4 Line profile classification

4.1 Classification scheme

What procedure should be used to classify the line profiles? Reipurth et al. (1996) review, briefly, existing classification schemes for the H ${\rm\alpha}$ line and propose a new one, which tries to reflect an underlying physical process as well as being able to deal with a greater variety of line profiles. Given that all Pa ${\rm\beta }$ and Br ${\rm\gamma }$ emission line profiles displayed in Figs. 1 and 2 fit in one of the classes of that classification scheme there is no need to introduce yet another one. Furthermore, this will allow a straightforward comparison between what is found for H ${\rm\alpha}$and what is found for these near infrared lines.

Figure 2: Br ${\rm\gamma }$ line profiles. Spectra are presented by the type of the line profile and for each type by alphabetical order. Spectra with error bars in each spectral point are from the UT9410 observing run. Stars for which more than one line profile is presented have the lines arranged by observing date

Figure 2: continued

Figure 2: continued

The Reipurth et al. (1996) classification scheme divides the line profiles into four main types: type I profiles are generally symmetric showing no evidence for absorption features or only very slight influence from those; type II profiles show two peaks with the intensity of the second peak exceeding half the strength of the main peak; type III profiles show two peaks with the intensity of the second peak being less than half the strength of the main peak and finally type IV profiles, which show an absorption feature beyond which no emission is seen. To types II, III and IV the letters B or R are appended, depending on the location of the secondary peak/absorption feature relative to the main peak: if blueshifted a B is appended, if redshifted an R is added. Note that type IV B correspond to normal P Cygni profiles and type IV R correspond to inverse P Cygni profiles (henceforth IPC).

4.2 Results

The classification of the Pa ${\rm\beta }$ and Br ${\rm\gamma }$(hereafter referred to as NIR lines) line profiles displayed in Figs. 1 and 2 is presented in Tables 2 and 3 respectively.

 

 

Table 3: Classification of the Br ${\rm\gamma }$ emission line profiles according to the classification scheme developed by Reipurth et al. (1996) for the T Tauri stars

Type	%	No. Stars	Star name
I	72%	18	BP Tau DG Tau DI Cep
			DL Tau DO Tau DR Tau
			FM Tau GG Tau GM Aur
			GW Ori Haro 6-13 HL Tau
			T Tau TW Hya UY Aur
			V1331 Cyg V773 Tau XZ Tau
II B	0%	0	-
II R	8%	2	RY Tau SU Aur
III B	0%	0	-
III R	0%	0	-
IV B	0%	0	-
II R	20%	5	BM And CW Tau DF Tau
			HP Tau RW Aur

Some of the NIR lines observed to be in emission are too noisy too allow a reliable classification of their profiles: DI Cep, DQ Tau, FX Tau and GH Tau in Pa ${\rm\beta }$ lines; and AA Tau, DK Tau, DS Tau, GK Tau and Haro 6-37 in Br ${\rm\gamma }$ lines. These were not considered for the statistics presented in Tables 2 and 3.

The main conclusions that can be taken from Tables 2 and 3 are the following: most line profiles are generally symmetric, especially the Br ${\rm\gamma }$ line profiles, where 72% of the profiles are classified as type I; both Pa ${\rm\beta }$ and Br ${\rm\gamma }$ line profiles lack blueshifted absorptions (only one star - CW Tau - displays any sort of absorption in the blue wing and only in Pa ${\rm\beta }$); nearly one third of the Br ${\rm\gamma }$ profiles have redshifted absorptions, with this number being even higher for the Pa ${\rm\beta }$ profiles, of which 44% display redshifted absorptions.

4.3 Remarks

From Figs. 1 and 2 it can be seen that some profiles classified as type I seem to show an absorption feature on or near the line centre. These absorptions could have lead us to classify those profiles either as type II B or II R. However, it should be noted that stars with no line emission also tend to show such an absorption in the "residual'' profile (e.g. Hubble 4, DI Tau). These absorptions seem to be due to the fact that the main sequence dwarves used to remove the photospheric component of the spectra, do not accurately represent the T Tauri stars' photospheric spectrum for Pa ${\rm\beta }$ and Br ${\rm\gamma }$ themselves, even though the strength of the metallic lines is well matched. That is, Pa ${\rm\beta }$ and Br ${\rm\gamma }$ are deeper in the T Tauri photospheric spectrum than in the main sequence dwarves. When we use WTTS, such as Hubble 4, as photospheric templates, the absorption we are referring to tend to disappear, strengthening our interpretation above. Despite this, we still decided to use the main sequence dwarves as photospheric templates. The reasons for this are discussed in FE99. Quantification of this "excess'' strength for the photospheric component and its interpretation is beyond the scope of this paper and will be tackled in future work.

  

 

Table 4: Pa ${\rm\beta }$ inverse P Cygni line profiles - Equivalent widths of the emission and of the absorption (measured below the continuum) component, central velocity of the absorption feature and line to continuum ratio (L/C) at the bottom of the absorption feature

Star	$EW_{\rm emission}$	$EW_{\rm absorption}$	$V_{\rm abs}$	L/C
	( ${\rm km\ s}^{-1}$)	( ${\rm km\ s}^{-1}$)	( ${\rm km\,s}^{-1}$)
BP Tau	$-147.3\pm0.7$	$4.8\pm0.1$	170	0.96
DO Tau	$-231.0\pm0.6$	$7.4\pm0.1$	175	0.93
DS Tau	$-118.5\pm0.4$	$2.5\pm0.1$	275	0.97
FM Tau	$-81.9\pm0.6$	$6.0\pm0.2$	150	0.92
FS Tau	$-127.4\pm0.7$	$8.3\pm0.4$	250	0.92
FP Tau	$-3.4\pm0.4$	$11.0\pm0.2$	45	0.93
GI Tau	$-56.6\pm0.6$	$11.0\pm0.3$	240	0.91
GK Tau	$-20.7\pm0.3$	$17.0\pm0.1$	115	0.88
GM Aur	$-298.5\pm0.6$	$13.2\pm0.2$	250	0.90
HP Tau	$-92.8\pm0.4$	$14.0\pm0.1$	200	0.91
IQ Tau	$-18.6\pm0.5$	$2.6\pm0.2$	200	0.97
RW Aur	$-332.6\pm0.4$	$2.8\pm0.1$	300	0.95
YY Ori	$-357\pm1$	$30.8\pm0.3$	280	0.81

For a few stars the Pa ${\rm\beta }$ and/or Br ${\rm\gamma }$ lines were observed more than once (refer to Sect. 9.4). The classification presented above refers to data obtained during the second observing run, i.e. December 1995, only. This is justified since, with the exception of the WTTS LkCa 7, all stars observed during UT94 were also observed during the 1995 campaign but with much higher signal-to-noise. The subject of variability in the line profiles will be left for Sect. 9.4, however a few remarks are relevant to the present discussion. The amount of variation present in the Pa ${\rm\beta }$ line profile of the stars observed twice during the 1995 campaign is such that the type of line profile did not change. However, for some stars, significant changes are seen in spectra taken 14 months apart, i.e. in October 1994 and in December 1995. These changes do not significantly alter the statistics presented, however they should be borne in mind.