of a few tens of AU.
38: Fluorescent H$_2$ along the cavity wall
39: represents the best evidence to date  for the action of 
40: low-density, wide-opening-angle outflows driving
41: cavities into the molecular medium
42: at scales $\le 100$ AU.
43: A southern region of emission consists of
44: two arcs, with shape and orientation
45: similar to the arcs of H$_2$ 2.12 {\rm $\mu m$}
46: and forbidden line emission crossing
47: the outflow associated with the embedded system T Tau S.
48: This region is located
49: near the centroid of forbidden
50: line emission at the blueshifted lobe of the
51:  N-S outflow.
52: 
53: \end{abstract}
54: 
55: \keywords{Stars: formation, pre-main sequence --- stars: individual (T Tau) 
56: --- stars: 
57: winds, outflows}
58: 
59: \newpage
60: 
61: \section{ Introduction}
62: T Tau is a multiple system of pre-main sequence stars, composed of
63: an optically-visible K0 star, T Tau N, and a heavily-extincted 
64: system, T Tau S \citep{D82}, which is itself binary \citep{K00, Kh00, D02}.
65:  The T Tau system is surrounded by
66: an infalling envelope of a few thousand AU \citep{C94}, which is in the
67: process of being disrupted by powerful outflows \citep{M96}.
68: Two bipolar outflows have been identified in the system
69: \citep{B88, B94}. One of the outflows runs 
70: NW-SE
71: and has been associated with the embedded pair T Tau S.
72: Analysis of the forbidden line emission indicates that
73: this outflow
74: is poorly collimated \citep{B94, S99}. Bright 
75: arcs
76: of forbidden line emission \citep{R95} and near
77: infrared H$_2$ line emission \citep{H96, H97}
78: are found crossing this outflow at scales of $\sim$ 2" - 14"  
79: ($\sim$ 280 - 2000 AU at the distance of Taurus,
80: 140 pc, Kenyon et al. 1994). 
81: Burnham nebula, $\sim$ 8" south of the
82: system, is associated with the blueshifted lobe of this outflow;
83: it continues in 
84: jets detected out to $\sim$ 0.7 pc \citep{R97}.
85: The other outflow runs  E-W and is associated
86: with T Tau N. It is less bright in the optical,
87: although high velocity forbidden line emission
88: is detected towards the west, ending in Herbig
89: Haro object HH155 located at
90: the edge of the reflection nebulosity NGC 1555 (Hind's
91: nebula), $\sim$ 35" west from the star. $^{12}$CO emission associated with the 
92: E-W
93: outflow shows blue and redshifted lobes of moderately
94: high velocity (Schuster et al. 1994), surrounded
95: by rings of $^{13}$CO emission at lower velocity;
96: these rings have been interpreted as arising at the wall of
97: the cavity  in the envelope opened by the E-W outflow
98: (Momose et al. 1996).
99: 
100: The structure of the outflows and molecular gas and dust 
101: very near T Tau (inner 2"x2")  is  
102: uncertain.  
103: Observations in the V, R, and I filters with the
104: WFPC2 camera on the Hubble Space Telescope
105: ($HST$) by Stapelfeldt  (1998, S98) showed a scattered light 
106: structure
107: whose morphology suggests an illuminated 
108: outflow cavity 
109: in the circumsystem envelope.  Spectroscopic mapping in the K (2.01-2.42 
110: {\rm 
111: $\mu$}m) 
112: and H (1.5-1.8 {\rm $\mu$}m) bands by \cite{K01} probed the inner
113: 2"$\times$2" of the T Tau system; unresolved Br$\gamma$ emission at
114: the positions of T Tau N and S was observed, but no infrared H$_2$ emission 
115: was 
116: detected.
117: 
118: In contrast, space-based observations in the UV of molecular hydrogen have 
119: indicated
120: structure on small scales near T Tau.
121: \cite{B81} detected extended H$_2$ fluorescent emission 
122: in T Tau N and its adjacent nebula in a low resolution large
123: aperture spectrum from the
124: IUE satellite.  
125: %In order to fluoresce with Lyman $\alpha$ emission,
126: %some H$_2$ must be in excited levels.
127: %Modeling the emission as arising in a shocked spherical shell 
128: %of material, Brown et al. estimated N(H$_2$)$\approx 
129: %5 \times 10^{18}$ ${\rm cm^{-2}}$ and T $\geq$ 2000 K for 
130: %the local H$_2$ column density and excitation temperature, respectively. 
131: \cite{V00} refined the
132: location of the H$_2$ emission by comparing data from the Goddard High
133: Resolution Spectrograph (GHRS) on $HST$ in G140L mode with all the 
134: available 
135: IUE 
136: spectra of T Tau.  The H$_2$ line fluxes were found to be
137: larger in the IUE spectra 
138: %than in the GHRS spectra, 
139: suggesting that the molecular hydrogen emission
140: extended beyond 0.2" from the star, the size of the Small Science 
141: Aperture in GHRS.  
142: In this paper,  we present new data from 
143: the Space Telescope Imaging Spectrograph
144: (STIS) on board $HST$ which
145: illustrate the spatial distribution of fluorescent H$_2$ emission in the T 
146: Tau
147: system, with important implications for the interaction of outflows and 
148: circumstellar
149: matter.
150: 
151: \section{Observations}
152: 
153: T Tauri was observed with $HST$ on 2000 February 7 in
154: program GO8317, using the STIS MAMA detector with the G140L 
155: grating. A 2" long slit with a PA=$30^{\circ}$ was employed, and the
156: data were taken in binned pixels mode with an exposure time of 1622 s.  
157: The spectral range covered from 1118 to 1715 {\AA} with a 
158: dispersion of 0.58
159: {\AA} per pixel resulting in a
160: %FWHM 
161:  resolution of $\sim $1
162:  {\AA} at 1500 {\AA}, and a plate scale of 0.024" per pixel.  
163: Standard CALSTIS pipeline procedures were used to reduce the data.
164: 
165: Figure 1 shows the position-dispersion image.  Faint emission
166: is seen extending roughly 1.6" $\sim$ 240 AU southward from the ultraviolet 
167: image of T Tau 
168: N.
169: Some extended emission is seen to the north, but to a much smaller extent.
170: The spectra of the extended emission were extracted using standard 
171: IRAF/STSDAS/X1D tasks,
172: although the extraction was difficult  because the spectra have low 
173: signal-to-noise 
174:  ($\sim$ 8 for the strongest lines) and because extended structures fill the
175: large aperture in the dispersion
176: direction. For the extraction,
177: the position of the background was taken outside of the range
178: corresponding to the geocoronal
179: Lyman    
180: $\alpha$ and O I lines and far from the extended
181: emission; the background was the same for all the spectra, averaged over 10 
182: pixels.
183: The signal along the extended emission is weak, roughly 20 times
184: weaker
185: than the stellar spectrum. Line emission accounts for about
186: 50\%
187: of the total signal.
188: We attempt to identify the strongest
189: emission features, recognizing that there may be weaker, more diffuse
190: emission that we cannot trace.
191: 
192: Figure 2 shows the spectrum of T Tau N, scaled by a factor of 0.2, and
193: spectra extracted in 10 pixel = 0.24" boxes at various offsets from 
194: the star along the slit orientation.
195: Lines from
196: the highly-ionized species Si IV 1393/1402, C IV 1458/1550, and He II 1640 
197: are present in the
198: stellar spectrum; the corresponding emission features 
199: in the spectra offset at $\pm$ 0.39" are consistent with the expected
200: wings of the instrumental point spread function for the stellar emission.
201: The other line features in the extended nebulosity
202: are consistent with fluorescent H$_2$ emission.
203: As shown in Figure 2, 
204: the strongest lines arising from
205: de-excitation
206: of the rotational-vibrational level (v',J'=1,4) of the electronic state
207: $^1\Sigma_u^+$ populated by the (1-2)P(5) transition can be identified.
208: There is also a suggestion that the strongest lines of the de-excitation of
209: the (v',J'=1,7) level, populated by the (1-2)R(6) transition are present.
210: 
211: Table 1 shows the observed fluxes of the fluorescent
212: lines due to the (1-2)P(5) transition in several offset
213: spectra. The fluxes are
214: measured above 
215: the estimated continuum.
216: The last column shows the ratios of the line fluxes at $\lambda 1446$ and 
217: $\lambda 1504$ \AA, 
218:  where the fluxes
219: have been corrected for reddening
220: with A$_V$=2.2 using the HD29647 
221: extinction law \citep{C03}.
222: The ratios are consistent with the optically thin ratio (0.73),
223: within the uncertainties given by their low S/N.
224: 
225: \section{Spatial distribution} 
226: \label{espacial}
227: 
228: To make the line identifications shown in Figure 2,
229: it is necessary to impose wavelength shifts to
230: the spectra of the extended emission.
231: If these shifts were interpreted as velocities, they would require motions
232: of $\sim$ 2000 {\rm km s$^{-1}$}  or more,
233: much larger than the velocities previously reported in the region
234: ($\sim$ 45 {\rm km s$^{-1}$}; 
235: \citep{B94, E98}.
236:  A more likely explanation for the wavelength shifts
237: is that they are due to spatial shifts of the emission regions
238: relative to the star within the large (2") aperture.
239: In this case, the wavelength shift $\Delta \lambda$
240: would imply a
241: displacement $ x = s \Delta \lambda / m $ from the axis
242: of the slit,
243: where $s$ is the plate scale and $m$ is the dispersion. 
244: All the offsets are measured from the position
245: of T Tau N.
246: 
247: Figure 2 indicates at the right the offsets $x$ required
248: to fit the H$_2$ lines
249: at each $y$ offset from the star.
250: Two spatially-distinct regions of H$_2$ emission can be identified.
251: One region extending to $y \sim 0.7$" southwest from the star,
252: which is shifted slightly westwards,
253: and a region covering
254: $0.7" < y < 1.6"$ from the star, which has
255: an eastward displacement.
256: These regions can be discerned in Figure 1. 
257:  The northern
258: region, labeled by N in the sketch below the spectrum in
259: Figure 1, can be seen as three features
260: extending to a distance of about 0.7" southward
261: from the star, at PA $\sim 210^{\circ}$,
262: corresponding to the H$_2$ line pairs at
263: 1490/1505 {\AA}, 1431/1446 {\AA}, and
264: 1547/1562 {\AA}, although in the latter case
265: the structure near the star is difficult to see due to the strong stellar
266: C IV 1550 {\AA} emission.
267: Further south, these emission features
268: seem to disappear; however, similar pairs of emission lines
269: can be seen shifted roughly 25 {\AA} blueward, extending nearly
270: vertically in the image to about 1.6" south.  This 
271: region of emission is labeled S in  Figure 1. 
272: The N region extends to the north of the star as well,
273: but it is difficult to identify
274: spectral features in it.
275: After taking into account the position angle of the slit,
276: we find that the spatial location of the upper N emission region 
277: appears to correspond to the
278: the brightest part of the southern optical scattered light structure imaged 
279: by S98, 
280: while the lower emission region roughly follows the outer border of the
281: scattered light nebula, although displaced from it.
282: 
283: To obtain more information about the spatial distribution
284: of the emission, we constructed a model and
285: compare the predicted
286: emission with the observed spectrum. If the spatial
287: distribution of the emission 
288: at a given offset $y$ is given by {\it w(x,y)}, and the intrinsic
289: emission at each point is $G_{\lambda}$, then the
290: observed flux at wavelength $\lambda$ is given by
291: the convolution of {\it w(x,y)} and  $G_\lambda$,
292: \begin{equation}
293: {\it F}_{\lambda}(y)=a \sum_{i=1}^{N}  G_{\lambda-\Delta 
294: \lambda_i}w(x(i),y)
295: \end{equation}
296: where $a$ is a scaling
297: constant,
298: {\it x(i)} is the spatial shift across the slit from its axis,
299: $\Delta \lambda_i$ is the wavelength shift corresponding to
300: $x(i)$, and $N$ is the number of pixels across the slit.
301: 
302: We used as template $G_{\lambda}$ the spectrum of 
303: HH 43, a low-excitation Herbig-Haro object,
304: which clearly shows the fluorescent lines we have identified
305: \citep{Sc83}.  The spectrum is an average of the three IUE large aperture
306: spectra with the highest exposure times, SWP31828, SWP23881, and SWP24924,
307:  taken from the Multimission Archive at the Space Telescope Science Institute,
308: which have been
309: processed under the NEWSIPS extraction pipeline.
310: The combined signal to noise of the resultant HH spectrum is
311: 8-13.
312: Since we are interested in the relative distribution
313: of the H$_2$ intensity, we 
314: normalized $w(x,y)$ to 1 and scaled
315: $G_{\lambda}$ with the constant {\it a}
316: to fit the observed spectrum
317: of the extended emission at offset $y$ = -0.39".
318: 
319: The predicted spectra at several offsets $y$ from the star
320: are shown in Figure
321: 3 compared to the observed spectra smoothed to the IUE resolution of 6 
322: {\AA}; 
323: the corresponding spatial map of the emission $w(x,y)$
324: is shown in Figure 4.  
325: 
326: The H$_2$ emission in the southern part of region N
327: (N1 in Figure 4)
328: appears to follow the brightness and spatial
329: distribution of the scattered light image 
330: of S98, with some indication that
331: the H$_2$ is somehow spatially more
332: concentrated than the optical region. The H$_2$ and
333: optical emission are similar from $\sim$ 0."45 to $\sim$ 0."65
334: from the star,  but around $\sim$ 0."45, the H$_2$ emission is $\sim$ 30\% 
335: narrower
336: than the optical scattered light.
337: Consistently, the H$_2$ emission in the northern part of
338: region N (N2 in Figure 4) is low, 
339: since the north part of the optical emission is two magnitudes fainter
340: than the
341: southern portion at the same distance from T Tau N (S98).
342: 
343: The resulting 
344: predicted spectra for region N1 are shown at -0.39" and -0.62" offsets in 
345: Figure 
346: 3.
347: The spectral agreement is reasonable considering the low S/N
348: of the spectra and the contamination by stellar emission at  
349: -0.39".  The principal failing of the model is the lack of predicted
350: emission near 1600-1620 {\AA}, which is identified in the spectrum of T Tau 
351: N as 
352: a 
353: mix of CII an Fe II by Valenti et al. (2000).
354: Fe II emission may arise from the $\sim$40 $\rm km~s^{-1}$
355: J-shocks discussed below as sources of Lyman $\alpha$ photons
356: for the lower emission region.
357: It is also likely that the broad
358: Lyman $\alpha$ emission profile of T Tau produces some of the
359: fluorescent sequences observed in Mira \citep{W02}
360: but not in low excitation HH objects.  Some of these sequences
361: include emission near 1610 {\AA}.
362: 
363: For region S, $y \le -0.85"$, 
364: the observed line profiles are too narrow to
365: resolve the emitting regions;
366: we found the best fits by introducing two narrow
367: structures
368: at slightly different positions, labeled S1 and S2 in Figure 4.
369: The agreement between predicted and observed spectra in Figure 3
370: is reasonably satisfactory except
371: for a feature near 1600-1620 {\AA} in the $y = -0.62"$ offset spectrum.
372: We should also emphasize again that it would be difficult for us to detect 
373: very
374: spatially-extended H$_2$ emission in our large aperture data.
375: 
376: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
377: \section{Physical conditions}
378: \label{physical}
379: 
380: 
381: \subsection{Estimates of Lyman $\alpha$ luminosities}
382: 
383: The total observed flux in the
384: strongest H$_2$ line we observe,
385: (1-7)P(5), $\lambda$= 1504.8 \AA, can be written as
386: (cf. Jordan et al. 1978)
387: \begin{equation}
388: F_{H_2} =3\times 10^{-16} \, N(2,5) A_{H_2} \, \bar J_{\nu}\, {\rm \; 
389: erg\;cm^{-2}\;s^{-1}},
390: \label{flujo}
391: \end{equation}
392: where $N$(2,5) is the column
393: density in the vibrational-rotational level (v",J"=2,5) of H$_2$,
394:  $A_{H_2}$ is the H$_2$ emitting area projected in the sky
395: in arcsec$^2$, and
396: $\bar J_{\nu}$ is the mean
397: intensity of the Lyman~$\alpha$ radiation capable of exciting
398: the (1-2)P(5) transition\footnote{From now on, the
399: bar notation refers to any quantity $\bar f=\int\phi_\nu f_\nu d_\nu$, with
400: $\phi_\nu$ the line profile of the H$_2$ line, centered at
401: $\nu_0 + \Delta \nu =\nu_0 + \nu_0\Delta v/c$
402: from the Lyman $\alpha$ line center at $\nu_0$.
403: For the  (1-2)P(5) transition, $\Delta v = +98 {\rm km \, s^{-1}}$.}, in
404: ${\rm \; erg\;cm^{-2}\;s^{-1} \; Hz^{-1}}$.
405: We have used the line parameters 
406: $B_{(2,5\to1,4)}=1.976\times 10^8$ erg$^{-1}$ cm$^2$ s$^{-1}$,
407: $A_{(1,4\to7,5)}=1.969\times 10^8$ s$^{-1}$ and
408: $\Sigma_j A_{(1,4\to j)}=1.7125\times 10^9$ s$^{-1}$ from \cite{A00}.
409: 
410:  If we assume that the H$_2$ emission
411: region N is irradiated by stellar Lyman $\alpha$
412: and that there is no extinction between the star and
413: the region, then
414: the mean intensity $\bar J_{\nu}$ reaching the region
415: is related to the specific
416: intensity $\bar I_{\nu}$ leaving the star by
417: \begin{equation}
418: \bar J_\nu \approx {1 \over 4\pi} { {\bar I_\nu A_{Ly \alpha}} \over  
419: d^{2}},
420: \end{equation}
421: where $A_{Ly \alpha}$ is the area of the Lyman $\alpha$ emitting region 
422: on the star
423: and $d$ the distance
424: between the star and the site where the H$_2$ molecules are
425: pumped. In turn, $\bar I_{\nu}$ can be expressed in terms of the 
426: flux $\bar F^{obs}_{Ly \alpha}$ of Lyman $\alpha$ observed at Earth,
427: as $\bar I_{\nu} =\bar  F^{obs}_{Ly \alpha} D^{2} / A_{Ly \alpha}$
428: (assuming isotropic emission)
429: where $D$ is the distance to the star.
430: Combining these expressions, we can write 
431: \begin{equation}
432: \bar J_\nu \approx 3.3 \, \times 10^9 { \bar F^{obs}_{Ly \alpha} \over 
433: \theta^{2}} {\rm 
434: erg 
435: \;cm^{-2}\;s^{-1}\;Hz^{-1}\;sr^{-1}},
436: \label{jotanu}
437: \end{equation}
438:  where $d/D$ 
439: has been approximated by the apparent angular separation
440: between the star and the H$_2$ emitting region
441: on the sky, $\theta$, in units of arcsec in eq.(\ref{jotanu}).  
442: 
443: Inserting in eq. (\ref{flujo})
444:  we obtain
445: \begin{equation}
446: F_{H_2}  \approx 10^{-6} 
447: { A_{H_2} \over \theta^{2}}
448: N(2,5) \bar F^{obs}_{Ly \alpha} {\rm \; 
449: erg\;cm^{-2}\;s^{-1}}
450: \label{flujo1}
451: \end{equation}
452: 
453:  We estimate a lower limit for $N$(2,5)
454: by taking the column density of the outermost layers
455: of the cavity where radiation of Lyman $\alpha$ can penetrate,
456: to a depth where $\tau_\nu \sim 1$.
457: The actual column density producing the emission may be larger than this if
458: the emitting region is not seen face-on.
459: This may be indeed the case, given the low inclination
460: to the line of sight, $\sim 10^{\circ}$, of the E-W
461: outflow estimated by \cite{M96}.
462: The optical depth for a transition line from an upper level {\it j} to a 
463: lower 
464: level {\it i} can be written as $
465: \tau_\nu=N_i B_{ij} h\nu_{ij} \phi_\nu (1-exp(-h\nu_{ij}/kT))/ 
466: 4\pi \sim N_i B_{ij} h\nu_{ij} \phi_\nu /4\pi$, for a temperature of a few 
467: thousand K,
468: with $\nu_{ij}$ the frequency of the transition.
469: We assume that the H$_2$ line is Gaussian in shape,
470: with a Doppler width
471: $\Delta \nu_D = \nu_{ij} \sqrt{ v_{th}^2 + v_{tur}^2} /c$.
472: The agreement with the spectrum of HH43 (\S \ref{espacial}) suggests
473: a low excitation situation, so we adopt a
474: temperature T$\sim$1000 K to determine the thermal velocity
475: $v_{th}$, and assume a turbulent velocity $v_{tur}\sim$10 km s$^{-1}$
476: as expected from an oblique shock interface (\S 5.1).
477: Assuming 
478: the line profile is given by $\phi_{\nu}\approx 1/\pi^{1/2}\Delta \nu_D$,  
479: we get a 
480: column density $N(2,5)\approx 7 \times10^{14}$ cm$^{-2}$
481: from the condition $\tau_\nu \sim 1$ for the (1-2)P(5) transition.
482: 
483: Let $Q$ be the ratio 
484: between $\bar F^{obs}_{Ly \alpha}$ and the total observed flux $F_{Ly\alpha}$. 
485:  We cannot use the 
486: stellar Lyman $\alpha$ line profile in the spectrum
487: because it has been heavily absorbed by circumstellar
488: and interstellar extinction. So, we
489: estimate the value of $Q$ assuming 
490: that the Lyman $\alpha$ profile is similar
491: to that of Mg II 2800 {\AA},
492: since both 
493: are resonant lines with extended wings formed by partial redistribution.
494:  We use the high resolution line profile of Mg II k $\lambda$ 2796.3 in T 
495: Tau N,
496: obtained with STIS/NUV-MAMA and echelle grating E230M 
497: in our $HST$ program GO8627
498: (Fig. 5, Saucedo et al., in preparation).
499: At $\Delta v$= 98 km s$^{-1}$ 
500: from the center of the Mg II 2796.3 {\AA}
501: line, we determine $ Q \sim 1.4 \times 10^{-12}$ Hz$^{-1}$.
502: As illustration, we show in Figure 5 the H$_2$ line profile $\phi_{\nu}$
503: at $\Delta v$= 98 km s$^{-1}$ used for the calculation
504: of $\bar F^{obs}_{Ly \alpha}$.
505: 
506: The emitting area $A_{H_2}$ can be estimated from the height of the
507: extraction box used to obtain the spectra and the width of the relative 
508: intensity
509: map {\it w(x,y)} as $\sim$ 0.2$\times$0.24 arcsec$^2$. 
510: Taking the distance $\theta$ between the star and the H$_2$ region as 
511: $\sim$ 
512: 0.5", 
513: an average flux of H$_2$ 1504.8 {\AA} $\sim 2.5 \times
514: 10^{-15}$ ${\rm erg\; cm^{-2}\;s^{-1}}$ in region N (cf. Table 1), 
515: the estimated values of $N(2,5)$ and $Q$, and eq. (\ref{flujo1}),
516: we can write for the luminosity of the stellar Lyman $\alpha$ line
517: \begin{equation}
518: 8 \times 10^{-3} L_\odot\leq 
519: L_{Ly \alpha} \leq  2.1 L_\odot
520: \label{lyalimits}
521: \end{equation}
522: where the limits arise from the uncertainty
523: in the reddening correction
524:  appropiate to the H$_2$ lines, since the
525: extinction along the line of sight of the nebulosity may not be the same as 
526: the
527: extinction of the star;
528: the lower limit comes from taking the observed
529: H$_2$ flux and the upper limit from dereddening this flux
530: by the extinction to T Tau N,
531:  $A_v=2.2$ \citep{C03}.
532: In addition, as discussed above, the actual value of $N(2,5)$
533: is likely to be higher because of projection effects,
534: making the required  Lyman $\alpha$ luminosity lower than
535: estimates in eq.(\ref{lyalimits}).
536: These limits for the Lyman $\alpha$ luminosity
537: are consistent with those obtained by scaling
538: by a factor of 30 the dereddened flux of the CIV
539: 1548 line in our stellar spectra, as suggested by Ardila et al. (2002).
540: With F$_{\rm CIV }\sim 9.2 \times 10^{-12}$ $\rm
541: erg\; cm^{-2}
542: \;s^{-1}$, the scaled Lyman
543: $\alpha$ luminosity would be $\sim$ 0.35 L$_\odot$.
544: 
545: The estimated maximum Lyman $\alpha$ luminosity (eq. [\ref{lyalimits}])
546: is $\le$ 60 \%
547: of the
548: accretion luminosity of T Tau N 
549:  estimated from the 
550: NUV spectrum obtained simultaneously with the FUV spectrum,
551: ${\rm L_{acc} = 3.2 L_\odot}$ \citep{C03}, 
552: so it can be accounted for by accretion energy.
553: We thus conclude that the  H$_2$ emission in 
554: region N is 
555: consistent with stellar  Lyman $\alpha$ pumping.
556: 
557: While the bright rims seen in region N are clearly illuminated
558: by T Tau N, 
559: the southern arcs 
560: S1 and S2  appear to be located (in projection) well inside
561: the envelope, roughly tracing the outer edge of the extended emission of
562: S98.
563: Lyman $\alpha$ from the visible star will be absorbed by dust in the
564: envelope so it cannot penetrate more than
565: one dust mean free path, $l\sim 3-4$ AU,
566: estimated from models of infalling envelopes \citep{W93} with typical envelope
567: parameters
568: from \cite{C94} ({$\rm \dot M=3\times10^{-6}\;M_\odot\;yr^{-1},\;
569: r_c=100\;AU,\;M=2\;M_\odot $}) and a dust opacity
570: at the  Lyman
571: $\alpha$ wavelength of 1538 cm$^2$ g$^{-1}$ \citep{O02}.
572:  It is therefore more likely that the required Lyman $\alpha$
573: emission comes from the shocks themselves,
574: as in the case of low excitation HH objects \citep{C95}.
575: 
576: If Lyman $\alpha$ 
577: is locally produced, then $\bar J_\nu
578: =\bar I_\nu /2$, so
579: we can write
580: \begin{equation}
581: L_{H_2} = 6.8 \times 10^{-6} N(2,5) Q L_{Ly \alpha}
582: \end{equation}
583: where $L_{H_2} = 4 \pi D^2 F$ is the H$_2$ luminosity.
584: Using similar values of $N(2,5)$ and $Q$ as in region N,
585: we get that the luminosity in Lyman $\alpha$ 
586: required to excite the H$_2$ fluorescence in region S is 
587: \begin{equation}
588: 3.4 \times 10^{-4} L_\odot\leq   
589: L_{Ly \alpha} \leq  0.1 L_\odot, 
590: \end{equation}
591:  where again, the lower limit is given by the observed values of the H$_2$ 
592: flux, 
593: and 
594: the upper limit by assuming that the extinction is the same as towards T 
595: Tau N, 
596: although 
597: there is no argument to support the assumption of an homogeneous extinction 
598: in 
599: the 
600: whole region. 
601:  These limits are consistent with values of
602: the Lyman $\alpha$ luminosity in low excitation HH objects;
603: for instance, \cite{C95} find $ L_{Ly \alpha} \sim 0.13 L_\odot$
604: in HH 47A.
605: 
606: \subsection{Estimates of temperature and Hydrogen column density}
607: 
608: Fluorescence
609: requires a significant population in excited levels which also can lead
610: to infrared H$_2$ emission lines.  Thus, the observed upper limits 
611: to the infrared emission line fluxes 
612: in observations covering region N \citep{K01} can provide an 
613: additional constraint
614: on molecular column densities. 
615: 
616: An upper limit for the  2.12 $\mu$m H$_2$ line
617: can be estimated from the equivalent width
618: of 0.15 \AA, which is the detection threshold for features 
619: in the \cite{K01} observations. Using the continuum flux
620: of $\sim 3.1 \times 10^{-9} {\rm erg\;cm^{-2}\;s^{-1}\mu m^{-1}}$,
621: we obtain that the flux at the 2.12 $\mu$m H$_2$ line
622: needed for a 2 $\sigma$ detection would be
623: $9.3 \times 10^{-14}$ $\rm erg \;cm^{-2}\;s^{-1}$.
624: 
625: The flux at 2.12 $\mu$m H$_2$ in the optically thin limit is given by
626: $F_{2.12} = I_{2.12} A_{H_2} / D^2 = N(1,3) A_{31} h\nu_{31}/4 \pi 
627: A_{H_2}/D^2$,
628: where $N(1,3)$ is the column density of the upper
629: level of the transition. With the Einstein value $A_{31}=3.66\times 
630: 10^{-7}s^{-1}$
631: \citep{G76} and the upper limit for the flux,
632: we obtain $N(1,3)\leq 2.24\times 10^{18}{\rm\;
633: cm^{-2}}$.
634: 
635: As mentioned before, we have only a lower limit to the actual N(2,5). 
636: With this lower limit and the upper limit for N(1,3),
637: we can obtain a lower limit to the excitation
638: temperature in region N through the Boltzmann relation
639: $ N(1,3)/N(2,5)
640: =1/1.57exp(-E_{(1,3\to 2,5)}/kT)$,
641: with $E_{(1,3\to 2,5)}/k \approx 6840$ \citep{H50}.
642: Using our estimates, we obtain 
643: $T \geq$
644: 800 K, consistent with a low velocity shock.  From 
645: Figures 6 and 7 in \cite{J78}, 
646: the population of the levels with (v"J"=2,5) in this temperature range is 
647: $\sim$ 
648: 12\% of
649: the populations with v"=2 and any J", and in turn, the population with 
650: v"=2 is 
651: $\sim$ 0.001\% of
652: the total population.  So, with $N(2,5) > $7$\times 10^{14}$ cm$^{-2}$, 
653: the column density of the H$_2$ molecule is
654: N(H$_2$)$ > 5.8 \times 10^{20}$ cm$^{-2}$.  
655: 
656: We have no measurements of H$_2$ 2.12 
657: $\mu$m in region S
658: to help constrain the 
659: temperature and total H$_2$ density.
660: Information on regions S1 and S2 can be obtained  from the [O I], [N 
661: II] 
662: and [S II] PV 
663: diagrams in \cite{B94}. From their Figure 5, 
664: regions S1 and S2 are located within 
665: one of the
666: multiple components identified in their work, component D,
667: which corresponds to the blueshifted lobe of the N-S outflow.
668: The line 
669: ratios [O I]/[S II], 
670: [N II]/[O I] and [N II]/[S II] around the location
671: of region S are 1.25, 0.35 and 0.28, respectively. 
672: Comparison with shock model predictions \cite{H87}
673: indicate a shock of velocity v=40 km s$^{-1}$.
674: For this shock velocity, the 
675: ratio of Lyman $\alpha$/[O~I] can be obtained from these models, yielding a 
676: value of $\sim$ 73; so, a rough estimate of the [O I] luminosity is $ 4.6 
677: \times 
678: 10^{-6} L_\odot\leq   L_{[O I]} \leq  3.5\times 10^{-4} L_\odot $.
679: Unfortunately, measurement of absolute fluxes for [O I] to compare with this 
680: prediction 
681: are not available (Solf 2002, private communication). 
682:  Inspection of the relative intensities of [O I]
683: along the slit with PA=0$^{\circ}$ centered
684: on the star in \cite{B94} indicates
685: that the region at $\sim $ 2 arcsec south is $\sim$ 4-5 times 
686: fainter than the star.
687: We can estimate the [O I] stellar luminosity from the mass
688: loss rate $\dot {\rm M_w}$, using the relationship 
689: log $\dot {\rm M_w}=-4.3 + {\rm log}(L_{[O\; I]}/L_\odot)$
690: \citep{H95}, assuming that outflow rate scales as $\sim$ 0.1 
691: the accretion rate \citep{C98}, and obtaining the mass accretion
692: rate from the accretion luminosity. This finally
693: leads us
694: to an estimate of $\rm L_{[O I]} \sim 2 \times 10^{-4} L_\odot$ for the region,
695: which is consistent with our expected limits of the [O I] luminosity.
696: 
697: \section{Discussion}
698: 
699: \subsection{The northern H$_2$ emitting region}
700: 
701: Region N appears to be aligned along the bright rims of the
702: reflection nebulosity around T Tau N, \S \ref{espacial};
703: which was plausibly interpreted by S98 as the
704: walls
705: of a cavity driven into the surrounding medium by the outflow from
706: this star.
707: The H$_2$ emission requires some heating to excite
708: molecules into the level which can fluoresce with Lyman $\alpha$.  
709: At distances of 
710: $\le 50$ AU where the fluorescent emission is observed, local
711: dust temperatures resulting from heating by the central star(s)
712: are predicted to be $\sim 100$~K \citep{C94},
713: whereas excitation temperatures must be $\ge 800$~K by the
714: argument of the previous section.  
715: The most likely explanation is that shock heating is responsible
716: for the excitation of H$_2$ in these regions, as also suggested
717: by the consistency of the observed spectrum with that of a low-excitation
718: Herbig-Haro object (\S \ref{espacial}).
719: 
720:  The impact of the wind emanating from T Tau N on the
721: molecular medium can naturally explain the needed excitation of H$_2$,
722: and it has already being suggested as a possible explanation
723: for the near infrared H$_2$ bright rims that seem to
724: coincide with the scattered light image at
725: larger scales \citep{H97}.
726: \cite{B94} 
727: argued that the E-W outflow is a highly
728: collimated, high velocity $\sim$ 200 ${\rm km s^{-1}}$
729: jet at 0.3" $\sim$ 40 AU from the star from analysis of the 
730: forbidden line emission.
731: Alternatively, as proposed previously by \cite{S95}, \cite{S98}, and more 
732: recently by \cite{S02}, the jet results from a density enhancement along the 
733: outflow
734: axis in a wide-angle wind. Our observations
735: support the second hypothesis in that a wide-angle
736: wind is required to produce the 
737: excitation along the walls of the cavity. Moreover,
738: we expect that an oblique shock
739: would be formed at the interaction of the wide-angle
740: wind and the molecular environment, as envisoned
741: by \cite{C80}. Only the normal component of the
742: velocity would be thermalized,  reducing
743: the effective pre-shock velocity.
744: The velocities required to produce
745: a low excitation spectrum as that emitted along the
746: cavity (\S \ref{espacial}) are $\le$ 60 km s$^{-1}$
747: \citep{H87}; higher velocities would result in high
748: ionization UV lines, such as CIV and Si IV, which are clearly absent in
749: our data.
750: Thus, the observed H$_2$ fluorescent emission at the cavity walls 
751: may be the clearest (though indirect) evidence to date for 
752: the presence of low-density, 
753: wide-opening-angle outflows driving 
754: cavities into the molecular medium in star-forming regions. 
755: 
756: Observations of other shock-excited features close to T Tau N
757: are needed to help constrain wide-angle wind properties.
758: 
759: \subsection{The southern H$_2$ emitting regions}
760: 
761: The double arcs of region S are located
762: as close as 0".3 (49 AU) to T Tau S. 
763: The shape and orientation of this double arc are very similar to the arcs
764: of H$_2$ 2.12 {\rm $\mu m$} and optical forbidden line emission
765: described by \cite{H96,H97},
766: interpreted as
767: terminal shocks of H$_2$, [S II] and [Fe II] \citep{H97,R95} in the N-S
768: outflow. 
769: Figure 4b shows
770: an enlarged view of the region around the T Tau
771: system, indicating the positions
772: of regions N and S and
773: the shocked structures described by \cite{H97},
774: as well as the scattered light image of S98.
775: As mentioned in \S \ref{physical},
776: the arcs of region S are located inside the
777: region of forbidden line emission labeled
778: D by \cite{B94}, very near the centroid.
779: The outlines of this region as well as the
780: position of the centroid are indicated
781: in Figure 4b. 
782: The mean velocity of region D
783: is v = -44 km s$^{-1}$,
784: enough to have a J-shock
785: which could excite the fluorescence as in
786: other low excitation HH objects \citep{C95}.
787: In the case of HH47A, 
788: \cite{C95} find that 90 \% of the UV line emission comes from a postshock 
789: distance of $\sim 2 \times 10^{14} {\rm cm} \sim $ 13 AU, 
790: which we could not resolve. Therefore, the coincidence
791: of the H$_2$ UV emission with the centroid of the [S II] emission of component
792: D in \cite{B94} is to be expected.
793: More sensitive near-infrared observations to constrain the 2.12 $\mu$m 
794: emission would provide better constraints on column densities in these
795: regions.
796: 
797: \section{Summary}
798: 
799:  We describe the spatial distribution of the H$_2$
800: fluorescent emission in the inner 2"x2" region
801: around the T Tau system. A northern
802: region of emission coincides with the walls of the
803: cavity of the envelope seen in scattered light.
804: Lyman $\alpha$ from T Tau N appears to be strong enough to
805: excite the fluorescence, but local
806: temperatures  $\sim$ 1000 K are required to mantain the
807: fluorescence; shock heating is requred
808: since stellar heating is not sufficient.
809: The required
810: temperatures and the spectrum
811: of the nebulosity, which resembles that of a low
812: excitation HH object, indicate a low velocity
813: shock. The shock may be produced by the interaction
814: of a wide-angle wind
815: emanating from T Tau N and the molecular
816: material around it. This represents the
817: clearest evidence for the presence of wide-angle
818: winds at scales $<$ 100 AU from the star.
819: A southern region of emission consists of
820: two arcs, with shape and orientation
821: similar to the arcs of H$_2$ 2.12 {\rm $\mu m$}
822: and forbidden line emission crossing
823: the outflow associated with the T Tau S system.
824: The arcs are near the centroid of the
825: forbidden line emission in the
826: blueshifted lobe of the outflow. The 
827: velocity of the outflow around the arcs
828: are consistent with J-shocks, which
829: could power the fluorescence as in
830: low excitation Herbig-Haro objects.
831: 
832: 
833: \vskip 0.5in
834: 
835: \noindent
836: Acknowledgments. 
837: We wish to thank K. Stapelfeldt for providing us the optical image of T 
838: Tau, M. Kasper for the infrared images and spectra of T Tau N, 
839: and K. B\"ohm and J. Solf for the original 
840: long slit spectrograms of T Tau N and its surroundings.  We also thank H. 
841: Abgrall, for the updated tables 
842: of Einstein coefficients and wavelengths for the UV H$_2$ transitions. 
843: This work was supported
844: by NASA through grants GO-08206.01-97A and GO-08317.01-97A from the
845: Space Telescope Science Institute, and by NASA Origins of Solar Systems
846: grant NAG5-9670.
847: The non HST ultraviolet data used in this paper were obtained from the
848: Multimission Archive at the Space Telescope Science Institute (MAST).
849: STScI is operated by the Association of Universities for Research in
850: Astronomy, Inc., under NASA contract NAS5-26555. Support for MAST for
851: non-HST data is provided by the NASA Office of Space Science via grant
852: NAG5-7584, and by other grants and contracts.
853:   
854: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
855: \clearpage
856: \begin{thebibliography}{}
857: 
858: \bibitem[Abgrall et al. (2000)]{A00}
859:  Abgrall, H.,
860: Roueff, E., \& Drira, I., 2000 A \& A.  Sup.  141, 297 
861: \bibitem[Ardila et al. (2002)]{A02}
862: Ardila, D., Basri, G., Walter, F., Valenti, J., \&Johns-Krull, C., 2002, 
863: ApJ, 
864: 566, 1100
865: 
866: 
867: \bibitem[B\"ohm \& Solf (1994)]{B94}
868: B\"ohm, K., \& Solf., J., 1994, ApJ 430, 277-290
869: 
870: \bibitem[Brown et al. (1981)]{B81}
871:  Brown, A., Jordan, C, Millar, T.  J., \& Gondhalekar, P.,
872: Wilson, R., 1981, Nature, 290,34 
873: 
874: \bibitem[Calvet et al. (1994)]{C94}
875:  Calvet, N., Hartmann, L., Kenyon, S.  J.,
876: \& Whitney, B.  A., 1994, ApJ 434,330  
877: 
878: \bibitem[Calvet  (1998)]{C98} Calvet, N., 1998, Eight Astroph. Conf., Accretion 
879: Processes in Astrophysical Systems: Some Like it Hot!, 431, 495-504
880: 
881: \bibitem[Calvet et al. (2003)]{C03}Calvet, N. et al., in preparation.
882: 
883: \bibitem[Cant\'o (1980)]{C80} Cant\'o, J.\ 1980, A \& A, 86, 327 
884: 
885: 
886: \bibitem[Curiel et al. (1995)]{C95}
887:  Curiel, S., Raymond, J., Wolfire, M., Hartigan, P.,
888: Schwartz, R., \& Nosenson, P., 1995, ApJ, 453, 322-331 
889: 
890: 
891: \bibitem[Dyck et al. (1982)]{D82}
892: Dyck, H., Simon, T., Zuckerman, B., 1982, ApJ 255, L103-106
893: 
894: \bibitem[Duch\^ene et al. (2002)]{D02}
895: Duch\^ene, G., Ghez, A., \& McCabe, C., 2002, ApJ 568, 771
896: 
897: \bibitem[Eisl\"offel \& Mundt (1998)]{E98}
898: Eisl\"offel, J., \& Mundt,
899: R., 1998, AJ 115, 1554 
900: 
901: \bibitem[Gautier et al. (1976)]{G76}
902: Gautier III, T.  N., Fink, U., Treffers, R., 
903: \& Larson, H.P., 1976, ApJ 207, L219-L133 
904: 
905: \bibitem[Gullbring et al. (2000)]{G00}
906: Gullbring, E., Calvet, N., Muzerolle, J., Hartmann, L., 2000, ApJ 544, 
907: 927
908: 
909: \bibitem[Hartigan et al. (1987)]{H87}
910: Hartigan, P., Raymond, J., \&
911: Hartmann, L., 1987, ApJ, 316, 323 
912: 
913: \bibitem[Hartigan et al. (1995)]{H95}
914: Hartigan, P., Edwards, S., \&
915: Ghandour, L., 1995, ApJ, 452, 736 
916: 
917: \bibitem[Herbst et al. (1996)]{H96}
918: Herbst, T., Beckwith, S., \& Krabbe, A.,
919: 1996, AJ 111,2403 
920: 
921: \bibitem[Herbst et al. (1997)]{H97} 
922: Herbst, T., Robberto, M., \& Beckwith, 1997 AJ 114, 744
923: 
924: \bibitem [Herzberg (1950)]{H50}
925: Herzberg, G., 1950, in \emph{Spectra of Diatomic Molecules}, Second
926: Edition, Ed.  Van Nostrand Reinhold Company 
927: 
928: \bibitem[Jordan et al. (1978)]{J78}
929:  Jordan, C., Brueckner, G. E., Bartoe, J.-D.  F., Sandlin, G.  D., \& van 
930: Hoosier, 
931: M.  E.,
932: 1978 ApJ 226, 687-697 
933: 
934: \bibitem[Kasper et al. (2001)]{K01}
935:  Kasper, M. E., Feldt, M.,
936: Herbst, T.  M., \& Hippler, S., 2001, astro.ph.12042K, to appear in ApJ, 
937: March
938: 2002 
939: 
940: \bibitem[Kenyon et al. (1994)]{K94}
941: Kenyon, S.  J., Dobrzycka, D., \& Hartmann, L., 1994, AJ, 108,1872 
942: 
943: \bibitem[K\"ohler et al. (2000)]{Kh00}
944: K\"ohler, R., Kasper, M., \& Herbst, T., 2000, Poster Proceedings of IAU Symp. 
945: 200 on The Formation of Binary Stars, eds. B. Reipurth \& H. Zinnecker, p. 63
946: 
947: \bibitem[Koresko (2000)]{K00}
948: Koresko, C., 2000, ApJ 531, L147
949: 
950: \bibitem[Li and Shu (1996)]{L96}
951: Li, Z.-Y., \& Shu, F., 1996, ApJ, 468, 261-268
952: 
953: \bibitem[Momose et al. (1996)]{M96}
954: Momose, M., Nagayoshi, O., Kawabe, R., Masahiko, H., \& Nakano, T., 1996, 
955: ApJ, 
956: 470, 1001-1014
957: 
958: \bibitem[Osorio et al. (2002)]{O02}
959: Osorio, M., D'Alessio, P., Calvet, N., \& Hartmann, L., 2002, ApJ, 
960: submitted
961: 
962: \bibitem[Reipurth et al. (1997)]{R97} Reipurth, 
963: B., Bally, J., \& Devine, D.\ 1997, AJ, 114, 2708 
964: 
965: \bibitem[Robberto et al. (1995)]{R95}
966: Robberto, M., Clampin, M., Ligori, S., Paresce, F., Sacc\'a, V., \&
967: Staude, 1995 A \& A, 296, 431 
968: 
969: \bibitem[Schwartz (1983)]{Sc83}
970: Schwartz, R., 1983, RMAA, 7,27-54 
971: 
972: \bibitem[Shang et al. (1998)]{S98}
973: Shang, H., Shu, F., \& Glassgold, A., 1998, ApJ, 493, L91-94
974: 
975: 
976: \bibitem[Shang et al. (2002)]{S02}
977: Shang, H., Glassgold, A., Shu, F., \& Lizano, S., 2002, ApJ 564, 853
978: 
979: \bibitem[Shu et al. (1995)]{S95} Shu, 
980: F. H., Najita, J., Ostriker, E.~C., \& Shang, H.\, 1995, ApJ, 455, L155 
981: 
982: \bibitem[Solf, B\"ohm, \& Raga(1988)]{B88} Solf, J., Bohm, 
983: K. H., \& Raga, A., 1988, ApJ, 334, 229 
984: 
985: \bibitem [Solf \& B\"ohm (1999)]{S99}
986: Solf, J., \& B\"ohm, K.  H., 1999, ApJ
987: 523,709 
988: 
989: \bibitem[Stapelfeldt et al. (1998)]{St98}
990: Stapelfeldt,
991: K., Burrows, C., Krist, J., Watson, A., Ballester, G., Clarke, J., 
992: Crisp., D.,
993: Evans, R., Gallagher, J., Griffiths, R., Hester, J., Hoessel, J., 
994: Holtzman, J.,
995: Mould, J., Scowen, P., Trauger, J., \& Westpal, J., 1998, ApJ 508, 74 
996: 
997: \bibitem[Valenti et al. (2000)]{V00} 
998: Valenti, J., Johns-Krull, C.M., \& Linsky, J., 2000, ApJS, 129,399 
999: 
1000: \bibitem[Whitney \& Hartmann (1993)]{W93}
1001: Whitney, B.  A., \& Hartmann, L., 1993, ApJ 402, 605-622 
1002: 
1003: \bibitem[Wood, Karovska, \& Raymond (2002) ]{W02}
1004:  Wood, B.~E., Karovska, M., \& Raymond, J.~C.\ 2002, ApJ, 575, 1057
1005: 
1006: \end{thebibliography}
1007: 
1008: 
1009: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1010: \clearpage
1011: \begin{deluxetable}{cccccccc}
1012: \tabletypesize{\scriptsize}
1013: \tablecolumns{8}
1014: \tablewidth{0pt}
1015: \tablecaption{Integrated Observed UV
1016: Fluxes coming from de-excitation of (1-2)P(5)}
1017: \tablehead{
1018: \colhead{ y offset }&  \multicolumn{6}{c}{{\rm Flux} ($\times 10^{-15}$ 
1019: {\rm 
1020: erg s$^{-1}$ 
1021: cm$^{-2}$ }) {\rm by 
1022: transition} }& \colhead{}}
1023: \startdata
1024:  "  & (1-6)R(3) & (1-6)P(5)  & (1-7)R(3) & (1-7)P(5) & (1-8)R(3) & 
1025: (1-8)P(5) &  
1026: $\frac{F_{(1-6)P(5)}}{F_{(1-7)P(5)}}\tablenotemark{a}$\\ 
1027:   & 1431.01 {\AA} & 1446.12 {\AA}  & 1489.56 {\AA} & 1504.75 {\AA} & 
1028:   1547.33 {\AA} & 1562.39 {\AA}  & \\
1029: \cline{1-8}
1030: 0.39   & 0.4 & 1.4 & 1.1 & 2.5 & 1.0 & 2.0 & 0.75\\
1031: -0.39  & 0.2 & 1.0 & 1.1 & 1.8 & 1.2 & 0.6 & 0.81\\
1032: -0.63  & 0.5 & 2.0 & 0.6 & 3.4 & 1.5 & 2.6 & 0.78\\ 
1033: -0.86  & 1.2 & 0.6 & 1.1 & 1.3 & 1.6 & 1.5 & 0.74\\ 
1034: -1.10  & 0.5 & 1.0 & 1.7 & 1.8 & 1.2 & 0.5 & 0.78\\ 
1035: -1.33  & 0.4 & 1.4 & 0.2 & 2.6 & 0.7 & 0.4 & 0.75\\ 
1036: -1.57  & 0.6 & 0.4 & 0.6 & 0.8 & 0.3 & 0.7 & 0.85\\
1037: \enddata 
1038: \tablenotetext{a}{Assuming a de-reddening correction with $A_v$=2.2} 
1039: \end{deluxetable} 
1040: 
1041: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1042: 
1043: \clearpage
1044: \begin{figure}
1045: \plotone{f1.ps}
1046: \figcaption{\scriptsize Position-dispersion image of T Tau N
1047: between 
1048: 1350 and 1650 \AA.  
1049: The  upper panel shows the
1050: observed spectrum. The isocontours have values of 0.65e-15, 0.8e-15 1.0e-15 and
1051: 3.0e-15 erg cm$^{-2}$ s$^{-1}$ \AA$^{-1}$ pix$^{-1}$.
1052: The extended
1053: emission is evident,
1054: specially at negative offsets from the star.
1055: The lower panel 
1056: shows the same isocontours,
1057: identifying some the features described in the text.
1058: Note the groups of periodic features, one close 
1059: to the stellar spectrum, marked N, and 
1060: another to the south, marked S,
1061: displaced redwards respect to the first. 
1062: The location of the extraction boxes used to obtain the envelope
1063: spectra in indicated in the upper panel.}
1064: \end{figure} 
1065: 
1066: 
1067: 
1068: 
1069: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1070: \clearpage
1071: \begin{figure}
1072: \plotone{f2.eps}
1073: \figcaption{\scriptsize Spectra of T Tau N and of the 
1074: extended emission at 
1075: several offsets from the star, indicated in Figure 1. The y offset is the 
1076: position along the 
1077: slit where the spectrum 
1078: was extracted. 
1079: Lines of Si IV, C IV and He II can be identified in the spectrum of T Tau 
1080: N, 
1081: as well as in 
1082: offsets {\it y}=+0.39", -0.39" and -0.62"; the emission from these ionized 
1083: elements 
1084: in the nebulosity is due to the stellar PSF wings. 
1085: Several H$_2$ fluorescent lines are present in 
1086: the stellar spectrum and in 
1087: the spectra from the nebulosity, with a much higher flux
1088: than expected from the stellar PSF, indicating local emission.
1089: The solid lines indicate fluorescent
1090: lines arising from the (1-2)P(5) transition,
1091: while the dot-dashed lines indicate lines from the (1-2)R(6)
1092: transition.
1093: The spectra of the offsets need to be shifted in order 
1094: to make the identifications, resulting in a spatial shift given by {\rm x} 
1095: offset (see \S 
1096: 2). It 
1097: can be seen that there are two different loci for the emission, one at 
1098: positive  and the 
1099: other at negative offsets.
1100: All offsets are measured relative to the
1101: position of T Tau N.}
1102: \end{figure}
1103: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%5
1104: \clearpage
1105: \begin{figure}
1106: \plotone{f3.eps}
1107: \figcaption{\scriptsize De-reddened spectra 
1108: and model predictions for 
1109: offsets along the slit south of T Tau N, following
1110: the extended nebulosity.  The thin
1111: line corresponds to the observations, and the thick
1112: line to the best fit for the simulated emission, using the spectrum
1113: of HH 43 as template.  The first three boxes show the contribution of the 
1114: fluorescence 
1115: following
1116: the optical envelope (region N in Figure 4). The emission at the wavelengths of 
1117: CIV 1548+1550 
1118: {\AA}, SiIV 
1119: 1396+1402 {\AA}, and HeII 1640 {\AA} correspond to the stellar PSF. The last 
1120: four boxes 
1121: show the contribution of the
1122: double arc-like structure (region S in Figure 4). }
1123: \end{figure}
1124: 
1125: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1126: \clearpage
1127: \begin{figure} 
1128: \plottwo{f4a.ps}{f4b.ps}
1129: \figcaption{\scriptsize Spatial distribution of the H$_2$ 
1130: emission.
1131: a): Isocontours of
1132: fluorescent H$_2$ emission (thick solid lines) superimposed on
1133: magnitude isocontours of
1134: 0.55 $\mu$m
1135: emission from Stapelfeldt et al. (1998) (relative intensity grayscale/thin 
1136: lines).
1137: The isocontours
1138: for the fluorescent features show relative intensity, with steps of 1 
1139: magnitude, corresponding to values of 1, 0.398 and 0.158.
1140: The regions discussed in the text, N1, N2, S1, and S2 
1141: are indicated.
1142: The mean centroid
1143: position of component D from B\"ohm \& Solf (1994) is also
1144: shown (open circle).
1145: The dotted box  shows an estimate for the location of region N2, since the 
1146: overextraction 
1147: of 
1148: the star in the optical image makes it difficult to obtain a brightness 
1149: distribution (S98). 
1150: b): View of a larger area around the region shown
1151: in the upper panel (indicated by the box), showing
1152: the observing slit and features discussed in the literature:
1153: Terminal H$_2$ shocks (thick lines), [S II] or [Fe II] shocks (dot-dashed 
1154: line),
1155: oblique H$_2$  shocks (dashed lines), from Herbst et al. (1997),
1156: associated to the N-S outflow.
1157: The extend of of component D from B\"ohm \& Solf (1994)
1158: is also shown (lobe shaped contour).}
1159: \end{figure}
1160: 
1161: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%5
1162: \clearpage
1163: \begin{figure}
1164: \plotone{f5.eps}
1165: \figcaption{\scriptsize De-reddened high resolution 
1166: line profile of Mg II k $\lambda$ 2796.3 in T
1167: Tau N,
1168: obtained with STIS/NUV-MAMA and echelle grating E230M
1169: in $HST$ program GO8627 (solid heavy line).
1170: The 
1171: vertical light solid line indicates
1172: the central velocity of the 
1173: radiation pumping transition (1-2) P(5).
1174: The H$_2$ line profile $\phi_\nu$ is indicated
1175: (dashed line).
1176: }
1177: \end{figure}
1178: 
1179: \end{document}
1180: